Hitachi Dynamic Link Manager Software User's Guide (for Solaris(R)) 3000-3-F05-50(E)
Relevant program products Hitachi Dynamic Link Manager version 6.5.1 For details about applicable OSs, see the Release Notes. Trademarks AIX is a trademark of International Business Machines Corporation in the United States, other countries, or both. HP-UX is a product name of Hewlett-Packard Company. HP StorageWorks is a trademark of Hewlett-Packard Company. Java is a registered trademark of Oracle and/or its affiliates.
Summary of amendments The following table lists changes in this manual (3000-3-F05-50(E)) and product changes related to this manual. Changes Location in this manual Virtual Storage Platform has been added as a storage system supported by HDLM. 1.1, 2.7.1, 2.8.1, 2.8.2, 3.1.2, 3.1.3, 3.1.4, 3.1.5, 3.5.1, 3.16, 6.7.2, C Oracle Solaris Cluster 3.3 has been added as cluster software supported by HDLM. 3.1.3, 3.1.5, 3.1.
Preface This manual describes the functions and use of the following program products. • Hitachi Dynamic Link Manager Intended readers This manual is intended for system administrators who use Hitachi Dynamic Link Manager (HDLM) to operate and manage storage systems.
problems that you might encounter. 6. Command Reference Chapter 6 describes all the HDLM commands. 7. Utility Reference Chapter 7 describes the HDLM utilities. 8. Messages Chapter 8 provides information about viewing messages output by HDLM. It also lists and explains the HDLM messages and shows the actions to be taken in response to each message. A. Sun Cluster 3.2 Commands Appendix A describes the Sun Cluster 3.2 commands. B.
Conventions: Abbreviations This manual uses the following abbreviations for product names.
Abbreviation Full name or meaning HP XP512 HP StorageWorks Disk Array XP512 HP XP1024 HP StorageWorks XP1024 Disk Array HP XP10000 HP StorageWorks XP10000 Disk Array HP XP12000 HP StorageWorks XP12000 Disk Array HP XP20000 HP StorageWorks XP20000 Disk Array HP XP24000 HP StorageWorks XP24000 Disk Array JDKTM JavaTM 2 SDK, Standard Edition JRE JavaTM 2 Runtime Environment, Standard Edition Lightning 9900 Series A generic term for: • HP XP48 • HP XP512 • Lightning 9900 Series Lightning 990
Abbreviation Full name or meaning UNIX A generic term for: • AIX • Solaris • Linux • HP-UX Virtual Storage Platform A generic term for: • Hitachi Virtual Storage Platform • Hitachi StorageWorks P9500 Disk Array VCS Veritas Cluster Server VxVM Veritas Volume Manager(TM) This manual also uses the following abbreviations.
Abbreviation Full name or meaning GMT Greenwich Mean Time GUI Graphical User Interface HBA Host Bus Adapter HDev Host Device HLU Host Logical Unit HTTP Hypertext Transfer Protocol I/O Input/Output IP Internet Protocol iSCSI Internet Small Computer System Interface LAN Local Area Network LDAP Lightweight Directory Access Protocol LDEV Logical Device LDoms Logical Domains LLT Low Latency Transport LU Logical Unit LUN Logical Unit Number MP Maintenance Pack NAS Network Atta
Abbreviation Full name or meaning SMTP Simple Mail Transfer Protocol SNMP Simple Network Management Protocol SP Service Pack SSL Secure Sockets Layer SVP Service Processor UFS UNIX File System VTOC Volume Table Of Contents WWN World Wide Name ZFS Zettabyte File System Conventions: Diagrams This manual uses the following conventions in diagrams: Conventions: Fonts and symbols Font and symbol conventions are classified as: vii
• General font conventions • Conventions in syntax explanations These conventions are described below. General font conventions The following table lists the general font conventions: Font Convention Bold Bold type indicates text on a window, other than the window title. Such text includes menus, menu options, buttons, radio box options, or explanatory labels. For example, bold is used in sentences such as the following: • From the File menu, choose Open. • Click the Cancel button.
Example font or symbol Convention database-name This font style marks a placeholder that indicates where appropriate characters are to be entered in an actual command. SD perm Bold code-font characters indicate the abbreviation for a command. Underlined characters indicate the default value. [ ] Square brackets enclose an item or set of items whose specification is optional. An item that is underlined is specified when other items are omitted.
Contents Preface i Intended readers .........................................................................................................i Organization of this manual .......................................................................................i Related Publications..................................................................................................ii Conventions: Abbreviations ....................................................................................
2.12 Collecting Audit Log Data..................................................................................... 42 2.12.1 Categories and Audit Events that HDLM Can Output to the Audit Log ... 43 2.12.2 Requirements for Outputting Audit Log Data ............................................ 46 2.12.3 Destination and Filtering of Audit Log Data.............................................. 47 2.12.4 Audit Log Data Formats ............................................................................. 48 2.
3.8.1 Precautions ..................................................................................................165 3.8.2 Configuring a Boot Disk Environment in Which HDLM Manages the Boot Disk and Mirroring the Environment by Using SVM ................................166 3.8.3 Placing the Boot Disks Under HDLM Management by Installing HDLM to a Mirrored Boot Disk Environment Incorporating SVM ..............................172 3.8.4 Uninstalling HDLM .....................................................
4. HDLM Operation 267 4.1 Notes on Using HDLM.......................................................................................... 268 4.1.1 Displaying Path Information ...................................................................... 268 4.1.2 When a Path Error is Detected ................................................................... 268 4.1.3 iostat Command.......................................................................................... 268 4.1.4 Storage System .................
5.2 Checking Error Information in Messages...............................................................345 5.3 What To Do for a Path Error ..................................................................................347 5.3.1 Examining the Messages.............................................................................348 5.3.2 Obtain Path Information..............................................................................348 5.3.3 Identifying the Error Path.................................
7.2 The DLMgetras Utility for Collecting HDLM Error Information......................... 438 7.2.1 Format......................................................................................................... 438 7.2.2 Parameters .................................................................................................. 438 7.2.3 List of Collected Error Information............................................................ 440 7.3 The dlmcfgmgr Utility for Managing the HDLM Configuration .........
8.10 8.11 8.12 8.13 8.14 KAPL10001 to KAPL11000 ................................................................................560 KAPL11001 to KAPL12000 ................................................................................596 KAPL13001 to KAPL14000 ................................................................................600 KAPL15001 to KAPL16000 ................................................................................
Chapter 1. Overview of HDLM HDLM is a software package that manages paths between a host and a storage system. HDLM is designed to distribute loads across multiple paths and will switch a given load to another path if there is a failure in the path that is currently being used, thus improving system reliability. This chapter gives an overview of HDLM and describes its features. 1.1 What is HDLM? 1.
1. Overview of HDLM 1.1 What is HDLM? With the widespread use of data warehousing and increasing use of multimedia data, the need for high-speed processing of large volumes of data on networks has rapidly grown. To satisfy this need, networks dedicated to the transfer of data, such as SANs, are now being used to provide access to storage systems. HDLM manages the access paths to these storage systems.
1.
1. Overview of HDLM 1.2 HDLM Features HDLM features include the following: The ability to distribute a load across multiple paths. This is also known as load balancing. When a host is connected to a storage system via multiple paths, HDLM can distribute the load across all the paths. This prevents one, loaded down path from affecting the processing speed of the entire system. For details on load balancing, see 2.7 Distributing a Load Using Load Balancing.
Chapter 2. HDLM Functions This chapter describes the various functions that are built into HDLM. Before the function specifications are explained though, this chapter will go into detail about the HDLM management targets, system configuration, and basic terms that are necessary to know to effectively operate HDLM. After that, the rest of the chapter focus on describing all the HDLM functions, including the main ones: load distribution across paths and path switching. 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.
2. HDLM Functions 2.1 Devices Managed by HDLM Below is a list of devices that can or cannot be managed by HDLM. The devices that can be managed by HDLM are called HDLM management-target devices. HDLM management-target devices: The following devices are from the storage systems listed in Section 1.1 What is HDLM?: • SCSI devices (sd or ssd devices) • Boot disks#1, #2 • Swap devices#1 • Dump devices#1 #1: If you want to use these disks as HDLM management-target devices, assign VTOC labels to them.
2. HDLM Functions 2.2 System Configuration HDLM manages routes between a host and a storage system by using the SCSI driver (sd or ssd driver). The host and storage systems are connected using SAN with fiber cables or SCSI cables. The cable port on the host is a host bus adapter (HBA). The cable port on the storage system is a port (P) on a channel adapter (CHA). A logical unit (LU) contained in a storage system is the target of input to, or output from, the host. You can divide an LU into multiple areas.
2. HDLM Functions Figure 2-1: HDLM System Configuration Table 2-1: HDLM System Components lists the HDLM system components. Table 2-1: HDLM System Components Components 8 Description HBA A host bus adapter. This serves as a cable port on the host. SAN A dedicated network that is used for data transfer between the host and storage systems CHA A channel adapter P A port on a CHA. This serves as a cable port on a storage system. LU A logical unit (a logical volume defined on the storage system).
2. HDLM Functions 2.3 LU Configuration After you have properly installed HDLM, the LU configuration will change as follows: Before the installation of HDLM: The host recognizes that an sd or ssd device is connected to each physical path. Thus, a single LU in the storage system is recognized as the same number of LUs as that of physical paths. After the installation of HDLM: An HDLM device that corresponds one-to-one with the Dev in an LU in the storage system is created above an sd or ssd device.
2. HDLM Functions Figure 2-2: LU Configuration Recognized by the Host After the Installation of HDLM Table 2-2: Components Recognized by the Host lists the components recognized by the host. Table 2-2: Components Recognized by the Host Components Description HLU An LU that the host recognizes via the HDLM driver. It is called a host LU. No matter how many physical paths exist, one host LU is recognized for one LU in the storage system.
2. HDLM Functions 2.4 Program Configuration HDLM is actually a combination of several programs. Because each program corresponds to a specific HDLM operation, it is important to understand the name and purpose of each program, along with how they are all interrelated. Figure 2-3: Configuration of the HDLM Programs shows the configuration of the HDLM programs. Figure 2-3: Configuration of the HDLM Programs Table 2-3: Function of HDLM Programs lists and describes the functions of these programs.
2. HDLM Functions Table 2-3: Function of HDLM Programs Program name Functions HDLM command Provides the dlnkmgr command, which enables you to: • Manage paths • Display error information • Set up the HDLM operating environment HDLM utility Provides the HDLM utility, which enables you to: • Collect error information • Add a new LU and delete an existing LU (reconfiguring an HDLM device dynamically) • Create an HDLM driver configuration definition file (/kernel/drv/ dlmfdrv.
2. HDLM Functions 2.5 Position of the HDLM Driver and HDLM Device The HDLM driver is positioned above the SCSI driver. Each application on the host uses the HDLM device (logical device file) created by HDLM, to access LUs in the storage system. Figure 2-4: Position of the HDLM Driver and HDLM Devices shows the position of the HDLM driver and HDLM devices.
2. HDLM Functions 2.6 Logical Device Files for HDLM Devices When you install HDLM, a logical device file to be used by HDLM will be created for each LU on a per-Dev (slice) basis. Setting this logical device file name in an application, such as volume management software, enables the application to access an LU by using the HDLM function. The logical device files existing before HDLM installation (the logical device files of an sd or ssd) will be deleted.
2. HDLM Functions Block logical device files for HDLM devices are created in /dev/dsk. Character logical device files for HDLM devices are created in /dev/rdsk.
2. HDLM Functions 2.7 Distributing a Load Using Load Balancing When the system contains multiple paths to a single LU, HDLM can distribute the load across the paths by using multiple paths to transfer the I/O data. This function is called load balancing, and it prevents a single, heavily loaded path from affecting the performance of the entire system. Note that some I/O operations managed by HDLM can be distributed to each path, while others cannot.
2. HDLM Functions Figure 2-5: Flow of I/O Data When the Load Balancing Function Is Not Used When the load balancing function is not used, I/O operations converge onto a single path (A). The load on that one physical path (A) will cause a bottleneck, which might cause problems with system performance.
2. HDLM Functions Figure 2-6: Flow of I/O Data When the Load Balancing Function Is Used When the load balancing function is used, I/O operations are distributed via multiple physical paths (A, B, C, and D). This helps to prevent problems with system performance and helps prevent bottlenecks from occurring. 2.7.1 Paths to Which Load Balancing Is Applied This section describes, for each type of storage system, the paths to which the load balancing function is applied.
2. HDLM Functions owner path also varies depending on the LU. A non-owner path is a path that uses a CHA other than the owner controller (a non-owner controller). The paths to be used are selected, in the order of owner paths and non-owner paths. To prevent performance in the entire system from deteriorating, HDLM does not perform load balancing between owner paths and non-owner paths.
2. HDLM Functions (2) When Using the Lightning 9900 Series, Lightning 9900V Series, Hitachi USP, Universal Storage Platform V/VM, Virtual Storage Platform, Hitachi AMS2000 Series, or Hitachi SMS All online paths are owner paths. Therefore, for the example in Figure 2-6: Flow of I/O Data When the Load Balancing Function Is Used, the load is balanced among the four physical paths A, B, C, and D.
2. HDLM Functions was distributed will be used. However, if a specified number of I/Os has been issued to a path, processing switches to the next path. If the I/O is not sequential with the previous I/O, these algorithms select the path to be used each time an I/O request is issued. • Extended Round Robin The paths are simply selected in order from among all the connected paths. • Extended Least I/Os The path that has the least number of I/Os being processed is selected from among all the connected paths.
2. HDLM Functions operation. For details on the set operation, see 6.6 set (Sets Up the Operating Environment).
2. HDLM Functions 2.8 Performing Failovers and Failbacks Using Path Switching When the system contains multiple paths to an LU and an error occurs on the path that is currently being used, HDLM can switch to another functional path, so that the system can continue operating. This is called a failover. If a path in which an error has occurred recovers from the error, HDLM can then switch back to that path. This is called a failback.
2. HDLM Functions A fatal error that might stop the system. Error A high-risk error, which can be avoided by performing a failover or some other countermeasure. For details on error levels, see 2.11.2 Filtering of Error Information. When the Thunder 9500V series, or Hitachi AMS/WMS series is being used, HDLM will select the path to be used next from among the various paths that access the same LU, starting with owner paths, and then non-owner paths.
2. HDLM Functions Figure 2-8: Path Switching (2) Automatic Failbacks When a path recovers from an error, HDLM can automatically place the recovered path back online. This function is called the automatic failback function. In order to use the automatic failback function, HDLM must already be monitoring error recovery on a regular basis. When the Thunder 9500V Series, or Hitachi AMS/WMS series is being used, HDLM selects the path to use from online owner paths, and then from online non-owner paths.
2. HDLM Functions Storage Platform V/VM, Virtual Storage Platform, Hitachi AMS2000 Series, or Hitachi SMS is being used, all the paths are owner paths. Therefore, if an owner path recovers from an error and HDLM automatically places the recovered path online, the path in use will not be switched to the recovered owner path. When intermittent errors# occur on paths and you are using the automatic failback function, the path status might frequently alternate between the online and offline statuses.
2. HDLM Functions Executing the online operation places the offline path online. For details on the online operation, see 6.5 online (Places Paths Online). After the path status is changed to online, HDLM selects the path to use in the same way as for automatic path switching. When the Thunder 9500V Series, or Hitachi AMS/WMS series is being used, HDLM selects the path to use from online owner paths, and then from online non-owner paths.
2. HDLM Functions error occurred on the path. The (E) in Offline(E) indicates the error attribute, which indicates that an error occurred in the path. (3) Status Transitions of a Path Figure 2-9: Path Status Transitions shows the status transitions of a path. Figure 2-9: Path Status Transitions Legend: Online operation: Online operation performed by executing the dlnkmgr command's online operation. Offline operation: Offline operation performed by executing the dlnkmgr command's offline operation.
2. HDLM Functions LU. #2 When the following conditions are satisfied, a path that has been determined to have an intermittent error also becomes subject to automatic failback: • All the paths connected to an LU are Online(E), Offline(E), or Offline(C). • All the paths connected to an LU have been determined to have an intermittent error. • The processing of continuous I/O issued to an LU is successful. #3 When an Online or Offline(E) path exists among the paths that access the same LU.
2. HDLM Functions 2.9 Intermittent Error Monitoring (Functionality When Automatic Failback Is Used) An intermittent error means an error that occurs irregularly because of, for example, a loose cable connection. I/O performance might decrease while an automatic failback is being performed to repair an intermittent error. This is because the automatic failback operation is being performed repeatedly (because the intermittent error keeps occurring).
2. HDLM Functions 2.9.3 Intermittent Error Monitoring Actions Intermittent error monitoring is performed on each path, and it automatically starts as soon as a path is recovered from an error by using the automatic failback function.
2. HDLM Functions (2) When an Intermittent Error Does Not Occur If an error does not occur in the path the specified number of times within the specified interval, the system determines that the path does not have an intermittent error. In such a case, the error monitoring will finish when the specified error-monitoring interval finishes, upon which the number of errors is reset to 0.
2. HDLM Functions and then manually place the path online. (3) When the Conditions for an Intermittent Error Are Changed During Error Monitoring When the conditions for an intermittent error are changed during error monitoring, the number of errors and the amount of time that has passed since the error monitoring started are both reset to 0. As such, the error monitoring will not finish, and it will start over by using the new conditions.
2. HDLM Functions If you want to check whether intermittent error monitoring is being performed for the path, check the IEP item displayed when the dlnkmgr command's view -path operation is executed with the -iem parameter. If a numerical value of 0 or greater is displayed in the above items, then intermittent error monitoring is being performed.
2. HDLM Functions When you disable the intermittent error monitoring function, information about paths not subject to automatic failback will be reset. If you do not want to reset the information about paths not subject to automatic failback when you turn the intermittent error monitoring function off, change the target paths to Offline(C). #2 The number of errors and the time since monitoring started is reset to 0, and then monitoring restarts in accordance with the changed monitoring conditions.
2. HDLM Functions 2.10 Detecting Errors by Using Path Health Checking HDLM can check the status of paths to which I/Os are not issued, at regular intervals, and detect errors. This function is called path health checking. Without path health checking, an error is not detected unless I/O is issued because the system only checks the path status when I/O is issued. With path health checking, however, the system checks the status of online paths at regular intervals regardless of whether I/O is issued.
2. HDLM Functions 2.11 Error Management For troubleshooting purposes, HDLM collects information and stores it into log files. The error information to be collected can be filtered out by error level, and then stored into the log files. Figure 2-13: Flow of Data When Collecting Error Information shows the data flow when collecting error information on a host where HDLM is running.
2. HDLM Functions Figure 2-13: Flow of Data When Collecting Error Information Logs might be collected in layers below HDLM, such as for the SCSI driver.
2. HDLM Functions details, see the Solaris documentation. 2.11.1 Types of Collected Logs HDLM collects information on detected errors and trace information in the integrated trace file, trace file, error logs, log for the dlmcfgmgr utility for managing the HDLM configuration, and syslog. You can use the error information to examine the status of an error and analyze the cause of the error. Table 2-6: Types of Error Information describes the types of error information.
2. HDLM Functions # When you want to configure the system so that HDLM messages are output to syslog, specify user for the facility in the /etc/syslog.conf file. The following shows an example where the system function name is user, and messages at the info level or higher are output to the /tmp/syslog.user.log file: user.info /tmp/syslog.user.log For details on error levels, see 2.11.2 Filtering of Error Information. 2.11.
2. HDLM Functions • Collects no error information. • Collects error information from the Error level and higher. • Collects error information from the Warning level and higher. • Collects error information from the Information level and higher. • Collects error information from the Information level and higher (including maintenance information).
2. HDLM Functions 2.12 Collecting Audit Log Data HDLM and other Hitachi storage-related products provide an audit log function so that compliance with regulations, security evaluation standards, and industry-specific standards can be shown to auditors and evaluators. The following table describes the categories of audit log data that Hitachi storage-related products can collect.
2.
2. HDLM Functions Table 2-9: Categories and Audit Events that Can Be Output to the Audit Log Category Explanation Audit event Severity# Message ID 1 StartStop Authentication 44 Startup and termination of the software Administrator or end-user authentication Startup of the HDLM manager was successful. 6 KAPL15401-I Startup of the HDLM manager failed. 3 KAPL15402-E The HDLM manager stopped. 6 KAPL15403-I Startup of the I/O information monitoring function was successful.
2. HDLM Functions Category Explanation Audit event Severity# Message ID 1 ConfigurationAccess Viewing or updating configuration information Initialization of path statistics was successful. 6 KAPL15101-I Initialization of path statistics failed. 3 KAPL15102-E An attempt to place a path online or offline was successful. 6 KAPL15103-I An attempt to place a path online or offline failed. 4 KAPL15104-W Setup of the operating environment was successful.
2. HDLM Functions Category Explanation Audit event Severity# Message ID 1 Processing of the 6 KAPL15022-I 3 KAPL15023-E 6 KAPL15024-I 3 KAPL15025-E 6 KAPL15026-I 3 KAPL15027-E dlmsetconf [-d] [-r] command was successful. Processing of the dlmsetconf [-d] [-r] command failed. Processing of the dlmsetconf [-d] -u command was successful. Processing of the dlmsetconf [-d] -u command failed. Processing of the dlmvxexclude [-d] command was successful.
2. HDLM Functions example, an HDLM utility is executed from external media.# #: The following audit log data is output: • Categories: StartStop, Authentication, and ConfigurationAccess • Severity: 6 (Critical, Error, Warning, or Informational) • Destination: syslog (facility value: user) Notes: • You might need to perform operations such as changing the log size and backing up and saving collected log data, because the amount of audit log data might be quite large.
2. HDLM Functions Severity Audit log data to output 1 Correspondence with syslog severity levels Alert 2 Critical Critical 3 Critical and Error Error 4 Critical, Error, and Warning Warning 5 6 Notice Critical, Error, Warning, and Informational 7 Informational Debug Filtering by category: The following categories can be specified: • StartStop • Authentication • ConfigurationAccess • All of the above For details on how to specify audit log settings, see 3.10.2 Setting Up the HDLM Functions.
2.
2. HDLM Functions Explanation Item# Host sending request Name of the host sending a request Port number sending request Number of the port sending a request Host receiving request Name of the host receiving a request Port number receiving request Number of the port receiving a request Common operation ID Operation serial number in the program Log type information Fixed to BasicLog Application identification information Program identification information Reserved area This field is reserved.
2. HDLM Functions 2.13 Integrated HDLM management using Global Link Manager By using Global Link Manager, you can perform integrated path management on systems running multiple instances of HDLM. For large-scale system configurations using many hosts running HDLM, the operational load for managing paths on individual hosts increases with the size of the configuration.
2.
2. HDLM Functions 2.14 Cluster Support HDLM can also be used in cluster configurations. For details on cluster software supported by HDLM, the supported Solaris version, and usable volume management software, see 3.1.5 Combinations of Cluster Software and Volume Managers Supported by HDLM. HDLM uses a path of the active host to access an LU. Paths are switched in units of physical paths. Therefore, if an error occurs in a path, all the other paths that run through the same physical path are switched.
Chapter 3. Creating an HDLM Environment This chapter describes the procedures for setting up an HDLM environment and the procedure for canceling those settings. Make sure that HDLM is installed and the functions have been set up. Set up volume groups and cluster software to suit your operating environment. 3.1 3.2 3.3 3.4 3.5 3.6 3.
3. Creating an HDLM Environment 3.1 HDLM System Requirements Check the following before installing HDLM: 3.1.1 Hosts and OSs Supported by HDLM You can install HDLM on Sun SPARC series computer on which an OS listed in Table 3-1: Applicable OSs for the host is running.
3. Creating an HDLM Environment • Hitachi Universal Storage Platform 100 • Hitachi Universal Storage Platform 600 • Hitachi Universal Storage Platform 1100 • Hitachi Universal Storage Platform V • Hitachi Universal Storage Platform VM • Hitachi Virtual Storage Platform • HP StorageWorks P9500 Disk Array • HP XP48/HP XP128/HP XP512/HP XP1024/HP XP10000/HP XP12000/ HP XP20000/HP XP24000 • Lightning 9900 Series • Lightning 9900V Series • SVS • Thunder 9500V Series# # Supports the Fibre Channel interface only.
3. Creating an HDLM Environment • HP XP10000/HP XP12000/HP XP20000/HP XP24000 • SVS (2) HBAs For details on the applicable HBAs, see the HDLM Release Notes.
3. Creating an HDLM Environment #4 You must apply MP2 or later when using the I/O fencing function. The I/O fencing function can be used only when Hitachi USP, Universal Storage Platform V/VM, or Virtual Storage Platform storage systems are connected in a Solaris 9 environment. Note that the only supported service group type is the failover service group. The parallel service group and hybrid service group are not supported. #5 You must apply MP1 or later when using the I/O fencing function.
3. Creating an HDLM Environment #1 When used with the Thunder 9500V series, Lightning 9900V series, Hitachi USP, Hitachi AMS2000/AMS/TMS/WMS/SMS series, HP XP512, HP XP48, Universal Storage Platform V/VM, or Virtual Storage Platform storage systems, the Array Support Library of VxVM is required. If the Array Support Library of VxVM is not installed, install it before installing HDLM. For details on how to install the Array Support Library, see the storage system documentation.
3. Creating an HDLM Environment OS Cluster Volume manager SVM 1.0#4 VxVM 4.1#5 VxVM 5.0#5 Sun Cluster 3.1#1 #5 None SVM 1.0#6 Sun Cluster 3.1 (9/04)#1 #5 None SVM 1.0#6 Sun Cluster 3.1 (8/05)#1 #5 None SVM 1.0#6 VxVM 4.1 VCS 4.1#5 #7 None VxVM 4.1 VCS 5.0#2 #3 #5 None VxVM 5.0 Solaris 10 None None SVM 1.0#4 #8 #9 VxVM 4.1#5 #9 VxVM 5.0#5 #9 Sun Cluster 3.1 (8/05) #1 #5 #9 None SVM 1.0#6 #10 VxVM 4.1 Sun Cluster 3.2 #1 #11 None#12 SVM 1.0#5 #6 #9 #10 VxVM 5.
3. Creating an HDLM Environment OS Cluster Sun Cluster 3.2 (2/08) #1 #11 #13 Volume manager None#12 SVM 1.0#5 #6 #9 #10 VxVM 5.0#5 #9 #14 Sun Cluster 3.2 (1/09)#11 #13 None SVM 1.0 #6 #10 VxVM 5.0 Sun Cluster 3.2 (11/09)#11 #13 None SVM 1.0 #4 #10 VxVM 5.0 Oracle Solaris Cluster 3.3#11 #13 None SVM 1.0#5 #6 #9 #10 VxVM 5.1#5 #9 VCS 4.1#2 #5 #9 None VxVM 4.1 VCS 5.0#3 None#15 VxVM 5.0#5 #9 #16 VCS 5.1#15 None VxVM 5.
3. Creating an HDLM Environment Does not support the I/O fencing function. #3 Does not support linkage with SFVS (Storage Foundation Volume Server). #4 Does not support the following SVM functions: • Multi-owner disksets • Diskset import • Automatic (top down) volume creation #5 Does not support the EFI label.
3. Creating an HDLM Environment #11 For a two-node configuration, the pathcount setting is only supported for the SCSI protocol (fencing protocol) of the storage device. For details on how to specify SCSI protocols for storage devices, refer to the Sun Cluster manual. #12 For the EFI label or ZFS, only two-node configurations are supported. #13 Only two-node configurations are supported. #14 You must apply MP1 or later. #15 You must apply MP1 or later when using the I/O fencing function.
3. Creating an HDLM Environment Program Volume Manager Remarks Sun Cluster Support for Oracle Parallel Server/Real Application Clusters 3.1 Required packages: • SUNWschwr • SUNWscor • SUNWscucm • SUNWudlm • SUNWudlmr Oracle9i 9.2.0.8.0 RAC is bundled with Oracle9i. Oracle UNIX Distributed Lock Manager 3.3.4.
3. Creating an HDLM Environment Configuratio n Configuration 2 Configuration 3 Configuration 4 66 Program Remarks Volume Manager ASM • ASM is bundled with Oracle RAC 10g. • ASM is used as the disk memory area for files and recovery files of the Oracle database. In Oracle RAC 10g, HDLM devices can be used following the same procedures as for disk devices. For details on how to install ASM, refer to the documentation for Oracle RAC 10g.
3. Creating an HDLM Environment Configuratio n Configuration 5 Configuration 6 Configuration 7 Program Remarks Cluster Oracle Clusterware 10.1.0.5.0 -- Volume Manager None (Specify an HDLM raw device) -- OS Solaris 9 -- Oracle RAC 10g Oracle 10g Database 10.2.0.2.0 -- Cluster Oracle Clusterware 10.2.0.2.0 -- Volume Manager None (Specify an HDLM raw device) -- OS Solaris 9 -- Oracle RAC 10g Oracle 10g Database 10.2.0.2.
3. Creating an HDLM Environment Table 3-6: Programs required to create an Oracle RAC 10g environment (For Solaris 10) Configuratio n Configuration 1 Configuration 2 Configuration 3 68 Program Remarks OS Solaris 10 -- Oracle RAC 10g Oracle 10g Database 10.1.0.4.0 -- Cluster Oracle Cluster Ready Services (CRS) 10.1.0.4.0 -- Volume Manager ASM OS Solaris 10 -- Oracle RAC 10g Oracle 10g Database 10.1.0.5.0 -- Cluster Oracle Clusterware 10.1.0.5.
3. Creating an HDLM Environment Configuratio n Configuration 4 Configuration 5 Configuration 6 Program Remarks • ASM is bundled with Oracle RAC 10g. • ASM is used as the disk memory area for files and recovery files of the Oracle database. In Oracle RAC 10g, HDLM devices can be used following the same procedures as for disk devices. For details on how to install ASM, refer to the documentation for Oracle RAC 10g. Volume Manager ASM OS Solaris 10 -- Oracle RAC 10g Oracle 10g Database 10.2.0.2.
3. Creating an HDLM Environment Configuratio n Configuration 7 Configuration 8 Configuration 9 70 Program Remarks Cluster Sun Cluster 3.1 8/05 and Oracle Clusterware 10.2.0.2.0 Volume Manager None (Specify an HDLM raw device by the device ID of Sun Cluster) -- OS Solaris 10 -- Oracle RAC 10g Oracle 10g Database 10.2.0.2.0 Only two-node configurations are supported. Cluster Sun Cluster 3.1 8/05 and Oracle Clusterware 10.2.0.2.0 Volume Manager VxVM 4.
3. Creating an HDLM Environment Configuratio n Configuration 10 Configuration 11 Program Remarks • ASM is bundled with Oracle RAC 10g. • ASM is used as the disk memory area for files and recovery files of the Oracle database. In Oracle RAC 10g, HDLM devices can be used following the same procedures as for disk devices. For details on how to install ASM, refer to the documentation for Oracle RAC 10g. Volume Manager ASM OS Solaris 10 -- Oracle RAC 10g Oracle 10g Database 10.2.0.3.
3. Creating an HDLM Environment Configuratio n Configuration 12 Configuration 13 Configuration 14 72 Program Remarks OS Solaris 10 -- Oracle RAC 10g Oracle 10g Database 10.2.0.3.0 Only two-node configurations are supported. Cluster Sun Cluster 3.2 and Oracle Clusterware 10.2.0.3.0 Volume Manager VxVM 5.0 cluster functionality# Allocates memory areas, shared among nodes, such as Oracle database files, SPFILE, REDO log files, Oracle Cluster Registry, and voting disks, to the VxVM 5.
3. Creating an HDLM Environment Configuratio n Configuration 15 Configuration 16 Configuration 17 Program Remarks OS Solaris 10 -- Oracle RAC 10g Oracle 10g Database 10.2.0.3.0 -- Cluster Oracle Solaris Cluster 3.3 and Oracle Clusterware 10.2.0.3.0 -- Volume Manager None (Specify an HDLM raw device from the Sun Cluster device ID) -- OS Solaris 10 -- Oracle RAC 10g Oracle 10g Database 10.2.0.3.0 Only configurations that consist of three or more nodes are supported.
3. Creating an HDLM Environment Configuratio n Configuration 18 Configuration 19 Configuration 20 74 Program Remarks OS Solaris 10 -- Oracle RAC 10g Oracle 10g Database 10.2.0.4.0 Only two-node configurations are supported. Cluster Oracle Solaris Cluster 3.3 and Oracle Clusterware 10.2.0.4.0 Volume Manager VxVM 5.
3. Creating an HDLM Environment # You must apply MP1 or later. Required patches Table 3-7: Patches required to create an Oracle RAC 10g environment (For Solaris 8), Table 3-8: Patches required to create an Oracle RAC 10g environment (For Solaris 9) show patches that are provided by Sun Microsystems Inc. and are required to create an Oracle RAC 10g environment.
3. Creating an HDLM Environment # When using Oracle RAC 10g 10.2.0.1.0, apply 108993-45, not 108993-19. Table 3-8: Patches required to create an Oracle RAC 10g environment (For Solaris 9) Target program Oracle RAC 10g Patch ID 112233-11 or later Timing for applying Apply the patch before installing Oracle RAC 10g. 111722-04 or later 113801-12 or later# # It is necessary only for a configuration where Sun Cluster 3.1 8/05 is used as the cluster.
3. Creating an HDLM Environment Change the value of MISSCOUNT to the following value or greater: (number-of-paths-connected-to-the-voting-disk x 60 seconds). If four or more paths are connected to a voting disk, change DISKTIMEOUT, which is the I/O timeout threshold value for a voting disk, to the following value or greater: (number-of-paths-connected-to-the-voting-disk x 60 seconds). • When using Oracle RAC 10g version 10.2.0.2.
3. Creating an HDLM Environment Table 3-11: Formula for Calculating DISKTIMEOUT Storage system type • • • • • Lightning 9900 series Lightning 9900V series Hitachi USP Universal Storage Platform V/VM Virtual Storage Platform • Hitachi AMS2000/AMS/WMS/ SMS series • Thunder 9500V series Number of paths connected to the voting disk Formula for obtaining the value of DISKTIMEOUT 3 or less You do not need to change the value of DISKTIMEOUT.
3. Creating an HDLM Environment Execution example: # ls -lL /dev/rdsk/c10t50060E8005271760d5s0 crw-r----1 root sys rdsk/c10t50060E8005271760d5s0 307, 1608 date/time /dev/ # In this example, the major number is 307 and the minor number is 1608. 2. Create an alias device file by executing the following command in each node. An alias device file corresponding to one disk slice should have the same name in all the nodes.
3. Creating an HDLM Environment 4.
3. Creating an HDLM Environment Configurati on Configuration 3 Configuration 4 Configuration 5 Program Remarks • ASM is bundled with Oracle RAC 11g. • ASM is used as the disk memory area for the Oracle database files and recovery files. In Oracle RAC 11g, HDLM devices can be used following the same procedures as for disk devices. For details on how to use ASM, refer to the documentation for Oracle RAC 11g. Volume Manager ASM OS Solaris 10 -- Oracle RAC 11g Oracle 11g Database 11.1.0.6.
3. Creating an HDLM Environment Configurati on Configuration 6 Configuration 7 Configuration 8 82 Program Remarks Cluster Sun Cluster 3.2 and Oracle Clusterware 11.1.0.6.0 Volume Manager None (Specify an HDLM raw device) -- OS Solaris 10 -- Oracle RAC 11g Oracle 11g Database 11.1.0.6.0 Only two-node configurations are supported. Cluster Sun Cluster 3.2 and Oracle Clusterware 11.1.0.6.0 Volume Manager ASM OS Solaris 10 -- Oracle RAC 11g Oracle 11g Database 11.1.0.6.
3. Creating an HDLM Environment Configurati on Configuration 9 Configuration 10 Configuration 11 Program Remarks • ASM is bundled with Oracle RAC 11g. • ASM is used as the disk memory area for the Oracle database files and recovery files. For the disk device used by ASM, specify the Sun Cluster device ID. For details on how to use ASM, refer to the documentation for Oracle RAC 11g. Volume Manager ASM OS Solaris 10 -- Oracle RAC 11g Oracle 11g Database 11.1.0.7.
3. Creating an HDLM Environment Configurati on Program Volume Manager VxVM 5.0 cluster functionality# Remarks Allocates memory areas, shared among nodes, such as Oracle database files, SPFILE, REDO log files, Oracle Cluster Registry, and voting disks, to the VxVM 5.0 cluster functionality volumes. For details on how to allocate memory areas, refer to the documentation for Oracle RAC 11g. # You must apply MP3 or later.
3.
3.
3. Creating an HDLM Environment # In this example, a device file for RAC whose major number is 307 and minor number is 1608 is created. 3. For the created alias device file, set the owner, group, and access permission mode by using the following command. The owner, group, and access permission mode to be set differs depending on the usage purpose of Oracle RAC 11g corresponding to that device. For details on this point, refer to the Oracle documentation.
3. Creating an HDLM Environment Only domains including I/O domains that manage the HBAs connected to HDLM-managed storage are supported. • HDLM cannot be installed in the guest domain. If the control domain and the I/O domains that manage the HBAs for HDLM-managed storage are separated: • HDLM can be installed in I/O domains. Only domains including I/O domains that manage the HBAs connected to HDLM-managed storage are supported. • HDLM cannot be installed in the control domain.
3. Creating an HDLM Environment 3.1.7 Memory and Disk Capacity Requirements This subsection explains the memory and disk capacity requirements. (1) Memory Requirements The host requires 11 MB of memory. (2) Disk Capacity Requirements Table 3-16: Disk capacity requirements for a host lists the disk capacity requirements for a host.
3. Creating an HDLM Environment up. 3.1.8 The Number of Paths Supported in HDLM Table 3-17: Number of LUs and Paths Supported in HDLM lists the number of paths per LU and the number of total physical paths that are Supported in HDLM. Table 3-17: Number of LUs and Paths Supported in HDLM Item Supported number Number of LUs 1 to 256 Number of physical paths per LU 1 to 64 Total number of physical paths 1 to 4096 When Sun Cluster 3.1, Sun Cluster 3.2, or Oracle Solaris Cluster 3.
3. Creating an HDLM Environment 3.2 Flow for Creating an HDLM Environment Set up the environment to use HDLM as follows. • The installation procedure includes the Sun Cluster setup procedure. • Execute the procedures described in 3.5.6 Preparations for an Upgrade Installation or Re-installation of HDLM and 3.5.7 Performing an Upgrade Installation or Re-installation of HDLM only when upgrading or re-installing HDLM. • Execute the procedures described in 3.
3.
3. Creating an HDLM Environment 3.3 HDLM Installation Types This subsection describes the following types of HDLM installation: new installation, upgrade installation, and re-installation. New installation of HDLM: Installing HDLM in a server on which HDLM has not been installed is called a new installation of HDLM. Upgrade installation of HDLM: Installing a newer version than the existing version without uninstalling the existing version is called an upgrade installation of HDLM.
3. Creating an HDLM Environment 3.4 Notes on Creating an HDLM Environment This section gives points to be noted when creating an HDLM environment. For points to be noted during operation of HDLM, see 4.1 Notes on Using HDLM. 3.4.1 Notes on Hardware Settings For details on the settings of the storage system, see storage system manual and maintenance documentation for the storage system.
3. Creating an HDLM Environment 3.4.2 Notes on Installation You cannot edit the HDLM driver configuration definition file (/kernel/drv/ dlmfdrv.conf). To change an HDLM-managed device to a non-HDLM-managed device, use the HDLM device unconfiguration function of the HDLM device. For details on the HDLM device unconfiguration function, see 3.5.5 Using the HDLM Device Unconfiguration Function When Performing a New Installation of HDLM or 4.5.
3. Creating an HDLM Environment • The HDLM driver unconfiguration definition file (/opt/ DynamicLinkManager/config/dlmfdrv.unconf) However, if both the conditions listed below are satisfied, the HDLM driver configuration definition file and the HDLM driver unconfiguration definition file are updated: • SVS is an HDLM management-target device. • You are upgrading from HDLM 5.6.1 or earlier.
3. Creating an HDLM Environment Integrated trace files after re-installation: /var/opt/hitachi/ HNTRLib2/spool/hntr2n.log (n indicates a file number) Trace files after re-installation: /var/opt/DynamicLinkManager/log/ hdlmtrn.log (n indicates a file number) When you re-install HDLM 5.6 or later after uninstalling HDLM 04-01, or when you perform an upgrade installation from HDLM 05-00 or later Trace files before uninstallation: /var/opt/hitachi/HNTRLib2/spool/hntr2n.
3. Creating an HDLM Environment • To allocate the HDLM device for each LU, execute the following command: zonecfg:zone1:device> set match=/dev/rdsk/ c5t50060E800436E211d0* • To allocate the HDLM device for each controller, execute the following commands, and then restart the non-global zone: zonecfg:zone1:device> set match=/dev/rdsk/c5t* zonecfg:zone1:device> end zonecfg:zone1> verify zonecfg:zone1> commit zonecfg:zone1> exit # For details on the commands, refer to the Solaris 10 documentation.
3. Creating an HDLM Environment • Configuring a boot disk environment • Setting up HDLM functions • Setting up an integrated trace# • Creating a file system • Executing an HDLM command • Using an HDLM device • Changing the configuration of the HDLM operating environment • Applying an HDLM service pack # In single-user mode, HDLM command activity is not logged to integrated trace files. When you install HDLM while the 32-bit kernel mode is being used, only the 32-bit version of HDLM will be installed.
3. Creating an HDLM Environment encapsulation while creating a disk group with an HDLM device or adding an HDLM device to a disk group, the KAPL08019-E or KAPL08022-E message might be output; however, this will not affect operations. Make sure that the version of HDLM is the same for all nodes on a cluster. HDLM and Sun StorageTek RAID Manager cannot be both present. Before installing HDLM, uninstall Sun StorageTek RAID Manager.
3. Creating an HDLM Environment 3.5 Installing HDLM When you install HDLM, Hitachi Network Objectplaza Trace Library (HNTRLib2) will also be installed. The file path of the integrated trace information file of Hitachi Network Objectplaza Trace Library (HNTRLib2) is /var/opt/hitachi/HNTRLib2/spool/ hntr2n.log (n indicates a file number). 3.5.
3. Creating an HDLM Environment 6. Unmount the device. If the device is mounted, unmount it. Also, if the device is set to be mounted automatically when the host starts, delete this setting from the /etc/vfstab file. The following is an example of this command. • Unmounting a device. First, check the current settings. # mount -p The current settings will be output as shown in Figure 3-2: Execution Results of the mount -p Command.
3. Creating an HDLM Environment Figure 3-3: An example of how to edit the /etc/vfstab file Comment the shaded portions out by placing a hash mark (#) at the beginning of the line, or delete the line. 7. Unregister the device from the volume management software. If the device is registered in VxVM, SDS, or SVM, cancel the registration. (2) Apply Solaris Patches See 3.1.1 Hosts and OSs Supported by HDLM, and then apply any necessary patches to the Solaris OS.
3. Creating an HDLM Environment (4) Set Up the /kernel/drv/sd.conf File When using an sd driver, depending on the HBA driver, you need to edit the / kernel/drv/sd.conf file to recognize the disks. For details on how to edit this file, refer to the HBA documentation. When using an ssd driver, you will not need to edit this file, since the connected disks will be automatically recognized. After editing the /kernel/drv/sd.
3. Creating an HDLM Environment To boot at the ok prompt, execute the following command: boot boot-device kernel/unix other-boot-options • To switch to the 64-bit kernel mode: If Solaris is running, execute the following command: # reboot -- "boot-device kernel/sparcv9/unix other-boot-options" To boot at the ok prompt, execute the following command: boot boot-device kernel/sparcv9/unix other-boot-options (6) Set Up the /etc/system File HDLM uses a maximum of 8 KB of the stack for the kernel thread.
3. Creating an HDLM Environment of "Error" or higher are to be output. Make sure that the /etc/syslog.conf file contains the following configuration entry. We recommend setting the output destination for *.err in the /etc/ syslog.conf file to the default /var/adm/messages directory. *.err;kern.debug;daemon.notice;mail.crit; /var/adm/messages When you collect audit log data, we recommend that you use any of the extended facilities (local0 to local7) as the exclusive output destination for the data.
3. Creating an HDLM Environment For details, refer to the VxVM documentation. 5. Remove any sd or ssd disk to be managed by HDLM from VxVM control. For details on how to remove any such sd or ssd device from VxVM control, refer to the VxVM documentation. If HDLM has been installed, remove any such sd or ssd device from VxVM control as directed in 3.13.3 Removing Devices from VxVM. (9) Set Up SDS and SVM When SDS or SVM uses the sd or ssd device to be managed by HDLM, clear the device settings.
3. Creating an HDLM Environment • If an sd or ssd device to be managed by HDLM has been set as a Quorum device, perform the following procedures to remove the Quorum device settings. 1. Use the following commands to check the device ID of the Quorum device: # /usr/cluster/bin/scconf -pv | grep Quorum # /usr/cluster/bin/scdidadm -L 2. If multiple Quorum devices exist, execute the following command until the number of Quorum devices becomes one.
3. Creating an HDLM Environment - The node is in the Offline status. - Both the current Quorum votes present and Quorum votes possible are 0. The following shows an example of executing the command: host2# /usr/cluster/bin/scstat -q 8. If the cluster is a two-node configuration, enter the install mode. If the cluster consists of three or more nodes, this step is not necessary so go to step 9. The following shows an example of executing the command: host2# /usr/cluster/bin/scconf -c -q installmode 9.
3. Creating an HDLM Environment 3.5.2 Preparation for Performing an Unattended Installation of HDLM An unattended installation is an installation in which HDLM is installed based on the responses predefined in the installation-information settings file instead of user responses. The following shows the processing flow of unattended installation: 1. Define information necessary for installing HDLM in the installation-information settings file.# 2. Use the installux.
3. Creating an HDLM Environment Notes • The license key file can be saved as a user-specified directory or a file name. However, this license key file is not deleted after installation. Delete this file manually after installation is completed if it is not necessary. • Available characters for the absolute path of the license key file are as follows: Upper case and lower case characters, numbers, and some special characters ( /, !, #, +, -, .
3. Creating an HDLM Environment In this example, the shaded portion shows information for each path of the device you are removing from HDLM management. Type in the appropriate device number, and then type quit to end the format command. 5. Insert the DVD-ROM. After a few seconds, the DVD-ROM will be automatically mounted, and the mount point of the DVD-ROM will be displayed in a list of files.
3. Creating an HDLM Environment After confirming that the KAPL09211-I message has been output, execute the following command to delete the installation-information settings file: # rm -r /any-directory/installation-information-settings-file If the driver_config key is set to n (so that the dlmsetconf utility is not executed) in the installation-information settings file, go to step 9.
3. Creating an HDLM Environment # pkginfo -l PKGINST: NAME: CATEGORY: ARCH: VERSION: BASEDIR: VENDOR: ... # 9. DLManager DLManager Dynamic Link Manager system sparc xx.x.x.xxxx / Make sure that the logical device file of the sd or ssd device is backed up.
3. Creating an HDLM Environment For details on how to perform this setting, see 3.5.5 Using the HDLM Device Unconfiguration Function When Performing a New Installation of HDLM. Note Set the boot disk in the post-migration environment as an HDLM-managed device if, before this step, you executed the procedure described in (2) Migration by Installing HDLM in the Existing Local Boot Disk Environment in 3.6.2 Procedure for Configuring a Boot Disk Environment.
3. Creating an HDLM Environment Path Health Checking : on(30) Auto Failback : on(1) Intermittent Error Monitor : off HDLM Manager Ver WakeupTime Alive x.x.x-xx yyyy/mm/dd hh:mm:ss HDLM Alert Driver Ver WakeupTime ElogMem Size x.x.x-xx yyyy/mm/dd hh:mm:ss 4096 Alive HDLM Driver Ver WakeupTime x.x.x-xx yyyy/mm/dd hh:mm:ss Alive License Type Expiration Permanent KAPL01001-I The HDLM command completed normally.
3. Creating an HDLM Environment these devices. 19. If you are using VxVM, execute the following command to confirm that the DEVICE column of the LUs managed by HDLM is displayed in the cWtXdYsZ format (for example, c6t0d10s2): # vxdisk list If the value is displayed in the cWtXdYsZ format, perform the following procedures: • Delete the /etc/vx/disk.info file. • Restart the host, or restart the VxVM configuration daemon by executing the following command: # vxconfigd -k 20.
3. Creating an HDLM Environment Item Description NewHDevName Indicates the HDLM 6.5.1 logical device file name (HDevName). If a hyphen (-) is displayed, the corresponding HDev does not exist after the HDLM 6.5.1 installation. DskName Indicates the storage system name. iLU Indicates the LU number in the storage system. 21. Based on the results of step 20, compare the logical device file names from before and after the HDLM 6.5.
3. Creating an HDLM Environment 2. Log in to Solaris as a user with root permissions. 3. Execute the format command to make sure that the sd or ssd device to be managed by HDLM is displayed. # format 4. If you want to remove a device from HDLM management, execute the format command to check information for each path of that device. The information you are checking here will be used in step 17.
3. Creating an HDLM Environment 6. Make sure that the key to be used for disk reservation has been registered with the device to be managed by HDLM.
3. Creating an HDLM Environment 0x40235c5200000002 0x40235c5200000001 0x40235c5200000003 Attempting to remove all keys from the disk... Scrubbing complete, use '/usr/cluster/lib/sc/scsi -c inkeys -d /dev/rdsk/c5t0d0s2' to verify success # 10. Make sure that the key registration was canceled for the device to be managed by HDLM. On any one node, perform step 6 again and confirm that no key is displayed.
3. Creating an HDLM Environment # pkgadd -d DVD-ROM-mount-point/HDLM_Solaris DLManager • If you are performing an unattended installation of HDLM, execute the installux.sh script or the installhdlm utility as shown below. When executing the installux.sh script # DVD-ROM-mount-point/installux.
3. Creating an HDLM Environment DynamicLinkManager directory exists. If the /var/opt/ DynamicLinkManager directory exists, execute the following command to delete it: # rm -rf /var/opt/DynamicLinkManager 14. Execute the following command to check whether HDLM 6.5.1 has been installed successfully. Make sure that VERSION appears as 06.5.1.0000. # pkginfo -l PKGINST: NAME: CATEGORY: ARCH: VERSION: BASEDIR: VENDOR: ... # DLManager DLManager Dynamic Link Manager system sparc xx.x.x.xxxx / 15.
3. Creating an HDLM Environment 17. Specify the devices that are to be excluded from management by HDLM, if there are any. This step is unnecessary if you want HDLM to manage all applicable devices. Devices that cannot be managed by HDLM are automatically removed from the HDLM driver configuration definition file. For details on devices that cannot be managed by HDLM, see 2.1 Devices Managed by HDLM.
3. Creating an HDLM Environment 20. With the reconfiguration option specified, start all nodes comprising the cluster in cluster mode, at the same time. Execute boot -r at the ok prompt. Note While a node is starting, the message below may be output to the console or syslog: Could not read symbolic link for: /dev/rdsk/c5t0d0s2 path not loaded This message is output because HDLM deleted the logical device file of an sd or ssd device managed by HDLM.
3. Creating an HDLM Environment name = view, completion time = yyyy/mm/dd hh:mm:ss # Nothing is displayed for Support Cluster even if you are using cluster software. However, the cluster-support functions are, in fact, operating without problem. 22. By checking the execution results of the view operation, verify the version of the installed HDLM. If HDLM Version is 6.5.1-00, the installed version of HDLM is correct. 23.
3. Creating an HDLM Environment Table 3-20: Information displayed by the dlmlisthdev utility for assisting HDLM transitions describes the items shown above. Table 3-20: Information displayed by the dlmlisthdev utility for assisting HDLM transitions Item Description OldHDevName Indicates the HDLM 04-00 logical device file name (HDevName). If a hyphen (-) is displayed, the corresponding HDev did not exist before the HDLM 6.5.1 installation.
3. Creating an HDLM Environment 30. Update the Sun Cluster global device namespace. Execute the following command: # /usr/cluster/bin/scgdevs 31. To use an HDLM device for a Quorum device, specify the Quorum device settings at any of the nodes that constitute the cluster. The following shows examples of executing the commands: • Check the device ID of the HDLM device. # /opt/DynamicLinkManager/bin/dlnkmgr view -path # /usr/cluster/bin/scdidadm -L • Set up the Quorum device.
3. Creating an HDLM Environment these devices. To use an HDLM-managed device with a file system created by using volume management software, perform step 34. To use an HDLM-managed device with a file system created without using volume management software, perform steps 35 to 37. 34. Register the disk device group. If you are using VxVM, see 3.13.5 Linking VxVM and Sun Cluster. If you are using SDS, see (2) To Use a Shared Diskset in 3.14.2 Registering HDLM Devices.
3. Creating an HDLM Environment dlmfdrv.unconf file (before editing) and Figure 3-5: Example of an /opt/ DynamicLinkManager/config/dlmfdrv.unconf file (after editing) show an example of editing the HDLM driver unconfiguration definition file (/opt/ DynamicLinkManager/config/dlmfdrv.unconf). Figure 3-4: Example of an /opt/DynamicLinkManager/config/dlmfdrv.unconf file (before editing) Figure 3-5: Example of an /opt/DynamicLinkManager/config/dlmfdrv.
3. Creating an HDLM Environment • # reboot -- -r • boot -r (at the ok prompt) 4. Execute the following command to verify that the settings were applied correctly: # /opt/DynamicLinkManager/bin/dlnkmgr view -drv Confirm that the LDEV that was removed from HDLM management is not shown. 3.5.6 Preparations for an Upgrade Installation or Re-installation of HDLM Before performing an upgrade installation or re-installation of HDLM, back up the HDLM-managed devices.
3. Creating an HDLM Environment • When you install HDLM while the 32-bit kernel mode is being used, only the 32-bit version of HDLM will be installed. When you install HDLM while the 64-bit kernel mode is being used, both the 32-bit and 64-bit versions of HDLM will be installed. Before installing HDLM, decide which kernel mode you will use to operate Solaris and, if needed, change to the desired kernel mode.
3. Creating an HDLM Environment The following shows an example of executing the command for mounting the DVD-ROM: # mkdir /cdrom/cdrom0 # mount -F hsfs -o ro /vol/dev/dsk/c0t6d0/ DVD-ROM-volume-name /cdrom/cdrom0 4. Perform the installation. • If you are not performing an unattended installation of HDLM, execute either of the following installation commands: When executing the installux.sh script # DVD-ROM-mount-point/installux.
3. Creating an HDLM Environment hdlm_license is not already saved, a message is output asking whether a user-specified license key file has been saved. Enter 1 if a user-specified license key file has been saved. Enter 2 to specify a license key directly. • Note that if any value other than 1 or 2 is entered three times, the installation is stopped. If 1 is entered, specify a user-specified license key file with the absolute path.
3. Creating an HDLM Environment Then, execute the following command at the ok prompt in all nodes: boot Note You must restart the host immediately after a successful upgrade installation or re-installation has been confirmed. 8. Execute the dlnkmgr command's view operation to display the status of the programs. The following shows an example of executing the command: # /opt/DynamicLinkManager/bin/dlnkmgr view -sys HDLM Version : x.x.
3. Creating an HDLM Environment load-balancing setting was previously on. The load-balancing algorithm options are Round Robin, Extended Round Robin, Least I/O, Extended Least I/Os, Least Blocks, and Extended Least Blocks. See 3.10.2 Setting Up the HDLM Functions, and select the appropriate load balancing algorithm as required. • Nothing is displayed for Support Cluster even if you are using cluster software. However, the cluster-support functions are, in fact, operating without problem. 9.
3. Creating an HDLM Environment A virtual disk backend is a location where virtual disk data is stored. A disk, disk slice, file, or volume (such as ZFS, SVM, and VxVM) can be used for the backend. In the control domain, register the install_disk file on a built-in disk as a virtual disk backend. The I/O domain recognizes this disk as a virtual disk. The device name of this virtual disk is /dev/[r]dsk/c0d0.
3. Creating an HDLM Environment To install HDLM to a host in which LDoms have been set: 1. Log in to the control domain. 2. Execute the ldm stop command to stop the guest domain.
3. Creating an HDLM Environment 3. Execute the ldm unbind command to unbind the resource of the guest domain. For example, to unbind a resource of the guest domain ldg2, execute the command as follows: # ldm unbind ldg2 4. Execute the ldm rm-vdisk command to remove virtual disks from the guest domain. For example, to remove virtual disks vdisk1 and vdisk2 from the guest domain ldg2, execute the command as follows: # ldm rm-vdisk vdisk1 ldg2 # ldm rm-vdisk vdisk2 ldg2 5.
3. Creating an HDLM Environment backend. 8. On the control domain, execute the ldm add-vdsdev command to export the virtual disk backends from the service domain on the I/O domain to which HDLM will be installed.
3. Creating an HDLM Environment Figure 3-7: Configuration Example (After Installation) In this configuration example, /dev/dsk/c4t50060E8010027A82d0s2 and /dev/ dsk/c4t50060E8010027A82d1s2 are HDLM devices.
3. Creating an HDLM Environment 3.6 Configuring a Boot Disk Environment This section describes how to configure a boot disk environment in which HDLM manages the boot disk. An environment with a boot disk located on the host is called a local boot disk environment. 3.6.
3. Creating an HDLM Environment • Building a new pre-migration environment: Perform the procedure described in (4) Migration by Building a New Pre-Migration Environment in 3.6.2 Procedure for Configuring a Boot Disk Environment, and then execute the procedure described in (5) Setting Up the Post-Migration Environment in 3.6.2 Procedure for Configuring a Boot Disk Environment. 3.6.
3. Creating an HDLM Environment Figure 3-8: Flow of setting up the boot disk environment when using an existing HDLM environment as the pre-migration environment 144 1. Allocate the LUs to be used by the boot disk in the post-migration environment. 2. Get Solaris to recognize the LUs to be used by the post-migration boot disk.
3. Creating an HDLM Environment 3. Set the LU slices to be used by the post-migration boot disk. 4. Execute the dlmsetconf utility as follows to create the HDLM driver configuration definition file (/kernel/drv/dlmfdrv.conf) and the HDLM driver unconfiguration definition file (/opt/DynamicLinkManager/config/ dlmfdrv.unconf): # /opt/DynamicLinkManager/bin/dlmsetconf This step and the next step set the LU to be used for the post-migration boot disk as an HDLM-managed LU. 5.
3. Creating an HDLM Environment Figure 3-9: Flow of setting up the boot disk environment when installing HDLM in the existing local boot disk environment 1. Allocate the LUs to be used by the boot disk in the post-migration environment.
3. Creating an HDLM Environment • To the slices of a single LU, allocate the Solaris file system, swap device, dump device, Sun Cluster global device file system, and replica of the SDS or SVM state database. • To the slices of multiple LUs, allocate the Solaris file system, swap device, dump device, Sun Cluster global device file system, and replica of the SDS or SVM state database. In this case, make sure that the physical path of each LU is the same. 2.
3. Creating an HDLM Environment Figure 3-10: Flow of setting up the boot disk environment when using an existing boot disk environment without HDLM installed as the pre-migration environment 1. Allocate the LUs to be used by the boot disk in the post-migration environment.
3. Creating an HDLM Environment • To the slices of a single LU, allocate the Solaris file system, swap device, dump device, Sun Cluster global device file system, and replica of the SDS or SVM state database. • To the slices of multiple LUs, allocate the Solaris file system, swap device, dump device, Sun Cluster global device file system, and replica of the SDS or SVM state database. In this case, make sure that the physical path of each LU is the same. 2.
3.
3. Creating an HDLM Environment 1. Install Solaris in the pre-migration environment. 2. Install Sun Cluster if required. For details, see your Sun Cluster documentation. 3. Install VxVM if required. For details, see your VxVM documentation. 4. Allocate the LUs to be used by the boot disk in the post-migration environment.
3. Creating an HDLM Environment Cluster, also initialize the Sun Cluster global device file system. The following shows an example of executing the command: # newfs /dev/rdsk/c8t50060E80004361D1d1s0 # newfs /dev/rdsk/c8t50060E80004361D1d1s6 # newfs /dev/rdsk/c8t50060E80004361D1d1s3 3. Mount the Solaris file system in the post-migration environment. If using Sun Cluster, also mount the Sun Cluster global device file system.
3. Creating an HDLM Environment 7. Set the rootdev parameter for the slice containing the / (root) file system of the post-migration boot disk to the /etc/system file of the post-migration boot disk. In the rootdev parameter, specify the physical device file name of the HDLM device. In the following example, the underlined part is the physical device file name displayed in the execution example in step 6. rootdev:/pseudo/dlmndrv@1/dlmfdrv@w50060e80004361d1,1:a 8.
3. Creating an HDLM Environment ... 30 node1:/dev/rdsk/c8t50060E80004361D1d1 rdsk/d30 9. /dev/did/ Edit the post-migration boot disk's /etc/vfstab file, changing the swap device definition from the pre-migration boot disk to the post-migration boot disk. An example is shown below. Before editing /dev/dsk/c0t0d0s1 - - swap - no - After editing /dev/dsk/c8t50060E80004361D1d1s1 no - - - swap - 10. Edit the post-migration boot disk's /etc/dumpadm.
3. Creating an HDLM Environment - Add the slice identifier of the / (root) file system (:a value from :a to :h) to the end. An example in which Sun Microsystems' HBA is used with Solaris 8 or Solaris 9 is presented below. For details on finding the boot device name when using an HBA other than from Sun Microsystems, and on setting the HBA in the OBP, see the documentation for the HBA you are using. Execute the dlnkmgr command's view operation with the -lu, -item, pd, type, and -hdev parameters specified.
3. Creating an HDLM Environment specified for the boot device. 12. Boot the system from the post-migration environment. • Perform steps 13 to 17 only if SDS or SVM is being used and you are migrating a replica of the SDS or SVM state database to the post-migration environment. In this case, keep the pre-migration environment recognized by Solaris. • Skip to step 18 if SDS or SVM is being used but you are not migrating a replica of the SDS or SVM state database to the post-migration environment.
3. Creating an HDLM Environment The following shows an example of executing the command: # metadb -d /dev/dsk/c0t0d0s7 /dev/dsk/c0t0d1s7 /dev/dsk/ c0t0d2s7 17. Make sure that the SDS or SVM state database replica that exists in the pre-migration boot disk has been deleted. The following shows an example of executing the command: 18. Back up the contents of both the pre-migration and post-migration environments.
3. Creating an HDLM Environment 3.7 Migrating from a Boot Disk Environment to the Local Boot Disk Environment This section describes how to migrate from a boot disk environment in which HDLM manages the boot disk to the local boot disk environment (boot disk configured on the host). In this section, the former is sometimes called the pre-migration environment or pre-migration, and the latter is sometimes called the post-migration environment or post-migration.
3. Creating an HDLM Environment Figure 3-12: Migration flow when coping the file system from a boot disk environment to the local boot disk environment 1. Boot the system from the pre-migration environment.
3. Creating an HDLM Environment Perform steps 1 to 16 in the pre-migration environment. 2. Allocate the LUs to be used by the boot disk in the post-migration environment. 3. Get Solaris to recognize the LUs to be used by the post-migration boot disk. 4. Set the LU slices to be used by the post-migration boot disk. • If using Sun Cluster, go to the next step. • If not using Sun Cluster, go to step 8. 5. Assign device IDs.
3. Creating an HDLM Environment # newfs /dev/rdsk/c0t0d0s0 # newfs /dev/rdsk/c0t0d0s6 # newfs /dev/rdsk/c0t0d0s3 11. Mount the Solaris file system in the post-migration environment. If using Sun Cluster, also mount the Sun Cluster global device file system. The following shows an example of executing the command: # mount /dev/dsk/c0t0d0s0 /mnt0 # mount /dev/dsk/c0t0d0s6 /mnt1 # mount /dev/dsk/c0t0d0s3 /mnt2 12.
3. Creating an HDLM Environment /dev/dsk/c0t0d0s6 /dev/rdsk/c0t0d0s6 /usr ufs 1 no /dev/did/dsk/d24s3 /dev/did/rdsk/d24s3 .devices/node@1 ufs 2 no global /global/ # Execute the scdidadm -l command to find out the device ID to specify as the mount device of the Sun Cluster global device file system. In the following example, d30 is the pre-migration boot disk, and d24 is the post-migration boot disk. # scdidadm -l ... 24 node1:/dev/rdsk/c0t0d0 d24 /dev/did/rdsk/ ...
3. Creating an HDLM Environment of the SDS or SVM state database to the post-migration environment. In this case, keep the pre-migration environment recognized by Solaris. • Skip to step 21 if SDS or SVM is being used but you are not migrating a replica of the SDS or SVM state database. • Skip to step 21 if SDS or SVM is not being used. 18. Check the state database settings. An example is shown below. 19. Add the post-migration boot disk to the SDS or SVM state database. An example is shown below. 20.
3. Creating an HDLM Environment For details on the required settings, see the documentation for the relevant application.
3. Creating an HDLM Environment 3.8 Configuring a Mirrored Boot Disk Environment Incorporating SVM A mirrored boot disk incorporating SVM can be managed by HDLM. The following configuration is supported: • OS: Solaris 10 • RAID level: Mirroring (no more than three mirrors) • Cluster: None 3.8.1 Precautions When you use the dlmsetconf utility, do not specify the -r option.
3. Creating an HDLM Environment performing the procedures in (2) Uninstalling HDLM from the Boot Disk Environment in 3.17.3 Uninstalling HDLM. 2. Upgrade Solaris. 3. Configure a boot disk environment again by performing the procedure in 3.8.2 Configuring a Boot Disk Environment in Which HDLM Manages the Boot Disk and Mirroring the Environment by Using SVM. When HDLM changes the path used to determine the logic device file name of an HDLM device, perform the procedures below.
3. Creating an HDLM Environment Figure 3-13: Configuring a boot disk environment in which HDLM manages the boot disk, from the local boot disk environment (2) Mirroring a Boot Disk Environment in Which HDLM Manages the Boot Disk by Using SVM Mirror a configured boot disk by using SVM. Figure 3-14: Mirroring a boot disk environment in which HDLM manages the boot disk shows mirroring a boot disk environment.
3. Creating an HDLM Environment Figure 3-14: Mirroring a boot disk environment in which HDLM manages the boot disk In this example, the LU device names of the boot disks are c8t50060E08000436D1d4 and c8t50060E08000436D1d5, as shown in figure above. In the examples of commands and settings here, assume that the following slices are allocated: • Slice 0: / (root) file system • Slice 1: Swap device and dump device • Slice 6: /usr file system • Slice 7: SVM state database 1.
3. Creating an HDLM Environment # metadb -a c8t50060E08000436D1d5s7 3. Decide the metadevice names. In this procedure, specify the settings as follows: • / (root) file system d10(mirror) ---- d11(submirror),d12(submirror) • swap d20(mirror) ---- d21(submirror),d22(submirror) • /usr d30(mirror) 4. ---- d31(submirror),d32(submirror) Create metadevices for / (root), swap, and /usr, and then mirror each of them (single mirror).
3. Creating an HDLM Environment 7. Execute the metaroot command. # metaroot d10 8. 9. Reboot the host. When the host starts, check the swap device and dump device. Then execute the mount command to confirm that / (root) and /usr have been correctly configured.
3.
3. Creating an HDLM Environment 131717 0 - c8t50060E08000436D1d4 000003 0001.0000.0000000000653600.0006 HITACHI .DF600F .77010152 0034 1C Online Non 44 0 - c8t50060E08000436D1d4 000004 0000.0000.0000000000653200.0007 HITACHI .DF600F .77010152 0043 0C Online Own 131670 0 - c8t50060E08000436D1d5 000005 0000.0000.0000000000653600.0007 HITACHI .DF600F .77010152 0043 1C Online Non 35 0 - c8t50060E08000436D1d5 000006 0001.0000.0000000000653200.0007 HITACHI .DF600F .
3. Creating an HDLM Environment Figure 3-15: Current boot disks and boot disks after the migration The LU device names of the current boot disks are c6t50060E08000436D1d4 and c6t50060E08000436D1d5, and the LU device names of the boot disks after the migration are c8t50060E08000436D1d6 and c8t50060E08000436D1d7, as shown in the above figure.
3. Creating an HDLM Environment Table 3-22: Metadevice names Boot disk Current File system /(root) Mirror d10 Sub mirror d11 d12 swap d20 d21 d22 /usr d30 d31 d32 After migration /(root) d100 d101 d102 swap d200 d201 d202 /usr d300 d301 d302 2. Create metadevices on the HDLM device, and then mirror them (single mirror).
3. Creating an HDLM Environment # newfs /dev/md/rdsk/d100 # newfs /dev/md/rdsk/d300 5. Mount the Solaris basic file system of the post-migration environment. For example: # mount /dev/md/dsk/d100 /mnt0 # mount /dev/md/dsk/d300 /mnt1 6. Copy the existing Solaris basic file system to the post-migration environment. For example: # ufsdump 0f - / | (cd /mnt0; ufsrestore rf -) # ufsdump 0f - /usr | (cd /mnt1; ufsrestore rf -) 7.
3. Creating an HDLM Environment 9. Execute the metaroot command, using the -R option to specify the mount point of the / (root) file system after the migration. # metaroot -R /mnt0 d100 10. In order to boot from the post-migration environment, determine the boot device name for the OBP. Only the path used by HDLM to determine the logic device file name of the HDLM device of the boot disk can be specified as the boot device.
3. Creating an HDLM Environment Figure 3-16: Booting from the post-migration environment 12. When the host starts, confirm that the swap device and dump device have been correctly configured. Then, use the mount command to confirm that / (root) and /usr have been correctly configured.
3. Creating an HDLM Environment largefiles/logging/xattr/onerror=panic/dev=1540064 on Mon Mar 9 11:00:25 2009 /usr on /dev/md/dsk/d300 read/write/setuid/devices/intr/ largefiles/logging/xattr/onerror=panic/dev=154012c on Mon Mar 9 11:00:25 2009 13. Set up a two-mirror configuration of / (root), swap, and /usr.
3. Creating an HDLM Environment d101: Submirror of d100 State: Okay Size: 25190400 blocks (12 GB) Stripe 0: Device Start Block Dbase State Reloc Hot Spare /dev/dsk/c8t50060E08000436D1d6s0 0 No Okay Yes d102: Submirror of d100 State: Okay Size: 25190400 blocks (12 GB) Stripe 0: Device Start Block Dbase State Reloc Hot Spare /dev/dsk/c8t50060E08000436D1d7s0 0 No Okay Yes ... 17. Execute the dlnkmgr command's view -path operation to confirm that I/O is issued to the LU set as the boot disk.
3. Creating an HDLM Environment .77010152 0043 1C Online Non 35 0 - c8t50060E08000436D1d7 000006 0001.0000.0000000000653200.0007 HITACHI .DF600F .77010152 0043 0C Online Own 131650 0 - c8t50060E08000436D1d7 000007 0001.0000.0000000000653600.0007 HITACHI .DF600F .77010152 0043 1C Online Non 35 0 - c8t50060E08000436D1d7 18. Back up the contents of the pre- and post-migration environments.
3. Creating an HDLM Environment Figure 3-17: Current boot disks and boot disks after the migration The LU device names of the current boot disks are c8t50060E08000436D1d4 and c8t50060E08000436D1d5, and the LU device names of the boot disk after the migration are c6t50060E08000436D1d6 and c6t50060E08000436D1d7, as shown in the above figure.
3. Creating an HDLM Environment Table 3-23: Metadevice names Boot disk File system /(root) Current Mirror d100 Sub mirror d101 d102 swap d200 d201 d202 /usr d300 d301 d302 After migration /(root) d10 d11 d12 swap d20 d21 d22 /usr d30 d31 d32 2. Create metadevices on the sd/ssd device, and then mirror them.
3. Creating an HDLM Environment # newfs /dev/md/rdsk/d10 # newfs /dev/md/rdsk/d30 5. Mount the Solaris basic file system of the post-migration environment. For example: # mount /dev/md/dsk/d10 /mnt0 # mount /dev/md/dsk/d30 /mnt1 6. Copy the existing Solaris basic file system to the post-migration environment. For example: # ufsdump 0f - / | (cd /mnt0; ufsrestore rf -) # ufsdump 0f - /usr | (cd /mnt1; ufsrestore rf -) 7.
3. Creating an HDLM Environment 9. Execute the metaroot command. Use the -R option to specify the mount point of the / (root) file system after the migration. For example: # metaroot -R /mnt0 d10 10. Reboot from the post-migration environment. Figure 3-18: Booting from the post-migration environment shows the disk that you will boot. Figure 3-18: Booting from the post-migration environment 11. When the host starts, confirm that the swap device and dump device have been correctly configured.
3.
3. Creating an HDLM Environment 15. Execute the metastat command to confirm that the two-mirror configuration has been correctly configured.
3. Creating an HDLM Environment 3.9 Checking the Path Configuration HDLM functions, such as load balancing and failover, are only available for HDLM management-target devices that have more than one active path. After you install HDLM or change the hardware configuration, check the structure and statuses of the paths. To check the path information, use the HDLM command's view operation. The following describes how to check path information by using the dlnkmgr command's view operation.
3. Creating an HDLM Environment making sure that an HDLM device corresponding to the sd or ssd device has been created.
3. Creating an HDLM Environment 3.10 Setting Up HDLM Functions HDLM includes functions like the load balancing function, the automatic failback function, the error logging function, the audit logging function, etc. You can set up these functions by using the dlnkmgr command's set operation. The following subsections describe these setup methods. 3.10.
3. Creating an HDLM Environment Table 3-24: The default and recommended values for each function Function Load-balancing Default value Recommended value on on The Extended Least I/Os algorithm is used. The recommended algorithm depends on the operating environment. on on 30-minute check interval The recommended checking interval depends on the operating environment. on 1-minute check interval The recommended checking interval depends on the operating environment.
3. Creating an HDLM Environment • exrr for the Extended Round Robin algorithm • lio for the Least I/Os algorithm • exlio for the Extended Least I/Os algorithm • lbk for the Least Blocks algorithm • exlbk for the Extended Least Blocks algorithm When you set this value to on, specify rr for round robin or exrr for extended round robin after the -lbtype option.
3. Creating an HDLM Environment If you set the number of times that the error needs to occur to 1, the above condition does not need to be satisfied. The following is an example of setting up automatic failback by using a command: # /opt/DynamicLinkManager/bin/dlnkmgr set -afb on -intvl 10 To enable the automatic failback function, set it to on. When on is specified, you can use the -intvl parameter to specify the checking interval.
3. Creating an HDLM Environment error in the -intvl parameter, and the number of times that the error needs to occur in the -iemnum parameter. When these parameters are omitted, the default values of 30 and 3 are used, respectively. (5) Setting the Error Log Collection Level The error log (the HDLM manager log (dlmmgrn.log (n indicates a file number from 1 to 16))) collection level can be set.
3. Creating an HDLM Environment Table 3-26: Values for the trace level setting Value Description 0 No trace is output. 1 Only error information is output. 2 Program operation summaries are output. 3 Program operation details are output. 4 All information is output. If an error occurs, you might have to set the trace level to 1 or higher to collect any trace information. The higher this value is set, the more information that will be output.
3. Creating an HDLM Environment (8) Setting the Number of Error Log Files The number of the error log files (the HDLM manager log (dlmmgrn.log (n indicates a file number from 1 to 16))) can be set. You can specify a value from 2 to 16 for the number of HDLM manager log files. By specifying both the log file size and the number of log files, you can collect up to 32000000 kilobytes (approximately 30GB) of error logs in total.
3. Creating an HDLM Environment to 1024000 kilobytes of trace data. The following shows an example command that is used to set the number of trace files: # /opt/DynamicLinkManager/bin/dlnkmgr set -systfn 10 Specify the number of trace files in numbers. (11) Setting Up Audit Log Data Collection If you want to collect audit log data, you must also specify the collection level for audit log data and the audit log categories.
3. Creating an HDLM Environment Value Explanation Audit log events of the StartStop, Authentication, and ConfigurationAccess categories are all collected. all This example shows how to enable the collection of audit log data: # /opt/DynamicLinkManager/bin/dlnkmgr set -audlog on -audlv 6 -category all Specify on if you want to collect audit log data, and off if you do not want to collect audit log data.
3. Creating an HDLM Environment # /opt/DynamicLinkManager/bin/dlnkmgr set -audfac local0 3.10.3 Checking the Updated Settings This chapter describes how to check the updated settings by using the dlnkmgr command's set operation after settings have been changed. When you change some settings, you can display information about all of HDLM function settings.
3. Creating an HDLM Environment 3.11 Setting up Integrated Traces When HDLM is used, the dlnkmgr command logs are output to the integrated trace information files of Hitachi Network Objectplaza Trace Library (HNTRLib2) (/var/ opt/hitachi/HNTRLib2/spool/hntr2n.log (n indicates a file number)). If a lot of integrated trace information is output, the older information might end up getting deleted in a very short amount of time.
3. Creating an HDLM Environment If HNTRLib2 has already been installed on a host when you install HDLM, the settings in the trace library will be inherited. If a different Hitachi product is using HNTRLib2 when you attempt to uninstall HDLM, HNTRLib2 will not be uninstalled 3.11.2 Displaying the Hitachi Network Objectplaza Trace Library setup Menu To display the Hitachi Network Objectplaza Trace Library setup menu: 1. Log in as a user with root permissions. 2.
3. Creating an HDLM Environment press the Enter key. A screen to set the size of the integrated trace file appears. The current value is displayed in Current Size(KB). 2. Enter the desired size in New Size(KB). You can specify a value from 8 to 8192 (KB). The default value is 256. Set this to a value larger than the buffer value set in step 2 of 3.11.5 Changing the Buffer Size Per Monitoring Interval Duration. The recommend value is 4096 for collecting an integrated trace.
3. Creating an HDLM Environment The default value is 4. You can specify a value from 1 to 16. The value set here becomes the maximum of n in /var/opt/hitachi/HNTRLib2/spool/ hntr2n.log. The recommended value for integrated trace collection is 8. If you do not want to change the number of integrated trace files, leave New Number(KB) blank, type !, and then press the Enter key to return to the Hitachi Network Objectplaza Trace Library setup menu. 3. Press the Enter key.
3. Creating an HDLM Environment A screen to set the monitoring interval appears. The current value is displayed in Current Span(sec). 5. Enter a shorter interval in New Span(sec). The default value is 10. You can specify a value from 1 to 300 (seconds). The recommended value is 5 for collecting an integrated trace. If you do not want to change the monitoring interval, leave New Span(sec) blank, enter !, and then press the Enter key to return to the Hitachi Network Objectplaza Trace Library setup menu.
3. Creating an HDLM Environment value is 0. If you do not want to change the monitoring interval, leave New Span(sec) blank, type !, and then press the Enter key to return to the Hitachi Network Objectplaza Trace Library setup menu. Note that when you specify a monitoring interval of 0, even if you specify the maximum number of messages in 7: Max messages per span, the amount of integrated trace information to be output will not be adjusted. 3. Press the Enter key.
3. Creating an HDLM Environment The new setting is applied and the Hitachi Network Objectplaza Trace Library setup menu appears again. 3.11.7 Finishing the Hitachi Network Objectplaza Trace Library Settings To close the Hitachi Network Objectplaza Trace Library setup menu when you are finished: 1. In the Hitachi Network Objectplaza Trace Library setup menu, type e, and then press the Enter key. You will be asked to if you wish to save the new settings. 2.
3. Creating an HDLM Environment # rm /opt/hitachi/HNTRLib2/mmap/hntr2mmap.mm 6. Execute the following command to start the integrated trace collection process: # /opt/hitachi/HNTRLib2/bin/hntr2mon -d & 7. Start the programs stopped in step 3. If you stopped programs other than HDLM in step 3, start them.
3. Creating an HDLM Environment 3.12 Creating File Systems for HDLM (When Volume Management Software Is Not Used) When volume management software is not used, the file system directly uses HDLM devices. This section describes how to mount an HDLM-managed device as a UFS file system by specifying the name of the logical device file for an HDLM device.
3. Creating an HDLM Environment Figure 3-19: An output example for the mount -p command Edit the /etc/vfstab file as appropriate based on the settings output. The shaded portion indicates the part of the /etc/vfstab file to set. An example of how to edit the /etc/vfstab file is shown below in Figure 3-20: An example of how to edit the /etc/vfstab file. Figure 3-20: An example of how to edit the /etc/vfstab file Add the shaded portion to the file.
3. Creating an HDLM Environment 3.13 Setting Up VxVM When you are using VxVM, VxVM accesses an LU by using an HDLM device. VxVM will automatically recognize HDLM devices. Perform setup as you would for any other device, by following the VxVM documentation. 3.13.1 Creating a Disk Group This subsection mainly describes the preparations for registering the HDLM device in the VxVM disk group. To register the HDLM device in the VxVM disk group, you must remove non-HDLM devices from VxVM.
3. Creating an HDLM Environment The following describes the procedure for checking whether the shaded HDLM devices are recognized. • vxdisk command Make sure that the devices are recognized by VxVM.
3. Creating an HDLM Environment in the format cUtXdYsZ, corresponding to the device name of the enclosure. Note The device name of the enclosure corresponding to the LU might vary before and after installing HDLM. Make sure that you perform step 1 to check the correspondence between the device name of the enclosure and the LU. If the vxdisk command displays the value of HDevName (cUtXdY) output by the dlnkmgr command's view -path operation, the HDLM device is recognized by VxVM. Go to step 2.
3. Creating an HDLM Environment be removed from VxVM management. To remove non- HDLM devices from VxVM management, see (2) Removing Devices from VxVM on a Controller Basis or (3) Removing Devices From VxVM on a Path Basis of 3.13.3 Removing Devices from VxVM. To apply the settings performed in procedure (2) or (3) in 3.13.3 Removing Devices from VxVM, restart the host. 3. Create a disk group by referring to the VxVM documentation.
3. Creating an HDLM Environment 3.13.3 Removing Devices from VxVM VxVM uses HDLM devices to access LUs managed by HDLM. Therefore, you must remove non-HDLM devices (such as sd or ssd devices), which correspond to the LU, from VxVM management. In this section, excluding devices from VxVM management is called removing devices. The following describes devices that must be removed from VxVM, and the procedure for removing such devices.
3. Creating an HDLM Environment Figure 3-21: Devices That Are Recognized by VxVM before They Are Removed from VxVM You must remove the following non-HDLM devices that are recognized by VxVM. If the following devices are being managed by VxVM, HDLM functions, such as load balancing and failover, will not work normally. • sd or ssd devices (c3t0d0, c4t1d0) When an HDLM device is created, the logical device file of the corresponding sd or ssd device will be deleted.
3. Creating an HDLM Environment actually created. Only one HDLM device is created for one LU#. However, for one LU, VxVM recognizes the same number of HDLM devices as the number of physical paths. # For logical device files of HDLM devices, the number of such files to be created is the same as the number of slices for one LU. Figure 3-22: Devices Recognized by VxVM after Removal of Devices from VxVM shows the devices that are recognized by VxVM after removal of the non-HDLM devices from VxVM.
3. Creating an HDLM Environment The settings files necessary for device suppression are created by selecting menu task 17 (Prevent multipathing/Suppress devices from VxVM's view) or 18 (Allow multipathing/Unsuppress devices from VxVM's view) from the vxdiskadm main menu, and then by selecting menu task 8 (List currently suppressed/non-multipathed devices) from the VolumeManager/Disk/ExcludeDevices or VolumeManager/Disk/ IncludeDevices menu.
3. Creating an HDLM Environment Figure 3-23: When HBA ports on the alternate path differ from those on the primary path connected to the Hitachi storage system To remove devices from VxVM, in the above configuration: 1. Use the dlmvxexclude utility to acquire the mapping between controller numbers and devices.
3. Creating an HDLM Environment execution result of the dlmvxexclude utility. For details on the dlmvxexclude utility, see 7.7 The dlmvxexclude Utility for Assisting Creation of the VxVM Configuration File. The following shows an example of executing the dlmvxexclude utility: The shaded portion, in the next step, is contained in the /etc/vx/ vxvm.exclude file. 2. Make an entry for the controller of the device you wish to remove from VxVM in the controllers section of the /etc/vx/vxvm.exclude file.
3. Creating an HDLM Environment Figure 3-24: Example of a description in the /etc/vx/vxvm.exclude (controllers section) The shaded portion indicates the specified entry. 3. Execute the following command to restart the host and apply the settings. # reboot 4. Execute the dlnkmgr command's view -path operation to identify the HDLM devices that must be recognized by VxVM.
3. Creating an HDLM Environment The devices displayed for HDevName (the shaded portion) are the HDLM devices that must be recognized by VxVM. In the above example, two devices, c5t0d0 and c5t0d1, must be recognized. 5. Execute the vxdisk command to list devices that are recognized by VxVM. The following shows an example of executing the command: If the HDLM devices output by the dlnkmgr command's view -path operation (the shaded portion) are displayed, the HDLM devices have been recognized by VxVM.
3. Creating an HDLM Environment VxVM of 3.13.3 Removing Devices from VxVM. 6. Execute the vxdmpadm command on the HDLM devices listed by the vxdisk command. A list of devices that access the same LU is displayed. Execute the vxdmpadm command on each of the devices listed by the vxdisk command.
3. Creating an HDLM Environment (3) Removing Devices From VxVM on a Path Basis If all the physical paths to one LU go through the same HBA, devices are removed on a path basis. The procedure for removing devices on a path basis is described below by using the configuration example shown in Figure 3-25: When the HBA Port of the Physical Path Connected to a Hitachi Storage System Is the Same As the One Connected to a Third-party Storage System.
3. Creating an HDLM Environment Figure 3-25: When the HBA Port of the Physical Path Connected to a Hitachi Storage System Is the Same As the One Connected to a Third-party Storage System To remove devices from VxVM, in the above configuration: 1.
3. Creating an HDLM Environment numbers and devices. The devices to be removed from VxVM are output to the paths section in the execution result of the dlmvxexclude utility. For details about the dlmvxexclude utility, see 7.7 The dlmvxexclude Utility for Assisting Creation of the VxVM Configuration File. The following shows an example of executing the dlmvxexclude utility: The shaded portion indicates the devices that need to be removed from VxVM. 2.
3. Creating an HDLM Environment Figure 3-26: Example of a description in the /etc/vx/vxvm.exclude (paths section) The shaded portion indicates the specified entry. 3. Execute the following command to restart the host and apply the settings. # reboot 4. Execute the dlnkmgr command's view -path operation to identify the HDLM devices that must be recognized by VxVM.
3. Creating an HDLM Environment that must be recognized by VxVM. In the above example, it is c4t0d0. 5. Execute the vxdisk command to list devices that are recognized by VxVM. The following shows an example of executing the command: If the HDLM devices output by the dlnkmgr command's view -path operation (the shaded portion) are displayed, the HDLM devices have been recognized by VxVM.
3. Creating an HDLM Environment Display of the vxdmpadm command Why suppression is unnecessary Only HDLM devices are displayed. Among the devices corresponding to HDLM-managed devices, the non-HDLM devices have been successfully removed from VxVM. Devices other than HDLM devices are DISABLED. Among the devices corresponding to HDLM-managed devices, the non-HDLM devices have been successfully removed from VxVM (DISABLED is displayed because the host has not been restarted after setup). 7.
3. Creating an HDLM Environment Example 1: The vxdisk command does not display HDLM devices but displays the sd or ssd devices corresponding to those HDLM devices. The vxdisk command displays the SCSI device c3t0d0, which corresponds to an HDLM device, instead of the HDLM device c5t0d0. Example 2: The vxdmpadm command outputs a non-HDLM device, and the STATE column shows ENABLED.
3. Creating an HDLM Environment The vxdmpadm command displays c3t0d0 and c4t1d0 in addition to the HDLM device c5t0d0, and the STATE column shows ENABLED. If sd or ssd devices have not been removed normally from VxVM, as shown in Examples 1 and 2, take the following action: To suppress sd or ssd devices: 1. Execute the vxdmpadm command. In the following example, HDLM device c5t0d0 has been created and a non-HDLM device is ENABLED.
3. Creating an HDLM Environment • Devices in /etc/vx/vxvm.exclude that do not include dlmndrv@ 4. For devices whose tXdY portions are the same, change the controller number described in /etc/vx/vxvm.exclude to one of the devices displayed by the vxdmpadm command. Figure 3-27: The /etc/vx/vxvm.exclude file (before editing) shows the /etc/ vx/vxvm.exclude file before editing. Figure 3-28: The /etc/vx/vxvm.exclude file (after editing) shows the /etc/vx/vxvm.exclude file after editing.
3. Creating an HDLM Environment Figure 3-28: The /etc/vx/vxvm.exclude file (after editing) Among the non-HDLM devices in the ENABLED state, there might be two or more devices whose tXdY portions in the logical device file name (cUtXdYs2 format) are the same. In this case, change the currently assigned controller number in / etc/vx/vxvm.exclude. 5. After modifying all the sd or ssd device entries that have not been suppressed in /etc/vx/vxvm.
3. Creating an HDLM Environment If there is more than one candidate for the controller number to be assigned in step 4, you might not be able to remove devices from VxVM in one operation. In this case, change the controller number you changed in step 4 to another controller number that was displayed using the vxdmpadm command, and then perform steps 5 and 6 again. 3.13.4 Introducing VxVM while Using HDLM To introduce VxVM in an environment in which HDLM is being used: 1.
3. Creating an HDLM Environment group to be registered in Sun Cluster into only one of the nodes. 5. If the VxVM disk group is not registered in Sun Cluster, execute the scsetup utility from the node into which you imported the VxVM disk group to be registered in Sun Cluster, and then register the VxVM disk group into a Sun Cluster disk device group. For details, refer to the explanation of system administration in the Sun Cluster documentation.
3. Creating an HDLM Environment Note After registering the VxVM disk group into a Sun Cluster disk device group, do not use the VxVM command to import or deport that VxVM disk group. If you register the VxVM disk group into a Sun Cluster disk device group where more than 3 nodes are configured, the following message might appear when changing the node that acts as the master for the disk group. If this message appears, ignore it. KAPL05023-E Could not process the IOCTL(00004D03). (0000000D:00000000) 6.
3. Creating an HDLM Environment 3.14 Setting Up SDS If SDS is installed on an HDLM host, SDS uses the logical device file name of the HDLM device to access the HDLM-managed device. Therefore, after you install HDLM, you need to register the logical device file name of the HDLM device. When uninstalling HDLM, unregister the logical device file name of the HDLM device and set the logical device file name of the sd or ssd device. 3.14.1 Notes When using SDS, keep the following in mind.
3. Creating an HDLM Environment or boot -r (at the ok prompt) For more information on SDS settings, refer to the SDS documentation. 3.14.2 Registering HDLM Devices If both HDLM and SDS are used on the same host, perform the following settings after installing HDLM. (1) To Use a Local Metadevice To use a local metadevice: 1. Create a replica of the state database. Create three replicas of the local state database on the disk (slice).
3. Creating an HDLM Environment metadevices created on a shared diskset. When using Sun Cluster, you use the device ID in the device name of the disk. Note When you register a disk into a shared diskset, the existing partitions may be reconfigured. For details, refer to the SDS documentation. To use a shared diskset: 1. Create three replicas of the local state database on the disk (slice). Create these replicas on all hosts that use the shared diskset.
3. Creating an HDLM Environment When using Sun Cluster (stripe type): host1# metainit -s diskset1 d1 1 2 /dev/did/dsk/d10s0 /dev/ did/dsk/d11s0 5. Create a file system on the shared metadevice of the shared diskset. The following shows an example of executing the commands: host1# newfs /dev/md/diskset1/rdsk/d1 6. Mount the created file system to the mount point. The following shows examples of executing the commands.
3. Creating an HDLM Environment 3.15 Setting Up SVM If SVM is installed on an HDLM host, SVM uses the logical device file name of the HDLM device to access the HDLM-managed device. Therefore, after installing HDLM, you need to register the logical device file name of the HDLM device in SVM. When uninstalling HDLM, unregister the logical device file name of the HDLM device and set the logical device file name of the sd or ssd device. 3.15.1 Notes When using SVM, keep the following in mind.
3. Creating an HDLM Environment or boot -r (at the ok prompt) For more information on SVM settings, refer to the SVM documentation. 3.15.2 Registering HDLM Devices If SVM is used, perform the following settings after installing HDLM, and register the SVM volume. (1) To Use a Local Volume To use a local volume: 1. Create a replica of the state database. Create three replicas of the local state database on the disk (slice).
3. Creating an HDLM Environment created on a shared diskset. When using Sun Cluster, you use the device ID in the device name of the disk. Note When you register a disk into a shared diskset, existing partitions may be reconfigured. For detail, refer to the SVM documentation. To use a shared diskset: 1. Create three replicas of the local state database on the disk (slice). Create these replicas on all hosts that use the shared diskset.
3. Creating an HDLM Environment When using Sun Cluster (stripe type): host1# metainit -s diskset1 d1 1 2 /dev/did/dsk/d10s0 /dev/ did/dsk/d11s0 5. Create a file system on the shared volume of the shared diskset. The following shows an example of executing the commands: host1# newfs /dev/md/diskset1/rdsk/d1 6. Mount the created file system to the mount point. The following shows examples of executing the commands.
3. Creating an HDLM Environment 3.16 Setting Up VCS When VCS is installed on an HDLM host, you can register HDLM devices, or VxVM disk groups containing HDLM devices, as resource groups. For details on registering resource groups, refer to the VCS documentation. Notes The I/O fencing function only supports the configurations shown in Table 3-33: Configurations that support the I/O fencing function. Table 3-33: Configurations that support the I/O fencing function OS VCS Solaris 9 4.
3. Creating an HDLM Environment If the number of paths differs depending on an HDLM device, use the maximum number for the calculation. #2 The I/O timeout value used for sending a notification to the sd or ssd driver depends on the timeout settings for switches or HBAs.
3. Creating an HDLM Environment 3.17 Uninstalling HDLM This section explains how to return the HDLM environment to the way it was before HDLM was installed, and describes each step of the process. 3.17.1 Overview of HDLM Uninstallation Return the HDLM environment to the way it was before HDLM was installed by following the flow shown below in Figure 3-29: Overview of HDLM uninstallation. Figure 3-29: Overview of HDLM uninstallation # Perform as required. 3.17.
3. Creating an HDLM Environment storage system. This information will enable you to keep the correspondence with the logical device file names after HDLM has been uninstalled. You can collect this correspondence information by outputting the results of the dlnkmgr command's view -path operation to a file. The following shows an example of executing the command: # /opt/DynamicLinkManager/bin/dlnkmgr view -path > /var/tmp/ old_view_path_list.
3. Creating an HDLM Environment 7. If HDLM devices are being used in Solaris 10 ZFS, execute the following command to export all ZFS storage pools that are using HDLM: # zpool export pool-name 8. Edit the /etc/vfstab file. In the /etc/vfstab file, delete or comment out the specifications of the logical device files for the HDLM devices (cUtXdYsZ) and the ZFS file system whose configurations you want to change, and then release the setting for automatic mounting. 9.
3. Creating an HDLM Environment As the utility instructs, enter the name of the VxVM disk device group you want to unregister. If the HDLM device has been configured as a Quorum device, perform the following operations to remove the Quorum device settings. 1. Check the device ID of the Quorum device. The following shows an example of executing the command: # /usr/cluster/bin/scconf -pv | grep Quorum # /usr/cluster/bin/scdidadm -L 2.
3. Creating an HDLM Environment 8. Confirm that the cluster node enters the maintenance mode. Make sure that the node that is in the maintenance mode has the following status: • The node is in the Offline status. • Both the current Quorum votes present and Quorum votes possible are 0. The following shows an example of executing the command: host2# /usr/cluster/bin/scstat -q 9. If the cluster is a two-node configuration, enter the install mode.
3. Creating an HDLM Environment (3) Remove VCS Settings If you are using VCS, before uninstalling HDLM, unregister the HDLM devices and VxVM disk groups containing HDLM devices, from the resource groups. For details on how to remove the VCS settings, refer to the VCS documentation. (4) Remove VxVM Settings If you are using both HDLM and VxVM on the same host, before you uninstall HDLM, perform the following to delete the HDLM devices from VxVM.
3. Creating an HDLM Environment # vxdg deport disk-group-name 3. To enable VxVM management, delete the entries for sd or ssd devices and the entries for HDLM devices from the /etc/vx/vxvm.exclude file. After uninstalling HDLM, the VxVM volumes that were used with HDLM devices can be used with sd or ssd devices. For details, see (1) VxVM Settings of 3.17.4 Settings Needed After HDLM Uninstallation.
3. Creating an HDLM Environment When using Sun Cluster: host1# umount /global/diskset1/d1 host1# metaclear -s diskset1 d1 2. Delete the disk from the shared diskset. The following shows an example of executing the command. When not using Sun Cluster: host1# metaset -s diskset1 -d -f c5t1d2 c5t1d3 When using Sun Cluster: host1# metaset -s diskset1 -d -f /dev/did/dsk/d10 /dev/did/ dsk/d11 3. Delete the shared diskset. The following shows an example of executing the command.
3. Creating an HDLM Environment When using a shared diskset 1. Delete the shared volume. The following shows an example of executing the command. When not using Sun Cluster: host1# umount /diskset1 host1# metaclear -s diskset1 d1 When using Sun Cluster: host1# umount /global/diskset1/d1 host1# metaclear -s diskset1 d1 2. Delete the disk from the shared diskset. The following shows an example of executing the command.
3. Creating an HDLM Environment Note When uninstalling HDLM on a host where a Device Manager Agent 5.0 or later is installed, do not execute any of the following commands of Device Manager Agent during uninstallation. Also, do not uninstall HDLM while executing any of the following Device Manager Agent commands: hbsasrv, HiScan, hdvmagt_account, hdvmagt_schedule, hldutil, TIC To uninstall HDLM: 1. Log in to Solaris as a user with root permissions. 2.
3. Creating an HDLM Environment Execute one of the following commands: • # reboot -- -r • boot -r (at the ok prompt) When Sun Cluster is being used: Execute the following command on one of the nodes: # /usr/cluster/bin/scshutdown -g0 -y Then, execute the following command at the ok prompt on all nodes: boot -r Note While a node is starting, the message below may be output to the console or syslog: Could not stat: ../../devices/pseudo/dlmndrv@1/ dlmfdrv@0,0:c,raw path not loaded.
3. Creating an HDLM Environment Figure 3-30: Configuration Example (Before Uninstallation) # A virtual disk backend is a location where virtual disk data is stored. A disk, disk slice, file, or volume (such as ZFS, SVM, and VxVM) can be used for the backend. In the control domain, register the install_disk file on a built-in disk as a virtual disk backend. The I/O domain recognizes this disk as a virtual disk. The device name of this virtual disk is /dev/[r]dsk/c0d0.
3. Creating an HDLM Environment Domain Name ldg1 Domain Type An I/O domain (including service domains) Boot Disk vol1@primary-vds0 Exporting Virtual Disk Backends vol1: /dev/dsk/ c4t50060E8010027A82d0s2 (For a local boot disk of the guest domain) vol2: /dev/dsk/ c4t50060E8010027A82d1s2 (For a data disk of the guest domain) ldg2 A guest domain vol1@ldg1-vds0 -- The following figure shows a configuration example for an LDoms environment before uninstalling HDLM.
3. Creating an HDLM Environment To uninstall HDLM from a host in which LDoms have been set: 1. Log in to the control domain. 2. Execute the ldm stop command to stop the guest domain. For example, to stop a guest domain ldg2, execute the command as follows: # ldm stop ldg2 3. 258 Execute the ldm unbind command to unbind the resource of the guest domain.
3. Creating an HDLM Environment For example, to unbind a resource of the guest domain ldg2, execute the command as follows: # ldm unbind ldg2 4. Execute the ldm rm-vdisk command to remove virtual disks from the guest domain. For example, to remove virtual disks vdisk1 and vdisk2 from the guest domain ldg2, execute the command as follows: # ldm rm-vdisk vdisk1 ldg2 # ldm rm-vdisk vdisk2 ldg2 5.
3. Creating an HDLM Environment virtual disk backends from the service domain on the I/O domain to which HDLM will be installed. For example, to export virtual disk backends vol1 and vol2 from the service domain ldg1, execute the command as follows: # ldm add-vdsdev /dev/dsk/c2t50060E8010027A82d0s2 vol1@ldg1-vds0# # ldm add-vdsdev /dev/dsk/c2t50060E8010027A82d1s2 vol2@ldg1-vds0# # /dev/dsk/c2t50060E8010027A82d0s2 and /dev/dsk/ c2t50060E8010027A82d1s2 are sd or ssd device names after exporting. 9.
3. Creating an HDLM Environment Figure 3-31: Configuration Example (After Uninstallation) In this configuration example, /dev/dsk/c2t50060E8010027A82d0s2 and /dev/ dsk/c2t50060E8010027A82d1s2 are sd or ssd devices. 3.17.
3. Creating an HDLM Environment later: 1. Execute the following command to import the disk group: # vxdg import disk-group-name 2. To use a volume that belongs to a disk group, execute the following command to activate the volume: # vxvol -g disk-group-name startall (2) SDS Settings Create meta devices. For details, refer to the SDS documentation. (3) SVM Settings Create volumes. For details, refer to the SVM documentation.
3. Creating an HDLM Environment 5. Set up a Quorum device at any of the nodes that constitute the cluster. Check the device ID of the sd or ssd device, which is used for the Quorum device. The following shows an example of executing the command: # /usr/cluster/bin/scdidadm -L Set up a Quorum device.
3. Creating an HDLM Environment 3.17.5 Uninstalling Hitachi Network Objectplaza Trace Library (HNTRLib2) When you uninstall HDLM, if the KAPL09019-E or KAPL09020-E message is output, follow the procedure below to uninstall HNTRLib2. To uninstall HNTRLib2: 1. Log in to Solaris as a user with root permissions. 2. Execute the following command to unregister the name of the bundled program products: # /opt/hitachi/HNTRLib2/etc/hntr2cancel "Hitachi Dynamic Link Manager" 3.
3. Creating an HDLM Environment files will not be deleted during uninstallation. In this case delete these files after uninstallation. 3.17.6 Uninstalling Hitachi Network Objectplaza Trace Library (HNTRLib) After you uninstall HDLM version 04-00 or earlier, check the manuals and Release Notes that came with the software installed on the host. If no applications other than HDLM are using HNTRLib, uninstall it by following the procedure below: To uninstall HNTRLib: 1.
3. Creating an HDLM Environment • Even if you attempt to uninstall HNTRLib2 when HNTRLib is installed, HNTRLib will not be uninstalled. If no other programs are using HNTRLib, delete it manually.
Chapter 4. HDLM Operation This chapter describes operating procedures for HDLM, including how to operate HDLM and the HDLM manager, and how to change the configuration of the operating environment. 4.1 4.2 4.3 4.4 4.
4. HDLM Operation 4.1 Notes on Using HDLM This section explains the notes on using HDLM and using the environment in which HDLM is installed. Make sure that you read this section. 4.1.1 Displaying Path Information • The path IDs displayed by using the dlnkmgr command's view operation differ depending on the order in which paths were detected when the host started. For this reason, you should always use the path name to specify a path.
4. HDLM Operation The shaded portion indicates mapping from combinations of sd or ssd device types and instance numbers to HDLM device logical device file names. 4.1.4 Storage System • Start up the storage system before you start up the hosts, so that Solaris can detect the storage system. • Do not change the vendor ID or product ID of a storage system. If you change either of them, HDLM will not be able to recognize the storage system. 4.1.
4. HDLM Operation 4.1.7 Starting the Host, System, and HDLM in a Solaris 10 Environment When starting the host If the host is started in single-user mode, the displayed path status indicates an error even when the actual path status is normal. Therefore, before you use HDLM devices in single-user mode, execute the procedure described in (1) Initializing HDLM When the Host Is Started in Single-User Modein 4.1.9 Operation in Single-User Mode. 4.1.
4. HDLM Operation 2. Execute the following HDLM command to place the paths online: # /opt/DynamicLinkManager/bin/dlnkmgr online -s Even if the paths are physically connected, this command may fail if executed immediately after the boot processing completes. In this case, wait at least 10 seconds and then re-execute the command. If still unsuccessful, take the action indicated in the displayed message. 3.
4. HDLM Operation • Changing the configuration of the HDLM operating environment • Applying an HDLM service pack # In single-user mode, HDLM command activity is not logged to integrated trace files. For applications • Install an application • Upgrade an application • Set or change the configuration 4.1.
4. HDLM Operation 4.2 HDLM Operations Using Commands This section explains how to use the HDLM command operations to utilize HDLM functions. For details on the command operations, see 6. Command Reference. 4.2.1 Notes on Using Commands • Execute the command as a user with root permissions. • To specify a parameter value containing one or more spaces, enclose the entire value in double quotation marks ("). 4.2.
4. HDLM Operation 4.2.3 Changing the Status of Paths This section explains how to change path statuses. (1) Changing the Status of Paths to Online To change the status of paths to online: 1. Check the current status of the paths. To place paths online by specifying an HBA port, CHA port, or single path check the path name or PATH_ID used to manage the path.
4. HDLM Operation (2) Changing the Status of Paths to Offline(C) To change the status of paths to Offline(C): 1. Check the current status of the paths. To change the status of a path to Offline(C) by specifying an HBA port, CHA port, or single path check the path name or PATH_ID used to manage the path.
4. HDLM Operation 4.2.4 Viewing LU Information This section explains how to display LU information by using the dlnkmgr command. To display LU information, execute the dlnkmgr command's view operation with the -lu parameter specified.
4. HDLM Operation Information). The execution result of this operation is displayed on a single line for each path. The following shows an example of executing the view operation: # /opt/DynamicLinkManager/bin/dlnkmgr view -drv PathID HDevName Device LDEV 000000 c6t5d0 sd74 9970/9980.15001.0670 000001 c6t5d1 sd75 9970/9980.15001.0671 000002 c6t5d2 sd76 9970/9980.15001.0672 000003 c6t5d3 sd77 9970/9980.15001.0673 000004 c6t5d4 sd78 9970/9980.15001.0674 000005 c6t5d5 sd79 9970/9980.15001.
4. HDLM Operation # /opt/DynamicLinkManager/bin/dlnkmgr clear -pdst KAPL01049-I Would you like to execute the operation? Operation name = clear [y/n]:y KAPL01001-I The HDLM command completed normally. Operation name = clear, completion time = yyyy/mm/dd hh:mm:ss # 3. Check to see whether the statistical information for all the paths has been initialized. The following shows an example of executing the command: # /opt/DynamicLinkManager/bin/dlnkmgr view -path 4.2.
4. HDLM Operation # /opt/DynamicLinkManager/bin/dlnkmgr view -sys -audlog Audit Log : off Audit Log Category : Audit Log Facility : KAPL01001-I The HDLM command completed normally. Operation name = view, completion time = yyyy/mm/dd hh:mm:ss # For details on the displayed items and their descriptions, see 6.7 view (Displays Information). (2) Setting Up the Operating Environment To set up the HDLM operating environment, execute the dlnkmgr command's set operation.
4. HDLM Operation KAPL01001-I The HDLM command completed normally. Operation name = set, completion time = yyyy/mm/dd hh:mm:ss # To check whether the settings have been applied, perform the procedure described in (1) Viewing the Operating Environment. 4.2.8 Viewing License Information This section explains how to display license information. To display license information, execute the dlnkmgr command's view operation with the -sys and -lic parameters specified.
4. HDLM Operation the -sys -lic parameter to make sure that the license is installed correctly. The following shows an example of executing the command: # /opt/DynamicLinkManager/bin/dlnkmgr set -lic KAPL01049-I Would you like to execute the operation? Operation name = set [y/n]: y KAPL01071-I A permanent license was installed. # 4.2.10 Viewing HDLM Version Information This section explains how to display HDLM version information.
4. HDLM Operation 4.2.11 Viewing HDLM Component Information This section explains how to display HDLM component information. To display HDLM component information, execute the dlnkmgr command's view operation with the -sys parameter specified. The following shows an example of executing the command is executed: # /opt/DynamicLinkManager/bin/dlnkmgr view -sys HDLM Version : x.x.
4. HDLM Operation 4.3 Starting and Stopping the HDLM Manager If an error occurs in the system, such as in an HDLM program, you may need to manually stop or start the HDLM manager to recover from the error. 4.3.1 Starting the HDLM Manager To start the HDLM manager, log in to Solaris as a user with root permissions and then execute the following command. # /etc/init.d/DLMManager start Lower-case characters can be used for the command name (dlmmanager). # /etc/init.
4. HDLM Operation # /opt/DynamicLinkManager/bin/dlnkmgr view -sys -msrv HDLM Manager Ver WakeupTime Dead KAPL01001-I The HDLM command completed normally. Operation name = view, completion time = yyyy/mm/dd hh:mm:ss # When the HDLM Manager column shows Dead, the HDLM manager is inactive.
4. HDLM Operation 4.4 HDLM Resident Processes Table 4-1: HDLM resident processes lists and describes the resident processes in HDLM. To monitor these processes, use the names below. Table 4-1: HDLM resident processes Process Description dlmmgr HDLM manager process hbsa_service Hitachi Command Suite Common Agent Component# process hntr2mon Hitachi Network Objectplaza Trace Library (HNTRLib2) process # You only need to monitor this process when HDLM is linked to Global Link Manager.
4. HDLM Operation 4.5 Changing the Configuration of the HDLM Operating Environment This section describes how to reconfigure the environment for running HDLM. If you perform any one of the following operations, you will also need to perform the operations that are described in this section: • Adding a new LU, or deleting or changing an existing LU • Adding or deleting a path for an existing LU • Editing the HDLM driver unconfiguration definition file (/opt/ DynamicLinkManager/config/dlmfdrv.
4. HDLM Operation dlmfdrv.conf). 3. Execute the dlmsetconf utility for creating the HDLM driver configuration definition file. Edit the HDLM driver unconfiguration definition file (/opt/ DynamicLinkManager/config/dlmfdrv.unconf) as required, and then execute the dlmsetconf utility. For details on how to edit the HDLM driver unconfiguration definition file (/ opt/DynamicLinkManager/config/dlmfdrv.unconf), see 3.5.5 Using the HDLM Device Unconfiguration Function When Performing a New Installation of HDLM.
4. HDLM Operation When using VxVM, execute the following command to confirm that the DEVICE column of the LUs being managed by HDLM is displayed in the cWtXdYsZ format (for example, c6t0d10s2): # vxdisk list If the value is displayed in the cWtXdYsZ format, perform the following procedures: 1. Delete the /etc/vx/disk.info file. 2.
4. HDLM Operation Execute the following command to display the driver configuration information, and then refer to the instance field of the HDLM driver obtained in step 1 (the driver name of the HDLM driver is dlmfdrv). Identify the path used to determine the HDLM logical device file name of the LU from the FilterPath property. # /usr/sbin/prtconf -vP The following shows an example of executing the command.
4. HDLM Operation Environment. 2. Change the path configuration. 3. Return to the HDLM boot disk environment, following the procedure described in 3.6 Configuring a Boot Disk Environment. (5) Dynamic Reconfiguration (DR) for Solaris You cannot use the Solaris dynamic reconfiguration (DR) function on HBAs that exist on a path managed by HDLM. To add, delete, or replace such HBAs, you need to perform a reconfiguration reboot of the host. 4.5.
4. HDLM Operation • Reconfiguring an LU that is not managed by HDLM, as an HDLM-managed device Reconfigure an LU that has already been added to the host, but is not managed by HDLM, as an HDLM-managed device. The following operations are not supported in the dynamic reconfiguration of the HDLM device. To enable the following operations, restart the host with the reconfiguration option specified.
4. HDLM Operation replaced, processing to unconfigure or reconfigure the HBA driver and either the sd or ssd driver might not complete until the host is restarted. When you modify disk device configurations (such as deleting or removing LUs in Solaris 8 or 9), or server-side hardware configurations, execute the dlmsetconf utility, and then restart the host, with the reconfiguration option specified.
4. HDLM Operation • If the major and minor numbers of the sd or ssd device of an LU to be newly added to HDLM matches the major or minor number of an sd or ssd device previously registered in HDLM, dynamic reconfiguration cannot be performed. You must restart the host. • Dynamic reconfiguration cannot be performed when using VxVM, because the host must be restarted in order to use (in VxVM) an LU that has been newly added to HDLM.
4. HDLM Operation Figure 4-1: Example of executing the format command 3. Execute the dlmsetconf utility for creating the HDLM driver configuration definition file to update the HDLM driver configuration definition file (/ kernel/drv/dlmfdrv.conf). # /opt/DynamicLinkManager/bin/dlmsetconf 4. Set up LUs not to be managed by HDLM. If you do not want HDLM to manage one of the newly recognized LUs, set up the LU so that it is excluded from HDLM management. For details, See 3.5.
4. HDLM Operation Specify settings as shown in (2) Removing Devices from VxVM on a Controller Basis or (3) Removing Devices From VxVM on a Path Basis of 3.13.3 Removing Devices from VxVM. 6. Force the HDLM driver to recognize the added LU. In environments supporting dynamic reconfiguration of HDLM devices, execute the dlmcfgmgr utility for managing the HDLM configuration. HDLM device will be reconfigured without restarting the host.
4. HDLM Operation 8. Register the added LU to the application. (3) Adding a New LU By Restarting the Nodes (When Using Sun Cluster) Note Have the HDLM drivers on all nodes in the cluster recognize the LU newly recognized by the node, and make the configuration the same for each node.
4. HDLM Operation 1. Set up the storage system and switch so that Solaris recognizes the LU. 2. Check the LU newly recognized by the node. You can check the LU by using the format command. Figure 4-2: Example of executing the format command shows an example of executing the format command. The shaded portion represents the LU newly recognized by the node. Figure 4-2: Example of executing the format command 3.
4. HDLM Operation Execute the following command on any node: # /usr/cluster/lib/sc/scsi -c inkeys -d the-name-of-the-logical-device-file-on-slice-2-of-the-LU-ne wly-recognized-by-the-node The following shows an example of executing the command: # /usr/cluster/lib/sc/scsi -c inkeys -d /dev/rdsk/ c3t50060E80033A9902d2s2 If the key is displayed, perform the steps beginning with step 5.
4. HDLM Operation 9. Make sure that the logical device files for the sd or ssd device are backed up. When you restart the node, with the reconfiguration option specified, after running the dlmsetconf utility in the next step (step 10), the logical device files of the sd or ssd device to be managed by HDLM will be deleted. Make sure that these files are backed up before you execute the dlmsetconf utility to restart the node.
4. HDLM Operation Devices from VxVM. 12. Stop all nodes in the cluster. If the key is displayed in step 4, and nodes are running in non-cluster mode in step 6, execute the following commands on all nodes: # shutdown -g0 -y -i0 When running nodes in cluster mode, execute the following command on any node: # /usr/cluster/bin/scshutdown -g0 -y 13. Start all nodes in the cluster simultaneously, in cluster mode. At the ok prompt, execute the boot -r command.
4. HDLM Operation Trace File Size(KB) : 1000 Number Of Trace Files : 4 Path Health Checking : on(30) Auto Failback : on(1) Intermittent Error Monitor : off HDLM Manager Ver WakeupTime Alive x.x.x-xx yyyy/mm/dd hh:mm:ss HDLM Alert Driver Ver WakeupTime ElogMem Size x.x.x-xx yyyy/mm/dd hh:mm:ss 4096 Alive HDLM Driver Ver WakeupTime Alive x.x.x-xx yyyy/mm/dd hh:mm:ss License Type Expiration Permanent KAPL01001-I The HDLM command completed normally.
4. HDLM Operation 17. Delete the device ID information for devices that do not exist. Execute the following command: # /usr/cluster/bin/scdidadm -C 18. Allocate the device ID.
4. HDLM Operation # /usr/cluster/bin/scdidadm -r 19. Notify the node of the most recent device ID information. Execute the following command: # /usr/cluster/bin/scdidadm -ui 20. Update the Sun Cluster global device namespace. Execute the following command: # /usr/cluster/bin/scgdevs 21. When using, as a Quorum device, an HDLM device of an LU newly recognized for the first time by HDLM driver, set up the Quorum device on any node in the cluster.
4. HDLM Operation results of the command, compare the correspondence with the correspondence that took place before the HDLM installation, and then re-register HDLM devices with the programs that will use them. If you want to use volume management software to create a file system on an HDLM-managed device, skip to step 24. 23. Perform registration for the disk device group. If you are using VxVM, see 3.13.5 Linking VxVM and Sun Cluster. If you are using SDS, see (2) To Use a Shared Diskset of 3.14.
4. HDLM Operation Figure 4-3: Example of executing the format command 3. If setup is performed for Quorum devices and disk device groups, on an LU newly recognized by the node, release these settings. When you unregister the SDS or SVM diskset, the diskset data will be deleted so first back up any necessary data. For details on how to remove settings for Quorum devices and VxVM disk groups, see (2) Remove Sun Cluster Settings of 3.17.2 Preparations for HDLM Uninstallation.
4. HDLM Operation The following shows an example of executing the command: # /usr/cluster/lib/sc/scsi -c inkeys -d /dev/rdsk/ c3t50060E80033A9902d2s2 Add the new LU to HDLM by following the steps in (3) Adding a New LU By Restarting the Nodes (When Using Sun Cluster)beginning with step 5. Even when you release the setup for Quorum devices and disk device groups, since the key used for disk reservation will remain registered, you will need to restart the node in non-cluster mode and unregister the key.
4. HDLM Operation 7. If you do not want HDLM to manage one of the LUs newly recognized by the node, set up the LU so that it is excluded from HDLM management. For details, see 3.5.5 Using the HDLM Device Unconfiguration Function When Performing a New Installation of HDLM. If you do not want to specify settings to exclude the LUs from HDLM management, skip to step 8. 8.
4. HDLM Operation indicates the path accessing the LU recognized for the first time by the HDLM driver. 11. Delete the device ID information for devices that do not exist.
4. HDLM Operation 12. Allocate the device ID. Execute the following command: # /usr/cluster/bin/scdidadm -r 13. Notify the node of the most recent device ID information. Execute the following command: # /usr/cluster/bin/scdidadm -ui 14. Update the Sun Cluster global device namespace. Execute the following command: # /usr/cluster/bin/scgdevs 15. When using, as a Quorum device, an HDLM device of an LU newly recognized for the first time by HDLM driver, set up the Quorum device on any node in the cluster.
4. HDLM Operation Check the correspondence between HDevName (the name of the logical device file of the HDLM device) and iLU (the internal LU number) from the execution results of the command, compare the correspondence with the correspondence that took place before the HDLM installation, and then re-register HDLM devices with the programs that will use them. If you want to use volume management software to create a file system on an HDLM-managed device, skip to step 18. 17.
4. HDLM Operation (1) Changing the Configuration by Restarting the Host 1. Back up the data in the file systems, volumes, and metadevices that use the HDLM device you are reconfiguring. 2. Execute the following command to unmount all mount points that use HDLM: # umount mount-point 3. If HDLM devices are being used in Solaris 10 ZFS, execute the following command to export all their ZFS storage pools: # zpool export pool-name 4. Edit the /etc/vfstab file.
4. HDLM Operation LU you are deleting, replace the LU. 9. Shut down the host. 10. Change the configuration. 11. Boot the host. 12. Make sure that Solaris recognizes the reconfigured operating environment. If you added a path: There is nothing to check. If you deleted a path: Execute the dlnkmgr command's view operation specifying the -lu parameter, and check that the deleted path is Offline(E). An execution example is shown below: # dlnkmgr view -lu ...
4. HDLM Operation The KAPL10242-I message appears when you execute the dlmsetconf utility. Type y in response. 15. Execute the following command to compare the HDLM driver unconfiguration definition file before and after execution of the dlmsetconf utility: # diff dlmfdrv.unconfbakn dlmfdrv.unconf Compare the old and new files to make sure that the reconfigured environment is recognized by HDLM. An example is shown below.
4.
4. HDLM Operation In the lines that define the LU that is to be removed from HDLM management, delete the hash mark (#) in front of the LDEV property line. Do not change anything else. 17. To remove an LU from HDLM management, apply the contents of the HDLM driver unconfiguration definition file to the HDLM driver configuration definition file. Use the following command to apply the settings in the HDLM driver unconfiguration definition file (/opt/DynamicLinkManager/config/ dlmfdrv.
4. HDLM Operation # /usr/cluster/bin/scgdevs 25. If you cleared the settings for a Quorum device in step 3, perform the following procedure: When using, as a Quorum device, an HDLM device of an LU newly recognized for the first time by HDLM driver, set up the Quorum device on any node in the cluster. The following shows an example of executing the command: Check the device ID of the HDLM device.
4. HDLM Operation in step 1. 31. If you released file system settings in steps 2 to 4, check the HDLM devices after reconfiguration, and then set up the file system again. (2) Deleting an LU via Dynamic Reconfiguration Note • You can delete an LU via dynamic reconfiguration only in Solaris 10. • You cannot delete an LU via dynamic reconfiguration if that LU is being used by an upper-level application. • You cannot delete only some of the paths to an LU; all paths must be deleted at one time.
4. HDLM Operation # /opt/DynamicLinkManager/bin/dlmcfgmgr -a The HDLM device will be deleted without restarting the host. 5. Execute the dlnkmgr command's view -path operation to check the path status after the configuration change. The following shows an example of executing the command: # /opt/DynamicLinkManager/bin/dlnkmgr view -path Paths:000006 OnlinePaths:000006 PathStatus IO-Count IO-Errors Online 27 0 PathID PathName DskName iLU ChaPort Status Type IO-Count IO-Errors DNum HDevName 000000 0000.
4. HDLM Operation 4.5.5 Specifying Whether a Logical Unit Is To Be Managed by HDLM (When Not Using Sun Cluster) This section describes the procedures for modifying the configuration of an HDLM-managed device. You must use the dlmsetconf utility for creating the HDLM driver configuration definition file to modify the configuration of an HDLM-managed device, therefore you should also see 7.6 The dlmsetconf Utility for Creating the HDLM Driver Configuration Definition File.
4. HDLM Operation # /opt/DynamicLinkManager/bin/dlmsetconf -u 4. Restart the host, with the reconfiguration option specified: # reboot -- -r (2) Changing a Non-HDLM-Managed Device to an HDLM-Managed Device To change a non-HDLM-managed device to an HDLM-managed device: 1. Check the path information of the LU to be managed by HDLM. Use the format command to check the information for each path of the LU to be managed by HDLM.
4. HDLM Operation Using a text editor, edit the definitions in the HDLM driver unconfiguration definition file (/opt/DynamicLinkManager/config/dlmfdrv.unconf) for the LU to be managed by HDLM. Based on the path information of the LU to be managed by HDLM checked in step 1 (shaded portion), find the correct LU in the HDLM driver unconfiguration definition file, and add a hash mark (#) in front of the LDEV property line corresponding to the LU.
4. HDLM Operation definition file (/kernel/drv/dlmfdrv.conf). Execute following the dlmsetconf utility to apply the settings in the HDLM driver unconfiguration definition file (/opt/DynamicLinkManager/config/ dlmfdrv.unconf) to the HDLM driver configuration definition file (/kernel/ drv/dlmfdrv.conf): # /opt/DynamicLinkManager/bin/dlmsetconf -u 4. Execute the dlmcfgmgr utility for managing the HDLM configuration, or restart the host, with the reconfiguration option specified.
4. HDLM Operation The shaded portion indicates the LDEV information about the LU that is to be removed from HDLM management. 3. If setup is being performed for Quorum devices and disk device groups in the LU that you want to remove from HDLM management, release the LU settings. When you unregister the SDS or SVM diskset, the diskset data will be deleted so first back up any necessary data.
4. HDLM Operation Figure 4-8: The /opt/DynamicLinkManager/config/dlmfdrv.unconf file (before editing) Figure 4-9: The /opt/DynamicLinkManager/config/dlmfdrv.unconf file (after editing) The shaded portion indicates the text to be edited. Note Do not edit anything other than deleting the hash mark (#) from the beginning of the LDEV property line. 5. Apply the settings in the HDLM driver unconfiguration definition file (/opt/ DynamicLinkManager/config/dlmfdrv.
4. HDLM Operation Figure 4-10: Result output by executing the dlmvxexclude utility Figure 4-11: Coding in the /etc/vx/vxvm.exclude file The shaded portion indicates the device not to be managed by HDLM and to be released from the VxVM disabled state.
4. HDLM Operation For details on the dlmvxexclude utility, see 7.7 The dlmvxexclude Utility for Assisting Creation of the VxVM Configuration File. 7. Stop all nodes in the cluster. On the desired one node, execute the following command: # /usr/cluster/bin/scshutdown -g0 -y 8. Start all nodes in the cluster simultaneously, in cluster mode. At the ok prompt, execute the boot -r command.
4. HDLM Operation HDLM Alert Driver Ver WakeupTime ElogMem Size x.x.x-xx yyyy/mm/dd hh:mm:ss 4096 Alive HDLM Driver Ver WakeupTime Alive x.x.x-xx yyyy/mm/dd hh:mm:ss License Type Expiration Permanent KAPL01001-I The HDLM command completed normally. Operation name = view, completion time = yyyy/mm/dd hh:mm:ss # • Even when cluster software is used, it is not displayed in Support Cluster. Nevertheless, no problem exists with the functionality for cluster support. 10.
4. HDLM Operation Execute the following command: # /usr/cluster/bin/scdidadm -C 13. Allocate the device ID. Execute the following command: # /usr/cluster/bin/scdidadm -r Note The following message might be output when the scdidadm -r command is executed: Unable to open '/dev/rdsk/c5t50060E80033A9902d2s2' This message is output when a logical device file for a non-HDLM-managed device exists. Delete the unnecessary logical device file. For details, refer to the Sun Cluster documentation. 14.
4. HDLM Operation Execute the following command to clear the install mode of a cluster when it consists of two nodes: # /usr/cluster/bin/scconf -c -q reset 17. Perform registration for the disk device group, create the file system, or set up the global mount as necessary. For details, refer to the Sun Cluster documentation.
4. HDLM Operation 3. If setup is being performed for Quorum devices and disk device groups in the LU to be managed by HDLM, release the LU settings. When you unregister the SDS or SVM diskset, the diskset data will be deleted so first back up any necessary data. For details on how to release settings for Quorum devices and VxVM disk groups, see (2) Remove Sun Cluster Settings of 3.17.2 Preparations for HDLM Uninstallation.
4. HDLM Operation The following shows an example of executing the command: # /usr/cluster/lib/sc/scsi -c scrub -d /dev/rdsk/ c3t50060E80033A9902d2s2 8. Make sure that the key for the LU to be managed by HDLM has been unregistered. Perform step 4 again on any node, and check that the key is not displayed. 9. Edit the HDLM driver unconfiguration definition file (/opt/ DynamicLinkManager/config/dlmfdrv.unconf).
4. HDLM Operation Figure 4-15: The /opt/DynamicLinkManager/config/dlmfdrv.unconf file (after editing) The shaded portion indicates the text to be edited. Note Do not make any modifications other than adding or deleting the hash mark (#) in front of the LDEV property line corresponding to the LU. 10. Apply the settings in the HDLM driver unconfiguration definition file (/opt/ DynamicLinkManager/config/dlmfdrv.unconf) to the HDLM driver configuration definition file (/kernel/drv/dlmfdrv.
4. HDLM Operation Note The following message might be output to the console window or syslog while the node is running: Could not read symbolic link for: /dev/rdsk/ c3t50060E80033A9902d2s2 path not loaded This message is output when HDLM deletes a logical device file for an sd or ssd device managed by HDLM. Perform the procedure for updating the device ID information (steps 17 to 20) so that this message is not output next time the node starts. For details, refer to the Sun Cluster documentation. 14.
4. HDLM Operation • Even when cluster software is used, it is not displayed in Support Cluster. Nevertheless, no problem exists with the functionality for cluster support. 15. Use the execution results from the view operation to check that the programs are operating normally. If HDLM Manager, HDLM Alert Driver, and HDLM Driver are all Alive, the programs are operating normally. 16. Check the status of the path for accessing the LU to be managed by HDLM.
4. HDLM Operation Execute the following command: # /usr/cluster/bin/scdidadm -r 19. Notify the node of the most recent device ID information. Execute the following command: # /usr/cluster/bin/scdidadm -ui 20. Update the Sun Cluster global device namespace. Execute the following command: # /usr/cluster/bin/scgdevs 21. When using an HDLM-managed device, as a Quorum device, set the Quorum device on any node in the cluster.
4. HDLM Operation that took place before the HDLM installation, and then re-register HDLM devices with the programs that will use them. If you want to use volume management software to create a file system on an HDLM-managed device, skip to step 24. 23. Perform registration for the disk device group. If you are using VxVM, see 3.13.5 Linking VxVM and Sun Cluster. If you are using SDS, see (2) To Use a Shared Diskset of 3.14.2 Registering HDLM Devices.
4. HDLM Operation Figure 4-16: Example of executing the format command 3. If setup is performed for Quorum devices and disk device groups in the LU to be managed by HDLM, release the LU settings. When you unregister the SDS or SVM diskset, the diskset data will be deleted so first back up any necessary data. For details on how to release settings for Quorum devices and VxVM disk groups, see (2) Remove Sun Cluster Settings of 3.17.2 Preparations for HDLM Uninstallation.
4. HDLM Operation If the key is displayed, place the LU under HDLM management by following the steps from step 5 of (2) Changing a Non-HDLM-Managed Device to an HDLM-Managed Device (When the Node Must Be Restarted) in 4.5.6 Specifying Whether a Logical Unit Is To Be Managed by HDLM (When Using Sun Cluster).
4. HDLM Operation Figure 4-18: The /opt/DynamicLinkManager/config/dlmfdrv.unconf file (after editing) The shaded portion indicates the text to be edited. Note Do not make any modifications other than adding or deleting the hash mark (#) in front of the LDEV property line corresponding to the LU. 6. Apply the settings in the HDLM driver unconfiguration definition file (/opt/ DynamicLinkManager/config/dlmfdrv.unconf) to the HDLM driver configuration definition file (/kernel/drv/dlmfdrv.
4. HDLM Operation 9. Delete the device ID information for devices that do not exist. Execute the following command: # /usr/cluster/bin/scdidadm -C 10. Allocate the device ID. Execute the following command: # /usr/cluster/bin/scdidadm -r 11. Notify the node of the most recent device ID information. Execute the following command: # /usr/cluster/bin/scdidadm -ui 12. Update the Sun Cluster global device namespace.
4. HDLM Operation Execute the following command: # /usr/cluster/bin/scgdevs 13. When using an HDLM-managed device, as a Quorum device, set the Quorum device on any node in the cluster.
4. HDLM Operation 16. Create the file system and set up the global mount as necessary. For details, refer to the Sun Cluster documentation.
Chapter 5. Troubleshooting This chapter describes how to properly use HDLM error information, and then how to resolve the problem, if an error occurs in HDLM. The resolutions for path errors, HDLM program errors, and other types of errors are all described separately. 5.1 Information collected by using the DLMgetras utility for collecting HDLM error information 5.2 Checking Error Information in Messages 5.3 What To Do for a Path Error 5.4 Actions to Take for a Path Error in a Boot Disk Environment 5.
5. Troubleshooting 5.1 Information collected by using the DLMgetras utility for collecting HDLM error information Immediately after an error occurs, execute the DLMgetras utility to collect HDLM error information, because restarting the machine may delete existing error information already collected by the DLMgetras utility. For details on the DLMgetras utility and the error information it collects, see 7.2 The DLMgetras Utility for Collecting HDLM Error Information.
5. Troubleshooting 5.2 Checking Error Information in Messages When you want to configure the system so that HDLM messages are output to syslog, specify user for the name of the system function defined in the /etc/syslog.conf file. In the following example, the system function name is user, and messages at the Information level or higher are output to the /etc/syslog.conf file: user.info /tmp/syslog.user.log You can check path errors by referring to the KAPL08xxx messages that are output to the syslog.
5. Troubleshooting • Host port number (hexadecimal) • Bus number (hexadecimal) • Target ID (hexadecimal) • Host LU number (hexadecimal) The path name is also the same as the PathName displayed by the dlnkmgr command's view operation. For details on the path name, see 6.7 view (Displays Information). DNum A Dev number. A hyphen (-) is displayed. This is the same as the DNum displayed by the view operation. For details on the view operation, see 6.7 view (Displays Information).
5. Troubleshooting 5.3 What To Do for a Path Error When a path error is detected, HDLM performs a failover on the path and outputs the KAPL08022-E message. This message indicates that an error has occurred in the components that make up the path. Figure 5-1: Error Location When the KAPL08022-E Message is Output indicates these components.
5. Troubleshooting Figure 5-2: Troubleshooting Procedure When a Path Error Occurs If an error occurs in a boot disk environment on a path connected to the boot disk, perform the procedure described in 5.4 Actions to Take for a Path Error in a Boot Disk Environment. The following shows the procedure for using the dlnkmgr command to handle a path error. 5.3.1 Examining the Messages Examine the message that is output to syslog in the host by using applications or tools for monitoring messages.
5. Troubleshooting environment. 5.3.3 Identifying the Error Path Check the obtained path information to find the path with the error. In the Status column, the error path has the status Offline(E) or Online(E). 5.3.4 Narrowing Down the Hardware That Might Have Caused the Error Check the DskName, iLU, ChaPort, and HBAPortWWN columns of the path with the error to narrow down the hardware that may be the cause of the error.
5. Troubleshooting 5.4 Actions to Take for a Path Error in a Boot Disk Environment This section describes the action to take if an error occurs in a boot disk environment on a path connecting to the boot disk. 5.4.1 Path Errors During Boot Processing The procedure for investigating the cause of an error, and the appropriate action to take, will differ according to the stage of the boot processing during which the path error occurred.
5. Troubleshooting If an error occurs on one or more of the paths to the boot disk HDLM outputs the KAPL08022-E message to the console, syslog, and error log after the HDLM manager has started.
5. Troubleshooting 5.5 What To Do for a Program Error The following describes what to do to handle errors that occur in an HDLM program. Figure 5-3: Troubleshooting Procedure When a Program Error Occurs shows the troubleshooting procedure. Figure 5-3: Troubleshooting Procedure When a Program Error Occurs The following shows the procedure for handling a program error by using the dlnkmgr command. 5.5.1 Examining the Messages Examine the messages that are output to the host syslog.
5. Troubleshooting 5.5.3 What To Do for the Program Error Follow the recommended actions for messages in 8. Messages. If the error occurs again after you thought that you had resolved the problem, use the dlnkmgr command's view operation to check the status of the HDLM program, and then do whatever is necessary to resolve the problem. For details on the view operation, see 6.7 view (Displays Information).
5. Troubleshooting 5.6 What To Do for Other Errors When the cause of an error may be related to HDLM but is neither a path error nor an HDLM program error, execute the DLMgetras utility to collect the HDLM error information, and then report the collected information to the HDLM vendor or maintenance company. For details on the DLMgetras utility and the information it collects, see 7.2 The DLMgetras Utility for Collecting HDLM Error Information.
Chapter 6. Command Reference This chapter describes the HDLM command (dlnkmgr) and its operations. 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.
6. Command Reference 6.1 Overview of the HDLM Command dlnkmgr This section describes how to specify the HDLM command dlnkmgr and its subcommands (called operations in HDLM). Command format Enter the command using the following format: dlnkmgr operation [parameter [parameter-value]] dlnkmgr The command name. operation The type of operation entered after dlnkmgr. parameter A value required for an operation. parameter-value A value required for an operation parameter.
6. Command Reference Operation Functions Displays I/O information at a specified interval. For details, see 6.8 monitor (Displays I/O Information at a Specified Interval). monitor Note: • Execute the command as a user with root permissions. • To specify a value that contains a space in its parameter, enclose the entire value in double quotation marks (").
6. Command Reference 6.2 clear (Returns the Path Statistics to the Initial Value) The dlnkmgr command's clear operation clears the statistics (I/O count and I/O errors) of all paths that are managed by the HDLM, and returns them to their initial value. 6.2.1 Format (1) To set the path statistics to 0 /opt/DynamicLinkManager/bin/dlnkmgr clear -pdst [-s] (2) To display the format of the clear operation /opt/DynamicLinkManager/bin/dlnkmgr clear -help 6.2.
6. Command Reference (2) To display the format of the clear operation -help Displays the format of the clear operation. Example # /opt/DynamicLinkManager/bin/dlnkmgr clear -help clear: Format dlnkmgr clear -pdst [-s] KAPL01001-I The HDLM command completed normally.
6. Command Reference 6.3 help (Displays the Operation Format) The dlnkmgr command's help operation displays the list of operations available for the dlnkmgr command, or the format of individual operations. 6.3.1 Format /opt/DynamicLinkManager/bin/dlnkmgr help [operation] [operation] ... 6.3.2 Parameter operation Specify the HDLM command operation whose format you want to know. You can specify multiple operation names.
6. Command Reference # Example 2 The following example shows how to display the formats of multiple operations. # /opt/DynamicLinkManager/bin/dlnkmgr help online offline help online: Format dlnkmgr online [-path] [-s] dlnkmgr online [-path] { -hba HBAPortNumber.
6. Command Reference help: Format dlnkmgr help { clear | offline | online | set | view | monitor } KAPL01001-I The HDLM command completed normally.
6. Command Reference 6.4 offline (Places Paths Offline) The dlnkmgr command's offline operation places online paths offline. Specify the paths to be placed offline by specifying an HBA port, CHA port, or single path. Instead of specifying a path, you can also specify a SCSI device name. There must always be at least one online path accessing each LU. Placing too many paths offline might prevent paths from being able to switch if an error occurs.
6. Command Reference -hba host-port-number.bus-number Use this parameter to place offline, at one time, all the paths that pass through a specific HBA port. The command will place offline all the paths connected to the HBA port that has the specified host port number and bus number. Specify the host port number and bus number of the target HBA port: the numbers are found in the PathName field displayed using the view operation. Enter a period between these two parameter values.
6. Command Reference KAPL01102-I All the paths which pass the specified HBA port will be changed to the Offline(C) status. Is this OK? [y/n]:y KAPL01103-I If you are sure that there would be no problem when all the paths which pass the specified HBA port are placed in the Offline(C) status, enter y. Otherwise, enter n. [y/n]:y KAPL01061-I 15 path(s) were successfully placed Offline(C); 0 path(s) were not.
6. Command Reference -chaid CHA-port-ID Use this parameter to place offline, at one time, all paths that pass through a specific CHA port. The command will place offline all paths connected to the CHA port that has the specified CHA-port-ID. To display the CHA port IDs, execute the following view operation: # /opt/DynamicLinkManager/bin/dlnkmgr view -cha For details about how to execute the view operation to display the CHA port IDs, see (5) To display CHA port information in 6.7.2 Parameters.
6. Command Reference device. For SCSI-device-name, specify the value of Device displayed by executing the view operation as follows: # /opt/DynamicLinkManager/bin/dlnkmgr view -lu For details on how to execute the view operation and display the SCSI device name, see (3) To display LU information in 6.7.2 Parameters. Only one case-sensitive parameter value can be specified.
6. Command Reference The following example shows how to display the format of the offline operation: # /opt/DynamicLinkManager/bin/dlnkmgr offline -help offline: Format dlnkmgr offline [-path] { -hba HBAPortNumber.
6. Command Reference # /opt/DynamicLinkManager/bin/dlnkmgr view -path -stname | grep 9500V | grep 0A The above command will display information pertaining to only those paths that pass through the specified CHA port.
6. Command Reference 6.5 online (Places Paths Online) The dlnkmgr command's online operation places offline paths online. Specify the paths to be placed online by specifying an HBA port, CHA port, or single path. Instead of specifying a path, you can also specify a SCSI device name. 6.5.1 Format (1) To place paths online /opt/DynamicLinkManager/bin/dlnkmgr online [-path] [-hba host-port-number.
6. Command Reference Use this parameter to place online, at one time, all the paths that pass through a specific HBA port. The command will place online all the paths connected to the HBA port that has the specified host port number and bus number. Specify the host port number and bus number of the target HBA port: the numbers are found in the PathName field displayed using the view operation. Enter a period between these two parameter values. For details on the view operation, see 6.
6. Command Reference 0 path(s) were not. Operation name = online # -cha -pathid AutoPATH_ID Use this parameter to place online, at one time, all the paths that pass through a specific CHA port. The command will place online all the paths that pass through the CHA port to which the path with the specified AutoPATH_ID is connected. Specify the current AutoPATH_ID of the target path, which is displayed by using the view operation. For details on the view operation, see 6.7 view (Displays Information).
6. Command Reference CHA port whose CHA-port-ID is 00001. # /opt/DynamicLinkManager/bin/dlnkmgr offline -chaid 00001 KAPL01104-I All the paths which pass the specified CHA port will be changed to the Online status. Is this OK? [y/n]:y KAPL01061-I 15 path(s) were successfully placed Online; 0 path(s) were not. Operation name = online # -pathid AutoPATH_ID Use this parameter to place a single path online. Specify the current AutoPATH_ID of the target path, which is displayed by using the view operation.
6. Command Reference -s Executes the command without displaying the message asking for confirmation of command execution from the user. Specify this parameter if you want to skip the response to the confirmation message: for example, when you want to execute the command in a shell script or batch file.
6. Command Reference Reference Executing the view operation together with standard UNIX commands enables you to filter the path information listed for a specific HBA port or CHA port. For details on the view operation, see 6.7 view (Displays Information). We recommend that you execute the following command and verify the information on the target paths before you execute the online operation to place online all the paths connected to a specific HBA port or CHA port.
6. Command Reference 6.6 set (Sets Up the Operating Environment) The dlnkmgr command's set operation sets the HDLM operating environment. 6.6.
6. Command Reference Table 6-2: Default and Recommended Values Item name Load balancing Default value Recommended value on on The Extended Least I/Os algorithm is used. The recommended algorithm depends on the operating environment. Error log collection level 3: Collects error information at the 3: Collects error information at the Information level or higher. Information level or higher.
6. Command Reference exlio: The Extended Least I/Os algorithm lbk: The Least Blocks algorithm exlbk: The Extended Least Blocks algorithm The type of algorithm specified by the -lbtype parameter remains stored in the system, even when you disable the load balancing function by specifying -lb off. Therefore, if you re-enable the load balancing function without specifying an algorithm, load balancing will be performed according to the algorithm that is stored in the system.
6. Command Reference hh:mm:ss # -elfs error-log-file-size Specify a value from 100 to 2000000 (in kilobytes) for the size of the error log files. The target log files for which a file size can be set are HDLM manager logs (dlmmgrn.log (n indicates a file number from 1 to 16)). The specified value is applied for HDLM manager logs. By specifying both the log file size and the number of log files, you can collect up to 32,000,000 kilobytes (approximately 30GB) of error logs in total.
6. Command Reference As the amount of log information to be output increases, the amount of time before existing information is overwritten becomes shorter. -systfs trace-file-size Specify the size of the trace file in kilobytes. Specify a value from 100 to 16000. By specifying both the trace file size and the number of trace files, you can collect up to 1024000 kilobytes of trace files in total.
6. Command Reference • When the checking interval has been specified before: The interval specified in the last time The explanation for the following sub-parameter describes how to specify the checking interval. -intvl checking-interval Specify the checking interval between path health checks. Specify a value from 1 to 1440 minutes depending on the user environment. When you change the checking interval, the new setting takes effect immediately.
6. Command Reference Operations. When you specify on, specify the checking interval by specifying the parameter immediately following on.
6. Command Reference number-of-times-error-is-to-occur] | off } Enables or disables intermittent error monitoring. on: Enabled off: Disabled Intermittent error monitoring can be set when automatic failback is set to on. When you use automatic failback, we recommend that you set intermittent error monitoring to on to prevent an intermittent error from reducing I/O performance.
6. Command Reference monitoring interval and the number of times that the error is to occur: -intvl error-monitoring-interval Specify the monitoring interval for an intermittent error. Use a value from 1 to 1440 minutes. The default is 30. During intermittent error monitoring, if changes are made in the settings of the intermittent error monitoring interval, the error count and the elapsed time measured since monitoring starts are set to 0, and monitoring starts by using the new settings.
6. Command Reference Intermittent error monitoring will then resume with the changed settings. If you set the automatic failback function to off while intermittent error monitoring is on, intermittent error monitoring will be disabled. Note, however, that if you use the view -sys operation to display the HDLM function configuration, Intermittent Error Monitor will be shown as on. When the automatic failback function is returned to on, intermittent error monitoring will once again be enabled.
6. Command Reference Setting AFB Executable operation IEM off off Result of operation on#2 off Change the IEM settings. • While a path is being monitored, the value of the error count and the elapsed monitoring time are reset to 0, and then intermittent error monitoring will restart.#1 • When a path is not being monitored, the IEM settings will take effect again when the path is recovered from the error status by performing an automatic failback. Set IEM to off. • IEM is disabled.
6. Command Reference # /opt/DynamicLinkManager/bin/dlnkmgr set -iem on -intvl 20 -iemnum 2 KAPL01049-I Would you like to execute the operation? Operation name = set [y/n]: y KAPL01001-I The HDLM command completed normally. Operation name = set, completion time = yyyy/mm/dd hh:mm:ss # -lic Specify this option for when a license is updated. The HDLM license is provided via a license key or license key file.
6. Command Reference A temporary license key cannot be installed by using the dlnkmgr command's set operation. Example 1 The following example shows how to update the license key when the license key file exists: # /opt/DynamicLinkManager/bin/dlnkmgr set -lic KAPL01049-I Would you like to execute the operation? Operation name = set [y/n]: y KAPL01071-I A permanent license was installed.
6. Command Reference Value (severity) Explanation 3 Critical-level and Error-level audit log data is collected. 4 Critical-level, Error-level, and Warning-level audit log data is collected. 5 6 Critical-level, Error-level, Warning-level, and Informational-level audit log data is collected. 7 -category [[ss] [a] [ca]|all] Specifies the categories of audit log data to be collected. Table 6-8: Values Indicating Audit Log Data Categories lists the values that indicate categories of audit log data.
6. Command Reference Value Corresponding facility value in the /etc/syslog.conf file local4 or 20 local4 local5 or 21 local5 local6 or 22 local6 local7 or 23 local7 -s Executes the command without displaying the message asking for confirmation of command execution from the user. Specify this parameter if you want to skip the response to the confirmation message. For example, when you want to execute the command in a shell script or batch file.
6.
6. Command Reference all } (Default Value all) Facility-Name { user | local0 - local7 } Facility-Number { 1 | 16 - 23 } (Default Value user) (Default Value 1) KAPL01001-I The HDLM command completed normally.
6. Command Reference 6.7 view (Displays Information) The dlnkmgr command's view operation displays HDLM program information, path information, LU information, HBA port information, CHA port information, and corresponding information about an HDLM device, sd or ssd device, and LDEV. 6.7.
6.
6. Command Reference 6.7.
6. Command Reference Parameter and program information to be displayed 396 Item Description Load Balance Settings for load balancing • Setting status: on: Enabled off: Disabled • Algorithm: When the setting status of load balancing is on, one of the following types of algorithms is used for load balancing is displayed in the parentheses following on.
6. Command Reference Parameter and program information to be displayed -msrv Item Path Health Checking • Settings for path health checking: on: Enabled off: Disabled • Checking interval: When the setting of the path health checking is on, the checking interval of path health checking is displayed within the parentheses following on. The time is in minutes.
6.
6. Command Reference Parameter and program information to be displayed Item Description Audit Log Category The categories of audit log data to be output are displayed. When more than one category is displayed, commas (,) are used as separators. ss: StartStop a: Authentication ca: ConfigurationAccess If all the above categories are specified, all is displayed. If the collection of audit log data is disabled, a hyphen (-) is displayed. Audit Log Facility The name of the audit log facility is displayed.
6. Command Reference # Example 2 The following example shows how to display information about the HDLM manager: # /opt/DynamicLinkManager/bin/dlnkmgr view -sys -msrv HDLM Manager Ver WakeupTime x.x.x-xx yyyy/mm/dd hh:mm:ss Alive KAPL01001-I The HDLM command completed normally.
6. Command Reference Example 6 The following example shows how to display information about the audit log settings: # /opt/DynamicLinkManager/bin/dlnkmgr view -sys -audlog Audit Log : off Audit Log Category : Audit Log Facility : KAPL01001-I The HDLM command completed normally.
6. Command Reference Use this parameter to display the model ID of the storage system in the product ID element of the DskName field. When this parameter is omitted, the command displays the product ID or emulation type of the storage system instead. For details about the information displayed for product IDs, see Table 6-14: Product ID Displayed by the view -path Operation. -iem Use this parameter to add IEP to path information and display information about intermittent errors.
6. Command Reference To display path information (by selecting a display item) -path -item When you specify the -path parameter together with the -item parameter, the command only displays the items specified by the value of the -item parameter. If you specify no value for the -item parameter, only the PathID and the Status fields are displayed.
6. Command Reference Selectable items Sub-parameters Type type IO-Count ic IO-Errors ie DNum dnu HDevName hd IEP iep # Because both PathID and Status are always displayed, you do not have to specify any parameters. In the sub-parameters following -path -item, you can list the paths (-hdev) and sort the list (-srt). If you omit both parameters, the command displays information for all the paths in ascending order of AutoPATH_IDs.
6. Command Reference Use this parameter to sort the path information in ascending order, according to the specified sorting key. The sorting keys are as follows: the first sorting key is the name of the storage system (DskName), the second sorting key is the value specified by the -srt parameter, and the third sorting key is AutoPATH_ID.
6. Command Reference To display a summary of path information -path -c When you specify the -path parameter together with the -c parameter, only the PathID, DskName, iLU, CP, Status, and Type fields are displayed from the path information managed by HDLM. The displayed contents are shortened so that each instance of path information is displayed on a single line. For details on what is displayed for in each item, see Table 6-12: Displayed Path Information.
6. Command Reference 11 or more characters in the product ID, the 8th and following characters are abbreviated to an ellipsis (...). The following shows the subsequent sub-parameters (following -path -c): -stname Use this parameter to display the model ID of the storage system in the product ID element of the DskName field. When this parameter is omitted, the command displays the product ID or emulation type of the storage system instead.
6. Command Reference Displayed path information Table 6-12: Displayed Path Information describes the displayed path information. The following explains the table headings: • No summary displayed: The user specifies only the -path parameter or the -path -item parameter. • Summary displayed: The user specifies the -path -c parameter.
6. Command Reference Displayed Item No summary displayed Description Summary displayed PathStatus Status of the displayed paths. The displayed status indicates the following: • Online: All paths are available. • Reduced: Some paths are not available. Reduced means that some paths might have an error status, in which case you should check the status of individual paths and resolve the problem for any paths that have an error status.
6. Command Reference Displayed Item No summary displayed Description Summary displayed iLU# iLU LU number of the storage system. This number combined with the storage system name (shown in DskName) identifies the LU that is accessed by a path. • For the Lightning 9900 Series, Lightning 9900V Series, Hitachi NSC55, Hitachi Universal Storage Platform 100, Hitachi Universal Storage Platform 600, or Hitachi Universal Storage Platform 1100, indicated by a hexadecimal number.
6. Command Reference Displayed Item No summary displayed Description Summary displayed IO-Count# -- Total I/O count for the path, indicated by a decimal number. The maximum value that can be displayed is 232 - 1 (4294967295). If the total I/O count reaches the maximum value, it will re-start from 0. To reset the IO-Count value to 0, execute the dlnkmgr command's clear operation. Executing the clear operation also resets the number of I/O errors (IO-Errors) to 0.
6. Command Reference Displayed Item Description Summary displayed No summary displayed HBAPortWW - N# A 16-digit hexadecimal number indicating the WWN information for an HBA connected to the storage system. This item is displayed only when you specify -iem together with the -hbaportwwn parameter. Legend: --: Not displayed # The path information is displayed only when a value is specified for the -path -item parameter.
6.
6. Command Reference (3) To display LU information When displaying LU information, if the -item parameter, -c parameter, or the -c -item parameter is specified at the same time as the -lu parameter, you can add and display items and display a summary of LU information. This section describes each parameter and the LU information and displayed items.
6. Command Reference 0002 c6t2d1 0003 c6t2d2 Product SerialNumber LUs sd62 sd65 sd63 sd66 000004 000010 000005 000011 Online Online Online Online : 9970/9980 : 15001 : 3 iLU HDevName Device PathID Status 0670 c6t5d0 sd74 000000 Online sd80 000006 Offline(E) 0671 c6t5d1 sd75 000001 Online sd81 000007 Offline(E) 0672 c6t5d2 sd76 000002 Online sd82 000008 Offline(C) KAPL01001-I The HDLM command completed normally.
6. Command Reference Item that can be added Sub-parameter Type type IO-Count ic IO-Errors ie DNum dnu IEP iep Physical Device pd All items are displayed all In the sub-parameters, -hdev or -pathid, you can filter the LU information to be displayed. If you omit both parameters, the command displays the information about all the LUs recognized by HDLM. For details on the contents of each displayed item, see Table 6-17: Displayed LU Information.
6. Command Reference To display a summary of LU information -lu -c When the -c parameter is specified with the -lu parameter, a summary of LU configuration information is displayed on one line. The total number of paths recognized by HDLM and the number of online paths are displayed for each LU. You cannot specify the -c parameter together with the -hdev or -pathid parameter. For details on the contents of each display item, see Table 6-17: Displayed LU Information.
6. Command Reference Omits the title for each information item. Example # /opt/DynamicLinkManager/bin/dlnkmgr view -lu -c Product S/N LUs iLU HDevName Paths OnlinePaths 9500V 0051 3 0001 c6t2d0 2 2 0002 c6t2d1 2 2 0003 c6t2d2 2 2 9970/9980 15001 3 0670 c6t5d0 2 1 0671 c6t5d1 2 1 0672 c6t5d2 2 1 KAPL01001-I The HDLM command completed normally.
6. Command Reference Displayed LU information Table 6-17: Displayed LU Information describes the displayed LU information. The following explains the table headings: • No summary displayed: The user specifies the -lu parameter or -lu -item parameter. • Summary displayed: The user specifies the -lu -c parameter or -lu -c -item parameter.
6. Command Reference Displayed item No summary displayed Description Summary displayed LU number of the storage system. This number combined with the storage system name (shown in DskName) identifies the LU that is accessed by a path. • For the Lightning 9900 Series, Lightning 9900V Series, Hitachi NSC55, Hitachi Universal Storage Platform 100, Hitachi Universal Storage Platform 600, or Hitachi Universal Storage Platform 1100, indicated by a hexadecimal number.
6. Command Reference Displayed item No summary displayed Description Summary displayed PathName#1 -- The path name, which indicates a path. When you modify the system configuration or replace a hardware item, you should check the path names to identify the path that will be affected by the change.
6. Command Reference Displayed item No summary displayed Description Summary displayed IO-Count#1 -- Total I/O count for a path. The maximum value that can be displayed is 232 - 1 (4294967295). If the total I/O count reaches the maximum value, it is reset, and the count is re-started from 0. To reset the IO-Count value to 0, execute the dlnkmgr command's clear operation. Executing the clear operation also resets the number of I/O errors (IO-Errors) to 0. For details on the clear operation, see 6.
6. Command Reference --: Not displayed #1 This information is displayed when one of the following conditions exist: - The user selected the item to be displayed by using the -lu -item parameter. - all was specified. - No value was specified for the parameter. #2 This information is displayed when one of the following conditions exist: - The user selected the item to be displayed by using the -lu -c -item parameter. - No value was specified for the parameter.
6. Command Reference Item Description OnlinePaths Number of available paths in the displayed paths, indicated by a decimal number. When the value of Paths equals the value of OnlinePaths, all paths are online. If the value of OnlinePaths is less than that of Paths, some paths might have an error status, in which case you should check the status of individual paths and take appropriate action for any paths that have an error status. Physical Device HBA port configuration information.
6. Command Reference CHA port. HDLM displays the unique CHA port ID assigned to each CHA port. This CHA port ID is used in monitor operations, and can also be used in Offline and Online operations. Table 6-19: Items Displayed as CHA Port Information lists and describes the displayed items.
6. Command Reference (Product), storage system serial numbers (S/N), and CHA port numbers (ChaPort). -t Omits the title for each information item. (6) To display corresponding information about an HDLM device, sd or ssd device, and LDEV -drv Displays the information about the PathID, the HDLM device, the sd or ssd device for the HDLM device, and the LDEV (the LU in the storage system).
6. Command Reference 000000 c6t5d0 sd74 9970/9980.15001.0670 000001 c6t5d1 sd75 9970/9980.15001.0671 000002 c6t5d2 sd76 9970/9980.15001.0672 000003 c6t5d3 sd77 9970/9980.15001.0673 000004 c6t5d4 sd78 9970/9980.15001.0674 000005 c6t5d5 sd79 9970/9980.15001.0675 000006 c6t5d0 sd80 9970/9980.15001.0670 000007 c6t5d1 sd81 9970/9980.15001.0671 000008 c6t5d2 sd82 9970/9980.15001.0672 000009 c6t5d3 sd83 9970/9980.15001.0673 000010 c6t5d4 sd84 9970/9980.15001.0674 000011 c6t5d5 sd85 9970/9980.15001.
6. Command Reference [ic] [ie] [dnu] [iep] [pd] | all ] [ -hdev HostDeviceName | -pathid AutoPATH_ID ] [-t] dlnkmgr view -lu -c [-t] dlnkmgr view -lu -c -item [slpr] [-t] dlnkmgr view -drv [-t] dlnkmgr view -hba [-srt pb] [-c | -portwwn] [-t] dlnkmgr view -cha [-srt cp] [-t] KAPL01001-I The HDLM command completed normally.
6. Command Reference 6.8 monitor (Displays I/O Information at a Specified Interval) The dlnkmgr command's monitor operation displays I/O information for each HBA or CHA port at a specified interval. HBA ports are specified by their HBA port ID, and CHA ports are specified by their CHA port ID. You can display HBA and CHA port IDs by executing the following view operation: • HBA port IDs /opt/DynamicLinkManager/bin/dlnkmgr view -hba For details, see (4) To display HBA port information in 6.7.2 Parameters.
6. Command Reference are 500 or more paths for the -intvl parameter, the time interval of the monitor operation might be several seconds to 20 seconds late for the value specified in the -intvl parameter. If the number of paths is 500 or more, specify an appropriate value for the -intvl parameter based on the values listed in Table 6-21: Recommended values for the -intvl parameter when there are 500 or more paths.
6. Command Reference Omits the title for each information item. Table 6-22: Items displayed as I/O information for each HBA port Item Description HbaID HBA port ID Port.Bus Port (host port number in hexadecimal) and Bus (bus number in hexadecimal) separated by a period Physical Device HBA port configuration information. The names of the physical device files under the / devices directory are displayed.
6. Command Reference format for the subsequent parameters are as follows: -intvl interval Specifies the interval at which I/O information is to be displayed, in the range from 1 to 60 (seconds). When this parameter is omitted, the default (3 seconds) is applied. -count count Specifies the number of times I/O information is to be displayed, in the range from 1 to 999. When this parameter is omitted, the default (10) applied. -t Omits the title for each information item.
6. Command Reference ChaPort StartTime : 1R : yyyy/mm/dd hh:mm:ss ElapsedTime IO-Count IO-Errors 00:00:30 1154 0 00:01:00 205 22 00:01:30 54 10 KAPL01001-I The HDLM command completed normally. Operation name = monitor, completion time = yyyy/mm/dd hh:mm:ss (3) To display monitor operation format -help Displays the monitor operation format.
Chapter 7. Utility Reference This chapter explains the utilities used by HDLM. 7.1 7.2 7.3 7.4 7.5 7.6 Overview of the Utilities The DLMgetras Utility for Collecting HDLM Error Information The dlmcfgmgr Utility for Managing the HDLM Configuration The dlminstcomp HDLM Component Installation Utility The dlmlisthdev Utility for Assisting HDLM Transitions The dlmsetconf Utility for Creating the HDLM Driver Configuration Definition File 7.
7. Utility Reference 7.1 Overview of the Utilities HDLM provides the following utilities: • The DLMgetras utility for collecting HDLM error information When an error occurs, this utility collects the files that contain information to be submitted to your HDLM vendor or maintenance company. For details on the DLMgetras utility, see 7.2 The DLMgetras Utility for Collecting HDLM Error Information.
7. Utility Reference Assisting Creation of the VxVM Configuration File. • The installhdlm utility for installing HDLM This utility performs a new installation, upgrade installation, or re-installation of HDLM in unattended installation mode. For details on this utility, see 7.8 The installhdlm Utility for Installing HDLM. For details on the procedure for performing an unattended installation, see 3.5.3 Performing a New Installation of HDLM (When Sun Cluster Is Not Being Used), 3.5.
7. Utility Reference 7.2 The DLMgetras Utility for Collecting HDLM Error Information This utility collects information that is needed to analyze HDLM errors that have occurred: information such as error logs, integrated trace files, trace files, definition files, core files, system crash dump files, and libraries. The collected information is archived in a file and saved to the directory that you specified.
7. Utility Reference Specify the output directory for the information that is to be collected by the DLMgetras utility for collecting HDLM error information. The collected information is compiled into three files shown in Function and output in the specified directory. You cannot specify multi-byte characters. -f name-of-a-file-for-defining-the-information-to-be-collected Use this parameter when you want to specify certain directories or files to be collected.
7. Utility Reference subdirectories. If no files are found in a specified directory, the utility does not perform file collection for that directory and does not create a directory for it in the destination directory. • Set up the specified file or directory so that it can be read by users with root permissions. The DLMgetras utility can only obtain information for a file or directory that can be read. -h Displays the format of the DLMgetras utility.
7. Utility Reference products other than HDLM. This file is output only when using the Hitachi Command Suite product that is the target for collecting error information. • getras.tar.Z This file contains HDLM information and system information. When you execute the DLMgetras utility, specifying the file for defining information to be collected, the getras.tar.Z file contains the information defined in the file for defining information to be collected. Table 7-1: Information Stored in the getras.tar.
7. Utility Reference Files Output directory#1 opt/DynamicLinkManager/ config Explanation dlmfdrv.unconf HDLM driver unconfiguration definition file dlmfdrv.unconf.bak Previous generation of the HDLM driver unconfiguration definition file. dlmfdrv.unconf.bak is created if the dlmfdrv.unconf file already exists when you attempt to create the dlmfdrv.unconf file. dlmfdrv.unconf_oldb ackup#3 Backup of the existing dlmfdrv.unconf file, taken when you perform an upgrade installation from HDLM 5.6.
7. Utility Reference Files Output directory#1 dlmfdrv.conf_oldbac kup#3 Explanation Backup of the existing dlmfdrv.conf file, taken when you perform an upgrade installation from HDLM 5.6.1 or earlier. dbackup#3 Backup of the existing dlmfdrv.conf.bak file, taken when you perform an upgrade installation from HDLM 5.6.1 or earlier. dlmndrv.conf HDLM nexus driver configuration definition file dlmndrv.conf.bak Backup file of the HDLM nexus driver configuration definition file bfa.conf, emlxs.
7. Utility Reference Files Output directory#1 gabtab llthosts llttab vxfendg Explanation I/O fencing function setup files vxfenmode#5 vxfentab mnttab File system mount information syslog.
7. Utility Reference Files Output directory#1 Explanation df-k.txt Mount information df-o.txt i node utilization status dlmgetomtrace.dmp In-memory trace information dlmls-la.txt List of directories and files related to the HDLM installation file information (output results from the ls -alR command) dlmmgr-cha.txt Output results of dlnkmgr view -cha dlmmgr-hba.txt Output results of dlnkmgr view -hba -portwwn dlmmgr-lu.txt HDLM LU information dlmmgr-lu-all.
7. Utility Reference Output directory#1 VxVM #6 Files prtconf-pv.txt Hardware configuration information of a host (output results of the prtconf -pv command) prtconf-vP.txt List of driver information prtconfMemory.txt Physical memory size prtvtoc.txt Partition configuration information of all disks (output results of the prtvtoc command) ps-ef.txt Process information psrinfo-v.txt CPU information showrev-p.txt List of installed patches svcs-a.
7. Utility Reference Files Output directory#1 Explanation VXVM_vxdmp.txt /dev/vx/dmp directory list VXVM_vxrdmp.txt /dev/vx/rdmp directory list VXVM_vxdsk.txt /dev/vx/dsk directory list VXVM_vxrdsk.txt /dev/vx/rdsk directory list VXVM_vxdisklist.txt VxVM disk list VXVM_vxdisklist-s.t xt Output results of vxdisk -s list VXVM_vxdiskpath.txt Output results of vxdisk path VXVM_vxdglist.txt List of disk groups VXVM_vxprint.
7. Utility Reference Output directory#1 var/adm/syslog VCS#6 var/VRTSvcs/log#6 SDS#6 448 Files syslog.log.txt (Not applicable) Explanation Log of error messages The output destination is either /var/vxvm/ vxconfigd.log or /var/adm/syslog/ syslog.log, depending on the debug option setting for the user. VCS information collection directory IOF_gabconfig-a.txt I/O fencing GAB information IOF_lltstat-nvv.txt I/O fencing LLT link details IOF_lltstat-p.
7. Utility Reference Files Output directory#1 etc/lvm#6 md.tab SDS or SVM input parameter file md.cf SDS or SVM configuration backup file mddb.cf Information on SDS or SVM status database copy storage locations mdlogd.cf Log collected from SDS or SVM (Not applicable) SVM#6 etc/opt/SUNWmd#6 etc/cluster#6 SVM information collection directory SVM_pkginfo.txt SVM package information SVM_metadb.txt Status database information SVM_metastat.txt Volume information SVM_metaset.
7. Utility Reference An output directory is created in the specified directory when you extract the getras.tar.Z file. #2 The value N depends on the setting in the dlmwebagent.properties file. The default value is 2. #3 The file might not be created depending on the device managed by HDLM. #4 In the actual file name, a file number is appended to Trace-file-prefix specified-in-the-Hitachi-Network-Objectplaza-Trace-Library-utility2. For example, the default will be hntr21.log to hntr216.log.
7. Utility Reference 7.3 The dlmcfgmgr Utility for Managing the HDLM Configuration This utility makes the HDLM driver recognize the contents of the HDLM driver configuration definition file (/kernel/drv/dlmfdrv.conf) dynamically without restarting the host. Table 7-2: Configuration Changes Applicable and Not Applicable to the dlmcfgmgr Utility for Managing the HDLM Configuration illustrates the configuration changes applicable and not applicable to the dlmcfgmgr utility for managing the HDLM configuration.
7. Utility Reference 7.3.2 Parameters -a Applies the contents of the HDLM driver configuration definition file (/kernel/ drv/dlmfdrv.conf) to HDLM driver. The added LU becomes valid after the command with this option is executed. The HDLM devices for the deleted LUs become unavailable after the command with this option is executed. -h Displays help about the usage of the command.
7. Utility Reference • You must restart the host after performing an upgrade installation or re-installation. LUs will not be added or deleted dynamically if you execute the dlmcfgmgr utility before restarting the host. • If the HDLM driver configuration definition file (/kernel/drv/ dlmfdrv.conf) contains the definition of a newly added LU, the dlmcfgmgr utility makes the HDLM driver recognize the new LU so that the LU is to be managed by HDLM.
7. Utility Reference occurs: -For the LU, you changed the LUN mapping, the mapping-destination storage port, or the HBA port on the recognizing host. -An upper-level program is using the LU you attempted to delete. • In Solaris 8 or 9, if the HDLM driver configuration definition file (/ kernel/drv/dlmfdrv.conf) does not contain HDLM device information, the dlmcfgmgr utility outputs the message and terminates. In Solaris 10, all LUs are deleted from the HDLM driver.
7. Utility Reference 7.4 The dlminstcomp HDLM Component Installation Utility This utility complements HDLM installation. Execute this utility when both of the following conditions are satisfied: • Linkage with Global Link Manager is used • When HDLM is installed, the KAPL09241-W message is output# # Execute this utility after installing the JDK. For details about the JDK version to be installed, see JDK required for linking with other Hitachi Command Suite products in 3.1.1 Hosts and OSs Supported by HDLM.
7. Utility Reference 7.5 The dlmlisthdev Utility for Assisting HDLM Transitions Use the dlmlisthdev utility when migrating to HDLM 6.5.1 from HDLM 04-00 or 04-01 installed in a Solaris 8 or Solaris 9 environment. The utility creates a correspondence table between the logical device file names of the HDLM devices used in version 04-00 or 04-01 and those of the HDLM devices created in version 6.5.1. 7.5.1 Format /opt/DynamicLinkManager/bin/dlmlisthdev input-file-name 7.5.
7. Utility Reference configuration in HDLM 6.5.1. 5. Create a list that shows the correspondences between the logical device file names of the HDLM devices in HDLM 04-00 and those in HDLM 6.5.1. As the input file for the command, specify the file that stores the output result of the dlnkmgr command's view -path operation executed before HDLM 04-00 was uninstalled.
7. Utility Reference The file that stores the output result of the dlnkmgr command's view -path operation must be created in a directory that is not deleted when HDLM is uninstalled. If not, the file will be deleted during uninstallation. The following are the directories that are deleted during uninstallation.
7. Utility Reference 7.6 The dlmsetconf Utility for Creating the HDLM Driver Configuration Definition File This dlmsetconf utility obtains device information retained by Solaris, and defines the storage system's sd or ssd device entries in the HDLM driver configuration definition file (/kernel/drv/dlmfdrv.conf) and the HDLM nexus driver configuration definition file (/kernel/drv/dlmndrv.conf). The sd or ssd device entries are sorted in ascending order of the LDEV information (for example, 9970/9980.15001.
7. Utility Reference In the subsequent explanations, the configuration definition file is indicated as the HDLM driver configuration definition file (/kernel/drv/dlmfdrv.conf) and the HDLM nexus driver configuration definition file (/kernel/drv/dlmndrv.conf). 7.6.1 Format /opt/DynamicLinkManager/bin/dlmsetconf [-d output-directory-name] [-r | -u] 7.6.
7. Utility Reference Depending on the contents of the HDLM driver unconfiguration definition file (/ opt/DynamicLinkManager/config/dlmfdrv.unconf) that defines the LU you want to remove from HDLM management, in the existing entries of the HDLM driver configuration definition file (/kernel/drv/dlmfdrv.conf), only the attribute value that specifies whether or not the LU is managed by HDLM is updated, and applied in the system when the reconfiguration option is specified to restart the host.
7. Utility Reference Example 4 The -u parameter is specified: Defines a non- HDLM-managed LU based on the contents of the definitions in the HDLM driver unconfiguration definition file (/opt/ DynamicLinkManager/config/dlmfdrv.unconf): # /opt/DynamicLinkManager/bin/dlmsetconf -u Note • Execution of the dlmsetconf utility causes sd or ssd devices to be defined in the HDLM driver configuration definition file (/kernel/drv/dlmfdrv.conf) in units of LUs.
7. Utility Reference definition file (/kernel/drv/dlmfdrv.conf) the utility adds an entry (that is, a definition) to the definition of the LU for which the path has been added. If any path has been deleted, the entry (definition) is deleted from the definition of the affected LU. Whether a path is added or deleted, existing instance numbers are inherited unless a new HDLM driver configuration definition file (/kernel/ drv/dlmfdrv.conf) file is created.
7. Utility Reference option specified, and then restart the host. • If the below message appears after you execute the dlmsetconf utility, it means that the device ID could not be acquired from the SCSI driver. If this happens, restart the host and then re-execute the dlmsetconf utility. KAPL10237-E The property information (devid) on a device (aa...aa) could not be collected. aa...
7. Utility Reference 7.7 The dlmvxexclude Utility for Assisting Creation of the VxVM Configuration File This utility outputs the file (vxvm_exclude.txt) having the same format as the VxVM management target exclusion file (/etc/vx/vxvm.exclude) that matches the definition in the HDLM driver configuration definition file (/kernel/drv/ dlmfdrv.conf) to the directory specified by the -d parameter.
7. Utility Reference # /opt/DynamicLinkManager/bin/dlmvxexclude -d /tmp Display example Example The entries for the following devices are output: • sd or ssd devices for HDLM-managed devices • HDLM devices that are not actually created Only one HDLM device is created for one HDLM-managed device (one slice). However, VxVM assumes that the number of HDLM devices created for one HDLM-managed device is the same as the number of physical paths.
7. Utility Reference • Do not use the output result of the dlmvxexclude utility as is for the vxvm.exclude file. If a device other than the HDLM-managed devices has already been invalidated in VxVM, then the invalidated setting might be canceled.
7. Utility Reference 7.8 The installhdlm Utility for Installing HDLM You can use the installhdlm utility to perform a new installation, upgrade installation, or re-installation of HDLM in unattended installation mode. You must predefine how the utility is to respond during installation in the installation-information settings file. For details on the procedure for performing an unattended installation, see 3.5.3 Performing a New Installation of HDLM (When Sun Cluster Is Not Being Used), 3.5.
7. Utility Reference Table 7-4: Keys that Can Be Defined in the [INSTALLATION_SETTINGS] Section Key name#1 Explanation#2 Required or optional? For new installati on For upgrade installation or re-installati on Maximum length (character s) installfile_location Specify the absolute path name of the directory on which the DVD-ROM is mounted. The default is /cdrom/cdrom0. Optional Optional 400 logdir Specify the absolute path name of the directory to which log data will be output.
7. Utility Reference Key name#1 Explanation#2 Specify whether to restart the host after installation by specifying either of the following values: y: Restart the system. n: Do not restart the system (default). If you use Sun Cluster, specify n. restart Required or optional? For new installati on For upgrade installation or re-installati on Optional Optional Maximum length (character s) 1 Legend: Optional: If you omit the key name and value, the default is used.
7. Utility Reference Table 7-5: Keys that Can Be Defined in the [ENVIRONMENT_SETTINGS] Section Key name#1 Explanation#2 Required or optional? Required for new installatio n Required for upgrade installati on or re-install ation Maximum length (character s) load_balance Specify whether to enable load balancing by specifying either of the following values: on: Enable load balancing (default). off: Do not enable load balancing.
7. Utility Reference Key name#1 Explanation#2 Required or optional? Required for new installatio n Required for upgrade installati on or re-install ation Maximum length (character s) trace_level Specify the output level for traces. You can specify a value from 0 to 4. The default is 0. Optional Optional 1 trace_file_size Specify the size of trace files (hdlmtr[1-64].log) in kilobytes. You can specify a value from 100 to 16000. The default is 1000.
7. Utility Reference Key name#1 Explanation#2 Required or optional? Required for new installatio n Required for upgrade installati on or re-install ation Maximum length (character s) intermittent_error_ monitor#3#4 Specify whether to enable intermittent error monitoring by specifying either of the following values: on: Enable intermittent error monitoring. off: Disable intermittent error monitoring (default).
7. Utility Reference For details on the settings functionality, see 6.6 set (Sets Up the Operating Environment).
7. Utility Reference 7.8.4 About the Log File During an unattended installation, information about installation progress is output to the log file installhdlm.log. The following describes the installhdlm.log file: • The installhdlm.log file is created in the directory specified by the logdir key in the installation-information settings file. Note that the directory must not be shared with hosts on which unattended installation is performed.
7. Utility Reference 7.9 installux.sh Utility for HDLM Common Installer This utility determines what OS HDLM will be installed on, and installs the corresponding version of HDLM from the DVD-ROM. This utility can also perform unattended installations via a parameter specification. For details about how to use this utility to install HDLM, see 3.5.4 Performing a New Installation of HDLM (When Sun Cluster Is Being Used), 3.5.6 Preparations for an Upgrade Installation or Re-installation of HDLM, or 3.5.
Chapter 8. Messages This chapter describes the format and meaning of the message IDs, and also the terms used in the messages and message explanations. For details on the meaning of the return codes output by HDLM when it receives a request from Global Link Manager and measures to take for them, see 8.14 Return Codes for Hitachi Command Suite Common Agent Component. 8.1 Before Viewing the List of Messages 8.2 KAPL01001 to KAPL02000 8.3 KAPL03001 to KAPL04000 8.4 KAPL04001 to KAPL05000 8.
8. Messages 8.1 Before Viewing the List of Messages This section explains the following information that is needed to locate messages and understand the explanations in the sections from 8.2 KAPL01001 to KAPL02000. • Format and meaning of the message IDs • Terms used in the messages and message explanations This information is explained below. 8.1.1 Format and Meaning of Message IDs Each message has a message ID. The following shows the format and meaning of message IDs.
8.
8. Messages 8.2 KAPL01001 to KAPL02000 KAPL01001-I The HDLM command completed normally. Operation name = aa...aa, completion time = bb...bb Details The HDLM command completed successfully. aa...aa: clear, help, monitor, offline, online, set, or view bb...bb: Year/month/day hour:minute:second Action None. KAPL01002-I The HDLM command started. Operation name = aa...aa Details The HDLM command was executed. aa...aa: clear, monitor, offline, online, set, or view Action None.
8. Messages Details aa...aa: clear, set, monitor,online, offline, or view bb...bb: Specified parameter Action Execute help operation-name of the HDLM command (dlnkmgr) to check the parameter, and then retry. For details on the help operation, see 6.3 help (Displays the Operation Format). KAPL01006-W A necessary parameter is not specified. Operation name = aa...aa Details The specified operation does not contain the necessary parameter. aa...
8. Messages KAPL01009-W A parameter value is invalid. Operation name = aa...aa, parameter = bb...bb, parameter value = cc...cc, Valid value = dd...dd Details aa...aa: monitor,offline, online, set, or view bb...bb: Parameter name cc...cc: Specified parameter value dd...dd: Specifiable parameter value range Action Specify a correct value for the parameter, and then retry. KAPL01012-E Could not connect the HDLM manager. Operation name = aa...
8. Messages KAPL01014-W No authority to execute the HDLM command. Operation name = aa...aa Details You do not have the administrator permissions necessary to execute the HDLM command. aa...aa: clear, monitor, offline, online, set, or view Action Execute the command as a user with root permissions. KAPL01015-W The target HBA was not found. Operation name = aa...aa Details The path having the port number and bus number specified in the -hba parameter could not be found. aa...
8. Messages the corresponding CHA_ID, and retry the operation. For details on the view operation, see 6.7 view (Displays Information). KAPL01018-W The target device was not found. Operation name = aa...aa Details The specified host device name could not be found. aa...aa: view Action Execute the view operation of the HDLM command (dlnkmgr view -path) to check the value displayed in HDevName. Specify a host device for the value of HDevName, and then retry. For details on the view operation, see 6.
8. Messages KAPL01021-E Cannot execute the HDLM command due to insufficient memory. Details Memory required for HDLM command processing could not be allocated. Action Terminate unnecessary applications to increase the amount of free memory, and then retry. KAPL01023-W The last Online path for the device cannot be placed Offline(C). Details The path specified for the offline operation cannot be set to Offline(C) because it is the last, available path to the LU from the host.
8. Messages PathID = aa...aa Would you like to continue the processing of the online operation? [y/n]: Details A path that cannot be placed Online was detected during multipath online processing. To ignore this path and perform online processing for the next path, enter y. To cancel processing, enter n. aa...aa: Path ID (decimal (base-10) number) Action If you want to continue processing of the online operation of the HDLM command for other paths, enter y. If you want to terminate processing, enter n.
8. Messages cc...cc: Duplicate parameter value Action Delete the duplicate parameter value, and then retry. KAPL01045-W Too many parameter values are specified. Operation name = aa...aa, parameters = bb...bb, parameter value = cc...cc Details aa...aa: monitor, offline, online, set, or view bb...bb: Parameter name cc...cc: Parameter value Action Execute help operation-name of the HDLM command (dlnkmgr) to check the parameter value, and then retry. For details on the help operation, see 6.
8. Messages Up the Operating Environment). KAPL01050-I The currently selected paths will be changed to the Online status. Is this OK? [y/n]: Details The online operation will be started. To continue the online operation, enter y. To cancel the operation, enter n. Action If you want to execute online processing, enter y. If you want to terminate processing, enter n. For details on the online operation, see 6.5 online (Places Paths Online).
8. Messages KAPL01053-I If you are sure that there would be no problem when the path is placed in the Offline(C) status, enter y. Otherwise, enter n. [y/n]: Details The offline operation will be started. To continue the offline operation, enter y. To cancel the operation, enter n. Action If you want to execute offline processing, enter y. If you want to terminate processing, enter n. For details on the offline operation, see 6.4 offline (Places Paths Offline).
8. Messages KAPL01056-I If you are sure that there would be no problem when all the paths which pass the specified aa...aa are placed in the Offline(C) status, enter y. Otherwise, enter n. [y/n]: Details This message re-asks the user whether they want to set all the paths to Offline(C). To set all the paths to Offline(C), enter y. To cancel the operation, enter n. aa...aa: CHA port, HBA Action If you want to execute offline processing for the paths that meet the specified requirements, enter y.
8. Messages KAPL01059-W Cannot specify the parameter aa...aa at the same time if you specify parameter bb...bb and parameter value cc...cc. Operation name = dd...dd Details A parameter value is conflicting with the value of another parameter. bb...bb: Parameter name cc...cc: Parameter value aa...aa: Parameter name dd...dd: view or set Action Execute help operation-name of the HDLM command (dlnkmgr) to check the parameter and parameter value, and then retry. For details on the help operation, see 6.
8. Messages Action None. For details on the online operation, see 6.5 online (Places Paths Online). For details on the offline operation, see 6.4 offline (Places Paths Offline). KAPL01063-I The target path(s) are already aa...aa. Details As a result of a previous online/offline operation, the specified path has already been set to Online/Offline(C). aa...aa: Online or Offline(C) Action Use the view operation of the HDLM command (dlnkmgr) to check the status of the path.
8. Messages Obtain a valid license key, and then retry. KAPL01071-I The permanent license was installed. Details The license was renewed and is registered as a permanent license. Action None. KAPL01072-I The emergency license was installed. The license expires on aa...aa. Details A license was renewed and is registered as an emergency license. aa...aa: Year (4 digits)/month (01-12)/day (01-31) Action Install a permanent license by the expiration day. KAPL01073-E The temporary license expired.
8. Messages Re-install HDLM. KAPL01076-I The permanent license has been installed. Details You need not install a license because a permanent license has already been installed. Action None. KAPL01078-W The operation terminated because the path configuration changed during execution of the HDLM command. Operation name = aa...aa Details aa...aa: monitor Action After the dynamic reconfiguration of the HDLM device has finished, re-execute the HDLM command.
8. Messages number-of-times-error-is-to-occur-for-intermittent-error-monitoring). KAPL01081-E The license key file is invalid. File name = aa...aa Details The format of the license key file is invalid. aa...aa: /var/tmp/hdlm_license Action Save the correct license key file in the designated, and then re-execute. /var/tmp/hdlm_license KAPL01082-E There is no installable license key in the license key file. File name = aa...aa Details There is no useable license key for HDLM in the license key file. aa..
8. Messages aa...aa: /var/tmp/hdlm_license Action If a license key file exists, delete it. /var/tmp/hdlm_license KAPL01088-W The specified parameter values cannot be specified at the same time. Operation name = aa...aa, parameter = bb...bb, parameter values = cc...cc Details aa...aa:view bb...bb: Parameter name cc...cc: Parameter values that cannot be specified at the same time Action Execute help operation-name of the HDLM command (dlnkmgr) to check which parameter can be specified, and then retry.
8. Messages aa...aa: Path ID bb...bb: Code showing reason for error Action Resolve the error in the relevant path, and then re-execute. KAPL01091-E The Offline path cannot be placed Online because memory is insufficient. path ID = aa...aa, details = bb...bb Details An attempt to recover the path failed because of insufficient memory. Online processing will be stopped if online processing is running for multiple paths. aa...aa: Path ID bb...bb: Code showing reason for error Action Re-execute.
8. Messages Details An attempt to recover the path failed because the correct disk label has not been set in the LU for the path to be put online. aa...aa: Path ID bb...bb: Code showing reason for error Action Use the format command to set the VTOC label or EFI label in the LU for the relevant path, and then re-execute the command. KAPL01094-E The Offline path cannot be placed Online because an error occurred. path ID = aa...aa, details = bb...
8. Messages aa...aa: Code showing the reason for the error Action Re-execute the command. If the same error occurs again, execute the DLMgetras utility for collecting HDLM error information, acquire the error information, and then contact your HDLM vendor or the company for which you have a service contract. KAPL01097-W All the current trace files will be deleted.
8. Messages Details Multiple paths will be collectively placed Offline(C) because the -hbaid or -chaid parameter was specified. To collectively place multiple paths Offline(C), enter y. To not collectively place them Offline(C), enter n. aa...aa: CHA, HBA Action If you want to execute the offline processing for the paths which pass the specified target, enter y. If you want to terminate the processing, enter n.
8. Messages want to terminate processing, enter n. KAPL01112-E An attempt to connect to the HDLM driver has failed. Operation name = aa...aa Details HDLM driver information must be collected to execute the given HDLM command, but the HDLM driver cannot be accessed. aa...
8. Messages KAPL01116-W The audit log configuration file is invalid. Restart the HDLM Manager, and execute the "dlnkmgr view -sys -audlog" command and check the setting. Details The audit log configuration file is invalid. Action Restart the HDLM Manager, and execute the "dlnkmgr view -sys -audlog" command, and then specify the desired setting by using the "dlnkmgr set -audlog" command or the "dlnkmgr set -audfac" command as necessary.
8. Messages KAPL01120-W A storage subsystem model ID could not be displayed. Details = aa...aa, bb...bb Details A storage system model ID could not be displayed. aa...aa: Storage recognition information bb...bb: Error code Action Execute the DLMgetras utility for collecting HDLM error information, and then contact your HDLM vendor or the maintenance company if there is a maintenance contract for HDLM. For details on the DLMgetras utility, see 7.
8. Messages 8.3 KAPL03001 to KAPL04000 KAPL03001-I HDLM API information - aa...aa Details This information is required for resolving problems. aa...aa: Trace information Action None. KAPL03003-E HDLM API Error information - aa...aa Details This information is required for resolving problems. aa...
8. Messages aa...aa: API trace error information Action Execute the DLMgetras utility for collecting HDLM error information, and then contact your HDLM vendor or the maintenance company if there is a maintenance contract for HDLM. For details on the DLMgetras utility, see 7.2 The DLMgetras Utility for Collecting HDLM Error Information. KAPL03007-E An error occurred during communication with the HDLM manager. (aa...aa) Details This information is required for resolving problems. aa...
8. Messages contract for HDLM. For details on the DLMgetras utility, see 7.2 The DLMgetras Utility for Collecting HDLM Error Information.
8. Messages 8.4 KAPL04001 to KAPL05000 KAPL04001-I HDLM manager started. Action None. KAPL04002-E Could not start the HDLM manager. Details The HDLM manager failed to start because the current environment is unsuitable for the HDLM manager to run in. Action Execute the DLMgetras utility for collecting HDLM error information, and then contact your HDLM vendor or the maintenance company if there is a maintenance contract for HDLM. For details on the DLMgetras utility, see 7.
8. Messages aa...aa: Option definition file name Action Check whether another program is currently using this file (for example, the file is being opened in a text editor), or whether the file has been inadvertently deleted. KAPL04009-E The option definition is invalid. Details HDLM manager could not start normally because some of the definitions in the option definition file were invalid.
8. Messages KAPL04012-E Could not create a communication pipe. RC = aa...aa Details HDLM manager could not start normally because it was unable to create a pipe file, which is used in communication with HDLM commands. aa...aa: OS error code (decimal (base-10) number) Action Execute the DLMgetras utility for collecting HDLM error information, and then contact your HDLM vendor or the maintenance company if there is a maintenance contract for HDLM. For details on the DLMgetras utility, see 7.
8. Messages Details An attempt to read the log information from the alert driver failed. aa...aa: API return code (decimal (base-10) number) Action Execute the DLMgetras utility for collecting HDLM error information, and then contact your HDLM vendor or the maintenance company if there is a maintenance contract for HDLM. For details on the DLMgetras utility, see 7.2 The DLMgetras Utility for Collecting HDLM Error Information. KAPL04021-I HDLM manager information - aa...
8. Messages DLMgetras Utility for Collecting HDLM Error Information. KAPL04024-C A critical error occurred in the HDLM manager. (aa...aa) Details This information is required for resolving problems. aa...aa: HDLM manager trace error information Action Execute the DLMgetras utility for collecting HDLM error information, and then contact your HDLM vendor or the maintenance company if there is a maintenance contract for HDLM. For details on the DLMgetras utility, see 7.
8. Messages Install a permanent license by the expiration day. KAPL04028-E The temporary license expired. Action Install a permanent license. KAPL04029-E The emergency license expired. Action Install a permanent license. KAPL04030-E The temporary license has already expired. Action Install a permanent license. KAPL04031-E The emergency license has already expired. Action Install a permanent license. KAPL04032-C A fatal error occurred in HDLM.
8. Messages KAPL04034-E An attempt to create the option definition file has failed. Details An attempt to re-create an option definition file (/etc/opt/ DynamicLinkManager/dlmmgr.xml) using the default values has failed. Action Remove unnecessary files to secure free capacity on the file system, or check the write permissions for the directory and file. KAPL04035-I The path health check will now start. Total number of paths = aa...aa Details aa...aa: Total number of paths Action None.
8. Messages aa...aa: Start or stop Action None. KAPL04045-I HDLM SNMP TRAP was sent. Trap ID = aa...aa, IP Address = bb...bb, Port Number= cc...cc, Community = dd...dd, Trap Data = ee...ee Details aa...aa: Trap ID bb...bb: Destination IP address of the trap cc...cc: Destination port number of the trap dd...dd: Community name given to the trap ee...ee: Transmission data Action None.
8. Messages 8.5 KAPL05001 to KAPL06000 KAPL05003-I The HDLM driver (filter component) was successfully attached to Disk (aa...aa), Partition (bb...bb). Details The path for the disk and partition was registered successfully in the core logic. aa...aa: minor number of the sd or ssd device (decimal number) bb...bb: slice number Action None. KAPL05008-E Could not allocate memory. (aa...aa:bb...
8. Messages aa...aa: Minor number of the sd or ssd device (decimal number) bb...bb: Slice number (decimal number) cc...cc: Error code (hexadecimal number) dd...dd: Address of table for managing the HDLM filter driver (hexadecimal number) Action Check whether the HDLM driver has started normally. If it has not started or contains an error, contact your HDLM vendor or the maintenance company, if there is a maintenance contract for HDLM, and report the error and detail code.
8. Messages KAPL05019-I The FO processing in the path (aa...aa) finished. The I/O request was processed in the path (bb...bb). Details aa...aa: Path ID (hexadecimal number) bb...bb: Path ID (hexadecimal number) Action None. KAPL05020-I Processing of IOCTL(aa...aa) will now start. Details An IOCTL request was received.. aa...aa: IOCTL code (hexadecimal number) Action None. KAPL05021-I Processing of IOCTL(aa...aa) completed normally.
8. Messages cc...cc: 0 (fixed) Action Check the message of the HDLM command (dlnkmgr) or the HDLM manager, and then take the appropriate action. If you do not know the appropriate action, execute the DLMgetras utility for collecting HDLM error information, and then contact your HDLM vendor or the maintenance company if there is a maintenance contract for HDLM and report the error and detail code. For details on the DLMgetras utility, see 7.2 The DLMgetras Utility for Collecting HDLM Error Information.
8. Messages KAPL05041-E The contents of the configuration definition file (dlmfdrv.conf) are invalid. (Instance = aa...aa) Execute the dlmsetconf command, and then reboot. (bb...bb:cc...cc:dd...dd) Details A definition in the configuration definition file for the HDLM filter driver is invalid. aa...aa:HDLM filter driver instance number bb...bb: Code indicating the error type cc...cc: Error information 1 dd...dd: Error information 2 Action 1. Make sure the path is connected to the disk correctly. 2.
8. Messages The instance number of the HDLM filter driver instance that you tried to attach is already being used. aa...aa: Instance number (decimal number) Action Contact the HDLM vendor, or the maintenance company if you have a maintenance contract for HDLM. KAPL05412-W An attempt to allocate a bus address for the HDLM device instance aa...aa failed. The HDLM device could not be attached.
8. Messages kernel/drv/dlmndrv.conf) of the filter driver or nexus driver is invalid. aa...aa: Driver name Action Check whether the path is correctly connected to the disk. Execute the dlmsetconf utility for creating the HDLM driver configuration definition file, and then reboot the host. KAPL05438-E The contents of the configuration definition file aa...aa.conf are invalid. Instance = bb...bb. Check whether the path is correctly connected to the disk. Execute the dlmsetconf command and reboot.
8. Messages aa...aa: Path ID bb...bb: Error information (internal code) Action Stop any applications that are using the HDLM devices, and then re-execute the dlmcfgmgr utility for managing the HDLM configuration. KAPL05901-E Failed to open aa...aa directory. errno = bb...bb Execute the DLMgetras utility to collect error information, and then contact your vendor or maintenance company. Refer to the HDLM User's Guide for instructions how to execute the DLMgetras utility.
8. Messages KAPL05904-E Failed to set device name. Execute the DLMgetras utility to collect error information, and then contact your vendor or maintenance company. Refer to the HDLM User's Guide for instructions how to execute the DLMgetras utility. Details An attempt by dlmsetdevname to report the logical device file name of the HDLM device to the filter driver has failed. Action Check whether the HDLM driver has started normally. KAPL05905-W No authority to execute the dlmsetdevname command.
8. Messages 8.6 KAPL06001 to KAPL07000 KAPL06003-I Initialization of the HDLM alert driver (aa...aa) was successful. Details aa...aa: Address of the table for managing the alert driver (hexadecimal number) Action None. KAPL06004-E Could not allocate memory. (aa...aa:bb...bb) Details An attempt to reserve memory to save alert information has failed. aa...aa: Program line (hexadecimal number) bb...bb: Target memory size (hexadecimal number) Action Check whether the HDLM driver has started normally.
8. Messages Action None. KAPL06009-I Invalid IOCTL (aa...aa) was received. The processing is canceled. Details A request to the alert driver was made with an invalid IOCTL code. aa...aa: IOCTL code (hexadecimal number) Action None. KAPL06010-E Could not process the IOCTL(aa...aa). (bb...bb:cc...cc) Details An IOCTL request was not processed normally. aa...aa: IOCTL code (hexadecimal number) bb...bb: IOCTL end code (hexadecimal number) cc...
8. Messages written is discarded. Review the actual memory size when another error does not occur. When the actual memory size is insufficient, increase the actual memory size. When the actual memory size is sufficient, execute the DLMgetras utility for collecting HDLM error information, and then contact your HDLM vendor or the maintenance company if there is a maintenance contract for HDLM. For details on the DLMgetras utility, see 7.2 The DLMgetras Utility for Collecting HDLM Error Information.
8. Messages 8.7 KAPL07001 to KAPL08000 KAPL07819-I Data for maintenance: aa...aa bb...bb cc...cc dd...dd. Details This message is generated by the core logic for maintenance. aa...aa: Detailed information 1 (decimal (base-10) number) bb...bb: Internal function number of the core logic (decimal (base-10) number) cc...cc: Detailed information 2 (decimal (base-10) number) dd...dd: Detailed information 3 (decimal (base-10) number) Action None.
8. Messages 8.8 KAPL08001 to KAPL09000 KAPL08019-E The path (aa...aa) detected an error (bb...bb). (cc...cc) Details An error occurred in the path. The error is most likely due to a disconnected cable. aa...aa: Path identifier (hexadecimal number) bb...bb: Error code (hexadecimal number) • When a path error is detected by a path health checking or the online operation: Displays 0x000F0000(Fixed). • When a path error is detected through an I/O error: Displays the OS error code. cc...
8. Messages ee...ee: Target ID (same as PathName of view -path) (hexadecimal number) ff...ff: HLU number (same as PathName of view -path) (hexadecimal number) gg...gg: Dev number (same as DNum of view -path) (decimal (base-10) number) hh...hh: Host device name (same as HDevName of view -path) Action There could be an error in the path. See 5.3 What To Do for a Path Error and restore the path displayed in the message to running status. KAPL08023-I A path was recovered. PathID = aa...aa, PathName = bb...bb.
8. Messages KAPL08027-E A path was excluded from the items subject to automatic failback. PathID = aa...aa Details A path was excluded from being subject to automatic failbacks because the system judged that an intermittent error was occurring in that path. aa...aa: Path ID (same as PathID of view -path) (decimal (base-10) number) Action An intermittent error has occurred. Check the path for any possible problems. For details on what to do, see 5.
8. Messages 8.9 KAPL09001 to KAPL10000 KAPL09002-E The disk does not have sufficient free space. Details The file system of the installation does not have sufficient free capacity to install HDLM. Action Delete unnecessary files to increase the amount of free capacity, and then retry. KAPL09008-W The license code is invalid. Action Double-check the license code, and then re-enter it. KAPL09009-E The license code is invalid. The HDLM installation program will now terminate.
8. Messages Manually cancel the PP name registration, and then uninstall Hitachi Network Objectplaza Trace Library 2. If the attempt to cancel the registration of the PP name and uninstall Hitachi Network Objectplaza Trace Library 2 fails again, contact your HDLM vendor or the maintenance company if there is a maintenance contract of HDLM. KAPL09020-E An attempt to uninstall Hitachi Network Objectplaza Trace Library 2 failed. Details An attempt to uninstall HNTRLib2 has failed.
8. Messages An attempt to copy a file related to HDLM to the directory of Hitachi Command Suite products other than HDLM has failed. Action If you detect this message while executing the HDLM installation program, re-execute the HDLM installation program. If you detect this message while executing the Hitachi Command Suite products other than HDLM installation program, re-execute the Hitachi Command Suite products other than HDLM installation program.
8. Messages Action See 3.1.1 Hosts and OSs Supported by HDLM and install HDLM in a supported OS system. KAPL09029-E This version of HDLM cannot be updated by installation. Uninstall the already installed version of HDLM. Action Uninstall the already installed version of HDLM. KAPL09030-E The installed version of HDLM is not supported for the present OS version. Uninstall the already installed version of HDLM. Action Uninstall the already installed version of HDLM.
8. Messages None. KAPL09077-I The temporary license was installed. The license expires on aa...aa. Details A temporary license was installed. aa...aa: Year (4 digits)/month (01-12)/day (01-31) Action Install a permanent license by the expiration day. KAPL09078-I The emergency license was installed. The license expires on aa...aa. Details An emergency license was installed. aa...aa: Year (4 digits)/month (01-12)/day (01-31) Action Install a permanent license by the expiration day.
8. Messages The emergency license has been installed. aa...aa: Year (4 digits)/month (01-12)/day (01-31) Action Install a permanent license by the expiration day. KAPL09082-W The temporary license expired. Action Enter a permanent license key. KAPL09083-W The emergency license expired. Action Install a permanent license. KAPL09086-W The entered license key is invalid. Action Enter a valid license key. KAPL09087-E The entered license key is invalid. Renewal of the license key will now stop.
8. Messages You need to install a license because the license information cannot be acquired. Action In the case that the license key file is being placed in the designated directory (/ var/tmp/hdlm_license): In this case, a license is installed from the license key file. No special operations are needed. In the case that the license key file is being placed in a directory other than above or saved with another file name: Enter 1 in response to the KAPL09200-I message, and then follow the output message.
8. Messages Place the correct license key file in the designated directory (/var/tmp/hdlm_license), in a user-specified directory or with a user-specified file name, and then re-install HDLM. Details The format of the license key file is invalid. aa...aa: /var/tmp/hdlm_license or the user-specified file Action Save the correct license key file in the designated directory (/var/tmp/ hdlm_license), in a user-specified directory or with a user-specified file name, and then re-execute the installation.
8. Messages Details An attempt to delete the license key file has failed. aa...aa: /var/tmp/hdlm_license Action If a license key file exists, delete it. /var/tmp/hdlm_license Store a correct license key file in the designated directory, and then re-execute. KAPL09133-E One or more required patches (aa...aa) could not be found. Please apply all required patches then install again. Details aa...aa: the number of the patch not yet applied Action Check the prerequisite patches described in 3.1.
8. Messages Action Please perform the utility for collecting HDLM error information (DLMgetras), and acquire obstacle information, and then contact your HDLM vendor or the maintenance company if there is a maintenance contract for HDLM and report the error and detail code. KAPL09140-E HDLM cannot be installed in this zone. Install it in the global zone. Details HDLM cannot be installed in a non-global zone. Install HDLM in a global zone. Action Install HDLM in a global zone. KAPL09142-E HDLM aa...
8. Messages KAPL09148-W Could not modify file. File name = aa...aa, Details = bb...bb. Please contact product support. Details An HDLM driver configuration definition file or HDLM driver unconfiguration definition file, aa...aa, could not be updated. aa...aa: One of the following file names: • /kernel/drv/dlmfdrv.conf • /opt/DynamicLinkManager/config/dlmfdrv.unconf • /kernel/drv/dlmfdrv.conf.bak • /opt/DynamicLinkManager/config/dlmfdrv.unconf.bak bb...bb: Internal code (decimal number) Action 1.
8. Messages KAPL09149-E An attempt to update the file has failed. File name = aa...aa, Details = bb...bb. Please contact product support. Details An HDLM driver configuration definition file or HDLM driver unconfiguration definition file, aa...aa, could not be updated. aa...aa: One of the following file names: • /kernel/drv/dlmfdrv.conf • /opt/DynamicLinkManager/config/dlmfdrv.unconf • /kernel/drv/dlmfdrv.conf.bak • /opt/DynamicLinkManager/config/dlmfdrv.unconf.bak bb...
8. Messages dlmfdrv.conf) by executing dlmsetconf utility. 2. When you want to specify that a device is not to be managed by HDLM, edit the HDLM driver unconfiguration definition file (/opt/ DynamicLinkManager/config/dlmfdrv.unconf) and re-execute dlmsetconf utility with -u option. 3. When you use VxVM, remove HDLM devices from VxVM. The dlmvxexclude utility assists this. 4. Restart the host with the reconfiguration option specified. KAPL09179-I Data for maintenance: aa...aa bb...bb Details aa...
8. Messages More than two parameters have been specified for the installhdlm utility for installing HDLM. Action Make sure that the parameters for installhdlm are appropriate, and then try again. KAPL09190-W The installation information settings file is not specified. Details The installation-information settings file is not specified for the second parameter in the installhdlm utility for installing HDLM.
8. Messages Action Follow the message output at the same time. KAPL09203-W There is no license key file. File name = aa...aa Details There is no license key file. aa...aa: Character string that the user entered Action Follow the message output at the same time. KAPL09204-E The entered absolute path is invalid. The HDLM installation will now terminate. Action Check the absolute path of the license key file, and then re-install HDLM. KAPL09205-E There is no license key file.
8. Messages KAPL09208-E There is no license key file. Renewal of the license key will now stop. File name = aa...aa Details There is no license key file. Renewal of the license key will now stop. aa...aa: Character string that the user entered Action Check the absolute path of the license key file, and then re-install HDLM. KAPL09209-E The entered value is invalid. Renewal of the license key will now stop. Action Re-execute the installation. KAPL09210-I aa...aa will now start. Details aa...
8. Messages Check the error message that was output just before this message, and then perform the action indicated in that error message. KAPL09213-W An error occurred during aa...aa processing. Details Although the aa...aa processing has ended, an error occurred during the processing. aa...aa: installhdlm, pkgadd, dlnkmgr, or dlmsetconf Action Check the error message that was output just before this message, and then perform the action indicated in that error message.
8. Messages You do not have write permission for the file system of the /var/tmp directory. Make sure you have access permissions for the file system of the /var/tmp directory. If the error code is 7: There is not enough free capacity for the file system of the /var/tmp directory. Make sure there is enough free capacity in the /var/tmp directory. If the error code is 10: The dlmsetconf command is not in the default location or you do not have execution permission for the dlmsetconf command.
8. Messages Action If the error code is 9001,-1: You do not have read permissions for the installation information settings file. Make sure that you have access permissions. If any other error code is output: Contact your HDLM vendor or the maintenance company if there is a maintenance contract for HDLM. KAPL09218-E aa...aa cannot be executed. Details A utility or command that installhdlm uses was not in the correct location, or you do not have a required permission. aa...
8. Messages maintenance contract for HDLM. KAPL09220-W The composition of the installation information settings file is invalid. Error Code = aa...aa,bb...bb Details In the installation information settings file, there is a line whose length exceeds 1,023 characters before [INSTALLATION_SETTINGS], or the first non-whitespace, non-comment line that appears is not [INSTALLATION_SETTINGS]. aa...aa: Error number (decimal number) that specifies the executed processing bb...
8. Messages Action Specify the revised installation information settings file and then retry the operation. KAPL09228-W The definition of the installation information settings file includes an invalid key value. Error Code = aa...aa,bb...bb, line = cc...cc Details The format of a key value is invalid. aa...aa: Error number (decimal number) that specifies the executed processing bb...bb: Return value (decimal number) of the executed processing cc...
8. Messages bb...bb: Return value (decimal number) of the executed processing cc...cc: Line number (decimal number) of an installation information settings file. Action Specify the revised installation information settings file and then retry the operation. KAPL09231-W The definition of the installation information settings file includes a duplicated key. Error Code = aa...aa,bb...bb, line = cc...cc Details A duplicated key is included. aa...
8. Messages aa...aa: Error number (decimal number) that specifies the executed processing bb...bb: Return value (decimal number) of the executed processing cc...cc: Line number (decimal number) of an installation information settings file. Action Specify the revised installation information settings file and then retry the operation. KAPL09234-W A folder or file specified by the installation information settings file does not exist. Name = aa...
8. Messages You do not have write permission for the installhdlm.log file. Modify your access permissions as necessary, and then retry the operation. KAPL09236-W An error occurred during the output of a log file. Details A memory shortage occurred at the output destination of the log file. Action Check the error message output before this message, and then perform the action indicated in that error message. KAPL09237-I A user operation ended installhdlm.
8. Messages Details The host will restart because restart was specified in the installation information settings file. Action None. KAPL09241-W An attempt to install an HDLM component failed. Error Code = aa...aa Details An attempt to install an HDLM component failed. aa...aa: Error number (decimal number) that specifies the executed processing Action Depending on the output error code, take the appropriate action as shown below.
8. Messages The HDLM component installation utility (dlminstcomp) finished successfully. Action None. KAPL09246-E An attempt to install an HDLM component failed because an HDLM component was operating. Error Code = aa...aa Details An attempt to install an HDLM component failed because an HDLM component was operating. aa...aa: Error number (decimal number) that specifies the executed processing Action Wait a while, and then execute the HDLM component installation utility (dlminstcomp).
8. Messages An SP cannot be applied because HDLM is not installed on the system. Action Check whether HDLM has been correctly installed. KAPL09504-E The language environments of HDLM and the Service Pack are different. Details The Japanese SP was applied to the English edition of HDLM, or vice versa. Action Acquire the SP that has the same language as the installed HDLM, and then try again. KAPL09505-E aa...aa cannot be applied to the installed bb...bb. Details This HDLM or SP (aa...
8. Messages Details A newer version of the SP (bb...bb) is already installed. Installation of SP (aa...aa) is stopped. aa...aa: The version of the SP to be installed bb...bb: The version of the installed HDLM or the version of the installed SP Action You do not have to install the SP. Keep using HDLM as is. KAPL09601-E Cannot install in this system. Install HDLM on a supported OS. Details HDLM cannot be installed on this system. Action Execute the installation in a supported OS.
8. Messages • If an installation is performed by using files copied from the DVD-ROM, re-copy all of the DVD-ROM files. After that, confirm the permission of the HDLM utility that caused the error. When the output information is an OS command: • Confirm that the OS command exists, and that you have the proper permission. KAPL09604-E A system error occurred. The installation will now stop. (command = aa...aa, error code = bb...bb) Details aa...aa: Command that failed bb...
8. Messages 8.10 KAPL10001 to KAPL11000 KAPL10001-W No parameter has been specified. Details No directory to which the collected information will be output has been specified. Action Check the parameters of the DLMgetras utility for collecting HDLM error information, and then retry. For details on the DLMgetras utility, see 7.2 The DLMgetras Utility for Collecting HDLM Error Information. KAPL10002-W Too many parameters have been specified. Details Four or more parameters have been specified.
8. Messages Action Check the parameters of the DLMgetras utility for collecting HDLM error information, and then retry. For details on the DLMgetras utility, see 7.2 The DLMgetras Utility for Collecting HDLM Error Information. KAPL10005-W The number of parameters is insufficient. Details The -f parameter exists but the file for defining the information to be collected does not exist. Action Check the parameters of the DLMgetras utility for collecting HDLM error information, and then retry.
8. Messages Details You do not have write permission for the specified directory, or the creation of a subdirectory of the specified directory failed. aa...aa: First parameter Action Check the following. 1. Check whether you have access permission for the specified directory. 2. Check whether the specified directory name is correct. 3. Check that the disk has sufficient free capacity. KAPL10009-W The specified directory already exists.
8. Messages Two or more file names or directory names exist in the file for defining the information to be collected. aa...aa: Line number of the file for defining information to be collected (decimal number) bb...bb: Indicated contents in a line Action After the DLMgetras utility for collecting HDLM error information terminates, check the contents of the file for defining the information to be collected. This file is shown in the message.
8. Messages bb...bb: Indicated contents in a line Action After the DLMgetras utility for collecting HDLM error information terminates, check the contents of the file for defining the information to be collected. This file is shown in the message. If the contents of the file are incorrect, correct them and then try to collect error information again. The DLMgetras will ignore the specified file and continue processing. For details on the DLMgetras utility, see 7.
8. Messages Details A root "/" cannot be specified in a directory to which collected information is output. Action Check the parameters of the DLMgetras utility for collecting HDLM error information, and then re-execute. For details on the DLMgetras utility, see 7.2 The DLMgetras Utility for Collecting HDLM Error Information. KAPL10017-W You lack privileges for executing the utility for collecting HDLM error information.
8. Messages KAPL10022-I The utility for collecting HDLM error information completed normally. Details Error information has been collected. Action None. For details on the DLMgetras utility, see 7.2 The DLMgetras Utility for Collecting HDLM Error Information. KAPL10030-I A user terminated the utility for collecting HDLM error information. Details Processing of the DLMgetras utility for collecting HDLM error information has been terminated because the user replied to the confirmation with an n response.
8. Messages The file to collect does not exist. aa...aa: File to collect Action None. KAPL10034-E The file could not be copied. Filename = aa...aa, Details = bb...bb Details Execution of the cq command failed. aa...aa: File name you tried to copy bb...bb: cq command output message Action An error occurred during copying of the file to be collected. The user environment might have been unstable. Check the system configuration. KAPL10035-E An attempt to archive the error information failed. Details = aa..
8. Messages company if you have a maintenance contract for HDLM. KAPL10049-I Error information collection command = aa...aa, Return value = bb...bb, Execution time = cc...cc Details A command was executed to collect information targeted for collection. aa...aa: Executed command bb...bb: Return value of the executed command cc...cc: year (4 digits)/month/date hour:minute:second Action None. KAPL10050-I The utility for collecting HDLM error information started. Start time = aa...aa (GMT aa...
8. Messages Action Check the capacity of the root file system or the file system including the directory specified by the -d option. Check the memory capacity that can be used. KAPL10204-I dlmsetconf completed successfully. Details The dlmsetconf utility for creating the HDLM driver configuration definition file completed successfully. Action None. KAPL10220-E An error occurred when collecting the HDLM support disk information.
8. Messages KAPL10224-E An error occurred when creating the configuration definition information. Details An attempt to create a configuration definition file has failed. Action Check the system configuration. KAPL10225-W The existing configuration definition file could not be opened. Errno = aa...aa (bb...bb) Details An attempt to open a configuration definition file has failed. aa...aa: Errno (decimal number) bb...bb: Errno string Action Remove the error by referencing Errno, and then retry.
8. Messages KAPL10229-E Internal processing error. Details An internal processing error occurred. Action Internal processing error. Contact your HDLM vendor or the maintenance company if there is a maintenance contract for HDLM. KAPL10230-E The aa...aa directory could not be opened. Errno = bb...bb (cc...cc) Details An attempt to open a directory has failed. aa...aa: Directory name bb...bb: Errno (decimal number) cc...cc: Errno string Action Remove the error by referencing Errno, and then retry.
8. Messages cc...cc: Errno string Action Remove the error by referencing Errno, and then retry. KAPL10235-W Get symbolic link file(aa...aa) status failed. Function = lstat, Errno = bb...bb (cc...cc) Details An attempt to acquire information about the symbolic link file has failed. aa...aa: File name bb...bb: Errno (decimal number) cc...cc: Errno string Action Remove the error by referencing Errno, and then retry. KAPL10236-E Could not allocate memory. Errno = aa...aa (bb...
8. Messages KAPL10238-W The device (aa...aa) could not be opened. Errno = bb...bb (cc...cc) Details An attempt to open an sd or ssd device failed. aa...aa: sd or ssd device name bb...bb: Errno (decimal number) cc...cc: Errno string Action Remove the error by referencing Errno, and then retry. KAPL10240-E The directory specified by the -d option does not exist. Details The directory specified in the -d option was not found. Action Check whether the specified directory exists.
8. Messages HDLM-managed sd or ssd device are deleted at reboot. To configure HDLM, enter y if you want to execute the dlmsetconf utility, enter n to stop. KAPL10243-W A lock error occurred during open processing for the physical path (aa...aa). Details An attempt to open a path failed because an exclusive open is being executed for another process. aa...aa: Physical path Action Comply with the following KAPL10244-I messages.
8. Messages KAPL10252-W An attempt to read the contents of symbolic link aa...aa failed. Function = readlink, errno = bb...bb (cc...cc) Details An attempt to acquire link destination information about the symbolic link file has failed. aa...aa: File name bb...bb: Errno (decimal number) cc...cc: Errno string Action See errno to eliminate the error, and then retry. KAPL10253-W An attempt to create symbolic link aa...aa to a file failed. Function = symlink, errno = bb...bb (cc...
8. Messages Processing will stop because an incorrect value was entered three or more times for the message confirming the execution of the dlmsetconf utility for creating the HDLM driver configuration definition file. Action When executing the utility, enter the correct value. KAPL10256-I The user terminated the operation. Details The user has interrupted processing. Action None. KAPL10257-W The output file (aa...aa) could not be opened. Errno = bb...bb (cc...
8. Messages If the disk indicated in Device is the LU that was dynamically deleted, no action is required. To prevent this message from being output, restart the host. If the disk indicated in Device is the LU that was not dynamically deleted, make sure that the disk is properly connected to the server. If there are any errors, remove the error, and then re-execute the command.
8. Messages Details An attempt to acquire HDevList failed because the dlnkmgr view -path command failed. Action Remove from the dlnkmgr view -path command error, and execute the command again. KAPL10299-E The input file is invalid. The input file is not the output result of the "dlnkmgr view -path" command. Details The content of a file specified in a parameter of the dlmlisthdev command differs from the output of the dlnkmgr view -path command. Action Check the input file, and then retry.
8. Messages KAPL10328-E The entered value is invalid. The HDLM-configuration definition utility processing will now stop. Execute the HDLM-configuration definition utility again. Details Processing will stop because an incorrect value was entered three or more times for the message confirming execution of the dlmcfgmgr utility. Action Re-execute the dlmcfgmgr utility. For details on the dlmcfgmgr utility, see 7.3 The dlmcfgmgr Utility for Managing the HDLM Configuration.
8. Messages Details Displays how to use dlmcfgmgr. Action Specify appropriate parameters, and then retry. For details on the dlmcfgmgr utility, see 7.3 The dlmcfgmgr Utility for Managing the HDLM Configuration. KAPL10369-I dlmcfgmgr aa...aa started: bb...bb Details Shows the execution option and the execution start time of the dlmcfgmgr utility. aa...aa: Parameter bb...bb: Date (character string) Action None. For details on the dlmcfgmgr utility, see 7.
8. Messages Details A message could not be acquired from the message catalog. aa...aa: Errno bb...bb: Errno string Action Remove the error by referencing ErrorCode(errno), and then retry. KAPL10373-I This operation will change the configuration of HDLM devices, and delete device files of sd/ssd devices managed by HDLM.Do you want to continue? [y/n]: Details This message confirms the execution of the dlmcfgmgr utility for managing the HDLM configuration.
8. Messages KAPL10376-E The device (aa...aa) could not be opened. ErrorCode = bb...bb (cc...cc) Details An attempt to open the device managing the HDLM filter driver has failed. aa...aa: Name of the device managing the HDLM filter driver bb...bb: Errno (decimal number) cc...cc: Errno string Action Take action in accordance with ErrorCode (errno), and then retry. KAPL10377-I The HDLM device configurations have not been changed. Details There was no change in the HDLM device configuration. Action None.
8. Messages Contact your HDLM vendor or the maintenance company if there is a maintenance contract for HDLM. KAPL10380-E A parameter is invalid. Execute the HDLM-configuration definition utility with valid parameter. Details An execution option of the dlmcfgmgr utility for managing the HDLM configuration is incorrect. Action Specify appropriate parameters, and then retry. For details on the dlmcfgmgr utility, see 7.3 The dlmcfgmgr Utility for Managing the HDLM Configuration.
8. Messages KAPL10384-E The value of the aa...aa property in the configuration definition file (bb...bb) is invalid. Execute the DLMgetras utility to collect error information, and then contact your vendor or maintenance company. Details The value of the property defined in the /kernel/drv/dlmfdrv.conf file is invalid. aa...aa: Property name bb...bb: Configuration definition file name Action Contact your HDLM vendor or the maintenance company if there is a maintenance contract for HDLM.
8. Messages Details The property value of the path described in the configuration definition file has the same value as a different LU path managed by dlmfdrv. aa...aa: LDEV name bb...bb: Property name cc...cc: LDEV name Action If you want to use the LU, reboot the host. KAPL10388-E The path (aa...aa) could not be added. (Function: bb...bb, Error Code = cc...cc (dd...dd)). Execute the DLMgetras utility to collect error information, and then contact your vendor or maintenance company.
8. Messages Check the capacity of the root file system or the file system including the file. Check the memory capacity which can be used. KAPL10391-E The output file (aa...aa) could not be opened. ErrorCode = bb...bb (cc...cc). Resolve the error by referencing ErrorCode. Details An attempt to open the output file has failed. aa...aa: Output file name bb...bb: Errno (decimal number) cc...cc: Errno string Action Remove the error by referencing ErrorCode (errno). KAPL10392-E The catalog file (aa...
8. Messages Action Contact your HDLM vendor or the maintenance company if there is a maintenance contract for HDLM. KAPL10396-W No authority to execute the dlmcfgmgr utility. Details You do not have permission to execute the dlmcfgmgr utility for managing the HDLM configuration. Action Retry as a user with root permissions. For details on the dlmcfgmgr utility, see 7.3 The dlmcfgmgr Utility for Managing the HDLM Configuration.
8. Messages KAPL10701-I usage: dlmvxexclude [ -d dir | -h ] Details Displays how to use dlmvxexclude. Action Specify appropriate parameters, and then retry. For details on the dlmvxexclude utility, see 7.7 The dlmvxexclude Utility for Assisting Creation of the VxVM Configuration File. KAPL10702-E A backup file (aa...aa) could not be created. Errno = bb...bb (cc...cc) Details An attempt to create a backup file has failed. aa...aa: Backup file name bb...bb: Errno (decimal number) cc...
8. Messages for Assisting Creation of the VxVM Configuration File. KAPL10705-E Internal processing error. Details An internal error occurred while the dlmvxexclude utility for assisting creation of the VxVM configuration file was executing. Action Check the system configuration and then retry. For details on the dlmvxexclude utility, see 7.7 The dlmvxexclude Utility for Assisting Creation of the VxVM Configuration File. KAPL10706-E Invalid argument. Details An invalid argument was specified.
8. Messages KAPL10709-E An error occurred when creating the file aa...aa. Details An attempt to create a file has failed. aa...aa: File name Action Check the system configuration. KAPL10710-E The existing configuration definition file aa...aa could not be opened. Errno = bb...bb (cc...cc) Details An attempt to open a configuration definition file has failed. aa...aa: Name of the configuration definition file bb...bb: Errno (decimal number) cc...
8. Messages Check the system configuration and then execute the dlmsetconf utility for creating the HDLM driver configuration definition file. KAPL10713-E The aa...aa directory could not be opened. Errno = bb...bb (cc...cc) Details An attempt to open a directory has failed. aa...aa: Directory name bb...bb: Errno (decimal number) cc...cc: Errno string Action Check the system configuration and then retry. KAPL10714-E Could not allocate memory. Errno = aa...aa (bb...bb).
8. Messages Details The specified file name is not the name of a directory. aa...aa: Directory name Action Specify the correct directory. KAPL10717-E The directory specified by the -d option does not exist. Details A non-existing directory was specified. Action Specify the correct directory. KAPL10718-E An error occurred when collecting the catalog message. Errno = aa...aa (bb...bb) Details A message could not be acquired from the message catalog. aa...aa: Errno bb...
8. Messages KAPL10720-W The entered value is invalid. Re-enter [y/n]: Details An incorrect value was entered for the message confirming execution of the dlmvxexclude utility for assisting creation of the VxVM configuration file. Action Enter y or n. For details on the dlmvxexclude utility, see 7.7 The dlmvxexclude Utility for Assisting Creation of the VxVM Configuration File. KAPL10721-E The entered value is invalid. The processing will now stop.
8. Messages Configuration File. KAPL10951-I The LU (LDEV = aa...aa) was successfully deleted. Details The LU was successfully deleted. aa...aa: LDEV name Action None. KAPL10952-W A nexus driver instance that manages a path to the LU does not exist. The LU (LDEV = aa...aa) will not be deleted. FilterPath = bb...bb Details This LU will not be deleted because a path exists without a nexus driver instance for the LU. aa...aa: LDEV name bb...bb: Filter path Action If you want to use the LU, reboot the host.
8. Messages KAPL10954-E Multiple paths have the same property (aa...aa) value for the LU (LDEV = bb...bb) that will be deleted. The LU configuration will not be changed. Details Multiple paths have the same property value for a LU registered in the filter driver information. aa...aa: Property name = property value bb...bb: LDEV name Action Contact your HDLM vendor or the maintenance company if there is a maintenance contract for HDLM.
8. Messages 8.11 KAPL11001 to KAPL12000 KAPL11901-I aa...aa has started. Details The operation has started on the host. aa...aa: Operation (character string) • Get Path Information • Get Option Information • Set Option Information • Clear Data • Get HDLM Manager Status • Get HDLM Driver Status • Get HDLM Alert Driver Status • Get SNMP Trap Information • Set SNMP Trap Information • Set LU Load Balance • Get Path Status Log Information • Get Local Time Action None. KAPL11902-I aa...aa has started.
8. Messages None. KAPL11903-I aa...aa has completed normally. Details The operation has completed normally on the host. aa...aa: Any of the following operations (character string) • Get Path Information • Get Option Information • Set Option Information • Clear Data • Get HDLM Driver Status • Get HDLM Manager Status • Get HDLM Alert Driver Status • Online • Offline • Get SNMP Trap Information • Set SNMP Trap Information • Set LU Load Balance • Get Path Status Log Information • Get Local Time Action None.
8. Messages • Clear Data • Get HDLM Driver Status • Get HDLM Manager Status • Get HDLM Alert Driver Status • Online • Offline • Get SNMP Trap Information • Set SNMP Trap Information • Set LU Load Balance • Get Path Status Log Information bb...bb: Error status returned from API (character string) Action Execute the DLMgetras utility for collecting HDLM error information, and then contact your HDLM vendor or the maintenance company if you have a maintenance contract for HDLM.
8. Messages KAPL11907-I XML reception - aa...aa Details This information is required for resolving problems. aa...aa: XML information (character string) Action None. KAPL11908-I XML transmission - aa...aa Details This information is required for resolving problems. aa...aa: XML information (character string) Action None.
8. Messages 8.12 KAPL13001 to KAPL14000 KAPL13601-W The audit log configuration file does not exist. Restart the HDLM Manager, and execute the "dlnkmgr view -sys -audlog" command and check the setting. Details The audit log configuration file does not exist. Action Restart the HDLM Manager, and execute the "dlnkmgr view -sys -audlog" command, and then specify the desired setting by using the "dlnkmgr set -audlog" command or the "dlnkmgr set -audfac" command as necessary.
8. Messages An internal error occurred during processing to read the audit log configuration file. Action Contact your HDLM vendor or the maintenance company if there is a maintenance contract for HDLM. KAPL13605-W An error occurred during processing to output the audit log configuration file. Details An internal parameter error occurred during output of the audit log data. Action Contact your HDLM vendor or the maintenance company if there is a maintenance contract for HDLM.
8. Messages 8.13 KAPL15001 to KAPL16000 KAPL15010-W The HDLM utility was executed by the user who does not have the authority. Command Line = aa...aa Details aa...aa: Name of the utility that the user executed KAPL15020-I A new LU was successfully added. Command Line = aa...aa Details aa...aa: Name of the utility that the user executed KAPL15021-E An attempt to add a new LU has failed. Command Line = aa...aa Details aa...
8. Messages aa...aa: Name of the utility that the user executed KAPL15026-I Information about exclusion from VxVm management was successfully output. Command Line = aa...aa Details aa...aa: Name of the utility that the user executed KAPL15027-E An attempt to output information about exclusion from VxVM management has failed. Command Line = aa...aa Details aa...aa: Name of the utility that the user executed KAPL15060-I DLMgetras was invoked. Command Line = aa...aa Details aa...
8. Messages dd...dd: Command that the user executed KAPL15104-W aa...aa path(s) were failed to place bb...bb. Command Line = cc...cc Details aa...aa: Number of paths that failed to be placed online or offline bb...bb: Online or Offline(c) cc...cc: Command that the user executed KAPL15105-I Setting up the operating environment succeeded. Command Line = aa...aa Details aa...aa: Command that the user executed KAPL15106-E Setting up the operating environment failed. Command Line = aa...aa Details aa...
8. Messages aa...aa: Command that the user executed KAPL15111-W The HDLM command was started or stopped by the user who does not have the authority. Command Line = aa...aa Details aa...aa: Command that the user executed KAPL15112-I I/O information monitoring function successfully started. Command Line = aa...aa Details aa...aa: Command line that the user executed KAPL15113-E Could not start the I/O information monitoring function. Command Line = aa...aa Details aa...
8. Messages 8.14 Return Codes for Hitachi Command Suite Common Agent Component When an operation requested of HDLM from Global Link Manager terminates abnormally, or terminates normally with a warning, HDLM outputs one of the return codes described below. 1002 Details There is no path on which the operation can be performed. Action Refresh the host information, check the path status, and then perform the operation again. 1003 Details No path was detected.
8. Messages Action Click Refresh to update the host information, check the path status, and then retry the offline operation. 1015 Details The Offline path cannot be placed Online. Action Remove the error in the path, and then retry. 1016 Details The target path(s) are already Online. Action Update the host information, and then check the path status. 1017 Details The target path(s) are already Offline(C). Action Update the host information, and then check the path status.
8. Messages maintenance contract for HDLM. For details on the DLMgetras utility, see 7.2 The DLMgetras Utility for Collecting HDLM Error Information. 1025 Details A parameter value is invalid. Action Refresh the host information, and then perform the operation again. If the same error occurs again, execute the DLMgetras utility for collecting HDLM error information, and then contact your HDLM vendor or maintenance company if you have a maintenance contract for HDLM.
8. Messages An attempt to recover the path failed because of insufficient memory. Online processing will be stopped if online processing is running for multiple paths. Action Re-execute. If the error occurs again, confirm that the host memory is sufficient. 1030 Details An attempt to recover the path failed because the LU for the path to be put online has been reserved by another path. Action Use the format command to confirm that the LU for the path to be put online is not reserved by another host.
8. Messages 1032 Details An attempt to recover the path failed because the correct disk label has not been set in the LU for the path to be put online. Action Use the format command to set the VTOC label or EFI label in the LU for the relevant path, and then re-execute the command. 1033 Details An attempt to acquire the HDLM version information failed. Action Re-execute the command.
8. Messages utility for collecting HDLM error information to collect the error information, and then contact your HDLM vendor or maintenance company if there is a maintenance contract for HDLM. For details on the DLMgetras utility, see 7.2 The DLMgetras Utility for Collecting HDLM Error Information. 1036 Details An attempt to acquire information about the HDLM version or SP version has failed. Action Re-execute the command.
8. Messages Action Refresh the host information, check the path status, and then perform the operation again. 1045 Details A parameter is invalid. Action Refresh the host information, and then perform the operation again. If the same error occurs even again, execute the DLMgetras utility for collecting HDLM error information, and then contact your HDLM vendor or maintenance company if you have a maintenance contract for HDLM. For details on the DLMgetras utility, see 7.
Appendixes A. Sun Cluster 3.2 Commands B. Functional Differences Between Versions of HDLM C.
A. Sun Cluster 3.2 Commands A. Sun Cluster 3.2 Commands Table A-1: Correspondence Between Sun Cluster 3.1 Commands and Sun Cluster 3.2 Commands shows the correspondence between Sun Cluster 3.1 commands and Sun Cluster 3.2 commands. For details about the command options, see the documentation for Sun Cluster 3.2.# # Note that there are no differences between Sun Cluster 3.2 commands and Oracle Solaris Cluster 3.3 commands. Table A-1: Correspondence Between Sun Cluster 3.1 Commands and Sun Cluster 3.
A. Sun Cluster 3.2 Commands Commands for Sun Cluster 3.1 Commands for Sun Cluster 3.2 scsetup clsetup scshutdown -g0 -y cluster shutdown -g0 -y scstat -q clquorum status scstat -D cldevicegroup status scswitch -F -D device-group-name cldevicegroup offline device-group-name scswitch -S -h node-name clnode evacuate node-name #1 Execute this command in step 13 in (10) Set Up Sun Cluster of 3.5.1 Preparations for a New Installation of HDLM and in step 14 in (2) Remove Sun Cluster Settings in 3.
B. Functional Differences Between Versions of HDLM B. Functional Differences Between Versions of HDLM This section explains the functional differences between newer and older versions of HDLM. B.1 Functional Differences Between Version 6.1 or Later and Versions Earlier Than 6.1 In a new installation of HDLM, the default load balancing algorithm has been changed from the Round Robin algorithm to the Extended Least I/Os algorithm. B.2 Functional Differences Between Version 6.
C. Glossary C. Glossary This glossary explains the terms used in this manual. A automatic failback A function for checking the status of paths at regular intervals, and automatically placing the status of a path recovered from an error into the Online status. If a path with an Offline(E) or Online(E) status recovers from an error, an automatic failback will place the path Online.
C. Glossary D Dev A logical division of an LU that HDLM controls and operates. A Dev is equivalent to a slice or a partition in Solaris. In Solaris, each LU can be divided into more than one Dev. Each Dev has a Dev number. Depending on the access method, Devs are classified into the following two types: block devices and character devices. (See also: Dev number) Dev number In UNIX, a device number usually indicates a combination of the major device number and minor device number.
C. Glossary failover A function for switching to another normal path if there is an error in the current access path, thereby enabling the system to continue to operate. FCode A program written by the interpreter language that is not dependent on the machine. The HBA that can connect to a boot disk might incorporate FCode for use. H HBA (Host Bus Adapter) A device that is an interface between hosts and external devices.
C. Glossary A host device name is assigned using the cUtXdY format. The variables used in cUtXdY indicate the following: U: Controller number reserved by HDLM X: Target ID or World Wide Name of the sd or ssd device for the HDLM device Y: LUN of the sd or ssd device for the HDLM device (See also: host device) host LU An LU that a host recognizes. Each host LU has a host LU number. (See also: LU, host LU number, host device) host LU number A number assigned to a host LU.
C. Glossary • The Extended Least Blocks algorithm local boot disk environment An environment in which a boot disk is located on the host. logical device file Device files under /dev/dsk and /dev/rdsk. Note that the term logical device used here is not an LDEV. LU (Logical Unit) A logical unit that is a logical volume defined in the storage system, and with which the host can perform input or output operations. (See also: host LU) N node A server in a cluster.
C. Glossary (See also: owner path, non-owner controller) owner path A path that passes through an owner controller for an LU in the Thunder 9500V Series, or Hitachi AMS/WMS series is in use. A path that passes through a controller other than an owner controller is called a non-owner path. When the Lightning 9900 Series, Lightning 9900V Series, Hitachi USP, Universal Storage Platform V/VM, Virtual Storage Platform, Hitachi AMS2000 Series or Hitachi SMS is in use, all paths are owner paths.
C. Glossary R reservation The reservation function enables a host to declare that it has exclusive rights to use a particular LU, and prevents other hosts from accessing that LU. Access permission for an LU that has been reserved is given only to the host that issued the reservation, so the LU cannot be accessed from multiple paths (coming from multiple hosts) simultaneously. However, because of this, load balancing is not possible.
Index Symbols /etc/system file setup 105 /kernel/drv/sd.
Index cUtXdYsZ 14 D default and recommended values 377 Dev 7, 8, 618 Dev number 618 diagram conventions vii directory to which collected information is output 438 disk group creating 209 disk label 100 Displayed program information 395 displaying information 393 LU information 414 path information 393, 401 path information (by selecting a display item) 393 program information 393, 395 summary of path information 393 displays I/O information at a specified interval 429 Distributing a Load Using Load Balanc
Index path error 347 HBA 8 HBA (Host Bus Adapter) 619 hbsa.tar.
Index INSTALLATION_SETTINGS 468 installhdlm 468 installhdlm.log 475 installing HDLM 110, 118 HDLM with Sun Cluster 118 HDLM without Sun Cluster 110 installing HDLM LDoms environment 136 installux.
Index O obtaining path information 348 program information 352 offline 23, 363 offline(C) status 27 offline(E) status 27 online 23, 370 online status 27 online(E) status 27 operating environment 376 setting 376 view [view operation] 399 operation name 478 operations 356 displaying format 360 of the HDLM command 356 organization of this manual i OS 56 overview command 356 HDLM 1 owner controller 18, 621 owner path 18, 622 P P 8 path 7, 622 checking the configuration 187 path error handling 347 path health
Index HDLM manager 283 in Solaris 10 270 statistics 358 clearing to the initial value 358 status transitions 28 stopping HDLM manager 283 storage systems 2 Sun Cluster setup 107 SVM setting up 239 VCS setting up 243 SVM settings Removing 252 switching kernel mode 104 symbol conventions vii syntax conventions viii syslog 39 system configuration 7 T taking actions for hardware error 349 target ID 412 TB meaning ix trace file 39 trace file size 380 trace level values 379 types of collected logs 39 types of e
