HP VMA SAN Gateway Installation and User Guide Document Number: AM456-9007B Rev 02 May 2012
LEGAL NOTICES Copyright 2011, 2012 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice. The only warranties for HP products and services are set forth in the express warranty statements accompanying such products and services. Nothing herein should be construed as constituting an additional warranty. HP shall not be liable for technical or editorial errors or omissions contained herein. Copyright © 2010-2012 Violin Memory, Inc. All rights reserved.
IN NO EVENT SHALL VIOLIN (OR ITS LICENSORS OR ANY OTHER THIRD PARTY IDENTIFIED HEREIN) BE LIABLE CONCERNING ANY USE OF THIS DOCUMENT, REGARDLESS OF THE FORM OF ANY CLAIM OR ACTION (WHETHER IN CONTRACT, NEGLIGENCE, STRICT LIABILITY OR OTHERWISE), FOR ANY DIRECT, INDIRECT, PUNITIVE, INCIDENTAL, RELIANCE, SPECIAL, EXEMPLARY OR CONSEQUENTIAL DAMAGES, INCLUDING WITHOUT LIMITATION, ANY LOSS OF DATA, LOSS OR INTERRUPTION OF USE, COST OF PROCURING SUBSTITUTE TECHNOLOGIES, GOODS OR SERVICES, OR LOSS OF BUSINESS, REV
HP VMA SAN Gateway Installation and User Guide AM456-9007B Rev 02
Table of Contents List of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 CHAPTER 1. HP VMA SAN Gateway Overview . . . . . . . . . . . . . . . . . . . .
Deploying vSHARE in a High Availability Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 CHAPTER 5. Operating vSHARE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 Verifying Connections and Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
VMA Utilities Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .214 APPENDIX D:. Standard System Configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 Single Gateway with 1–2 3000-Series Arrays, Non-Redundant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Taiwan BSMI Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271 Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
List of Figures Figure 1.1 HP VMA SAN Gateway System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Figure 2.1 HP VMA SAN Gateway Deployment Flowchart. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Figure 2.2 HP VMA SAN Gateway Back Pane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Figure 2.
Figure D.3 Multiple redundant gateways with 1 to 2 VMA 3200 Series arrays, each highly available. . . . . . . . . . . . . . 237 Figure F.1 Australian C-Tick Label . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Preface This document describes how to install, configure, and manage the VMA-series SAN Gateway. Document Conventions The table lists the icons that indicate special kinds of information in this guide, with an example of each icon. Icon Usage Caution: A Caution icon emphasizes information that helps you avoid improperly configuring the system. Note: A Note icon draws your attention to significant information. A Web icon indicates tasks that can be performed in the VMA Web Interface.
Text Formatting The following table summarizes the font conventions used in this guide. Font Usage Example Bold Object labels in the VMA Web Interface, such as column headings. Italic Provides emphasis and identifies variables and document titles. Monospace Commands, prompts, parameters, parameter options, file names, and command examples. User Name login: Command Syntax Conventions The following conventions define the command syntax in this guide.
Notice Description FCC Class A Compliance "This device complies with Part 15 of the FCC Rules. Operation is subject to the following two conditions: (1) This device may not cause harmful interference, and (2) this device must accept any interference received, including interference that may cause undesired operation." This equipment has been tested and found to comply with the limits for a Class A digital device, pursuant to Part 15 of the FCC Rules.
Getting Help Contacting HP Before you contact HP Be sure to have the following information available before you contact HP: • Technical support registration number (if applicable). • Product serial number • Product model name and number • Product identification number • Applicable error message • Add-on boards or hardware • Third-party hardware or software • Operating system type and revision level.
After registering, you will receive email notification of product improvements, new driver versions, firmware updates, and other product resources. Reference Documents • HP VMA-series VMA Array Installation and Service Guide Documentation feedback HP welcomes your feedback. To make comments and suggestions about product documentation, send a message to docsfeedback@hp.com.
AM456-9007B Rev 02 HP VMA SAN Gateway Installation and User Guide
CHAPTER 1 HP VMA SAN Gateway Overview Introduction to the HP VMA SAN Gateway The HP VMA-series storage products are designed for I/O intensive server applications such as high performance databases. Flash memory offers tremendous performance advantages over traditional hard disk drive systems enabling data centers to consolidate hard disk drives/spindles and reduce the number of servers/ CPUs and associated software licenses.
servers to share the high performance flash VMA Arrays. Sharing reduces costs and allows hosts to cooperate. The HP VMA Arrays provide high capacity flash storage with RAID protection using a special flash-optimized algorithm (vRAID). This algorithm minimizes latency and provides data integrity very efficiently. HP VMA SAN Gateways support vSHARE, a LUN management solution for SAN block storage. • HP VMA SAN Gateways with vSHARE provide for flexible LUN provisioning, LUN reservations, and full-speed LUNs.
provided on the HP VMA SAN Gateways and VMA Arrays to enable the systems to be configured and monitored as a single system. SNMP and callhome e-mail alerts are sent in the event of hardware failures or other critical events. The internal architecture of the vSHARE system is designed for minimum latency and multipathing the HP VMA SAN Gateway and the VMA Arrays to improve availability.
any number of flashlets. There is no need for flashlets to be contiguous and hence LUNs can be packed very well with no wasted space. • Full-Speed LUNs: Each flashlet and hence each LUN is striped across all of the VIMMs in a flash VMA Array. By doing this, the performance of the LUN is only limited by the performance of the whole VMA Array. • Shared LUNs: Each LUN can be accessed using Fibre Channel and can be shared by many servers in a cluster.
vCLUSTER management software enables you to manage the HP VMA SAN Gateways and VMA Arrays in the cluster as a single system using either the CLI or the VMA Web Interface.
AM456-9007B Rev 02 HP VMA SAN Gateway Installation and User Guide
CHAPTER 2 Installing and Upgrading the HP VMA SAN Gateway Deployment Overview VMA-series SAN Gateway deployment occurs in five distinct phases. Each phase of the deployment process must be completed before you can proceed to the next phase of the deployment. • Phase 1: HP VMA SAN Gateway Installation includes the unpacking, racking, and cabling of HP VMA SAN Gateway hardware.
The diagram below shows the five phases of HP VMA SAN Gateway configuration and the discrete steps required within each phase of an implementation. Figure 2.
• Unpacking HP VMA SAN Gateway Hardware Components: The first step in the hardware installation phase is to unpack the HP VMA SAN Gateway shipping box and confirm that all components are present. • Rack-mounting the HP VMA SAN Gateway Chassis: The second step is to mount the HP VMA SAN Gateway chassis in the rack. • Connecting to Memory Arrays with PCIe Cables: The third step is to connect each HP VMA SAN Gateway to one or more Memory Arrays using PCIe cables.
Phase 4, HP VMA SAN Gateway Configuration, includes the following procedures: • Configuring the Master HP VMA SAN Gateway: The first step is to configure the master node of the HP VMA SAN Gateway cluster and define the global cluster parameters, which are inherited by each node subsequently added to cluster. • Configuring additional HP VMA SAN Gateways: Connect to each remaining HP VMA SAN Gateway node to set its hostname and local system parameters.
HP VMA SAN Gateways running vSHARE can be deployed in a High Availability (HA) configuration, where two HP VMA SAN Gateways provide active-active access to a Memory Array. Data is accessible via both HP VMA SAN Gateways. If one of the Gateways fails, or a link along one of the paths to the Array goes down, data is available seamlessly via the other Gateway. All vSHARE configuration procedures are described in CHAPTER 4, Configuring vSHARE on page 65.
Required Installation Tools The following tools are required during the installation process: • Null Modem Cable or KVM (Keyboard/Video/Mouse) access to console. • Network requirements: The HP VMA SAN Gateway uses Gbit Ethernet for management and cluster network connections. Front View The front panel has Power and Reset buttons that allow you to start, stop, or reset the HP VMA SAN Gateway. Caution: Using the Power button to turn off the system power removes the main power but keeps standby power.
Rear View The rear panel provides the interfaces for input and output devices, power supplies, cluster management, network traffic, and one or more Memory Arrays. Figure 2.2 HP VMA SAN Gateway Back Pane Figure 2.
Serial Console Port The serial port may be used for CLI access. Gbit Ethernet Ports Two Gbit Ethernet ports provide management access to the HP VMA SAN Gateway cluster. Table 2.1 Rear Panel Interfaces Rack-Mounting HP VMA SAN Gateways The next step in the hardware installation of an HP VMA SAN Gateway is to rackmount the HP VMA SAN Gateway chassis. The HP VMA SAN Gateway may be racked as you would a normal server.
In typical configurations, either a single stand-alone or two redundant paired VMA SAN gateways are connected to one or two VMA Arrays using PCIe cables. Figure 2.3 HP VMA SAN Gateway and Memory Array Back Panels If one HP VMA SAN Gateway were to fail, the other HP VMA SAN Gateway would immediately become the primary owner of that Memory Array. Connecting Power The next step in the hardware installation of a HP VMA SAN Gateway is to connect to a power source.
The HP VMA SAN Gateway chassis has two single-phase power supplies installed prior to shipping. The IEC-C14 male receptacles accept two IEC-C13 female connectors. Figure 2.4 HP VMA SAN Gateway Power Supplies Each HP VMA SAN Gateway features redundant power supplies. Connect the AC power cord to either power receptacle on the HP VMA SAN Gateway or both.
A typical HP VMA SAN Gateway system will have three management interfaces. Figure 2.5 Network Interfaces • 1-2 & 4-5: Management Ethernet Ports (Eth1 is on right, Eth4 is on left) • 3: iLO Management Port In this example, 8GB Fibre SAN ports are the primary ports for failover management. These interfaces may used as individual ports configured for different networks or they may be bonded together to run on the same subnet for increased throughput and reliability.
The settings for the modem are as follows: Port Setting Name Value Bits per second 9600 Data bits 8 Parity None Stopbits 1 Flowcontrol None Table 2.2 Modem Settings KVM Connection If a null modem cable is not available, a KVM (USB keyboard, VGA video monitor, and USB mouse) may be used to directly plug into the public interface for configuration. Connecting Data Ports to Network Switches In a typical configuration, two HP VMA SAN Gateways are connected to two switches.
throughput and availability. All the Ethernet or Fibre Channel ports on the HP VMA SAN Gateway may be connected to the same switch. Additional FC ports can be added to an HP VMA SAN Gateway by using HP supplied HBAs (part number AJ764A), each of which supports two data ports. The maximum number of HBAs supported is four.
AM456-9007B Rev 02 HP VMA SAN Gateway Installation and User Guide
CHAPTER 3 Configuring and Managing Clusters vCLUSTER Overview Cluster management software provides for the central management of multiple HP VMA SAN Gateways and VMA Arrays as nodes within a single system. Using vCLUSTER, multiple HP VMA SAN Gateways and VMA Arrays may be managed from the CLI or VMA Web Interface.
for each node (HP VMA SAN Gateway in the cluster), the cluster management VIP address, and the VMA Array IP address. • Configuring Network Connections: The second step of Phase 2 is to configure support on the cluster and network switches for bonded interfaces and VLANs. Configuring Network Settings The first step of Phase 2, System Network Configuration, is to set up and define the network settings for the HP VMA SAN Gateway cluster.
Defining Cluster Name and Host Names Names for the cluster and each HP VMA SAN Gateway may be set via the serial console port on the Master Gateway or via the cluster management VIP address: • Cluster name to identify the VMA-series SAN Gateway. • Host names for each HP VMA SAN Gateway in the cluster. Typically, these would be given obvious names such as “vmg-n1” and “vmg-n2” or something similar.
To separate cluster network traffic from cluster management traffic, you can configure virtual LAN (VLAN). Network switches must also be configured to provide a private VLAN if it will be used for the cluster network traffic. For best performance and minimal impact on the customer network, HP recommends configuring a private network for use by the cluster through VLAN tagging.
Logging into the HP VMA SAN Gateway Once you have connected to the KVM console or serial port, you are prompted to log in. Enter the username admin (bold shows user input): tc500-1973fa login: admin The admin user does not have a default password. Once logged in, you are prompted whether to use the configuration wizard, as shown below: Unauthorized Access Prohibited. Usage of the HP VMA SAN Gateway is subject to the HP License agreement which is included under this product's Web Interface Help section.
The first HP VMA SAN Gateway added to the HP VMA SAN Gateway cluster is the master node. The master node is the default gateway for all cluster management tasks. The cluster management VIP directs all cluster management traffic to the public interface—generally eth1 (or eth0 if a bonded interface is used)—on the master node. Should the master node fail, the standby node replaces it and becomes the master node, and cluster management traffic is automatically redirected to the new master node.
DNS server name(s) Specify a primary and, optionally, a secondary DNS server. Domain name The name of your local domain. Hostname A local hostname for the HP VMA SAN Gateway. This should be a unique host name for each gateway. Admin password Setting a password is highly recommended. Timezone Set the clock timezone by specifying the zone and subzones. (Use "?" to display zone names interactively.) NTP server Optional and recommended.
Cluster name The name of the cluster used by DNS. Cluster management IP address & netmask The virtual IP address (VIP) and netmask for management of the cluster. The address and netmask are assigned to the master node of the cluster, and redirected as needed whenever another node becomes the master. Table 3.1 HP VMA SAN Gateway Settings The fastest and easiest way to configure the master node of a HP VMA SAN Gateway cluster is to use the configuration wizard.
previously configured). After the last step, the wizard repeats your settings and lets you return to any step if you want to make more changes. To rerun the Configuration Wizard, use the command configuration jumpstart from the CLI(confg) prompt. To configure the Master HP VMA SAN Gateway using the configuration wizard: 1. Connect and log in as admin. The configuration wizard prompts you to use it for the initial configuration of the HP VMA SAN Gateway.
When you complete the list of questions, the configuration wizard displays your responses and prompts you to accept or reject the settings: To change an answer, enter the step number to return to. Otherwise hit to save changes and exit. Choice: █ 5. To save your responses, and exit the configuration wizard press the Enter key. The configuration wizard displays the HP VMA SAN Gateway cluster settings and exits to the CLI. You have entered the following information: 1. 2. 3. 4. 5.
24. Cluster management IP address: 10.3.19.61 25. Cluster management IP netmask: 255.255.252.0 To change an answer, enter the step number to return to. Otherwise hit to save changes and exit. Choice: █ 6. To return to the configuration wizard after exiting, enter the command configuration jump-start in the CLI. This command can be used anytime to change the configuration settings. Getting Help During Configuration Additional usage advice is provided at the beginning of the wizard.
Type "?" for help with the current step. For example, when setting the timezone you can type "?" to display the possible values, along with general help: ... Step 10: Set clock timezone? [yes] Step 11: Zone to use? [UTC] America Step 12: Sub-zone 1 to use? [Caribbean] ? Caribbean Central North South Press '?' for help, Ctrl+B to go back to the previous step. Default value is in square brackets: press Enter to accept it. Press Ctrl+R to clear default to enter empty string.
cluster settings and enters the CLI. If you changed the master node’s hostname or the cluster’s name, the command prompt will show their new names. For example: Choice: Configuration changes saved. Cluster ID: Cluster name: Management IP: Cluster master IF: Cluster node count: Local name: Local role: 00500-0008-0105 cluster1 10.3.19.61/22 eth0 2 gateway01 master Local state: Master address: Master state: online 10.3.7.84 (ext) 169.254.72.
• Standby node: The second node added to the cluster is automatically designated as the standby node. Should the master node fail, the standby node replaces it and becomes the master node, and cluster management traffic is automatically redirected to the new master node. • Normal nodes: Each additional HP VMA SAN Gateway added to the cluster is defined as a normal node. During the course of vCLUSTER operation, a normal node can assume the role of standby or master.
A typical configuration of a normal node HP VMA SAN Gateway using the configuration wizard might look like the following example: Step Step Step Step Step Step Step 1: 2: 3: 4: 5: 6: 7: Configure as master/stand-alone? [yes] no Public interface name? [eth0] Public interface IP address? [10.1.9.101] Public interface netmask? [255.255.252.
Cluster ID: Cluster name: Management IP: Cluster master IF: Cluster node count: Local name: Local role: Local state: Master address: Master state: 00500-0008-0105 cluster1 10.1.9.100/22 eth0 2 gateway02 standby online 10.1.9.102 (ext) 169.254.9.
Note: VLAN and interface bonding configuration are not part of the configuration wizard and must be configured using the CLI. If you specified either a VLAN ID or the bonded interface (eth0) for the “Public interface name” or “Cluster interface name” questions, then additional steps are required. See Network Bond Commands on page 150 for information on how to set up bonding. See VLAN Commands on page 151for information on how to configure a VLAN.
Cluster Management VIP A typical HP VMA SAN Gateway cluster consists of one or more HP VMA SAN Gateways, one of which is designated as the master node and one more which is designated as the standby node. All other nodes in the vCLUSTER are considered “normal” nodes. The cluster management VIP, a virtual IP address, provides access to the cluster, enabling administrators to manage and monitor all nodes of the management cluster.
In the VMA Web Interface, the Gateways page lists all of the HP VMA SAN Gateways (nodes) in the cluster, their status, and role (master node, standby node, or normal node). Using controls in the Gateways page, you may view additional information, stop the HP VMA SAN Gateway (remove it from the cluster), or reboot it. If you stop a HP VMA SAN Gateway in this page, it can only be restored to service by using the CLI. For more information, see Gateways Page on page 188.
An asterisk after a node’s ID number designates the local node where you issued the command. Show Cluster Global Command Without the brief option, the show cluster global command displays additional information about each HP VMA SAN Gateway in the cluster. Instead of using an asterisk to designate the local node ID, the label <--- (local node) is used to identify the local node. gateway02 [cluster1: standby] # show cluster global Cluster ID: 00500-0008-0175 Cluster name: cluster1 Management IP: 10.3.
Hostname: vmg-vc1-n1 Node Role: standby Node State: online Node internal address: 169.254.151.82, port: 56345 Node external address: 10.3.19.101 Recv. Heartbeats from: 1 Send Heartbeats to: 1 gateway02 [cluster1: standby] # █ Configuration File Management Changes made to the configuration of an HP VMA SAN Gateway cluster take effect immediately, but those changes can be lost if they are not saved to a configuration file.
Note that an asterisk (*) appears before the command prompt to remind you that changes have been made to the HP VMA SAN Gateway cluster, which have not yet been saved.
The no-switch option can be used to save an interim configuration which you will finish later. Saving with the no-switch option does not change a configuration’s unsaved status (indicated by an asterisk in the CLI), because recent changes are still not stored in the active configuration file. Saving the Configuration File to a Remote Server You can create a copy of the active Gateway configuration file and store it on a remote server for recovery purposes.
• hostname is the name of the remote server • path is the location in which to store the file on the remote server 4. Check the remote server to make sure that the file was uploaded successfully. 5. Repeat steps 1-4 for the Standby node, using its configuration file. To download the configuration file from the remote server: Use the fetch command if you need to recover a configuration file from the remote server.
Show Configuration The show configuration commands enable you to view the commands required to bring a fresh system up to match a given configuration. • show configuration • show configuration running • show configuration files • show configuration full • show configuration running full • show configuration unsaved The output of each of these commands begins with a short header that contains the name and version number of the configuration in a comment.
The show configuration running full command displays all commands, including any default settings, to reproduce the state of the current running configuration. show configuration unsaved The show configuration unsaved command displays commands that have been configured, but not yet saved in the configuration file. User passwords cannot be recovered with a show configuration command (or similar command) unless the passwords are encrypted.
An active configuration file may not be deleted or renamed, nor may it be the target of a move or copy. It may be the source of a copy, in which case the original remains active. In the VMA Web Interface, select one or more configuration files in the Current Configurations section of the Cluster Administration page and click the Delete button. For more information, see Cluster Administration Page on page 193.
Software Upgrades Upgrading the HP VMA SAN Gateway software requires that you obtain the latest software release, run a command to distribute and install the software on each HP VMA SAN Gateway, and then reboot the HP VMA SAN Gateway cluster. There are three options for upgrading the HP VMA SAN Gateway cluster: immediate, rolling, and staged. For immediate upgrades, the HP VMA SAN Gateways are upgraded simultaneously and then restarted.
CHAPTER 4 Configuring vSHARE HP VMA SAN Gateways running vSHARE can be deployed in a High Availability configuration, where a pair of HP VMA SAN Gateways provide redundant access to a VMA Array. See Deploying vSHARE in a High Availability Configuration on page 82. Understanding vSHARE vSHARE is a solution for block storage management.
• Targets: A target identifies the I/O port on the HP VMA SAN Gateway through which the initiator may access the LUNs stored in a container. Access to the LUNs in a container may be restricted by target port. vSHARE System Architecture vSHARE is ideal for applications that require rapid access to and processing of large amounts of data including enterprise databases, data warehouses, application servers, and Cloud-scale analytics.
vSHARE Configuration Overview vSHARE configuration must be understood within the context of an HP VMA SAN Gateway deployment. vSHARE configuration occurs during the fifth phase of an HP VMA SAN Gateway deployment. During the first three phases (1) HP VMA SAN Gateway system hardware is assembled and cabled and the network is defined; (2) one or more VMA Arrays is configured; and (3) the HP VMA SAN Gateway cluster is configured.
administrators to define sophisticated rules for controlling access to the LUNs by initiator groups, initiators, or target ports.
The vSHARE flowchart shows the steps required to configure the HP VMA SAN Gateway for vSHARE block storage within the larger context of configuring the HP VMA SAN Gateway. PHASE 5: vSHARE Configuring Containers Configuring Initiator Groups and Initiators Configuring Target Ports 1 2 3 Exporting LUNs Creating LUNs 4 5 Figure 4.3 vSHARE Deployment Flowchart In general, vSHARE configuration is a five-step process consisting of the following procedures.
• Step 5: Exporting LUNs: The LUNs must be exported to the initiator groups or initiators via target ports. Only those initiator groups or initiators to which the LUN is exported may access the LUN. Access may be further restricted to a specific target port. Step 2, Configuring the Target Ports, and Step 3, Configuring Initiator Groups and Initiators, are, strictly speaking, optional steps.
Command Parameters device The parameter indicates the array to be initialized and is a string of the form: ataVIOLIN_MEMORY_ARRAY_xxxxxxxxxxxxxx type The type parameter defines media as block (vSHARE). name The name parameter is used to name the partition. The default is to use the last 14 characters of the VMA Array serial number. To use a different name, add this parameter followed by a custom name of choice. The name can be up to 14 alphanumeric characters in length.
A message confirms that the VMA Array was successfully initialized. media init completed successfully SM076 [NJ-Lab-vSHARE00: standby] (config) # Viewing Containers Once you have initialized the VMA Array, you can see the container created using the show containers command. 1. Log on as an administrator into the HP VMA SAN Gateway attached to Port 1 of the VMA Array. Before a block media device can be used for block storage, it must be enabled.
Configuring Target Ports The second step in configuring a vSHARE HP VMA SAN Gateway is to configure the target ports. The target ports control access to a particular LUN, which may be useful for both security and bandwidth management. In a vSHARE system, the VMA Arrays are the targets and the host system (for example, one or more database servers or application servers) are the initiators. The storage systems have storage target devices, LUNs, which the hosts access through the HP VMA SAN Gateway.
Syntax Description [node ] Displays targets on node ID. [hostname ] Displays targets on hostname. [protocol fc] Displays targets using protocols FC. [id ] Displays a specific target. [sessions] Displays targets including sessions. [detail] Displays in detail. Table 4.
2. To add additional initiators to an initiator group, use the igroup addto command. (config) # igroup addto name initiators wwn.21:01:00:1b:32:ba:18:65 You can also add two Fibre Channel initiators to the same initiator group.. (config) # igroup addto name initiators wwn.21:00:00:1b:32:9a:18:65 wwn.21:01:00:1b:32:ba:18:65 Fibre Channel initiator identifiers (WWN) are fixed by the HBA port.
Syntax Description name The name is alphanumeric only. initiator_name The initiator_name must be prefixed with "wwn." (FC initiators). Table 4.2 igroup create Command Examples The following examples demonstrate how the igroup create and igroup addto commands may be used to manage initiator groups and initiators. The command adds the three initiators to an initiator group named FINANCE: igroup create name FINANCE initiators wwn.20:20:20:20:20:20:20:20 wwn.20:20:20:20:20:20:20:21 wwn.
inside a VMA Array. In many systems, a LUN may be referred to as a volume or logical unit. On the host system (initiator), the LUN appears as a local SCSI disk. The host may format and partition the LUN. The target, or storage system, sees the contents of the LUN as a set of blocks of arbitrary data. Every LUN is created within a container. A container is an addressable partition of one or more VMA Arrays.
Typically, fewer than 100 LUNs would be created within a container; however, the creation of up to 1024 LUNs is supported per container. lun create Command The lun create command enables you to create a LUN within a specified storage container. Syntax [no] lun create container name size [ | equal] [quantity ] [nozero] [readonly] [startnum ] [blksize 512 | 4096] [offline] Syntax Description The table shows syntax for LUN commands.
blksize Defines the logical block or sector size for the created LUNs; the default is 512, can be set to 512 or 4096. Not all systems can handle 4096byte sizes. Use 512 if you are in doubt. There is another word here I can’t read. offline Creates the LUNs in offline mode. Table 4.3 lun create Command lun set Command The lun set command enables you to set a LUN as read-only. The lun set command is the only command usable after a LUN has been created except the no lun create … readonly command.
Exporting LUNs The fifth step in configuring the vSHARE HP VMA SAN Gateway is to export the LUNs to an initiator. The LUN must be exported before it may be accessed over block protocols. Access to LUNs may be restricted to a specific initiator group, initiator, or target port when the LUN is exported using the lun export command. • The optional [igroup | initiator] parameter identifies the initiator group or initiator that may access the LUN.
Syntax Description lun_name The lun_name could be wildcard specified, such as FINANCE* container name lunid Designates the LUN ID assigned to an exported LUN. By default, LUN IDs are automatically assigned by vSHARE. User-defined LUN IDs may be specified to track specific LUNs. igroup The name of the igroup. If no igroup or initiator is specified, all initiators will have access to the LUN. initiator WWN (Fibre Channel) name of the initiator.
Optimizing Connectivity to Storage Arrays for Windows Depending on the Windows operating system, client machines using vSHARE should use the Microsoft® Multipath I/O (MPIO) driver. Install the appropriate driver for your operating system: Windows Server 2008 R2 and above Install the Microsoft MPIO driver. MPIO is not installed by default. Use the following link for installation instructions. Installing and Configuring MPIO for Windows Server 2008 R2: http://technet.microsoft.com/en-us/library/ee619752(WS.
Port A x4 PCIe HP VMA Memory Array Shared Partition vSHARE Metadata Memory Gateway 1 vSHARE Metadata Figure 4.4 shows an illustration of a vSHARE HA configuration. Port B Memory Gateway 2 x4 PCIe Figure 4.4 vSHARE High Availability Configuration In this configuration, two HP VMA SAN Gateways are connected to a VMA Array.
3. From the HP VMA SAN Gateway CLI on the cluster master, enter Config mode. > enable # configure terminal (config) # 4. Initialize the HP VMA Array for vSHARE (block) storage. (config) # media init device ata-_ type block name Strad201 force This command generates the container Strad201 and creates three partitions, vtmsa1, vtmsa2, and vtmsa3.
Configuring Interface Bonding In a High Availability vSHARE configuration, management traffic and cluster traffic must both share the same physical links. The Gigabit interfaces on each HP VMA SAN Gateway, eth1 and eth2, should be configured as a bonded interface, eth0. A bonded network interface provides port/cable redundancy. The following is an example of configuring simple round-robin balance mode for the bonded interface, which does not require any switch changes.
vian HP VMA SAN Gateway 2). To a host server, each of these paths appears as a separate device (for example, /dev/sdb, /dev/sdc, and so on.) Memory Gateway 1 HBA-A1 Port A Port B HP VMA memory array Memory Gateway 2 x4 PCIe x4 PCIe HBA-A1 HBA-A2 HBA-A2 vSHARE Client /dev/sdc /dev/sdd /dev/sdb /dev/sde /dev/mapper/mpath1: /dev/sdb /dev/sdc /dev/sdd /dev/sde Figure 4.
In an HA configuration, where there are two HP VMA SAN Gateways with access to the HP VMA Array, you can perform a staged or non-disruptive software upgrade, which results in no loss in data availability. During a staged software upgrade, the cluster is split into two clusters, A and B, with half the HP VMA SAN Gateways in Cluster A and half in Cluster B. The HP VMA SAN Gateway(s) in Cluster B are upgraded first, followed by those in Cluster A.
When the LUN has been taken out of HA mode, the show luns command displays the status of the LUN as Single and indicates the port from which it is online.
CHAPTER 5 Operating vSHARE Verifying Connections and Performance Verifying Target Port Activity and Status The show targets and show interfaces commands enable you to verify the activity and status of target ports. Viewing Port Activity show targets [node ] [hostname ] [protocol fc] [id ] [sessions] [detail] The show targets command lists all of the target ports and their addresses (wwn), filtered by various options.
The sessions option displays the number of SCSI sessions active on a target and the initiator and the assigned LUN ID for each session. SM075 [NJ-Lab-vSHARE00: master] (config) # show targets hostname SM076 sessions Node Hostname Protocol Target Enab Address protocol fc node 3 ---- --------------- -------- ---------- ---- --------------------------------3 SM076 fc hba-a1 yes wwn.21:00:00:1b:32:8f:24:5d Connected Sessions 3 SM076 fc : : : : : : wwn.21:00:00:1b:32:82:1e:50 wwn.
For detailed information about this command, see Show Targets Command on page 104. Viewing Interface Statuses show interfaces [] [configured | brief] The show interfaces command may be used to view information about a specific interface or, if one is not named, all interfaces. By default, detailed information about the interface and its runtime state is given.
Viewing Containers The show containers command returns information about the storage containers on an HP VMA SAN Gateway cluster including the node, container name, the number of LUNs, the space allocated, and the space free.
health Display status information and lifetime remaining for media devices. For more information on the health option, see Show Media Health Commands on page 98. id Display information for a specified media device or all media (id all). Table 5.1 Show Media Options The show commands described in this section can be used in Config or Enable mode. All other media management commands require Config mode. (See Command Modes on page 117.
show media all The show media all command displays all media devices installed in an HP VMA SAN Gateway. This command shows a summary line for each device location, giving the size and status (online/offline) of the installed media device. Location Media ID Model Size Status -----------------------------------------------------------------------unknown-00 Fender1002 VMA Array 6895.75G online unknown-01 Strad130 VMA Array 432.
show media block id all The show media all block id all command displays the location, media ID, model, size, and status (online/offline) of all block storage media. SM075 [NJ-Lab-vSHARE00: master] # show media block id all Location Media ID Model Size Status -----------------------------------------------------------------------unknown-00 Fender1002 VMA Array 6895.75G online unknown-01 Strad130 VMA Array 432.
Manufacturer serial no: 6C057CWX00134 Device ID: ata-VIOLIN_MEMORY_ARRAY_6C057CWX00134 Violin model: VMA Array Violin serial no: Strad134 Violin Part number: V1010 Violin Revision: Media Block Enable Commands Before a media device can be used for block storage, it must be enabled. This section describes commands for enabling and disabling block storage media devices on a vSHARE HP VMA SAN Gateway. The enable commands described in this section can be used in Config mode only.
In the following example, the media block id enable command is used to enable a single block storage media device: * gateway02 [cluster1: master] (config) # media block id ? all Change block media state for all block devices VTMS0101-2 VTMS0101-4 * gateway02 [cluster1: master] (config) # media block id VTMS0101-2 enable * gateway02 [cluster1: master] (config) # █ To enable all block storage media devices on a local HP VMA SAN Gateway, use the all option as described in media block id all enable o
In the following example, the media block id all enable command enables all storage media devices on the local HP VMA SAN Gateway: SM076 [NJ-Lab-vSHARE00: standby] (config) # media block id all enable Enabling 23108R00000043 Enabling Strad134 no media block id all enable To disable the use of all media devices in the HP VMA SAN Gateway, use the no form of the above command in Config mode.
The show media health commands display the location, media ID, model, life remaining, and status (or health) of media devices on the local HP VMA SAN Gateway or on all of the HP VMA SAN Gateways in the cluster. The commands can be used in Enable and Config modes. Understanding Media Health Statuses The status of the media device is defined by one of four values: ok The device is running correctly and has plenty of expected life remaining. unknown The status cannot be determined. warning A S.M.A.R.
show media health The show media health command returns information about the health of storage media devices on the local HP VMA SAN Gateway. SM075 [NJ-Lab-vSHARE00: master] # show media health Location Media Id Model Life remain Health ------------------------------------------------------------------------unknown-00 Fender1002 VMA Array 99.0% ok unknown-01 Strad130 VMA Array 96.0% ok To view information about storage media on other nodes in the cluster, use the global option.
Media Read and Write Statistics The show stats media commands enable you to view read and write rates (in MB per second) for both the current ten-second sample and an average of the samples taken over the last five minutes. The show stats media command displays read and write statistics for block storage media devices in one or more nodes in the cluster. The scope of each command is the same as the corresponding show media command, described above.
vSHARE Block Storage Management Commands Managing Containers The show containers command returns information about the storage containers on an HP VMA SAN Gateway cluster including the node, container name, the number of LUNs, the space allocated, and the space free.
The show luns session command returns information about the sessions connected to LUNs including the target port and the session ID. Use the name option to identify a specific LUN.
LUN Set Commands [no] lun set container name readonly The lun set command enables you to set a LUN as read-only. The lun set command is the only command usable after a LUN has been created except the no lun create … readonly command. Managing Targets Show Targets Command show targets [node ] [hostname ] [protocol fc] [id ] [sessions] [detail] The show targets command displays all Fibre Channel targets.
The LUN Status page enables you to view information about containers and the LUNs within those containers. Figure 5.1 LUN Status Page Viewing Container Status The LUN Status page displays high level information about the containers in the vSHARE system including the container name, its address, and the number of LUNs in that container. Viewing LUN Status To view information about the LUNs within a specific container, select that container in the LUN Status page.
Managing LUNs The LUN Management page displays tools which enable you to add or remove LUNs from containers and to export LUNs to specific targets. Figure 5.2 LUN Management Page The LUN Management Page is divided into two basic areas: the Container area and the LUNs area. • The Container area displays information about the containers created on the attached VMA Arrays including the name, the percentage of space used, and the number of LUNs. The Containers dropdown list enables you to select a container.
The Create LUN window appears. 4. Define the number of LUNs in the # LUNs to Make space. 5. Define the name of the LUNs in the LUN Name space. If the number of LUNs to be created is greater than one, the name of each of the newly created LUNs will be appended with an index number beginning with 1 up to the number of LUNs created. 6. Define the size of the LUNs. • To define the size of the LUN based on the space available, select the “Use all available space equally” button.
Exporting LUNs To export LUNs: 1. Select vSHARE > Manage LUNs in the VMA Web Interface. The LUN Management page appears. 2. Select a container in the Container ID drop-down list in the Container area. 3. Click on a LUN in the LUNS area to highlight it. 4. Click the Add Export button in the Exports area. 5. The Add Export dialog box appears. 6. Select the initiators to which the LUNs are exported. 7. 8. • To export to all initiators, select the All Initiators option button.
Managing Initiators In a vSHARE HP VMA SAN Gateway environment, the hosts (for example, database servers or application servers) that access LUNs are called initiators and the HP VMA SAN Gateways themselves are called targets. The Initiator Management page enables you to define initiator groups and add or remove initiators to those groups. Figure 5.3 Initiator Management Page vSHARE enables you to control access to LUNs on an initiator-by-initiator basis or by defining initiator groups (igroups).
2. Select one or more initiator groups in the Initiator Groups table. 3. Click the Remove Checked iGroup(s) button. The initiator group is removed from the Initiator Group table. Adding or Removing Initiators to Initiator Groups Fibre Channel initiators are identified by World-Wide Names (WWN). Fibre Channel initiators are fixed by the appropriate HBA port. To add an initiator to an initiator group: 1. Select vSHARE > Manage Initiators in the VMA Web Interface. The Initiator Management page appears.
Fibre Channel Target Ports The Fibre Channel table in the Target Management page displays the node, target name, hostname, speed, and address of each Fibre Channel target port. If using Fibre Channel, the target ports are automatically configured when you create the storage containers on the VMA Array.
AM456-9007B Rev 02 HP VMA SAN Gateway Installation and User Guide
APPENDIX A: Command Line Interface Web: For information about using the VMA Web Interface as an alternative to the CLI, see APPENDIX B:VMA Web Interface Reference on page 177. Using the Command Line Interface The CLI can be used to perform all of the functions provided in the VMA Web Interface, as well as other functions that are not available in the VMA Web Interface.
within the CLI command set. For example, the commands to display the system date and time or the hostname can be abbreviated as: sh clo show clock sh h show hosts Additional letters can be included but none can be skipped; for example, the show clock command can be typed as sho clo or sh clock or various other combinations, but not shw clk.
provides a list of available commands corresponding to the current mode. (Modes are described in Command Modes on page 117.) You can also query for options of a specific command by typing in the command, following it with a space, and adding a question mark. After displaying a list of options, the command line echoes the string and puts the cursor after it, ready for more input. For example, in Standard mode you can enter cli ? at the command line and see the following output.
For example: > cli session ? auto-logout paging prefix-modes progress terminal x-display Configure keyboard inactivity timeout for automatic logout Configure the ability to view text one screen at a time Configure the CLI's prefix modes feature for this session Configure progress updates for long operations Set terminal parameters Set the display to use for X Windows applications > cli session paging ? enable Enable paging > cli session paging enable ? > cli session terminal ? length Set the number o
At that point, pressing Tab again displays a list of possible completions, which might be keywords or values, or both. h Completes the help keyword. sh Completes the show keyword (but s does not, because more than one command starts with the letter “s”). show Lists options of the show command that can immediately follow the show keyword. sh Completes the show keyword and lists options of the show command.
Enable Mode The enable command moves the CLI to Enable mode. This mode has commands to view all state information and take certain kinds of actions, such as rebooting the system or configuring some system parameters, but it excludes commands that configure the cluster. Its commands are a superset of those in Standard mode. The disable command moves the CLI from Enable mode back to Standard mode. The exit command (in Enable mode) closes the CLI.
of the prompt string indicates what command mode the CLI is in: > for standard mode, # for Enable mode, or (config) # for Config mode. For example, if the hostname of the master node is gate1 and the cluster name is vmgCluster then the prompts for each of the CLI modes are: Standard mode: gate1 [vmgCluster: master] > Enable mode: gate1 [vmgCluster: master] # Config mode: gate1 [vmgCluster: master] (config) # The role can be master, standby, normal, or unknown.
TB terabytes (10244 = 1,099,511,627,776 bytes) PB petabytes (10245 = 1,125,899,906,842,624 bytes) and so on for E (exabytes), Z (zettabytes), and Y (yottabytes). Single-letter abbreviations such as k, M, or G are sometimes used to conserve space, or for units other than bytes. CLI Command Descriptions An alphabetical list of commands is provided in Quick Reference to Commands on page 173.
An IPv4 network prefix specifying a network. This is used in conjunction with a netmask to determine which bits are significant. For example, "192.168.0.0". ID of a node (HP VMA SAN Gateway) belonging to a cluster. This is a numerical value greater than zero. A TCP or UDP port number. An extended regular expression as defined by the "grep" man page. (The value you provide here is passed on to "grep -E".
If you omit the ":password" part, you may be prompted for the password in a follow up prompt, where you can type it securely (without the characters being echoed). This prompt will occur only if the "cli default prompt empty-password" setting is true; otherwise, the CLI assumes you do not want any password. If you include the ":S" character, this is taken as an explicit declaration that the password is empty, and you will not be prompted in any case.
Cluster A. While Cluster B is being upgraded, Cluster A provides access to storage, and vice-versa. Upgrading via USB Drive To upgrade the software from a USB flash drive: 1. Load the new software image onto the USB drive. 2. Insert the USB drive into a USB port on the HP VMA SAN Gateway. 3. From the CLI, type the following at the command prompt: image fetch usb Where image_filename is the name of the software image file.
mgmt-core (for mgmtd alone), mgmt-back (for other back end components), and mgmt-front (for front end components, utilities, and tests). [no] logging local override Enable or disable all class-specific overrides to the local log level. Overrides are enabled by default. The no variant which disables them will leave them in configuration, but disable them such that the logging level for all classes is determined solely by the global setting.
Allow this system to receive log messages from another host. This is disabled by default. If enabled, only log messages matching or exceeding the minimum severity specified with the logging local command are logged, regardless of what is sent from the remote host. logging format standard logging format welf no logging format welf Sets the format in which log messages should be set. The choices are standard and welf. The default is standard, and the no variant resets the format to this.
occur at the same time it was previously scheduled for. Naturally, if the autorotation was based on size, this will delay it somewhat as it reduces the size of the active log file to zero. logging files delete oldest [] Force the immediate deletion of the specified number of the oldest log files. logging files upload {current, } Upload a log file to a remote host. The word current specifies the current log file.
remote-dir
Set the remote directory for uploading logs. remote-site Set the remote site for uploading logs, in URL format. user Change the user/password used with the remote server. protocol Options are email, FTP, HTTP, and SCP. Note: Files are uploaded from all gateways simultaneously.User Accounts and Local Authentication Note: There are two defined system accounts: one with full privileges (admin), and one with privileges for reading all data and performing all actions, but not for changing any configuration (monitor). [no] username Create or remove a user account. New users are created initially with admin privileges and disabled. To enable a user account, just set a password on it (or use the nopassword command to enable it with no password required for login).
recovered after it is set. These two commands are thus named for Cisco compatibility. The nopassword command means that no password is required to log in. The disable command configures the account so that no one can log into that account. The disable password command leaves the account as a whole the same, but forbids login with a password. It is assumed that ssh key access will be used instead. To re-enable the account, the user must un-disable it, and put a password on it.
may be required to be specified in two, three, or four words, depending on the city. The possible forms this could take include: • An offset from GMT. This will be in the form GMT-offset GMT, GMT-offset GMT+<1-12>, or GMT-offset GMT-<1-14>. • UTC. This is almost identical to GMT. The default is GMT-offset GMT. show clock Display the current system time, date, and timezone.
Disable or re-enable an NTP server or peer. Servers and peers start enabled; disabling is just a way of making them temporarily inactive without losing their configuration. [no] ntp enable [no] ntp disable Enable or disable NTP overall. Note that the latter is the Cisco command; the former is just a pair of aliases added to increase usability, as otherwise it may be hard for a user to figure out how to enable NTP if they are not aware of no commands and only see a way to disable it.
• If an email domain is specified using this command, it is always used. If the hostname has any dots in it, everything to the right of the first dot is stripped and the email domain is appended. • Otherwise, if the hostname has dots in it, it is used as is. • Otherwise, the currently-active system domain name is used. This can come either from the resolver configuration, or from state dynamically instantiated by DHCP.
For example: (config) # email notify event ? process-crash A process in the system has crashed process-exit A process in the system unexpectedly exited liveness-failure A process in the system was detected as hung cpu-util-high CPU utilization has risen too high cpu-util-ok CPU utilization has fallen back to normal levels paging-high Paging activity has risen too high paging-ok Paging activity has fallen back to normal levels disk-space-low Filesystem free space has fallen too low disk-space-ok Filesystem f
unexpected-cluster-join unexpected-cluster-leave unexpected-cluster-size unexpected-shutdown interface-up interface-down user-login user-logout cache-faults hwmon-file-sys-error hwmon-mce-error media-device-health-warn media-device-state-change media-device-lifetime-warn media-device-unknown-type media-system-swap-state A node has unexpectedly joined the cluster A node has unexpectedly left the cluster The number of nodes in the cluster is unexpected Unexpected system shutdown An interface's link state has
Add or remove an email address from the list of addresses to send email notifications of events. [no] email notify recipient class info [no] email notify recipient class failure Enable or disable the sending of informational or failure events to the specified recipient. Each event type is classified as either informational or failure.
Network diagnostic tools ping and traceroute. Invokes standard binary, passing command line parameters straight through. tcpdump [] Network diagnostic tool tcpdump. Invokes standard binary, passing command line parameters straight through. Runs in foreground, printing packets as they arrive, until user presses CTRL+C. slogin [] Invokes the SSH client. The user is returned to the CLI when SSH finishes. telnet [] Invokes the telnet client.
If both prompts are enabled, and the configuration was unsaved, the user will be prompted twice. The noconfirm parameter suppresses both of these confirmations if it is specified. reload force If the system is busy performing another operation requiring the management subsystem (which is almost any management operation), the regular reload [halt]command will block until it is finished. If rebooting the system is urgent, the reload force command will do it immediately.
Enable or disable the specified alarm. There are three alarms that can be enabled or disabled: cpu_util_indiv (CPU utilization too high), paging (paging activity too high), and fs_mnt (filesystem low on free space).
Show Stats CPU show stats cpu Display some basic stats about CPU utilization: the current level, the peak over the past hour, and the average over the last hour. Show Stats FC show stats fc [on (hostname |global)] [continuous] [detail] [port ]* Shows fibre channel statistics values. Zero, one, or more than one port can be specified. If no host is specified, FC ports at the local host are shown. Use the global parameter to show all the fc port statistics values at the cluster.
The custom option should be a reserved report name (that is, no reports should be named that) to leave room in the command set for later allowing the user to specify manually which series to export. If the filename is specified, it must come just after the report name. If the after and/or before parameters are specified, they may come in either order relative to each other. Stats Reports show files stats Display a list of statistics report files.
Note: These configuration commands only work when you are in Config mode (which you enter by using the conf t command—see Enable and Config Modes on page 144). configuration new [factory [keep-basic]] Create a new configuration file under the specified name. If no additional parameters are specified, active licenses from the current running configuration are copied over into it. If the factory parameter is specified, nothing is copied over; it has the factory defaults.
On a system with clustering, only save the configuration on the local box instead of attempting to save on all cluster members. configuration delete configuration move configuration copy Delete, move (rename), or copy a configuration file. These do not affect the current running configuration. The active configuration file may not be deleted or renamed, nor may it be the target of a move or copy.
Run the initial-configuration wizard. The wizard is automatically invoked whenever the CLI is launched when the active configuration file is fresh (that is, not modified from its initial contents). This command invokes the wizard on demand. show configuration [full] Display a list of CLI commands that will bring the state of the a fresh system up to match the current persistent state of this system. A short header is included, containing the name and version number of the configuration, in a comment.
command is idempotent, and will not continue to cycle through all of the available locations.) boot bootmgr password boot bootmgr password 0 boot bootmgr password 7 no boot bootmgr password Configure or remove a password to control access to boot manager parameters. Similar to the username * password ...
Exit Config mode and return to enable mode. exit Exit the current mode. From Config mode, go to enable mode. From enable or standard mode, log out of the system. Note that to go from enable mode to standard mode, the disable command must be used. help Display a general CLI help message. Web Proxy Settings web proxy host [port ] no web proxy If desired on a http or ftp download, a proxy can be specified. If no port is specified, the default is 1080.
file tcpdump delete Delete the specified tcpdump output file. file tcpdump upload Upload the specified tcpdump output file to the specified URL. Only scp pseudoURLs are supported for the destination. See Key to Command Parameters on page 120 for the scp:// URL format. Debug Dumps show files debug-dump Display a list of debug dump files. show files debug-dump Display a summary of the contents of a particular debug dump file.
show cluster global Display global cluster run state. show cluster global brief Display global cluster run state in brief. show cluster local Display local cluster run state. show cluster local error-status Display error status of local node. show cluster master Display run state information about master node. show cluster node Display information about node with specified node id. show cluster standby Display run state information about standby node.
no cluster master address vip Reset cluster master virtual address to default. cluster name Set the name describing the cluster. This is the cluster equivalent of the hostname. no cluster name Reset the cluster name to default. Network Configuration Commands This section describes the CLI commands for network configuration.
links. See Dual Gateways with 1–2 3000-Series Arrays, Highly Available on page 233 for more information. interface ip address no interface ip address Set or clear the IP address and netmask of this interface. interface mtu no interface mtu Set the maximum transition unit (MTU) of this interface. The no variant resets the MTU to its default.
[no] interface zeroconf The command may be used to enable or disable use of zero configuration networking (zeroconf) on the specified interface. This randomly chooses a unique Link Local IPv4 address from the 169.254/16 block. zeroconf is an alternative to DHCP. show interfaces [] configured show interfaces [] brief Display information about the specified interface, or all interfaces if one is not named. Either configured or brief or neither may be specified.
link-agg Link Aggregation mode link-agglayer2+3 Link Aggregation Layer 2 + 3 mode link-agglayer3+4 Link Aggregation Layer 3 + 4 mode balance-tlb Adaptive transmit balancing balance-alb Adaptive load balancing VLAN Commands A VLAN, or virtual LAN, is a set of nodes that are grouped together into a single logical network, regardless of their physical location. Multiple VLANs may share the same physical Ethernet links.
Name Resolution Commands hostname no hostname Set or clear the system hostname. [no] ip name-server [no] ip domain-list [no] ip host The commands configure DNS servers, specifies which domain(s) to try unqualified hostnames in, and defines hostname/IP mappings for /ETC/HOSTS. show hosts Show all values configured by the group of commands above: hostname, name servers, domain name list, and static host mappings.
no arp Remove a static entry from the ARP cache. Note that this cannot be used to remove dynamic entries; use clear arp-cache for that. clear arp-cache Clear dynamic entries in the ARP cache. Note that this does not delete static ARP entries configured with the arp ... command. show arp Display the contents of the ARP cache. This should contain all of the staticallyconfigured ARP entries, as well as any that the system has picked up at dynamically.
determines how the remote user mapping behaves. If the authenticated user name is valid locally, no mapping is performed. The setting has the following three possible behaviors: • remote-first: If a local-user mapping attribute is returned and is a valid local user name, map the authenticated user to the local user specified in the attribute. Otherwise, if the attribute is not present or not valid locally, use the user specified by the default-user command. (This is the default behavior.
radius-server host {hostname , ip-address} [auth-port ] [timeout ] [retransmit ] [key ] no radius-server host {hostname , ip-address} [auth-port ] Add a RADIUS server to the set of servers used for authentication. Some of the parameters given may override the configured global defaults for all RADIUS servers. The auth-port defaults to 1812 and is used for authentication requests.
Sets (or resets to 0) a global communication value for all TACACS+ servers. Can be overridden in a tacacs-server host command. Defaults to 2. Sets the number of times the client will attempt to authenticate with any TACACS+ server. To disable retransmissions set it to zero. (Number of times to search the TACACS list). Range is 0-5. tacacs-server key no tacacs-server key Sets (or clears) a global communication value for all TACACS+ servers. Can be overridden in a tacacs-server host command.
SSH Configuration SSH Server [no] ssh server enable Enable or disable the ssh server. If the ssh server is disabled, the CLI is only accessible over the serial console. Note that this does not terminate existing ssh sessions; it will only prevent new ones from being established. ssh server host-key generate Regenerate new host keys for the ssh server. This generates three keys: RSA for sshv1, RSA for sshv2, and DSA for sshv2.
SSH Client ssh client user identity generate [passphrase ] Generate a new identity (private and public keys) for the specified user name. The given user name must correspond to a valid local user account. When the keys are generated, the private key is written to the user’s ssh directory in an appropriately named file (id_dsa). This identity can be used when the user uses the slogin command to connect from the system to another host.
Remove a public key from the specified user’s authorized key list. The key identifier can be found by using 'show ssh client'. show ssh client Display information about SSH client identities (public/private keys) and the per user list of authorized keys for the users. Banner At various login points, some legal and welcome text can be displayed.
• terminal ... commands are clones of a subset of the cli session ... commands, and are only present for Cisco compatibility. All of these commands are INTERACTIVE (not available from the Scheduler). Note that some settings, such as the terminal length and width, are inherently session-specific, and there are no corresponding commands to set the defaults in configuration.
... commands, which have no confirmations in any case, since they are explicitly dealing with configuration. [no] cli default confirm reload Enable or disable confirmations of rebooting or halting the system using the 'reload' command. This confirmation is in addition to any separate confirmations that may be displayed for unsaved changes.
Display CLI settings: the inactivity timeout, whether or not paging is enabled, the terminal size and type. For settings which have configured defaults, both those and the current session settings are displayed. show terminal Display current terminal width and length, whether auto-detected or overridden, as well as the current terminal type. This is a subset of the information displayed by show cli, and is only present for Cisco compatibility.
Enable (or disable) the listen interface restricted list for snmpd. If enabled and at least one non-DHCP interface is specified in the list, the snmp connections are only accepted on those specified interfaces. When disabled, snmp connections are accepted on any interface. [no] snmp-server listen interface Provides a mechanism to add and remove interfaces to the "listen" list. If the interface is also running as a DHCP client, it will be as if the interface was not added.
Display all SNMP configuration options. VMA Web Interface Configuration [no] web enable Enable or disable the VMA Web Interface. [no] web http enable Enable or disable HTTP access to the VMA Web Interface. This setting is only meaningful if the VMA Web Interface as a whole is enabled. web http port no web http port Set the port number for HTTP. The default is 80. The no command resets it to the default, but does not disable HTTP.
web auto-logout no web auto-logout Control the length of user inactivity required before the VMA Web Interface will automatically log out a user. The command no web auto-logout disables the automatic logout feature. web session renewal no web session renewal Control the length of time before web session cookies are automatically regenerated. web session timeout no web session timeout Control the maximum lifetime of a web session cookie.
Remove all state associated with the specified . If the job has not executed, the timer event is canceled. If the job was executed, the results are deleted along with all other job state. job date-time [] job date-time [] Set the time for the job to execute. If the time specified is in the past, the job will not execute and will be in the "inactive" state.
List all jobs that currently exist in the system. show jobs Lists the following information about the specified job: Job Status Name Comment Absolute range Commands .. ... Any command results are also displayed. If an error string exists for the job, it is displayed. Media Management Commands This section describes commands for managing and monitoring block storage (vSHARE) media devices.
Enable or disable a specified block storage media device. Use the block option to enable for vSHARE./ The can be found in the output of the show media all or show media block id all command. Caution: When a media device is disabled, all applications using that device immediately stop using it, and space from other media devices is allocated for them. This may be a disruptive operation.
Display the same information as for a single HP VMA SAN Gateway, for every HP VMA SAN Gateway in the cluster. Gateway IDs are labeled Module <#> in the display. The show media global command is an alias for show media [block] id all global. To include non-cache or non-storage media, use show media all global (or show media global all). The health and detail options can also be used in combination with the global option, in any order.
show stats media all cache id [continuous] [detail] [global] show stats media [block] id all [continuous] [detail] [global] [all] show stats media [block] id [continuous] [detail] [global] [all] Display read and write statistics for block storage or cache media devices in one or more nodes in the cluster. The scope of each command is the same as the corresponding show media command, described above. The command show stats media is an alias for show stats media [block] id all.
The show containers command displays all containers created within a VMA Array enabled for block storage. Targets Show Targets Command show targets [node ] [hostname ] [protocol fc] [id ] [sessions] [detail] The show targets command displays all Fibre Channel. Use the hostname, protocol, and id parameters to filter the targets returned. The command returns the node, hostname, target port, status (enabled or not), and address (WWN).
igroup addto Commands [no] igroup addto initiators [initiator_name …] The igroup addto command adds one or more initiators to a group. LUNs Show LUNs Command show luns container name sessions The show luns command enables you to view LUNs. LUNs may be filtered by container, name, and session parameters. To view a list of containers, enter show luns container ?. The name parameter enables you to view only those LUNs which are prefixed by a particular name.
The show targets command displays all Fibre Channel. Use the hostname, protocol, and id parameters to filter the targets returned. The command returns the node, hostname, target port, status (enabled or not), and address (WWN). The detail parameter returns the network bindings for each target. Quick Reference to Commands This section lists the commands that are documented in this guide; it does not include every command that is available in the CLI.
S: Standard Mode E: Enable Mode C: Config Mode no Variant E, C C [no] S, E, C Command See section starting on page # file tcpdump page 145 ftp-server page 145 help page 144 C [no] hostname page 152 C [no] igroup page 171 C [no] interface page 148 C [no] ip host page 152 C [no] ip domain-list page 152 C [no] ip name-server page 152 C [no] ip route page 152 E, C [no] job page 165 E, C [no] logging page 123 C [no] luns page 172 C [no] media page 167 netw
S: Standard Mode E: Enable Mode C: Config Mode no Variant Command See section starting on page # E, C show files debug-dump page 146 E, C show files stats page 140 S, E, C show files system page 135 E, C show files tcpdump page 145 E, C show ftp-server page 145 S, E, C show hosts page 152 E, C show igroups page 171 S, E, C show images page 143 E, C show interfaces page 148 S, E, C show ip page 152 S, E, C show jobs page 165 E, C show log page 123 S, E, C show logging
S: Standard Mode E: Enable Mode C: Config Mode no Variant Command See section starting on page # E, C show web page 164 E, C slogin page 135 E, C [no] snmp-server page 162 E, C [no] ssh client page 158 C [no] ssh server page 157 E, C [no] stats page 137 C [no] tacacs-server page 155 E, C tcpdump page 135 S, E, C telnet page 135 C [no] telnet-server page 145 S, E, C [no] terminal page 159 traceroute page 135 S, E, C C [no] username page 128 E, C [no] web p
APPENDIX B: VMA Web Interface Reference Understanding the VMA Web Interface The VMA Web Interface is a web-based graphical user interface (GUI) that may be accessed by connecting to the Master HP VMA SAN Gateway (master node) of the HP VMA SAN Gateway cluster.
Organization of the VMA Web Interface The VMA Web Interface is organized in into five basic areas: the message bar, menu bar, shortcuts menu, copyright bar, and the control page. Menu Bar Shortcuts Menu Control Page Message Bar Copyright Bar Figure B.1 Web Interface Organization Message Bar Each page in the VMA Web Interface has a message bar at the top, which shows the page’s location in the menu tree, various status messages, your user name, and the Logout and Help buttons.
Page help Below the Help button on the right side of the message bar, the Page Help button in the main (unshaded) part of the page provides information about the current page. Breadcrumbs The breadcrumbs enable you to navigate between the control pages. Menu Bar The Menu Bar organizes VMA Web Interface control pages into a hierarchical tree structure and provides navigation for the tool.
Category Tools Help Pages • Upgrade • Diagnostics • Documentation • About • License • Acknowledgements Table B.1 Web Menu Categories and Pages A light gray area just above the menu shows the cluster’s name, status, and memory allocation. Shortcut Menu A shortcut provides a link to a frequently-used control page. The Shortcuts menu is located on the left side of the page, below the menu. For users with monitor privileges, the shortcuts in the Shortcuts menu are the only pages available.
• A PAGE HELP button in the upper right corner of the central page, just below the Help button in the message bar, displays a brief explanation of the current page (or hides the explanation if it is already displayed). VMA Web Interface Requirements The VMA Web Interface has requirements for browsers, display resolution, Adobe Flash Player, JavaScript, and cookies. Supported Web Browsers The software is tested with current versions of Firefox, Internet Explorer, Safari, and Chrome.
JavaScript JavaScript must be enabled for the VMA Web Interface. Adobe Flash Player Adobe Flash Player version 8 (or above) is required for viewing charts in the VMA Web Interface. Cookies Enable cookies for login and session management. Abbreviations for Large Numbers The following abbreviations are used for large numbers throughout the VMA Web Interface: B bytes KB kilobytes MB megabytes GB gigabytes TB terabytes PB petabytes and so on for E (exabytes), Z (zettabytes), and Y (yottabytes).
Login and Logout To log in and view, you must enter a valid username and password in the Login page. Figure B.2 Login Page By default, the Status page is displayed upon login. For more information, see Cluster Status Page on page 184. Accessing the VMA Web Interface To access VMA Web Interface, open the VMA Web Interface URL (http:// ) using one of the supported web browsers and log into the VMA Web Interface.
Logout Page When you click the Logout button in any page of the VMA Web Interface, a Logout page appears. Figure B.3 Logout Page On the Logout page the shortcuts, Help button, and hyperlinks initially take you to the Login page, where you must log in again before going to the requested location in the VMA Web Interface. Cluster Management The Cluster section of the menu bar displays five pages: Cluster Status Page, Gateways Page, Alerts Page, Logs Page, Versions Page.
The Cluster Status page displays two or more HP VMA SAN Gateway panels, which may be expanded or collapsed in order to view or hide information about the VMA Arrays in the cluster. Figure B.
• Number of VIMMs in each state (active, spare, booting, etc.
Using controls in the Board Status panel, you may view information about individual VIMMs or view VIMMs by their properties. To view information about specific VIMMs, move the cursor over its number (for general status) or over the appropriate Alarm, Temp, or Lifetime indicator box. Viewing VIMMs by RAID Group To view VIMMs by RAID group, select a RAID Group number at the top of the Board Status panel.
Gateways Page The Gateways page lists all the HP VMA SAN Gateways (nodes) in the cluster, their status (online/offline), and their role (master node, standby node, normal node). Figure B.7 Gateways Page The Master Gateway is the current master node in the cluster. The master node works together with a standby node to manage the availability of the cluster. When there are more than two nodes in the cluster, all other nodes are designated as normal nodes.
Viewing Gateway Details The View Gateway page shows the selected HP VMA SAN Gateway’s internal and external IP addresses, its role and status, uptime, software version, and the name of its installed image. The page also shows how much DRAM the HP VMA SAN Gateway contains. Figure B.8 Gateway Details Page View shows more details about an individual HP VMA SAN Gateway. Alternatively, you can show the details by clicking on the Gateway ID in the Gateways column.
Alerts Page The Alerts page shows a listing of current and past system events that triggered an alert. Select the types of alerts to display (informational, warning, or error) and click the APPLY button. Alerts are emailed selectively to recipients, based on the type of alert (see Alert Recipients Page on page 200). Figure B.
Logs Page The Logs page shows a listing of logged events. It allows you to view current and historical logs in paginated format. Log files are rotated once a day. Figure B.10 Logs Page You can use the Prev and Next hyperlinks, or the numbered navigation pages, to navigate to a desired log page within the file. To see the most recent log entries, click the last available page in the log and scroll to the bottom.
VMA SAN Gateway has been running (Uptime) and identifies the boot HP VMA SAN Gateway and next HP VMA SAN Gateway to be booted (Boot/Next). Figure B.11 Versions Page To view details about an HP VMA SAN Gateway, click the Gateway ID button. For more information see Viewing Gateway Details on page 189.
Cluster Administration Page The Cluster Administration page allows an administrator to manage the HP VMA SAN Gateway cluster. Figure B.12 Cluster Administration Page The Cluster Administration page is organized into three sections: System, Software, and Configuration, System AM456-9007B Rev 02 The System section displays the Reboot and Shutdown buttons.
Software The Software section displays the software version number and allows you to choose which boot image the HP VMA SAN Gateway should use (the Switch Boot Partition button). Two boot images are installed for each HP VMA SAN Gateway in separate partitions. Configuration The Configuration section allows you to save your current configuration setting to a specified file (the Save button).
Caution: Changing the cluster management address is not recommended. You can use controls in the Network page to enable routing by global default gateways and to set a global default gateway, or you can disable the use of global default gateways. DNS Settings Page The DNS Settings page shows the list of DNS servers available to the HP VMA SAN Gateway cluster. If one or more IP addresses for DNS servers were set up as part of the initial out-of-the-box setup, they are listed here. Figure B.
NTP Settings Page The NTP Settings page shows the NTP settings for time synchronization. You can enable or disable NTP synchronization and add or remove an NTP server. Figure B.15 NTP Settings Page Web Administration Page VMA Web Interface configurations can be changed using the Web Admin page or the web commands in the CLI.
The Web Administration page allows you set the following parameters for the VMA Web Interface: Figure B.16 Web Administration Page Enable Web Management: This check box enables (or disables) VMA Web Interface access to the HP VMA SAN Gateway cluster. Note: If you disable access by un-checking the Enable Web Management check box, you will not be able to access the VMA Web Interface again; therefore the only way to restore access is to use the “web enable” command in the CLI.
Web Session Renewal (secs) Time in seconds before the VMA Web Interface session cookie expires. Web Session Timeout (secs) Time in seconds before the VMA Web Interface session cookie times out. Feature Licenses Page The Feature Licenses page shows which features are installed on the HP VMA SAN Gateway cluster and allows you to add or remove features. Figure B.17 Feature Licenses Page Adding a new feature requires a license key, which you must obtain from an HP Customer representative.
Users Page The Users page displays the name, role, e-mail address, and password status of user accounts and contains controls for adding or removing users to the HP VMA SAN Gateway cluster. Figure B.18 Users Page Using controls in the Users page, you may add, edit, and delete users and manage user passwords. • Adding Users • Changing Passwords • Editing Users • Deleting Users Adding Users The Add User light box allows you to add a user account and set a role for the user.
Changing Passwords The User Management: Change Password dialog box allows you to set or reset a password for the specified user. Note: Passwords can be no longer than 8 characters. If you attempt to enter a longer password, you will receive multiple errors and will be unable to login. Editing Users This light box allows you to change an existing user’s information and role. Deleting Users The Delete User confirmation box allows you to delete a user account from the HP VMA SAN Gateway cluster.
If Receive Failures is set to true, the recipient will get notification of failure events such as file system errors, process crashes, and unexpected shutdowns. If Receive Infos is true, the recipient will get notification of less urgent informational events. A list of all events that can trigger alerts can be found on the CLI by using the show email events command. If Receive Details is true, the notifications include detailed information; otherwise only summaries are sent.
Call Home The Call Home page allows you to enable the Auto Support feature and set parameters for fault reporting. Figure B.20 Call Home Page Auto Support enables the HP VMA SAN Gateway cluster to report alerts and critical events to a specified email address. If Auto Support is enabled, you will need to specify the SMTP server and the email domain name. In the Fault Reporting section of the Call Home page, you can specify the SMTP server and mail domain name to use for sending call-home messages.
Tools: Upgrade The Tools: Upgrade page allows you to upgrade the HP VMA SAN Gateway software running on all nodes in the HP VMA SAN Gateway cluster. Figure B.21 Upgrade Page For more information, see Configuration File Management on page 57. Note: If the node count does not match the expected number during an upgrade, you will be prompted with an option to force the upgrade.
Tools: Diagnostics The Tools: Diagnostics page provides information about diagnostics used while testing the system. Figure B.22 Diagnostics Page Help Pages The following pages are available in the Help section of the menu tree: Help: Documentation, Help: About, Help: License, and Help: Acknowledgements. Help: Documentation The Help: Documentation page provides a hyperlink to this guide as a PDF file. Another link lets you download the Adobe Acrobat Reader for viewing and printing the PDF file.
Help: License The Help: Licenses page provides a complete copy of the HP license. Help: Acknowledgements The Help: Acknowledgements page lists copyright and licensing information for the open source components of the product. Portions of this software product utilize open source copyrighted material; attribution to copyright holders of such materials along with applicable licensing terms are included here and also available on the HP Web site.
Viewing Container Status The LUN Status page displays high level information about the containers in the vSHARE system including the container name, its address, and the number of LUNs in that container. Viewing LUN Status To view information about the LUNs within a specific container, select that container in the LUN Status page. The container expands displaying high level information about those LUNs including its name, size, and the number of active sessions.
Creating LUNs To create a LUN: 1. Select vSHARE > Manage LUNs in the VMA Web Interface. The LUN Management page appears. 2. Select a container in the Container ID drop-down list in the Container area. Every LUN is created within a container. Note the size and percentage used of the container may restrict the number and size of the LUNs created within the container. 3. Click the Create LUN button in the LUNs area. The Create LUN window appears. 4. Define the number of LUNs in the # LUNs to Make space.
Zeroing-out data could take a long time for large LUNs. 9. To set the LUNs online, select the Online check box. LUNs may be online or offline. 10. To define the LUNs as read-only, select the Read-only check box. 11. Click the OK button. The LUNs are created in the container. Exporting LUNs To export LUNs: 1. Select vSHARE > Manage LUNs in the VMA Web Interface. The LUN Management page appears. 2. Select a container in the Container ID drop-down list in the Container area. 3.
9. • To assign a user-defined LUN ID to exported LUNs, select the Value option button and enter a value in the space. User-defined LUN IDs may make it easier differentiate between LUNs. If you do assign user-defined LUN ID, HP recommends that you specify a number below 255 as some operating systems (for example, Windows) will only discover LUN IDs between 0 and 254. • To automatically assign an ID to the exported LUNs, select the Auto option button. Click the OK button.
The Name of the New Group dialog box appears. 3. Enter a name in the dialog box and click the OK button. The new initiator group is displayed in the Initiator Group table. To delete an initiator group: 1. Select vSHARE > Manage Initiators in the VMA Web Interface. The Initiator Management page appears. 2. Select one or more initiator groups in the Initiator Groups table. 3. Click the Remove Checked iGroup(s) button. The initiator group is removed from the Initiator Group table.
The Target Management page displays tools which enable you to view information about Fibre Channel target ports. Figure B.26 Target Management Page vSHARE supports both Fibre Channel ports. Every target is either a specific port on a hardware Fibre Channel host bus adapter (HBA). Fibre Channel Target Ports The Fibre Channel table in the Target Management page displays the node, target name, hostname, speed, and address of each Fibre Channel target port.
AM456-9007B Rev 02 HP VMA SAN Gateway Installation and User Guide
VMA Utilities APPENDIX C: Understanding the VMA Utilities The VMA Utilities are tools designed to enable you to monitor the performance of HP VMA Array systems connected to HP VMA SAN Gateways. Using the VMA Utilities, you may retrieve detailed information about the HP VMA Array and data transfer counts that enable you to configure the system for optimal performance. Running the VMA Utilities VMA Utilities commands can be issued in following ways: • As The root user on a Linux computer.
VMA Utilities Reference varray The varray returns information about an array. Syntax varray Example The varray utility returns information similar to the following when run on a Linux host: # varray HP, Inc. Version: G5.0.
Syntax vcounts [ ] Example The vcounts utility returns information similar to the following when run on a Linux host: # vcounts HP, Inc. Version: vtms-linux-utils-D4.5.
IRQ calls for errors The total of DMA errors returned as well as PCIe link loss errors. Completed I/O bytes The total bytes read/written from/to a HP VMA Array. Completed read bytes The total bytes read from the HP VMA Array. Completed write bytes The total bytes written to a HP VMA Array. Completed I/O's The total I/O read / write requests from and to HP VMA Array. This is not the individual DMA descriptors completed, but for each of the user requested I/Os.
Requested DMA writes Incremented for each write DMA descriptor added to the descriptor ring. Note that a single I/O may result in multiple DMA descriptors to complete a single I/O request. Flash partial page reads Incremented when a DMA descriptor for read is less than a flash page (4kB) in size. On a DRAM-based system, this will always be 0.
The Windows example returns the following information. Part # Displays the part number of the main board. Serial # Displays the serial number of the main board. Board Ver Displays the version of the main board. Mfg. Date Indicates the manufacturing date of the main board. Mgmt. MAC Displays the MAC address of the management interface. Linux Example The veeprom utility returns information similar to the following example when run on a Linux host: # veeprom HP, Inc. Version: vtms-linux-utils-D4.5.
vincident -a The vincident script collects useful information from the HP VMA SAN Gateway and HP VMA Array, such as version/timestamp of the current kernel, CPU information, partition information, HP VMA Array configuration, and HP VMA Array logs. Once collected, this information can be sent to Customer Support for analysis to determine the source of performance issues, such as ECC errors. The vincident script is installed on the host server as part of the VMA Utilities package.
Example # vincident -a Gathering information from host... Gathering information from target... Target IP = 10.1.10.212 This could take a couple of minutes, please be patient... Once collected, use the following command to upload the log: #file debug-dump upload vincident.xxxxx scp:// vinfo The vinfo utility displays the HP VMA Array type and version.
Returns The vinfo utility returns the following information. AM456-9007B Rev 02 Host Driver Displays the host system vtms device driver version. Driver Date Displays block mode only. When set to 1, allows READA read-ahead I/Os to be accepted. Target S/W Displays the software / firmware version running on HP VMA Array. Memory Displays the size in bytes of usable system capacity. For flash VIMMs, this value changes based on formatted capacity.
vmesg The vmesg utility displays kernel log messages.
Example # vmesg Linux version 2.6.35-7EIsmp (root@eng-builds-sw.eng.vmem.int) (gcc version 4.1.2 20080704 (Red Hat 4.1.2-46)) g6os G5.0.
vpartial The vpartial utility displays the number of read/write I/O requests processed and the number of partial 4kB flash pages. Syntax vpartial [ ] Example The vpartial utility returns information similar to the following example when run on a Linux host: # vpartial HP, Inc. Version: vtms-linux-utils-D4.5.
Completed write I/O's Displays the total I/O write requests to a HP VMA Array. This is not the individual DMA descriptors completed, but for each of the user requested I/Os. Unaligned host buf writes Displays the total I/O write requests to a HP VMA Array, but only incremented when an unaligned host address required special buffer byte copying to service the DMA request.
Parameters Table C.1 vring Utility Parameters Parameter Description -p Specifies that a partition is to be selected. Specifies the name of the partition. Example The vring utility returns information similar to the following example when run on a Linux host: # vring HP, Inc. Version: vtms-linux-utils-D4.5.4, 05/23/2011 INDX 0 1 2 3 4 ...
Example # vspeedtest Checking speed on array device id ataVIOLIN_MEMORY_ARRAY_6C067CWX00132 1555.131 Read MB/s, 379670 IOPS vstat The vstat utility displays the status of the connection and the ready status of a HP VMA Array. Syntax vstat [ ] Example The vstat utility returns information similar to the following example when run on a Linux host: # vstat HP, Inc. Version: vtms-linux-utils-D4.5.
-- Target Status -Status LED: ON Alarm LED: OFF PWR_A LED: ON PWR_B LED: OFF ready: 1 formatting: 0 format_done: 0 paused: link: lid_ajar: raid_rebuild: write_buffer: linkwidth: maxlinkwidth: cur_payload: max_payload: cur_read_req: dma_active: io_pend: 0 1 0 0 1 8 8 128 1024 512 569 525 Returns The vstat utility returns the following information. 228 Status LED Indicates whether the Status LED is on or not. Alarm LED Indicates whether the Alarm LED is on or not.
paused Indicates the pause interval for I/Os. link Indicates whether the PCIe connection is online or offline. lid_ajar Indicates whether the lid is closed or not. raid_rebuild Indicates the status of a RAID group rebuild. write_buffer Indicates whether flash write buffering is enabled or disabled. linkwidth Indicates how many active PCIe lanes are available. maxlinkwidth Indicates the maximum number of active PCIe lanes. cur_payload Indicates the size of the PCIe payload.
Example The vupdate_tz utility returns information similar to the following example when run on a Linux host: # vupdate_tz HP, Inc. Version: vtms-linux-utils-D4.5.4, 05/23/2011 tz_secs=28800 tz_altsecs=25200 tz_minwest=420 tm_isdst=1 The output shows that this HP VMA Array is 300 minutes (5 hours) from GMT. vvimms Use this command to display the VMA Array VIMMs.
Example # # vvimms HP, Inc. Version: G5.0.2, 08/23/2011 Device: Index: VIMM 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 ...
AM456-9007B Rev 02 HP VMA SAN Gateway Installation and User Guide
APPENDIX D: Standard System Configurations Single Gateway with 1–2 3000-Series Arrays, Non-Redundant Figure D.1 shows a configuration where a single HP VMA SAN Gateway is connected to a group of up to two VMA Arrays using PCIe cables. In the example, the HP VMA SAN Gateway is connected to Port 1 on each VMA Array in x8 mode. Figure D.
In the example, Gateway 1 is connected to Port A on each Array, and Gateway 2 is connected to Port B on each Array. Note: It is important that the Gateways be cabled to the Arrays using identical slots for HA pairs. That is, if PCIe Slot 1 on Gateway 1 is connected to Port A on Array 1, then PCIe Slot 1 on Gateway 2 must be connected to Port B on Array 1. Figure D.
The following is an example of configuring simple round-robin balance mode for the bonded interface, which does not require any switch changes.
For each HA pair, each HP VMA SAN Gateway is connected to the same port on each VMA Array (that is, one HP VMA SAN Gateway in the pair is connected to Port 1 on each of the VMA Arrays, and the other HP VMA SAN Gateway is connected to Port 2). For the Single HP VMA SAN Gateway configuration, the HP VMA SAN Gateway is connected to Port 1 on each VMA Array.
Figure D.
AM456-9007B Rev 02 HP VMA SAN Gateway Installation and User Guide
APPENDIX E: SNMP Usage for HP VMA SAN Gateway Version G5.1.x The examples in this appendix show Release G5.1.0 SNMP MIB items accessed from a standard SNMP client. This assumes a standard HP VMA SAN Gateway with SNMP enabled using a default community string of “public” and the SNMP v2c protocol. An enterprise manager should be configurable to track these items and report changes in the enterprise. Note: You must use the MIB associated with the product version.
SNMP Configuration on the HP VMA SAN Gateway Use the snmp and snmp-server commands to configure SNMP. It defaults to on. *lab-ib-srv1 [lab-ib-srv1: master] (config) # show snmp SNMP enabled: yes SNMP port: 161 System location: System contact: Read-only community: public Traps enabled: yes Trap community: public Interface listen enabled: yes No Listen Interfaces. Trap sinks: 10.1.1.1 Enabled: yes Type: traps version 2c Community: public (default) 10.1.1.
Events for Which Traps Will be Sent New Traps in Release G5.1.x Versions of the Gateway array-data-plane-ready: Array data plane state changed. array-led-change: Array LED state changed. array-pcie-link-down: Array PCIE link down. array-pcie-link-up: Array PCIE link up. array-psu-state: Array PSU state changed. array-raid-rebuild: Array RAID rebuild state changed. chassis-temperature-high: High chassis temperature detected.
media-device-state-change: Unexpected change in media device state. media-device-unknowntype: Detected a media device of unknown type. media-system-swap-state: Detected a change in system swap state. memusage-high: Memory usage has risen too high. netusage-high: Network utilization has risen too high. unexpected-cluster-join: A node has unexpectedly joined the cluster. unexpected-cluster-leave: A node has unexpectedly left the cluster.
Trap Information Table Trap Name Threshold Value Unit Description Recommendation Severity array-data-plane-ready State Change 0 or 1 A VMA Array data plane available state change has occurred. The old value and new values are supplied. A value of 0 = not ready, 1 = ready. If the new value is 0, then check the VMA Array alarms for more information. High / Low array-led-change State Change State One or more of the LEDs on a VMA Array has changed state.
cpu-util-high 98 Percent an HP VMA SAN Gateway has detected that a CPU has exceeded utilization above the threshold level Please contact HP customer support if the system is not under full data traffic load. Low disk-space-low 0 Percent Free an HP VMA SAN Gateway disk space has crossed the threshold of percent of bytes free. Please contact HP customer support. Medium interface-down Event Triggered A network interface on an HP VMA SAN Gateway has lost link.
vimm-state-change State Change State One or more VIMMs have changed state on a VMA Array. A comma separated list of VIMMs are provided for each state: admin_down, booting, active, spare, failed, present, or alarmed. The set value for each VIMM list / state change is “true” for set or “false” for clear. Check the VMA Array for alarms to determine severity of state change.
media-device-health-warn 10 Percent an HP VMA SAN Gateway media device (VMA Array) has one or more health attributes in a warning or critical state. Normalized SMART attributes are used. The key attributes for HP Arrays are Avail Reserved Space and Media Wearout Indicator. SMART attr 232: Avail Reserved Space is the percentage of reserved blocks available (unused). This threshold is 10%. SMART attr 233: Media Wearout Indicator tracks the number of erase cycles for flash as a percentage of life remaining.
netusage-high 10485760 Bytes / sec an HP VMA SAN Gateway has detected that a network interface has exceeded a bytes per second threshold. Please verify your network configuration. For 10 GbE iSCSI, the threshold should be adjusted much higher. Low unexpected-cluster-join Event Triggered an HP VMA SAN Gateway has unexpectedly joined a vCLUSTER after the 180 seconds cluster startup time has elapsed. Check the HP VMA SAN Gateway log for any errors.
Configuring and Testing Traps Use the snmp command to configure traps, which can be sent to multiple hosts: lab-stein6-acma [hw-stein6: master] > enable lab-stein6-acma [hw-stein6: master] # conf terminal lab-stein6-acma [hw-stein6: master] (config) # snmp ? community contact Set the read-only community string Set a value for the syscontact variable in MIB-II enable Enable SNMP-related functionality host Configure hosts to which to send SNMP traps listen location Configure SNMP server interface acces
This will send a test trap to all configured trap sinks. This will be the 'testTrap' notification from the TMS-MIB. This trap is only ever sent on request from the user; it is not triggered automatically. This trap is not available for enable or disable through configuration; it is always enabled, meaning it will always be sent when requested by the user. Spare VIMMs To query the spare VIMMs in the system: [martin@lab-ib-cn2 mibs]$ snmpwalk -Ou -M +. -m +./G5.0.1_VIOLIN-MEMORY-MIB.
Temperatures: per VIMM and Chassis To query the temperatures (reported in degrees celsius) of the chassis and the VIMMs: [martin@lab-ib-cn2 mibs]$ snmpwalk -Ou -M +. -m +./G5.0.1_VIOLIN-MEMORY-MIB.txt -v 2c -c public lab-ib-srv1 violinmemory | grep -i temp enterprises.violin-memory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.temp."lab-fender098"."vimm00" = INTEGER: 39 enterprises.violin-memory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.temp."lab-fender098".
enterprises.violin-memory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.temp."lab-fender098"."vimm29" = INTEGER: 37 enterprises.violin-memory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.temp."lab-fender098"."vimm30" = INTEGER: 46 enterprises.violin-memory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.temp."lab-fender098"."vimm31" = INTEGER: 50 enterprises.violin-memory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.
enterprises.violin-memory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.temp."lab-fender098"."vimm65" = INTEGER: 44 enterprises.violin-memory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.temp."lab-fender098"."vimm66" = INTEGER: 46 enterprises.violin-memory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.temp."lab-fender098"."vimm67" = INTEGER: 46 enterprises.violin-memory.products.memoryGateway.appliance.media.arrayVimmTable.arrayVimmEntry.
Performance Stats To query performance statistics for the system: [test@host1 mibs]$ snmpwalk -Ou -M +. -m +./G5.1.0_VIOLIN-MEMORY-MIB.txt -v 2c -c public lab-vs3-n2 violin-memory | grep ata-VIOLIN enterprises.violin-memory.products.memoryGateway.appliance.media.mediaDevTable.mediaDevEntry.mediaDevIdx."ataVIOLIN_MEMORY_ARRAY_2110CR00000304" = STRING: ata-VIOLIN_MEMORY_ARRAY_2110CR00000304 enterprises.violin-memory.products.memoryGateway.appliance.media.mediaDevTable.mediaDevEntry.devId.
Fibre Channel (vSHARE) Information To query Fibre Channel information for a vSHARE configuration: enterprises.violin-memory.products.memoryGateway.vShare.globalTargetFcTable.globalTargetFcEntry.wwn.1."hba-a1" = STRING: wwn.21:00:00:24:ff:26:6a:8c enterprises.violin-memory.products.memoryGateway.vShare.globalTargetFcTable.globalTargetFcEntry.wwn.1."hba-a2" = STRING: wwn.21:00:00:24:ff:26:6a:8d enterprises.violin-memory.products.memoryGateway.vShare.globalTargetFcTable.globalTargetFcEntry.wwn.5.
Fibre Channel Performance Statistics (vSHARE) Example The following is an example of querying Fibre Channel performance statistics for a vSHARE configuration. $ snmpwalk -c public -v 2c 10.1.10.137 VIOLIN-MEMORYMIB::statsTargetFcTable.statsTargetFcEntry -m VIOLIN-MEMORY-MIB.txt VIOLIN-MEMORY-MIB::statsTargetFcIdx."hba-a1" = STRING: hba-a1 VIOLIN-MEMORY-MIB::statsTargetFcIdx."hba-a2" = STRING: hba-a2 VIOLIN-MEMORY-MIB::txFrames."hba-a1" = STRING: 1222433 VIOLIN-MEMORY-MIB::txFrames.
Event: unexpected-cluster-join: A node has unexpectedly joined the cluster.
Event: media-device-health-warn: A media device has a health status warning. vmemHealthWarning NOTIFICATION-TYPE OBJECTS { id, location, type, vmemModel, vmemSerialNo, output } STATUS current DESCRIPTION "vmemHealthWarning" ::= { vmemNotifTraps 0 11 } Event: media-device-lifetime-warn: A media device has a low estimated life remaining.
Event: media-device-state-change: Unexpected change in media device state. vmemStateChange NOTIFICATION-TYPE OBJECTS { id, oldTypeString, newTypeString, devPath, blockSz, numBlocks, offset, vmemModel, devId, expected } STATUS current DESCRIPTION "vmemStateChange" ::= { vmemNotifTraps 0 13 } Event: media-device-unknown-type: Detected a media device of unknown type.
Event: vimm-temperature-high: High VIMM temperature detected. vmemVimmTempRisingError NOTIFICATION-TYPE OBJECTS { id, value, node } STATUS current DESCRIPTION "vmemVimmTempRisingError" ::= { vmemNotifTraps 0 16 } Event: vimm-temperature-ok: VIMM temperature returned to normal levels. vmemVimmTempRisingClear NOTIFICATION-TYPE OBJECTS { id, value, node } STATUS current DESCRIPTION "vmemVimmTempRisingClear" ::= { vmemNotifTraps 0 17 } Event: chassis-temperature-high: High chassis temperature detected.
Event: chassis-temperature-ok: Chassis temperature returned to normal levels. vmemChassisTempRisingClear NOTIFICATION-TYPE OBJECTS { id, value, node } STATUS current DESCRIPTION "vmemChassisTempRisingClear" ::= { vmemNotifTraps 0 21 } Event: lid-ajar-time-rising-error: Chassis lid ajar time has exceeded a configured threshold.
Event: array-pcie-link-up: Array PCIE link up. vmemArrayPcieLinkUp NOTIFICATION-TYPE OBJECTS { id, container } STATUS current DESCRIPTION "vmemArrayPcieLinkUp" ::= { vmemNotifTraps 0 31 } Event: array-pcie-link-down: Array PCIE link down. vmemArrayPcieLinkDown NOTIFICATION-TYPE OBJECTS { id, container } STATUS current DESCRIPTION "vmemArrayPcieLinkDown" ::= { vmemNotifTraps 0 32 } Event: array-data-plane-ready: Array data plane state changed.
Event: array-raid-rebuild: Array RAID rebuild state changed. vmemArrayRaidRebuild NOTIFICATION-TYPE OBJECTS { id, container, state } STATUS current DESCRIPTION "vmemArrayRaidRebuild" ::= { vmemNotifTraps 0 34 } Event: vimm-state-change: Array VIMM state changed. vmemArrayVimmStateChange NOTIFICATION-TYPE OBJECTS { id, container, vimms, set, stateTypeString } STATUS current DESCRIPTION "vmemArrayVimmStateChange" ::= { vmemNotifTraps 0 35 } Event: array-psu-state: Array PSU state changed.
Event: array-led-change: Array LED state changed. vmemArrayLedChange NOTIFICATION-TYPE OBJECTS { id, container, alarmLedChanged, alarmLedOld, alarmLedNew, powerALedChanged, powerALedOld, powerALedNew, powerBLedChanged, powerBLedOld, powerBLedNew, statusLedChanged, statusLedOld, statusLedNew } STATUS current DESCRIPTION "vmemArrayLedChange" ::= { vmemNotifTraps 0 37 } Event: fc-port-state-change: Fibre channel port state changed.
Event: array-fan-state-change: Array fan state changed. vmemArrayFanChange NOTIFICATION-TYPE OBJECTS { id, container, fan1Changed, fan1Old, fan1New, fan2Changed, fan2Old, fan2New, fan3Changed, fan3Old, fan3New, fan4Changed, fan4Old, fan4New, fan5Changed, fan5Old, fan5New, fan6Changed, fan6Old, fan6New } STATUS current DESCRIPTION "vmemArrayFanChange" ::= { vmemNotifTraps 0 39 } Event: events-login: A user has logged in event.
Event: events-logout: A user has logged out event.
AM456-9007B Rev 02 HP VMA SAN Gateway Installation and User Guide
APPENDIX F: AM456-9007B Rev 02 Compliance Information HP VMA SAN Gateway Installation and User Guide 267
Regulatory Information For your protection, this product has been tested for conformance to various national and international regulations and standards. The scope of this regulatory testing includes electrical and mechanical safety, electromagnetic emissions, immunity, acoustics and hazardous materials. Regulatory Model Number For the purpose of regulatory compliance certifications and identification, this product is assigned a regulatory model number.
• Always wear a grounded wrist strap when handling printed circuit boards. • Treat all assemblies, components and interface connections as static-sensitive. • Avoid working in carpeted areas, and keep body movement to a minimum while removing or installing boards, to minimize buildup of static charge. Lithium Battery Caution WARNING! Observe the correct polarity when changing the lithium battery. There is a danger of explosion if battery is installed incorrectly.
Instead, it is your responsibility to dispose of your waste equipment by handing it over to a designated collection point for the recycling of waste electrical and electronic equipment. The separate collection and recycling of your waste equipment at the time of disposal will help to conserve natural resources and ensure that it is recycled in a manner that protects human health and the environment.
Japan Radio Frequency Interference VCCI Korea RFI Statement Canada RFI Statement This Class A digital apparatus complies with Canadian ICES-003. Cet appareil numérique de la classe A est conforme à la norme NMB-003 du Canada. Australia C-Tick Label Figure F.
AM456-9007B Rev 02 HP VMA SAN Gateway Installation and User Guide
Index A show cluster show cluster global 56 show cluster global brief 55 admin role 63 access privileges 63, 183 cluster management VIP 54 commands in Config mode 118 obtaining password from master node 50 command line interface (CLI) ambiguous command 114 password, setting 63 asterisk for unsaved changes 58 asterisk command completion find options with ? 115 find values with ? 116 tab completion 116 after module ID 56 before CLI prompt 58 command descriptions 120 B command list display with tab
active configuration files 57 error messages changes unsaved to 119 ambiguous commands 114 deletion of 62 begin with % in CLI 119 management of 57 global commands 118 reverting to saved configuration 62 message bar in Web Interface 178 saving 57–59 expected lifetime of media devices 98 show configuration commands 61–62 warning or critical status 99 switching 58 configuration wizard F additional Memory Gateways 49–52 Fibre Channel getting help 47–49 network connectivity 35, 39 Master Gatew
See initiator groups Master Gateway See master node initiator groups configuration of 74–75 master node 54, 188 INTERACTIVE commands 122 configuring cluster 41–49 connecting to Violin Web Interface through 177 interfaces statistics, viewing 185 bonding of 39, 148, 150–151 mDNS 40 commands 149–150 ifname CLI parameter 120 public interface 42, 148 verifying activity and status 91 media enable or disable all media 97 media health status 99 IP address CLI parameter 120 read and write statistics 101
N routing commands 152 netmask CLI parameter 120 S network connectivity 39–40 S.M.A.R.T.
initial value 63 show configuration 62 standby nodes 54 statistics privileges admin role 63, 118, 183 monitor role 63, 118, 180, 183 unpriv role 63, 117, 183 media read and write statistics 101 storage media read and write statistics 101 privileges and roles 63 show media options 92 user management 63 status of storage media 99 V vCLUSTER T configuration wizard, and 44 target ports IP addressing 38 configuring 73 vcounts command 214–217 Fibre Channel ports 73 verifying port activity and status
switches, configuration of 39 VLAN tagging 40 vpartial command 224–225 vring utility command 225–226 vSHARE architecture 66 configuration 67–70 vstat command 227–229 vupdate_tz command 229 vzero command 231 W write memory example 58 write terminal 61 Z zeroconf 150 zeroing-out data 77, 78, 107, 208 278 AM456-9007B Rev 02 HP VMA SAN Gateway Installation and User Guide
