HP 3PAR System Reporter 2.9 Software User’s Guide Abstract This guide provides the information you need to install and use HP 3PAR System Reporter to monitor performance, create charge back reports, and plan storage resources for HP 3PAR Storage Systems, and is intended for system and storage administrators who monitor and direct system configurations and resource allocation for HP 3PAR Storage Systems.
© Copyright 2011 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.
Apache HTTP Server Copyright & License Licensed under the Apache License Version 2.0, January 2004, http://www.apache.org/licenses/ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION 1. Definitions. "License" shall mean the terms and conditions for use, reproduction, and distribution as defined by Sections 1 through 9 of this document. "Licensor" shall mean the copyright owner or entity authorized by the copyright owner that is granting the License.
wherever such third-party notices normally appear. The contents of the NOTICE file are for informational purposes only and do not modify the License. You may add Your own attribution notices within Derivative Works that You distribute, alongside or as an addendum to the NOTICE text from the Work, provided that such additional attribution notices cannot be construed as modifying the License.
When we speak of free software, we are referring to freedom, not price. Our General Public Licenses are designed to make sure that you have the freedom to distribute copies of free software (and charge for this service if you wish), that you receive source code or can get it if you want it, that you can change the software or use pieces of it in new free programs; and that you know you can do these things.
Accompany it with a written offer, valid for at least three years, to give any third-party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or, Accompany it with the information you received as to the offer to distribute corresponding source code.
for this. Our decision will be guided by the two goals of preserving the free status of all derivatives of our free software and of promoting the sharing and reuse of software generally. Documentation For the latest version of this document, go to http://www.hp.com/go/3par/, navigate to your product page, click Support for your product, and then click Manuals.
Contents 1 Introduction.............................................................................................18 User Interfaces.......................................................................................................................18 Related Documentation............................................................................................................18 Typographical Conventions...................................................................................................
About Apache HTTP Server Authorization and Access Control......................................................49 Removing the System Reporter Components................................................................................50 Removing the System Reporter Tools on Windows..................................................................50 Removing the System Reporter Tools on Linux.........................................................................51 Removing the MySQL Database (Optional).....
Object Selection................................................................................................................85 Select PDIDs.................................................................................................................85 Include Chunklets..........................................................................................................85 Select Cage IDs............................................................................................................
CPU Performance Reports..................................................................................................110 Link Performance Reports..................................................................................................110 Adaptive Optimization Reports...............................................................................................111 7 Configuring the System Reporter Policy Settings..........................................
charttab..........................................................................................................................132 values:.......................................................................................................................132 example:....................................................................................................................132 charttype...............................................................................................................
value:........................................................................................................................144 example:....................................................................................................................144 selusrcpg........................................................................................................................144 selsnpcpg................................................................................................................
pd_space_time................................................................................................................153 example:....................................................................................................................153 pd_space_group..............................................................................................................154 example:....................................................................................................................
regiondensity example.................................................................................................160 regmoves........................................................................................................................160 regmoves example......................................................................................................161 vvtier..............................................................................................................................
Selecting an HP 3PAR Storage System for Adaptive Optimization................................................179 Configuring Adaptive Optimization.........................................................................................179 Tier Definition (CPG Name and GiB)..................................................................................180 Schedule........................................................................................................................182 Measurement Hours....
The The The The The The statlink Table.............................................................................................................212 statlinkinv Table.........................................................................................................213 statldrg Table.............................................................................................................213 statldrg_timeinv Table.....................................................................................
1 Introduction This guide provides the information you need to install and use HP 3PAR System Reporter to monitor performance, create charge back reports, and plan storage resources for HP 3PAR Storage Systems. An optionally licensed component called Adaptive Optimization is also included with System Reporter that can be used to monitor and automatically optimize utilization of storage resources.
Typographical Conventions This guide employs the following typographical conventions: Typeface Meaning Example ABCDabcd Used for dialog elements such as titles, When prompted, click Finish to button labels, and other . complete the installation. ABCDabcd Used for paths, filenames, and screen Open the file output. \gui\windows\setup.exe Advisories To avoid unexpected configuration or operational problems, be sure to observe the notes and cautions in this guide.
2 Overview and Features This chapter provides a brief overview of the features and capabilities offered by System Reporter. About System Reporter System Reporter monitors performance and the usage of storage resources and allows you to generate charts and graphs that report useful statistics for planning and configuring the operation of HP 3PAR Storage Systems.
• Alert Rules -- Used to add, modify and remove alert rules and to specify the parameters for generating email reports. • Scheduled Reports -- Used to add, modify, and remove scheduled reports for generation at a specified time. • Adaptive Optimization -- Configures the optionally licensed Adaptive Optimization components that are used to monitor and optimize storage utilization. See for details on how to configure the policy settings for your site.
Table 1 Available Reports (continued) Report Name Description Report Metrics Cross Reference VLUN Performance Displays VLUN (volume-LUN) performance. “Common Performance Metrics” (page 108). LD Performance Displays logical disk performance. “Common Performance Metrics” (page 108) PD Performance Displays physical disk performance. “Common Performance Metrics” (page 108) VV Cache Performance Displays VV (virtual volume) cache performance.
• Object Selection – Selects the list of objects that are to be included in the reporting data. For example, you can select the VVs, ports and hosts for a VLUN. • Format Selection – Selects the format of the report including whether the report should show charts and/or tables and the type and size of charts. See for complete details on how to build customized reports.
together in a single web page. You can do this by creating an html file that includes multiple iframe elements, each of which includes a standard web report query. See “Web Queries” (page 126)for details on how to instigate web queries. Using the Database Schema to Build Custom Reports The databases used by System Reporter are standard relational databases capable of supporting SQL queries. The default (built-in) database is SQLite (see http://www.sqlite.
3 Installation, Configuration, Update, and Removal This chapter describes how to install and configure as well as remove the HP 3PAR System Reporter components. Before You Begin Before installing HP 3PAR System Reporter components, select a system on which to run the System Reporter sampler and Web server. This system must use Windows® Server 2003, Windows® Server 2008 or Red Hat® Enterprise Linux 5.
Table 2 InServ Storage Server CD Contents (continued) Directory /Windows/ CLI/ Filename Description installer.exe System Reporter installer executable for Microsoft Windows. setup.exe HP 3PAR InForm OS CLI installer executable for Microsoft Windows. System Requirements HP 3PAR System Reporter supports the use of the HP 3PAR InForm OS, versions 2.2.4 through 2.3.1 MU1 and beyond.
Table 4 HP 3PAR Storage System Requirements (continued) Component Supported Operating Systems Additional Requirements Adaptive Optimization InForm OS version 2.3.1 MU1 and beyond. Not supported when using SQLite database. On-Line Help (User’s Guide) Adobe Acrobat Reader NOTE: The optionally licensed Adaptive Optimization component is only supported by InForm OS versions 2.3.1 MU 1 and beyond. Adaptive Optimization is not supported for SQLite databases; only Oracle and MySQL are supported.
Figure 1 System Reporter Sizing Spreadsheet 1. Click on the appropriate type of database that you are using for System Reporter at your site. You will notice that SQlite only supports one HP 3PAR Storage System for use with System Reporter. Please consider another database if you plan on using System Reporter to monitor more than a single HP 3PAR Storage System. 2. Modify the Sampling Parameters in red to reflect the approximate values for your installation.
Figure 2 HP 3PAR Storage System and User Interface Components • Database. The HP 3PAR System Reporter database is capable of supporting SQL queries. Four types of databases are supported: Microsoft SQL Server, SQLite, MySQL and Oracle (see “Installing the Database Server” (page 32)). • Sampler components.
• Web server components. The HP 3PAR System Reporter Web server has customized scripts that access the database to generate reports to standard browsers as clients, enabling users to access the reports and configure System Reporter without installing any additional software. • Excel client. The customized HP 3PAR System Reporter Excel client has built-in Web queries to the Web server. Use of the Excel client is optional, but offers you additional options.
Figure 3 Installation Process Flowchart Installing the Web Server HP 3PAR System Reporter requires Apache HTTP Server, version 2.2 or later. A copy of the installer executable for Windows of Apache HTTP Server, version 2.2, is located on the HP 3PAR System Reporter CD or you can download the most recent version from www.apache.org. Red Hat Linux typically includes the Apache Web Server.
NOTE: System Reporter 2.9 and beyond supports secure connections on Apache servers using the SSL protocol. For details on implementing the SSL protocol on Apache servers, please consult the following links: http://httpd.apache.org/docs/2.2/ssl/ssl_intro.html http://httpd.apache.org/docs/2.2/ssl/ssl_howto.html http://httpd.apache.org/docs/2.1/ssl/ssl_faq.
the sampler or the Web server can time out. Furthermore, database compaction (see “Compacting Databases” (page 113)) cannot be done in parallel with sample insertion so there could be a large gap in samples during database compaction. Sampling of multiple HP 3PAR Storage Systems is not supported with an SQLite database, either the Microsoft SQL Server, MySQL or the Oracle database servers should be used instead.
Table 5 Database Comparison (continued) Comment SQLite Microsoft SQL MySQL Oracle Scale to large number of No concurrent users? Yes Yes Yes Performance for sample insertion and deletion? Good High Good Adequate for small databases Installing and Setting Up the Microsoft SQL Server Database (Optional) To install the Microsoft SQL Server Database, you must create the database and then set up a valid login and user. Consult the appropriate Microsoft SQL Server documentation for details.
1. Select Administrative Tools→Data Sources (ODBC). The ODBC Data Source Administrator screen appears: Figure 5 ODBC Data Source Administrator Screen 2. 3. Select the System DSN tab. Click Add. The Create New Data Source screen appears: Figure 6 Selecting the data source for the SQL driver 4. Choose SQL Server Native Client 10.0. Any SQL server ODBC driver will work. However, the SQL Server Native Client ODBC is supplied with the Microsoft SQL Server client and is considered to be the most efficient.
Figure 7 Create a New Data Source to SQL Server Screen Note that the screens appear somewhat differently depending on whether you have chosen to use the default version of the SQL ODBC. 6. Click Next.
7. Enter the user's log in and password information, then click Next. Figure 9 Specifying the MSSQL Database Name 8. 9. Check the Change the default database check box, and enter the name of the MSSQL database that has been created for System Reporter. Click Next. Figure 10 Checking the Details for the Database 10. Click Finish.
Figure 11 ODBC Microsoft SQL Server Setup Screen. 11. Check the details for the selected database, then click Test Data Source. The screen must show the following (Figure 12 (page 38)): Figure 12 SQL Server ODBC Data Source Test Screen If the screen does not show the appropriate information, please run the configuration again. Installing and Setting Up the MySQL Database (Optional) To use MySQL, you must install MySQL Database Server, version 5.1, following the instructions provided in this section.
NOTE: Beginning with the 2.2 release, System Reporter can use a MySQL database on a remote server. The MySQL installation and configuration steps are the same for the remote server. See section for configuring System Reporter to use a remote MySQL database. This type of configuration is highly recommended. After downloading the MySQL executable, use the MySQL Server Setup Wizard to set up and configure MySQL as follows: 1. Open the MySQL executable. 2. When prompted, click Next. 3.
10. Confirm that Modify Security Settings is selected, enter a root password, and then retype the password to confirm. CAUTION: Be sure to record the password for later reference. 11. Make sure that Create An Anonymous Account is not selected and then click Next. 12. Click Execute. 13. When prompted, click Finish. CAUTION: System Reporter only supports and creates MyISAM tables. Conversion of these tables to use any other type of storage engine is not supported.
These users have different privilege levels that correspond with optional Apache HTTP Server privilege levels, as described in “Installing or Upgrading the System Reporter Tools” (page 43). • cliuser is the default user name for the sampler and policy change user. This user can create and edit database tables and change database sampling policies (see “Editing Sampling Policies” (page 114)). • webuser is the default user name for the Web reports user. This user has database read privileges only.
10. After clicking Finish, check the settings in the confirmation screen and click OK. 11. Click Exit after the database has been created. Creating the Oracle Users for System Reporter Use the instructions that follow to create two users: cliuser and webuser. These users have different privilege levels that correspond with optional Apache HTTP Server privilege levels, as described in “About Apache HTTP Server Authorization and Access Control” (page 49).
2. 3. 4. 5. 6. 7. Start the Net Configuration Assistant. Select the Local Net Service Name Configuration radio button and click Next. Select the Add radio button and click Next. Enter the service name for the database according to the directions and click Next. Select TCP protocol and click Next. Enter the host name for the database server along with the appropriate port number and click Next. 8. Select Yes, perform a test and click Next. The test may fail if the default user and password are not set.
3. 4. 5. 6. 7. When prompted, use the Browse... button to select and enter the full path to the InForm CLI executable cli.exe (if it is different from the default) then click Next. When prompted, use the Browse... button to select and enter the full path to the location where you installed Apache HTTP Server (if it is different from the default) then click Next. Use the Browse...
Alerts” (page 116)) you may leave all entries in this screen blank. If you want to be able to configure System Reporter to generate email alerts, you must enter the IP address or name of the SMTP server, the user name and password if the SMTP server requires authentication, and the SMTP originator name (the mailbox name that the email appears to be from). This information is stored in the sampleloop_config.tcl file in the System Reporter installation directory. Click Next to continue.
Installing System Reporter Tools on Linux System Reporter tools on Linux consists of two RPM packages available on the CD: • sampleloop (sampleloop-2.9-1.i386rpm) consists of the files needed to install the sampleloop daemon. • sysrptwebsrv (sysrptwebsrv-2.9-1.i386.rpm) consists of the files needed to install the System Reporter web server code. Both these packages must be installed after logging in to a terminal session on the server as the root user.
6. Install the sysrptwebsrv package by mounting the System Reporter CD, changing to the directory on the CD that includes the rpm packages and running: rpm --install sysrptwebsrv-2.9-1.i386.rpm 7. Edit the /var/www/cgi-bin/3par-rpts/config.tcl and /var/www/cgi-bin/ 3par-policy/config.tcl files following the directions in those files. If previous versions of the file were saved, you may keep the previous settings and simply replace the files with the saved versions.
• Days to keep Daily Data -- Specifies the number of days to retain daily data. • Days to keep Adaptive Optimization Data -- Specifies the number of days to retain Adaptive Optimization data. 10. Click Submit Query. NOTE: Since performance data for LDs can increase the database size substantially you should check the Skip LD Performance data box unless you intend to configure Adaptive Optimization for the system.
NOTE: Use of the HP 3PAR System Reporter Excel client requires that you have Microsoft® Excel® 2003 or 2007 software already installed on your local machine. See for additional details. The HP 3PAR System Reporter Excel client is contained in a single Microsoft® Excel® workbook with the file name 3PAR System Reporter.xls. After downloading this file, you can save it in any convenient location. To download the HP 3PAR System Reporter Excel client: 1.
distinction between users with database and sampling policy read access and those with database sampling policy edit privileges. Note that, when adding Directory directives to the httpd.conf file, you must also create a password file {password_file}, as described in your Apache documentation.
6. 7. 8. 9. Using Windows Explorer, go to the location where you installed the HP 3PAR System Reporter tools (for example, C:\Program Files\3par\System Reporter) and delete the entire directory. Locate and remove the inservstats.db file if the file was in a location other than the System Reporter folder. Also remove the file inservstats.db-journal if present in the same folder.
4 Getting Started This chapter describes how to start using HP 3PAR System Reporter from the primary web interface or Excel client. NOTE: For information on using a Web browser to edit database sampling policies, add or remove storage servers from the database, schedule reports or optimize storage utilization with the optional Adaptive Optimization feature, see “Configuring the System Reporter Policy Settings” (page 112).
Figure 14 The System Reporter Main Window The HP 3PAR System Reporter Main Menu (Figure 14 (page 53)) is divided into the following areas: • Title Pane which shows the version of System Reporter. • Main Menu tabs which provide options for generating quick reports, accessing scheduled reports and creating customized reports. (See for details). • Extras Menu area has links to: ◦ Excel Client used to create and access custom reports.
◦ User Guide (this document in PDF format). ◦ Sampling Status (the time since the most recent high-resolution sample for each system being sampled). NOTE: Instead of opening a new window, browsers can be configured to open a new tab instead. For Firefox, open the options window (Tools→Options), click on the Tabs tab, and select the option that opens new pages in a new tab.
The list of available objects can be filtered by typing an expression pattern in the text input area above the list of available objects and clicking on the Filter button. CAUTION: Filtering large numbers of objects may exceed the limit for your server, in which case you should reduce the number of objects being filtered and try again.
Figure 15 System Reporter Excel Client Setting the Security Level for Excel The HP 3PAR System Reporter Excel client uses macro code to implement its functionality. In order for this code to be executed, you must set the proper security settings depending on the version of Excel that you are using. Setting the Security for Excel 2003 In most cases, when accessing the System Reporter client for Excel 2003, a Security Warning screen appears indicating that macros have been disabled (Figure 16 (page 56)).
Figure 17 Setting the Security Level 3. On the Security Level tab, set the security level to Medium or Low (Medium is the default and suggested setting) and then click OK. NOTE: The HP 3PAR System Reporter Excel client will not work if the Microsoft® Excel® security level is set to High or Very High. 4. If you changed the security level setting, close the Excel workbook and then reopen it.
1. Click the Options button next to the security alert. The Security Alert - Macro screen appears (Figure 18 (page 58)). Figure 18 Macro Security Alert for Excel 2007 2. 3. Choose the Enable this content radio button. Click OK. If you do not see a Security Warning when accessing the Excel client, it is likely that the main Excel Macro Settings are not set up properly and need to be changed. In this case, follow these steps: 4. 5. 6. 7.
Figure 19 Trust Center Screen 8. If it is not already set, from the list of Macro Settings, choose Disable all macros with notification. NOTE: The HP 3PAR System Reporter Excel client will not work if the Macro Setting is set to Disable all macros without notification or Disable all macros except digitally signed macros. 9. Click OK, then click OK again to return to the System Reporter Excel client main screen. 10. Close the Excel client, then launch it again from the System Reporter main menu. 11.
1. After opening the HP 3PAR System Reporter Excel client workbook with the correct security level setting, click Initialize. The Connection Parameters dialog box appears (Figure 20 (page 60)). Figure 20 Connection Parameters Dialog Box 2. In the Server Hostname box, type a host name and click OK. The host name is the name of the Web server where the Apache HTTP Server and HP 3PAR System Reporter Web server scripts are installed (see “Before You Begin” (page 25)).
Changing the Excel Client Server and Resetting the Workbook The HP 3PAR System Reporter Excel client workbook includes a custom 3PAR menu that is accessible from the Microsoft® Excel® menu tab called Add-Ins. This menu tab enables you to change the Web server, reset the workbook, and delete menu sheets. Changing the Web Server To change the Web server used for querying the System Reporter database: 1. Click the Add-Insmenu tab, choose the 3PAR pull down menu, and then select Set server name....
Deleting Data Sheets Data sheets remain until you manually delete them. Manually deleting a data sheet causes any embedded charts associated with that data sheet to disappear as well. To delete a data sheet: 1. At the bottom of the Microsoft® Excel® window, click a data sheet tab to enable the sheet to be deleted. 2. Point to the name of the data sheet tab and then right-click. 3. On the shortcut menu that appears, click Delete. 4. In the confirmation dialog that appears, click OK.
Figure 22 Security Alerts Multiple Issues To enable macros and the data connection, choose the Enable this content buttons. NOTE: If it is ever necessary, you can obtain an entirely new copy of the HP 3PAR System Reporter Excel client workbook by following the instructions provided in “Installing the Excel Client (Optional)” (page 48).
5 Accessing and Creating Reports This chapter describes how to access quick reports, scheduled reports, and build custom reports that are specifically tailored for your operating environment. The menu selections and controls available for creating custom reports from either the web interface or the Excel client are functionally identical, so the information in this chapter applies to both with explanations provided for any minor variations.
Figure 23 Quick Reports Window Accessing Scheduled Reports Scheduled reports are generated using pre-selected parameters at scheduled times and are stored in sub-directories whose structure is reflected in the tree shown in Figure 24 (page 66). Scheduled reports are stored with a name that is assigned when the report is scheduled. Each time the report runs, a new subdirectory is created for that instance of the report using the timestamp as the name.
Figure 24 Scheduled Reports Window NOTE: See“Scheduling Reports” (page 122) for complete details on scheduling a report. Creating Custom Reports You can create customized reports by selecting the Custom Reports tab from the Main Menu or by choosing the Excel Client from the Extras Menu. NOTE: The web interface and Excel client for creating custom reports are identical except that the Excel client charting functions are accessed separately from the main interface and provide some additional flexibility.
Building the Report Menu The first step in creating a report is to choose the appropriate options for the type of report that you want to generate. Figure 25 Custom Reports Window NOTE: The layout and placement of controls may vary slightly for the Excel client but operate in the same manner as described for the web interface.
Table 6 Sample Resolution Option Purpose Daily Choose the Daily data table, which covers the largest time range, but has the coarsest resolution with only one sample per day. Hourly Choose the Hourly data table, which covers a shorter time range than the daily data table, but has one sample per hour. High-Res Choose the High-Resolution data table, which spans an even smaller time range, but has the finest granularity.
Table 8 Select System Option Purpose -All Systems- The report will include all the systems that are included in the database. list of specific system names Select one or more systems to include in the report. Multiple systems can be selected. NOTE: When selecting systems for System Reporter where there are multiple HP 3PAR Storage Systems available for reporting, consider limiting the number of systems that you select to avoid having to wait for longer report generations.
Figure 26 Report Menu for Daily VLUN Performance The top of the report menu includes a section that lists the current selection of systems of domains. In this example, the menu includes only system named unicorn and all domains. For most reports, the controls in the report menu are divided into four sections: • 1. Report Selection: controls the type of report that will be generated, see “Report Selection” (page 76). • 2.
Use the controls in the menu to select the desired report parameters and then click Generate Report. See for a listing of the controls that are available for each report and then consult the relevant sections later in this chapter for complete details. CAUTION: Filtering large numbers of objects may exceed the limit for your server, in which case you should reduce the number of objects being filtered and try again.
Figure 27 Query Results If necessary, use the scrollbox on the right to view all charts and/or tables. Controls Available by Report Table 10 (page 73) lists the set of controls that are valid for generating each report. The entries have the following meanings: 72 • A blank entry indicates that the control is not valid for report. • A Y entry indicates that the control is valid for all report types. • A T entry indicates that the control is valid for the report but only for Versus Time report types.
Table 10 Controls for Each Report Report: Summay r PD CPG LD Section and Space Space Space Control VV PD Space Perf Port Perf LD Perf VLUN Perf VV Node CPU Cache Cache Perf Perf Perf Link Perf Adaptive Optimization Main Menu “Sample Y Resolution” (page 67) Y Y Y Y Y Y Y Y Y Y Y Y Y “Select Systems” (page 68) Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y “Select Domains” (page 69) Y Report Selection “Standard Report Types” (page 76) Y Y Y Y Y Y “Compa
Table 10 Controls for Each Report (continued) Report: Summay r PD CPG LD Section and Space Space Space Control VV PD Space Perf Port Perf LD Perf VLUN Perf VV Node CPU Cache Cache Perf Perf Perf Link Perf Adaptive Optimization “End Time” (page 85) T T T T T T T T H H H H T T T “Count” (page 85) Y T Object Selection “Select PDIDs” (page 85) Y Y “Include Chunklets” (page 85) Y Y “Select Cage IDs” (page 86) Y Y “Select Disk Types” (page 86) Y Y “Select Disk Speed” (page 8
Table 10 Controls for Each Report (continued) Report: Summay r PD CPG LD Section and Space Space Space Control VV PD Space Perf “Select VVs” (page 89) Y “Select Prov Types” (page 89) Y “Select VV Types” (page 89) Y Port Perf “Select Hosts” (page 91) Y “Select Port Types” (page 90) Y “Select Port Types” (page 90) Y “Select Port Rates” (page 90) LD Perf Y VLUN Perf VV Node CPU Cache Cache Perf Perf Perf Y Y Link Perf Adaptive Optimization Y Y “Select Ports (n:s:p)” (page 90) “Select
Table 10 Controls for Each Report (continued) Report: Summay r PD CPG LD Section and Space Space Space Control VV PD Space Perf Port Perf LD Perf VLUN Perf VV Node CPU Cache Cache Perf Perf Perf Link Perf Adaptive Optimization “Chart Y Pixels” (page 96) Y Y Y Y Y Y Y Y Y Y Y Y Y “Time Labels” (page 96) T T T T T T T T T T T T “Legend Position” (page 96) Y Y Y Y Y Y Y Y Y Y Y Y “Time Format” (page 96) T T T T T T T T T T T T “Select Space Unit” (pa
Table 11 Standard Report Types (continued) Report Type Purpose The beginning of the time range is specified by the Begin/At Time control (see “Begin/At Time” (page 84)) and the end of the time range is specified by theEnd Time control (see “End Time” (page 85)). See “Histogram Versus Time Report Example” (page 80). Histogram At Time Display histogram of metrics at the time specified by the Begin/At Time control (see “Begin/At Time” (page 84)).
Figure 29 Versus Time with Compare=System Report Example If you want to compare the performance of a particular class of objects (for example, systems), then you can use the Compare control (see section “Compare” (page 81)) available in some reports if the Report Type is Versus Time. Figure 29 shows an example where systems are being compared.
Figure 30 Group By System:Host At Time Report Example The At Time report can be used in conjunction with the Group By and Order By controls to do more detailed comparisons at a specific point in time. Figure 30 shows an At Time report with System and Host selected in the Group By control and IOPs/s selected in the Order By control. This allows us to easily compare the performance of hosts on different systems.
Figure 31 Histogram Versus Time Report Example The Histogram Versus Time report provides a histogram or service times for the aggregate of the selected objects as shown in Figure 31 “Histogram Versus Time Report Example” (page 80). There is a separate line per service time bucket.
Figure 32 Histogram At Time Report Example The Histogram At Time report provides a histogram or service times and IO sizes at a specific time for the aggregate of the selected objects as shown in Figure 32 “Histogram At Time Report Example” (page 81). Select Summary Columns This control is valid only for the Summary report and selects which columns are displayed in the Summary report.
Select Peak control (see section “Max. Number” (page 82)) can be used to select 16 objects to compare. Max. Number Specifies the maximum number of graphs that will be generated for a given report. This option is only valid when a compare is requested and when set to none, this option is ignored. Select Peak For some reports, when the number of objects to be compared (see section “Compare” (page 81)) exceeds 16, the Select Peak control can be used to select which 16 objects to compare.
Table 12 Group By (continued) Option Purpose PortNode Group by port’s node number (The n in n:s:p). n:s:p Group by the port (n:s:p). Order By This control is only valid for At Time reports and specifies the order of the X (category) axis for the reports. The menu options available depend on the report selected and are listed with each report. One or more Order By options can be selected (use CTRL+CLICK to select multiple options) and the default if none are selected is System.
Table 13 Order By (continued) Option Purpose SelChunkSize Order by total of the selected chunklets for PD space reports. IOPs/s Order by total IOPs. Bandwidth Order by total Bandwidth. Service Time Order by total service time. IO Size Order by total IO Size. Queue Length Order by queue length. Busy Percentage Order by the percentage of time that the VLUN is busy (i.e., has at least one outstanding IO operation).
Table 14 Begin/At Time Option Purpose earliest Begin with the earliest sample available. a list of specific time values Begin at the specified time value. The values in the menu are populated based on the actual samples in the selected data table. a list of relative time values Begin at the specified time before the current time.
Table 18 Include Chunklets Option Purpose -All Chunklets Types- Show all the chunklet types in the report. list of specific chunklet types Select one or more chunklet types to include in the report. Multiple chunklet types can be selected using (CTRL+CLICK). See Table 44 (page 106) for a list of chunklet types and their description. Select Cage IDs This is a filtered multi-selection control (see “Filtered Multi-Selection Control” (page 54)) used to select one or more Cage IDs to include in the report.
Table 22 Include PDs in States Option Purpose -All PD States- Include all PD in the report. valid Include PDs in a valid state in the report. Not valid Include PDs not in a valid state in the report. Select Usr CPGs This is a filtered multi-selection control (see “Filtered Multi-Selection Control” (page 54)) used to select one or more Usr CPGs to include in the report. This control is available for VV Space reports. NOTE: For InForm OS releases prior to 2.3.
Select LDs This is a filtered multi-selection control (see “Filtered Multi-Selection Control” (page 54)) used to select one or more LDs to include in the report. This control is available for LD Space and LD Performance reports. Table 26 Select LDs Option Purpose -All LDs- The report will include all the LDs that are included in the database. list of specific LDs Select one or more LDs to include in the report. Multiple LDs can be selected.
Table 29 Select LUN Usage Option Purpose -All LUNs- The report will include all the LUNs that are included in the database. list of specific LUNs Select one or more LUNs to include in the report. Select VVs This is a filtered multi-selection control (see “Filtered Multi-Selection Control” (page 54)) used to select one or more VVs to include in the report. This control is available for VV Space, VLUN Performance and VV Cache reports.
Table 32 Select VV Types Option Purpose -All VV Types- The report will include all the VV Types that are included in the database. list of specific VV Types Select one or more VV Types to include in the report. Multiple VV Types can be selected using (CTRL+CLICK). VV Types include: • base - Base volume (not a copy) • pcopy - Physical (full) copy • vcopy - Virtual copy (snapshot) Select Port Types This control is only available for Port Performance reports.
Select Hosts This is a filtered multi-selection control (see “Filtered Multi-Selection Control” (page 54)) used to select one or more Hosts to include in the report. This control is available for VV Space and VLUN Performance reports. Table 36 Select Hosts Option Purpose -All Hosts- The report will include all the Hosts that are included in the database. list of specific Hosts Select one or more Hosts to include in the report. Multiple Hosts can be selected.
Report Information For each report except for the Summary Report, you can use the Report Information pull-down list to choose between displaying query results in the form of a chart, a table, or both. Tables display query results in tabular format, as in the example shown in Figure 33 (page 92). Figure 33 Example of a Table Chart Lib Allows you to optionally select the charting package used by System Reporter.
You can choose from the following chart types: • Stacked Bars Figure 34 Chart Using Stacked Bars • Stacked Vert Bars Creating Custom Reports 93
Figure 35 Chart Using Stacked Vertical Bars • Stacked Area Figure 36 Chart Using Stacked Areas • 94 Lines Accessing and Creating Reports
Figure 37 Chart Using Lines • Bars Figure 38 Chart Using Bars Creating Custom Reports 95
• Vert Bars Figure 39 Chart Using Vert Bars Chart X Pixels You can explicitly specify the length in pixels of the X (category) axis of charts. If this is left blank the X axis length will be automatically chosen. If the length specified is too small, the chart may not be drawn. Chart Y Pixels You can explicitly specify the length in pixels of the Y (value) axis of charts. If this is left blank the Y-axis length will be automatically chosen.
Table 41 Time Format Option Purpose Auto Time format is chosen automatically based on the sample resolution. If resolution is Daily, the format includes the month, date, hours and minutes, for example: 11-29 08:32. If the resolution is hourly, the format includes the date, hours and minutes, for example: 29 08:32. If the resolution is High-Res, the format include date, hours, minutes and seconds, for example: 29 08:32:05.
Figure 40 Change Chart Type Menu for the Excel Client 4. 5. Choose the appropriate selection from the list of chart types, then select on the chart that you want to use to display the report data. Click OK. CAUTION: The Excel Chart controls are configured for the default Chart Type. If you change the default Chart Type, you may also need to modify other chart controls for optimal viewing.
1. Right-click in any data cell and click Data Range Properties in the shortcut menu that appears. Figure 41 Data Range Properties Shortcut Menu The External Data Range Properties dialog box appears. 2. Select the Refresh every checkbox and enter the desired value. Figure 42 External Data Range Properties Dialog Box 3. Click OK.
It is also possible to retain a chart for further reference. This can be done by moving the chart and the data sheet linked to that chart to new worksheets. See the instructions that follow for a detailed description of how to complete this procedure. To retain a chart for further reference 1. On the chart to be retained, right-click to bring up the shortcut menu. 2. On the shortcut menu, click Location (for Excel 2003) or Move Chart (for Excel 2007).
NOTE: The Visual Basic code for the Excel client is password protected to prevent accidental modification. For customers who would like to add their own code, the password is “password”. Please observe caution when considering using another password other than the default, since subsequent access would not be possible if knowledge of the altered password were to become unavailable for some reason.
To create a custom chart with the Percent spare used column in Excel 2003: 1. On the Dly PDSpace (1) sheet, click and drag to select the Time column (header cell and all data cells) and then CTRL+CLICK and drag to select the Percent spare used column. 2. On the Microsoft Excel 2003 main menu, click Insert and then Chart to bring up the Chart Wizard. 3. Use the chart wizard to create a new chart. a.
6 Interpreting Report Metrics This chapter describes the metrics that are shown for each of the various reports generated through System Reporter. Summary Report Figure 43 (page 103) shows a typical example of the metrics that are provided by a Summary Report. Figure 43 Summary Report Metrics Table 43 (page 103) describes all of the metrics available for Summary Reports.
Table 43 Summary Report Metrics (continued) Metric Description TPVVs Shows all TPVVs. Virtual Copies Shows all virtual copies. Host Ports Number of Host Ports on which VLUNs are exported. FC Host Ports Number of host ports. iSCSI Host Ports Number of iSCSI host ports. Disk Ports Number of disk ports. RCIP Ports Number of RCIP ports. RCFC Ports Number of RCFC ports. Free Ports Number of free ports. FC 15K RPM PDs Number of FC 15K physical disks.
Table 43 Summary Report Metrics (continued) Metric Description PD Total IOSz KByte/s Total (reads + writes) average IO size in KBytes to PDs. PD Total Queue Length Total (reads + writes) queue length to PDs. Free PD Capacity GiB Free PD space available. Shown by default. Used PD Capacity GiB Used PD space. Total PD Capacity GiB Total PD space available. Shown by default. Free (FC 10,15K), (NL 7k), (SSD) Free space for a specified drive in GIBs.
Space Reports This section describes the metrics that are shown for the Physical DIsk (PD), Common Provisioning Group (CPG), Logical Disk (LD), and Virtual Volume (VV) space reports used to determine storage capacities and usage. PD Space Reports PD (Physical Disk) Space reports show the amount of physical disk space composed of chunklets according to the usage types shown in Table 44 (page 106).
Table 45 CPG Space Usage Types (continued) Usage Type Description Admin Free Space GiB Amount of admin free space available for CPG allocation. Snap Free Space GiB Amount of raw snap space available for CPG allocation. User Free Space GiB Amount of user free space available for CPG allocation. Total Free Space GiB Total amount of free space available for CPG allocation. Admin Space GiB Amount of remaining Admin space available for currently allocated CPGs.
Table 47 VV Space Report Metrics (continued) Metric Description Snap Rsvd Space Space that is reserved for Snap Space for the selected VVs (Snap Used Space + Snap Free Space). This is zero for Virtual Copy (snapshot) VVs. Snap Vcopy Space Snap Space used by Virtual Copy (snapshot) VVs only. This is zero for all VVs that are not Virtual Copies (snapshots). Note that this value is calculated on the HP 3PAR Storage System by running the updatesnapspace command. System Reporter does not run this command.
Table 48 Performance Metrics for PD, Port, LD and VLUN Performance Reports (continued) Metric Description Total Svct ms Average total (read + write) service time in millisec. Read IOSz KBytes Average size of read operations in KBytes. Write IOSz KBytes Average size of write operations in KBytes. Total IOSz KBytes Average size of read and write operations in KBytes. Queue Length Queue length at the sample time.
Table 50 Node Cache Performance Report Metrics Metric Description Read Hits Number of reads that hit in the cache. Read Misses Number of reads that miss in the cache. Read Total Total number of reads. Not shown in charts. Write Hits Number of writes for which the page is already in cache AND is dirty (i.e., has previously written data that has not yet been flushed to disk). Write Misses Number of writes that miss in the cache.
Table 52 Link Performance Report Metrics Metric Description Transfers/s Transfers per second on the link. Bandwidth Bandwidth used on the link. Transfer size Average size of the transfers on the link. Adaptive Optimization Reports Table 53 (page 111) describes the metrics that are shown for Adaptive Optimization Reports. Table 53 Adaptive Optimization Report Metrics Metric Description IO Rate Density Shows IO rate density for a given region.
7 Configuring the System Reporter Policy Settings This chapter describes how to edit HP 3PAR System Reporter policies, such as editing the s ampling policies, adding or removing storage servers from the database, and scheduling reports. These actions can only be performed using a Web browser and cannot be performed using the System Reporter Excel client. The end of the chapter also describes how to edit the Apache HTTP Server configuration files, which hold the Apache HTTP Server usernames and passwords.
Table 54 Sampling Policies (continued) Policy Default Default number of hours to keep hi-res samples (in force by default for all systems 24 unless specified otherwise on an individual basis) Default number of days to keep hourly samples (in force by default for all systems 7 unless specified otherwise on an individual basis) Default number of days to keep Adaptive Optimization hourly samples (in force 7 by default for all systems unless specified otherwise on an individual basis) Default number of days t
} ) NOTE: Please note that System Reporter does not support using Microsoft SQL with Linux systems. Editing Sampling Policies To edit a System Reporter database sampling policy: 1. Access HP 3PAR System Reporter by pointing your browser at the Web server where Apache HTTP Server and the HP 3PAR System Reporter Web server scripts are installed. The 3PAR System Reporter window appears. 2. 3. In the Extras Menu area, select Policy Settings.
Figure 46 Connection Dialog Box NOTE: See for more instructions on setting Apache HTTP Server to require a username and password in order to edit sampling policies. 6. In the New Value box, type a new value and then click Submit Query. The Sampling Policies window appears with the new value. 7. When finished editing policies, close the Sampling Policies window.
2. Click on Policy Settings in the Extras Menu area. The System Reporter Policies window appears. 3. 4. 5. 6. 7. Choose the InServ Systems tab. Click Remove InServ. Choose the storage server from the drop-down list and then click Submit Query. An alert window appears confirming that the system was removed. Click OK to return to the Sampling Policies window. NOTE: If you want to temporarily disable sampling of a storage server, you do not need to remove it from the list of systems being sampled.
Adding an Alert Rule To add an alert rule: 1. Point your browser at the Web server where Apache HTTP Server and the HP 3PAR System Reporter Web server scripts are installed. The 3PAR System Reporter main window appears. 2. Click on Policy Settings in the Extras Menu area. The 3PAR System Reporter Policies window appears. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. Choose the Alert Rules tab. Click Add Alert. Choose the Data Table to which the rule applies from the drop-down list.
Removing an Alert Rule To remove an alert rule: 1. Point your browser at the Web server where Apache HTTP Server and the HP 3PAR System Reporter Web server scripts are installed. The 3PAR System Reporter main window appears. 2. Click on Policy Settings in the Extras Menu area. The 3PAR System Reporter Policies window appears. 3. 4. 5. 6. 7. 8. Choose the Alert Rules tab. Click Remove Alert. Select the alert rule to remove from the drop-down list. Click Submit Query.
Table 55 Data Table (continued) Data Table Description statpd_ssd150sum Represents the aggregation of all the SSD PDs of speed 150 in the HP 3PAR Storage System. statld Logical Disk performance table. statvlun VLUN performance table. statcpu CPU performance table. Resolution The resolution of data table on which the rule operates. This also implicitly determines how often the rule is evaluated because the rule is evaluated after each sample at the specified resolution.
Table 57 Metric (continued) Metrics Description Data Tables nl statpd_ssd150sum, statld, statvlun read_kbps Read KBytes/sec statport, statport_nodesum, statport_syssum, statpd, statpd_nl7, statpd_fc10, statpd_fc15, statpd_ssd150, statpd_sum, statpd_nl7sum, statpd_fc10sum, statpd_fc15sum, statpd_ssd150sum, statld, statvlun nl nl nl nl nl nl write_kbps Write KBytes/sec statport, statport_nodesum, statport_syssum, statpd, statpd_nl7, statpd_fc10, statpd_fc15, statpd_ssd150, statpd_sum, statpd_nl7s
Table 57 Metric (continued) Metrics Description Data Tables idle_perc Percentage idle time for the node statcpu total_utilperc Total utilization per second. For instance, when the PD utilization exceeds a given high (say 90%) utilization meaning that 90% of the time there is some IO operation outstanding at the PD.
Figure 48 Example of an alert email The email shows that a single VLUN instance had an average total service time of 2098.800680 millisec during a high-resolution sampling interval. Scheduling Reports You can have a predefined report run according to a predetermined schedule by choosing the Scheduled Reports tab from the System Reporter Policies window (Figure 49 (page 122)).
Adding a Scheduled Report To add a scheduled report: 1. Point your browser at the Web server where Apache HTTP Server and the HP 3PAR System Reporter Web server scripts are installed. The 3PAR System Reporter main window appears. 2. Click Policy Settings in the Extras Menu area. The System Reporter Policies window appears. 3. Choose the Scheduled Reports tab. The Scheduled Reports window appears. 4. 5. Click Add Scheduled Report. Specify the report directory.
Removing a Scheduled Report To remove a scheduled report: 1. Point your browser at the Web server where Apache HTTP Server and the HP 3PAR System Reporter Web server scripts are installed. The 3PAR System Reporter main window appears. 2. Click Policy Settings in the Extras Menu area. The System Reporter Policies window appears. 3. Choose the Scheduled Reports tab. The Scheduled Reports window appears. 4. Click Remove Scheduled Report. The Select Scheduled Report to Remove window appears. 5. 6. 7.
-rwxr-xr-x 1 apache apache 6717686 Jan 21 23:49 inserv_perf.exe -rwxr-xr-x 1 apache apache 1211700 Jan 21 23:49 chart.exe ls -ltra /var/www/cgi-bin/3par-policy -r--r--r-- 1 apache apache 1208 Sep 22 2010 config.tcl drwxr-xr-x 3 apache apache 4096 Jan 21 17:31 . -rwxr-xr-x 1 apache apache 6683099 Jan 21 23:49 inserv_policy.exe You must make sure that the files are owned by Apache. If necessary, you can use the chown command to change permissions for the files.
8 Web Queries This chapter documents the queries that the HP 3PAR System Reporter interfaces make to the System Reporter Web server. About Web Queries System Reporter Web queries have the following basic format: http:///cgi-bin/3par-rpts/ inserv_perf.exe?report=[&
where is the name of your server. The following is an example of the table returned: Figure 50 System Reporter Version Report This query returns a 3x3 table with the column headers Category, Major, and Minor. The two rows after the header contain major and minor numbers for Product and ReportData. The Product version numbers are informational only and are the numbers assigned to the product when it is released. The Product version number is not necessarily related to the ReportData version.
Report Options Options are included in the query URL by using the following syntax:
DiskType Compares selected disk types. This value is only available for the pd_perf_time, ld_space_time and pd_space_time reports. DiskSpeed Compares selected disk types. This value is only available for the pd_perf_time and pd_space_time reports. RaidType Compares selected RAID types. This value is only available for the ld_space_time report. n:s:p Compares selected port positions. For this value, an escape sequence may be required to retain spacing and special characters (for example: n%3As%3Ap).
Compares selected disk magazine position. This value is only available for the pd_space_time report. example: The following example specifies that all logical disks are compared: &compare=LD comparesel When the compare option (see “compare” (page 128)) is used, up to 16 objects can be compared. However, if there are more than 16 objects selected, then the comparesel option is used to select a metric. The16 objects with the highest peak values of the metric are chosen to be compared.
The virtual size. table Use the table option to specify a table from which data is returned when using one of the objects_present reports (see “Objects Present Reports” (page 167)). This option can only be specified once in a given query string. values: Specify one of the following: ldspace Returns values from the ldspace database tables. pdspace Returns values from the pdspace database tables. vvspace Returns values from the vvspace database tables. statpd Returns values from the statpd database tables.
example: The following example specifies that the report is to be reloaded every 2 minutes: &refresh=2 charttab Use the charttab option to specify whether to post the results in tabular format, as charts, or both. The number of charts and/or tables used to post the results depend on the report type. This option can only be specified once in a given query string. values: Specify one of the following: chart Specifies that results will be displayed in the form of one or more charts.
Specifies that charts will use stacked vertical bars. Stacked charts stack data on top of each other so you can see the total as well. See Figure 35 (page 94)for an example of a chart using stacked vertical bars. stackedarea Specifies that charts will use stacked areas instead of stacked bars. SeeFigure 36 (page 94) for an example of a chart using stacked areas. bars (Default). Specifies that bar charts will be used. See Figure 38 (page 95) for an example of a bar chart.
graphlegpos Use the graphlegpos option to specify the position of the chart legend. If not defined, the legend position is selected automatically based on the charttype. This option can only be specified once in a given query string. value: bottom The legend is placed below the chart. right The legend is placed to the right of the chart. example: The following example specifies that the chart legend should be to the right of the chart.
example: The following example specifies that all the time axis points should be labelled: &alllabels=1 timeform Use the timeform option to specify a time format (either relative or absolute) when using a report that requires a time range. This option can only be specified once in a given query string.
If is non-negative it represents a point in time, expressed as the number of seconds since January 1, 1970, GMT. For example, to express 2005/07/13 11:00:34 as a value, use the value 1121277634. If is negative, it represents a point in time seconds before the current time. NOTE: You can use the sample_times report (see “sample_times” (page 177)sample_times on page 8.78 ) to see tsecs values paired with times expressed using an easier-to-read format.
Domain Group results by domain. Host Group results by hosts. CPG Group results by Common Provisioning Group (CPG) name. UsrCPG Group results by Common Provisioning Group (CPG) name for the User Space of VVs. SnpCPG Group results by Common Provisioning Group (CPG) name for the Snap Space of VVs. VV Group results by Virtual Volume (VV) name. VVType Group results by Virtual Volume (VV) Type. ProvType Group results by virtual volume (VV) provisioning type. LD Group results by Logical Disk (LD) name.
PortType Group results by port type. example: The following example specifies that results should be grouped by system: &groupby=System The following example specifies that results should be grouped by system, drive magazine, and physical disk ID: &groupby=System&groupby=Mag&groupby=PDID orderby Use the orderby option to determine how to order results. The orderby option can be used multiple times in the same query string.
PortNode Order results by port node number (The n part in n:s:p). RaidType Order results by RAID level (for example, RAID 0, RAID1, RAID 5 and RAID 6). LDUsage Order results by logical disk usage. See Table 28 (page 88) for a list of these usage types. DiskType Order results by physical disk type. DiskSpeed Order results by physical disk speed. CageID Order results by drive cage ID number. Mag Order results by magazine position within the drive cage (0 through 7).
Order results by CPU context switches/sec. VirtSize Order results by VV virtual size. TotalRsvdSize Order results by VV reserved space size. TotalRawSize Order results by VV total raw reserved space size. LDSize Order results by total LD size. LDFreeSize Order results by total free LD size. LDRawSize Order results by total raw LD size. TotalAccesses Order results by total number of accesses to the VV. ReadHit% Order results by total read hit percentage to the VV.
example: The following example specifies that the histogram should include all accesses: &histbegin=1 ldspaceunit Use the ldspaceunit option to determine the unit of measure for logical disk space reports. This option can only be specified once in a given query string. values: Specify one of the following: MiB (2^20 bytes) (Default). Use megabytes (1,048,576 bytes) as the logical disk space unit.
See “ldspaceunit” (page 141). GiB (2^30 bytes) See “ldspaceunit” (page 141).
value: Specify one or more of the following: --All cage IDs-(Default). Does not exclude cages based on their cage IDs. This value consists of a drive cage ID.
SPR_AVAIL_CLEAN Include only spare available (unused) chunklets that have been cleaned (initialized). SPR_AVAIL_DRTY Include only spare available (unused) chunklets that are still being cleaned (and are uninitialized). SPR_AVAIL_FAIL Include only spare available (unused) chunklets that have failed.
selsnpcpg Use the selusrcpg option to select a Common Provisioning Group (CPG) that is used for the snap space of VVs. Usage and values are similar to selcpg (see“selcpg” (page 144)). seldisktype Use the seldisktype option to select a disk type. value: Specify one or more of the following: --All Disk Types-(Default). This value does not exclude disks based on their type. FC This value allows you to include only Fibre Channel type disks. NL This value allows you to include only Near Line type disks.
The following example specifies that only 10K RPM disks are to be included: &seldiskspeed=10 seldomain Use the seldomain option to select a domain according to the domain name. value: Specify one or more of the following: --ALL Domains-(Default). Does not exclude any domain based on domain names. This value consists of a storage server domain name.
The following example specifies including only VLUNs exported to host winhost1: &selhost=winhost1 The following example specifies including only VLUNs exported to host winhost1 or winhost2: &selhost=winhost1&selhost=winhost2 selld Use the selld option to select logical disks according to their LD names. value: Specify one or more of the following: --All LDs-(Default). Does not exclude logical disks based on their names. This value consists of a logical disk name.
Includes only logical disks used for logging. example: The following example includes logical disks of all usage types: &sellduse= The following example specifies only logging logical disks: &sellduse=log selnode Use the selnode option to select controller nodes according to their numbers (0-7). value: Specify one or more of the following: --All Nodes-(Default). Specifies that nodes are not to be excluded based on their names.
example: The following example shows two ways to specify that no ports are to be excluded based on location: &selnsp= OR &selnsp=--All%20Ports-- The following example specifies that only port 0:1:1 will be included: &selnsp=0%3A1%3A1 selpdid Use the selpdid option to select physical disks according to the physical disk ID. value: Specify one or more of the following: --ALL PDIDs-(Default). Do not exclude PDs based on their ID number. This value consists of a physical disk ID number.
example: The following example shows two ways to specify that physical disks are not to be excluded based on their states: &selpdstate= OR &selpdstate=--All%20PD%20States-- The following example specifies that only physical disks with a current state of Not valid should be used: &selpdstate=Not%20valid selporttype Use the selporttype option to select ports according to their type. values: Specify one or more of the following: --All Port Types-(Default). All port types are selected.
Specify this value to include RAID 1 logical disks. 5 Specify this value to include RAID 5 logical disks. 6 Specify this value to include RAID 6 logical disks. example: The following example specifies inclusion of all logical disks regardless of RAID level: &selraidtype= The following example includes RAID 1 and RAID 5 logical disks only: &selraidtype=1&selraidtype=5 selsys Use the selsys option to select a particular storage system. values: Specify one or more of the following: --All Systems-(Default).
example: The following example shows two ways to specify that volumes be included regardless of name: &selvv= OR &selvv=--All%20VVs-- The following example includes volumes with names vol1, vol2, and vol3 only: &selvv=vol1&selvv=vol2&selvv=vol3 selvvtype Use the selvvtype option to specify virtual volumes according to their types. values: Specify one or more of the following: --All VV Types-(Default). Does not exclude virtual volumes based on their types. base Select base VVs.
Report Types Report types are identified with names that typically use the following format:
selcageid=--All%20Cage%20IDs--&seldisktype=--All%20Disk%20Types-&report=pd_space_time pd_space_group Plots the total number of chunklets of different types at a specified time summed over (grouped by) any combination of parameters (for example, system, PDID, etc.) example: The following example shows the entire query string for a pd_space_group report: http://localhost/cgi-bin/3par-rpts/ inserv_perf.
example: The following example shows the entire query string for a vv_space_time report: http://localhost/cgi-bin/3par-rpts/ inserv_perf.
example: The following example shows the entire query string for a pd_perf_group report: http://localhost/cgi-bin/3par-rpts/ inserv_perf.exe?charttab=chart&charttype=stackedbars&category=daily&begintsecs=& selsys=--All%20Systems--&selpdid=--All%20PDIDs--&selnsp=--All%20Ports--& groupby=System&orderby=System&report=pd_perf_group pd_svt_hist Plots the service time histogram of all the physical disks in the selected systems at a specific time.
example: The following example shows the entire query string for a port_perf_group report: http://localhost/cgi-bin/3par-rpts /inserv_perf.exe?category=daily&charttab=chart&charttype=stackedbars&refresh=&begin tsecs=&selsys=--All%20Systems--&selporttype=--All%20Port%20Types--&selnsp=-All%20Ports--&groupby=System&orderby=System&report=port_perf_group port_svt_hist Plots the service time histogram of all the ports in the selected systems at a specific time.
example: The following example shows the entire query string for a ld_perf_group report: http://localhost/cgi-bin/3par-rpts/ inserv_perf.exe?charttab=chart&charttype=stackedbars&category=daily&begintsecs= &selsys=--All%20Systems--&selld=--All%20LDs--& groupby=System&orderby=System&report=ld_perf_group ld_svt_hist Plots the service time histogram of all the logical disks in the selected systems at a specific time.
example: The following example shows the entire query string for a vlun_perf_group report: http://localhost/cgi-bin/3par-rpts/ inserv_perf.exe?charttab=chart&charttype=stackedbars&category=daily&begintsecs=& selsys=--All%20Systems--&selhost=--All%20Hosts--&selvv=--All%20VVs--& selnsp=--All%20Ports--&groupby=System&orderby=System&report=vlun_perf_group vlun_svt_hist Plots the service time histogram of all the VLUNs (exports) in the selected systems at a specific time.
tsecs=&selsys=--All%20Systems--&selvv=--All%20VVs-&groupby=System&orderby=System&report=vv_cmp_group cpu_perf_time Returns a chart and/or table of performance CPU metrics versus time, limited by any options selecting systems or nodes. example: The following example shows the entire query string for a cpu_perf_time report: http://localhost/cgi-bin/3par-rpts/ inserv_perf.
regmoves example The following example shows the entire query string for a regmoves report: http://localhost/cgi-bin/3par-rpts/ inserv_perf.exe?charttab=chart&charttype=stackedbars&begintsecs=-10800& selsys=3par1&report=regmoves vvtier Returns a report showing the amount of data in each tier for each VV space (usr, snp and adm) at the selected time. The amount of data is shown in both absolute terms (MiB) and as a percentage.
Figure 51 Valid Options for Summary, Space and Adaptive Optimization Reports (I) 162 Web Queries
Figure 52 Valid Options for Summary, Space and Adaptive Optimization Reports (II) Options Available by Report 163
Figure 53 Valid Options for Summary, Space and Adaptive Optimization Reports (III) Figure 54 Valid Options for Performance Reports (I) 164 Web Queries
Figure 55 Valid Options for Performance Reports (II) Options Available by Report 165
Figure 56 Valid Options for Performance Reports (III) Figure 57 Valid Options for Performance Reports (IV) 166 Web Queries
Objects Present Reports Objects present reports return the objects present in a table over a specified time for use in user interface controls. For example, to allow the user to select a display of PD space usage at a particular time, you would need to know which times, systems, and PD IDs are present in a table. Sort order of columns is ascending, starting with the left column, then the next to the right, and so on.
Table 58 Objects Present Report table values (continued) Report table values objects_present&“objectpres= domain_name” (page 171)objectpres= domain_name on page 8.69 ldspace vvspace statld statvlun statvvcmp nl nl nl nl nl objects_present&“objectpres= host_name” (page 171)objectpres= host_name on page 8.69 statvlun nl nl objects_present&“objectpres= ld_name” (page 172)objectpres= ld_name on page 8.
objects_present&objectpres=cageid http://localhost/cgi-bin/3par-rpts/ inserv_perf.exe?report=objects_present&objectpres=cageid&table=pdspace& category=daily output: Returns a row for each drive cage ID number on each system. Output columns are: • System: the system name • CageID: drive cage ID numbers The following is example output of the query: objectpres= cpg_name Returns a list of common provisioning group (CPG) names.
objectpres= diskspeed Returns a list of disk speeds. example: The following example shows the entire query string for: objects_present&objectpres=diskspeed http://localhost/cgi-bin/3par-rpts/ inserv_perf.exe?report=objects_present&objectpres=diskspeed&table=statpd& category=daily output: Returns a row for each disk speed on each system. Output columns are: • System: the system name • DiskSpeed: list of disk speed objectpres= disktype Returns a list of disk types.
objectpres= domain_name Returns a list of domain names. example: The following example shows the entire query string for: objects_present&objectpres=disktype http://localhost/cgi-bin/3par-rpts/ inserv_perf.exe?report=objects_present&objectpres=domain_name&table=statvlun& category=daily output: Returns a row for each domain present over the time range on each system. Output columns are: • System: the system name • Domain: domain names objectpres= host_name Returns a list of host names.
objectpres= ld_name Returns a list of logical disk names. example: The following example shows the entire query string for: objects_present&objectpres=disktype http://localhost/cgi-bin/3par-rpt/ /inserv_perf.exe?report=objects_present&objectpres=ld_name&table=ldspace& category=hourly output: Returns a row for each logical disk present over the time range for each system.
objects_present&objectpres=node http://localhost/cgi-bin/3par-rpts/ inserv_perf.exe?report=objects_present&objectpres=node&table=statcpu& category=hourly output: Returns a row for each node present over the time range for each system. Output columns are: • System: the system name • Node: controller node number The following is example output of the query: objectpres=pdid Returns a list of physical disk IDs.
objectpres=port Returns a list of ports. example: The following example shows the entire query string for objects_present&objectpres=port http://localhost/cgi-bin/3par-rpts/ inserv_perf.exe?report=objects_present&objectpres=port&table=statvlun& category=daily output: Returns a row for each port present over the time range for each system.
objectpres= vv_name Returns a list of virtual volume names. example: The following example shows the entire query string for: objects_present&objectpres=vv_name http://localhost/cgi-bin/3par-rpts/ inserv_perf.exe?report=objects_present&objectpres=vv_name&table=statvlun& category=daily output: Returns a row for each virtual volume present over the time range for each system.
output: Returns a row for each volume present over the time range (excluding snapshot volumes) for each system. Output columns are: • System: system name • VV: virtual volume name The following is example output of the query: objecctpres= porttype Returns a list of port types. example: The following example shows the entire query string for: objects_present&objectpres=porttype http://localhost/cgi-bin/3par-rpts/ inserv_perf.
sample_times Returns valid times of database samples. example: The following example shows the entire query string for: sample_times http://localhost/cgi-bin/3par-rpts/ inserv_perf.exe?report=sample_times&category=daily&table=pdspace output: Returns a row for each time sample in the given table for the given category (resolution). Output columns are: • time: string with format yyyy-mm-dd hh:mm:ss • tsecs: Unix seconds, the number of seconds since 01/01/1970, GMT.
9 Using Adaptive Optimization This chapter provides information about the optionally licensed Adaptive Optimization feature that works in conjunction with System Reporter to optimize storage utilization across multiple tiers. NOTE: Adaptive Optimization is only supported by InForm OS versions 2.3.1 MU 1 and beyond.
Selecting an HP 3PAR Storage System for Adaptive Optimization To begin using Adaptive Optimization on an HP 3PAR Storage System array, you must either add it configured to include collection of data (region-level data) for sampling purposes (see “Adding Storage Systems” (page 47)) or change the configuration to include the data. To change the configuration, click on the InServ Systems tab in the System Reporter Policies window.
3. Click the Adaptive Optimization tab. The Adaptive Optimization Configuration window appears (Figure 60 (page 180)). Figure 60 Adaptive Optimization Configuration Main Window The following information is provided for each Adaptive Optimization configuration: • Id -- A unique identification number assigned automatically to the configuration when it is created. • InServ -- Name of the HP 3PAR Storage System for which the configuration is implemented.
NOTE: When you specify two tiers, for the purposes of analysis and data movement only the order of the tiers matters, not the actual tier number. For example specifying tiers 0 and 1 is equivalent to specifying 0 and 2 or 1 and 2. The lower numbered tier is considered the faster tier and the higher numbered tier is considered the slower tier.
except in the case where the CPG is the default for snapshot or base TPVV allocation as noted in the following caution. CAUTION: For some CPGs, especially those that use small PDs such as SSDs, the additional buffer space in the CPG may be too large to be left unused. To minimize the buffer space you can use the CLI to set the CPG grow size to the smallest possible value.
not be optimal. We can achieve optimal results by scheduling adaptive optimization execution for 11 am Monday through Friday with a measurement interval of 3 or 4 hours. That way only performance measurements during the important period for the application will be considered. NOTE: The measurement hours selected should not exceed the retention time for adaptive optimization samples set in the sampling policies (see “Editing Sampling Policies” (page 114)).
10. When done, click Submit Query. An alert window appears confirming that Adaptive Optimization configuration has been added. 11. Click OK to return to the Sampling Policies window. Changing an Adaptive Optimization Configuration To change an Adaptive Optimization configuration: 1. From the Adaptive Optimization Configuration window, click Change next to the configuration that you want to change. The Change Adaptive Optimization Configuration window appears. 2. 3. 4. Make the necessary changes.
Figure 61 Region IO Density Showing Total Space Figure 62 Region IO Density Showing Total IO Access Each region is put into the appropriate histogram bucket on the X-axis based on its IO access rate during the specified time period. The Y-axis (value) for the first chart is the total space for the regions in the histogram bucket, and the Y-axis (value) for the second chart is the total IO accesses/min for the regions in the histogram bucket.
Figure 63 Region IO Density without Adaptive Optimization You can see that much of the space is used for regions with very few IO accesses and most of the IO accesses are handled by a very small number of regions. Adaptive Optimization would work very well in this situation since the busiest regions would be moved to the fastest tier and the most idle regions would be moved to the slowest tier.
Figure 65 VV Tiers Report Adaptive Optimization Space Moved Report The Space Moved report shows the space moved over a selected time interval for each Adaptive Optimization configuration (Figure 66 (page 187)). Figure 66 Space Moved Report The space moved between tiers represents a performance overhead for Adaptive Optimization since it takes system resources and disk bandwidth to move data between tiers.
report shows the space moved over the selected time interval for each Adaptive Optimization configuration. The showvvcpg CLI extension Adaptive Optimization may move regions of a VV into different CPGs. In order to more easily see how much of each VV is in each tier, the CD for System Reporter 2.8 includes a CLI script file, showvvcpg.tcl, that adds a new CLI command when it is sourced from the CLI. Once you start the CLI and connect to the HP 3PAR Storage System array, you can source the showvvcpg.
esx-sysrpt esx-ssd-r1 128 - 7680 - - esx-sqa esx-fc-r1 384 - 44544 - - esx-sqa esx-ssd-r1 128 - 1536 - - esx-sw esx-fc-r1 224 - 249088 - - esx-sw esx-ssd-r1 288 - 16512 - - esx-rdm-mysqlserver esx-fc-r1 384 1024 44032 - - esx-rdm-mysqlserver esx-ssd-r1 128 - - - - esx-rdm-dev-oracle esx-fc-r1 384 1024 78848 - - esx-rdm-dev-oracle esx-ssd-r1 128 - 7168 - - esx-rdm-sysreporterfield esx-fc-r1 320 - 470400 - - esx-rdm-sysreporterfield esx-ssd-r1 448 - 896 - - ----------------------------------------------------
10 Database Schema This chapter documents the HP 3PAR System Reporter database schema. Using The Database Schema The description of the database schema is provided as a convenience to those wishing to query the database directly instead of using the web queries described in “Web Queries” (page 126).
Table 60 sampleinserv_5 Table Column Type Comment name VARCHAR(64) NOT NULL IP address of the system to sample. usessl INTEGER NOT NULL Determines whether CLI connections to the InServ will use SSL. 0 - Do not use SSL 1 - Use SSL cliport INTEGER NOT NULL Currently unused. In future this will be used to explicitly specify the port number to use for the CLI connection to the system. noldperf INTEGER NOT NULL 0 - Collect LD performance data 1 - Do not collect LD performance data.
Table 61 system_4 Table (continued) Column Type Comment system_model VARCHAR(32) NOT NULL System hardware model, for example: InServ S400X InServ T400 CHSIZEMB INTEGER NOT NULL Indicates the chunklet size. The alertconfig Table The alertconfig_2 table contains a row for each alert configuration rule. Table 62 alertconfig_2 Table Column Type Comment alertid INTEGER PRIMARY KEY (AUTO_INCREMENT) A unique identifier for each rule.
Table 62 alertconfig_2 Table (continued) Column Type Comment condition_code VARCHAR(32) Defines an additional condition that must be met for certain alerts to be generated. See “Condition” (page 121). condition_val DOUBLE PRECISION NOT NULL Defines a value used in conjunction with condition_code. See “Condition Value” (page 121). recipient VARCHAR(64) The email address to which alerts for this rule should be sent.
Table 65 adprovconfig_1 Table Column Type Comment adprovid INTEGER PRIMARY KEY A unique identifier for each adaptive optimization configuration. (AUTO_INCREMENT) sysname VARCHAR(32) NOT NULL The system name. t0_cpg_name VARCHAR(32) NULL Tier 0 CPG name. If there is no tier 0, this should be the empty string. t1_cpg_name VARCHAR(32) NULL Tier 1 CPG name. If there is no tier 1, this should be the empty string. t2_cpg_name VARCHAR(32) NULL Tier 2 CPG name.
Table 67 ldrgmoves_1 Table Column Type Comment sched_tsecs INTEGER NOT NULL Time in secs when the LD region move was scheduled. This is the Unix time and is the number of seconds since 0:00:00 Jan 1, 1970 GMT. sys_uid INTEGER NOT NULL A unique identifier for the system instance that refers to the uniq_id for the system instance in the system_3 table. adprovid INTEGER PRIMARY KEY (AUTO_INCREMENT) A unique identifier for each adaptive optimization configuration.
Table 68 recentsample_1 Table Column Type Comment resolution VARCHAR(32) NOT NULL The sample resolution, one of hires, hourly or daily. tsecs INTEGER NOT NULL Sample time in secs of the most recent complete sample of the resolution. This is the Unix time and is the number of seconds since 0:00:00 Jan 1, 1970 GMT. The Sample Data and Inventory Tables The sample data tables include the space and performance samples of various objects in the systems.
Table 69 pdspace__3 Table (continued) Column Type Comment NRM_AVAIL_CLEAN INTEGER NOT NULL Number of normal unused chunklets that have been initialized. These are free for use in new LDs. NRM_AVAIL_DRTY INTEGER NOT NULL Number of normal unused chunklets that have not yet been initialized. These will be free for use in new LDs after the system has had time to initialize them. NRM_AVAIL_FAIL INTEGER NOT NULL Number of normal unused chunklets that have failed.
The ldspace Table The ldspace__3 table holds logical disk (LD) space data samples. Table 71 ldspace__2 Table Column Type Comment tsecs INTEGER NOT NULL Sample time in secs. This is the Unix time and is the number of seconds since 0:00:00 Jan 1, 1970 GMT. sys_uid INTEGER NOT NULL A unique identifier for the system instance refers to the uniq_id for the system instance in the system_3 table. dom_name VARCHAR(32) NOT NULL The domain name that the LD belongs to.
Table 72 ldspaceinv__2 Table Column Type Comment tsecs INTEGER NOT NULL Sample time in secs. This is the Unix time and is the number of seconds since 0:00:00 Jan 1, 1970 GMT. sys_uid INTEGER NOT NULL A unique identifier for the system instance refers to the uniq_id for the system instance in the system_3 table. dom_name VARCHAR(32) NOT NULL The domain name that the LD belongs to. If the LD is not assigned to a domain, this column contains the value “-”.
Table 73 vvspace__3 Table (continued) Column Type Comment provv_type VARCHAR(32) NOT NULL The provisioning type of the VV (see the Prov column for the showvv CLI command in InForm OS 2.3.1 or later). vv_type VARCHAR(32) NOT NULL The VV Type (see the Type column for the showvv CLI command for InForm OS 2.3.1 or later). copyof VARCHAR(32) NOT NULL The name of the VV that this VV is a copy of. vv_state VARCHAR(32) NOT NULL The state of the VV.
Table 73 vvspace__3 Table (continued) Column Type Comment usr_rawrsvd_mb DOUBLE PRECISION NOT NULL Raw (physical space including RAID overhead) user space reserved for the VV in MB. usr_used_mb DOUBLE PRECISION NOT NULL User space used by the VV in MB. usr_zn DOUBLE PRECISION NOT NULL User zones in the VV. usr_free_zn DOUBLE PRECISION NOT NULL Free user zones in the VV. The vvspaceinv Table The vvspaceinv__3 table holds the inventory for the vvspace__3 table.
Table 75 cpgspace__1 Table Column Type Comment tsecs INTEGER NOT NULL Sample time in secs. This is the Unix time and is the number of seconds since 0:00:00 Jan 1, 1970 GMT. sys_uid INTEGER NOT NULL A unique identifier for the system instance refers to the uniq_id for the system instance in the system_3 table. dom_name VARCHAR(32) NOT NULL The domain name that the CPG belongs to. If the CPG is not assigned to a domain, this column contains the value “-”.
The cpgspaceinv Table The cpgspaceinv__1 table holds the inventory for the cpgspace_1 table. Table 76 cpgspaceinv__1 Table Column Type Comment tsecs INTEGER NOT NULL Sample time in secs. This is the Unix time and is the number of seconds since 0:00:00 Jan 1, 1970 GMT. sys_uid INTEGER NOT NULL A unique identifier for the system instance refers to the uniq_id for the system instance in the system_3 table. dom_name VARCHAR(32) NOT NULL The domain name that the CPG belongs to.
Table 77 Common Performance Columns (continued) Column Type Comment rbytes DOUBLE PRECISION NOT NULL Total number of bytes read since the object was first created or restarted d_rbytes DOUBLE PRECISION NOT NULL Difference between rbytes value from last sample for the same object. rerror DOUBLE PRECISION NOT NULL Total number of read errors since the object was first created or restarted d_rerror DOUBLE PRECISION NOT NULL Difference between error value from last sample for the same object.
Table 77 Common Performance Columns (continued) Column Type Comment whist0 DOUBLE PRECISION NOT NULL Number of writes in service time bucket number 0. d_whist0 DOUBLE PRECISION NOT NULL Difference between whist0 value from last sample for the same object. ... ... ... whist31 DOUBLE PRECISION NOT NULL Number of writes in service time bucket number 31. d_whist31 DOUBLE PRECISION NOT NULL Difference between whist31 value from last sample for same object.
Table 78 Common Performance Metrics for a Sample Row (continued) Metric SQL Expression Write KBytes/sec CASE WHEN d_now > 0 THEN d_wbytes * 1000.0/d_now ELSE 0 END Total KBytes/sec CASE WHEN d_now > 0 THEN (d_rbytes + d_wbytes)* 1000.0/d_now ELSE 0 END Read service time (millisec) CASE WHEN d_rcount > 0 THEN d_rticks / (d_rcount * 1000.0) ELSE 0 END Write service time (millisec) CASE WHEN d_wcount > 0 THEN d_wticks / (d_wcount * 1000.
Table 79 statpd__4 Table (continued) Column Type Comment port_s INTEGER NOT NULL The PCI slot in the node that the disk’s primary port is connected to (the “s” part of n:s:p). port_p INTEGER NOT NULL The port number in the PCI slot in the node that the disk’s primary port is connected to (the “p” part of n:s:p). disktype VARCHAR(32) NOT NULL The disk type. Currently, the disk types are • FC - Fibre channel drive. • NL - Nearline drive.
Table 81 statld__5 Table (continued) Column Type node INTEGER NOT NULL The node that owns the LD cpg_name VARCHAR(32) NOT NULL (common performance columns) Comment The name of the CPG that the LD belongs to. If the LD is not in a CPG, this column contains the value “----”. The rest of the columns are the common performance columns described in “Common Performance Table Columns” (page 203)). The statldinv Table The statldinv__4 table holds the inventory for the statld__5 table .
Table 83 statport__4 Table (continued) Column Type Comment • iscsi - iSCSI host ports. • free - ports that are not connected. (common performance columns) The rest of the columns are the common performance columns described in “Common Performance Table Columns” (page 203)). The statportinv Table The statportinv__4 table holds the inventory for the statport__4 table. Table 84 statportinv__4 Table Column Type Comment tsecs INTEGER NOT NULL Sample time in secs.
Table 85 statvlun__3 Table (continued) Column Type Comment host_wwn VARCHAR(128) NOT NULL The host WWN (world wide name) or iscsi name. port_n INTEGER NOT NULL The node that the port is connected to (the “n” part of n:s:p). port_s INTEGER NOT NULL The PCI slot in the node that the port is connected to (the “s” part of n:s:p). port_p INTEGER NOT NULL The port number in the PCI slot in the node that the port is connected to (the “p” part of n:s:p).
Table 87 statvvcmp__2 Table (continued) Column Type Comment vv_name VARCHAR(32) NOT NULL The VV name. rdhit DOUBLE PRECISION NOT The number of read cache hits for the VV. NULL d_rdhit DOUBLE PRECISION NOT Difference between rdhit value from last sample for same VV. NULL rdmiss DOUBLE PRECISION NOT The number of read cache misses for the VV. NULL d_rdmiss DOUBLE PRECISION NOT Difference between rdmiss value from last sample for same VV.
Table 89 statcpu__1 Table (continued) Column Type Comment d_usert DOUBLE PRECISION NOT Difference between usert value from last sample for same CPU. NULL syst DOUBLE PRECISION NOT Jiffies (0.01 sec) in system mode. NULL d_syst DOUBLE PRECISION NOT Difference between syst value from last sample for same CPU. NULL idlet DOUBLE PRECISION NOT Jiffies (0.01 sec) in idle mode. NULL d_idlet DOUBLE PRECISION NOT Difference between idlet value from last sample for same CPU.
Table 91 statlink__2 Table Column Type Comment tsecs INTEGER NOT NULL Sample time in secs. This is the Unix time and is the number of seconds since 0:00:00 Jan 1, 1970 GMT. sys_uid INTEGER NOT NULL A unique identifier for the system instance refers to the uniq_id for the system instance in the system_3 table. node INTEGER NOT NULL The node number node_to INTEGER NOT NULL The node number of the destination of the link. queue VARCHAR (32) NOT NULL The name of the Queue.
Table 93 statldrg_hourly_2 Table Column Type Comment tsecs INTEGER NOT NULL Sample time in secs. This is the Unix time and is the number of seconds since 0:00:00 Jan 1, 1970 GMT. sys_uid INTEGER NOT NULL A unique identifier for the system instance refers to the uniq_id for the system instance in the system_3 table. ldid INTEGER NOT NULL The logical disk ID of the LD region. ldoffmb INTEGER NOT NULL Offset in MiB of the LD region. vvid INTEGER NOT NULL The VV ID for the LD region.
11 Troubleshooting This chapter describes how to troubleshoot and solve common problems that may occur while setting up or using System Reporter. Troubleshooting Overview The first step when troubleshooting System Reporter is to identify the component that is causing problems (see for the structure of System Reporter). • If there are problems in sampling components see “Troubleshooting the Sampling Components” (page 215).
When reporting a problem with sampling, please provide the entire log directory. CAUTION: If you get an error in a sampling component that causes the service or daemon to stop, save the entire log directory before restarting the service or daemon so that you do not lose the error information in the logs. Installation Configuration On Windows, you can see the parameters with which the sampler was installed by typing the following command in that directory (which should contain the winserv.
Errors incurred by the CLI sampler process. Errors incurred by the CLI sampler process may include loss of network connectivity to the HP 3PAR Storage System, a timeout because the CLI process took too long to get the data, or authentication failure on the CLI connection (the user was removed or password changed). In both cases, the sampleloop process restarts the inservsample process so, if the error is not persistent, sampling can resume.
Figure 67 Microsoft Office Excel Internal Error If you experience a problem with a recovered copy of the workbook (Figure 68 (page 218)), try starting with a fresh copy of System Reporter workbook. Figure 68 Microsoft Excel Recovery Error Extracting Data from the Database with sysbck Sometimes it is necessary to extract some of the data from the System Reporter database. For example, you may want to send that data to HP to help in analyzing a performance problem.
-dbhost For mysql: name or IP address of host. Default is localhost. For oracle: hosts string. Ignored if dbtype is sqlite. -db Name of the database. For SQLite this is the full file name, including the path, of the file that contains the database. -dbpwfile Name of file whose first line is the username and password for the database. Ignored if dbtype is sqlite.
From MS SQL to SQLite: sysbck -dbtype mssql -dbhost istat -db cliuser -dbpwfile dbpwfile -dbouttype sqlite -scripttype both -backupfilename "c:/test/test_20100713_1322PM" -begintime "07/13/2010 12:45:00" -endtime "07/14/2010" -inservname s074:s138 • From MS SQL to MySQL: sysbck -dbtype mssql -dbhost istat -db cliuser -dbpwfile dbpwfile -dbouttype mysql -scripttype both -backupfilename "c:/test/test_20100713_1322PM" -begintime "07/13/2010 12:45:00" -endtime "07/14/2010" -inservname s074:s138 • From MySQL
• From SQLite to MS SQL: sysbck -dbtype sqlite -db cliuser -dbouttype mssql -scripttype both -backupfilename "c:/test/test_20100713_1322PM" -begintime "07/13/2010 12:45:00" -endtime "07/14/2010" -inservname s074:s138 • From SQLite to Oracle: sysbck -dbtype sqlite -db cliuser -dbouttype oracle -scripttype both -backupfilename "c:/test/test_20100713_1322PM" -begintime "07/13/2010 12:45:00" -endtime "07/14/2010" -inservname s074:s138 The following examples show how to use sysbck without any optional parame
NOTE: When executing sysbck.exe from any other directory on Windows, ensure that the PATH environment variable is set to the System Reporter root directory path (by default, this path is C:\Program Files\3par). Dumping the Entire Database When you want to extract a large part of the database, it may be more efficient, both in time and disk space, to dump the entire database instead of using sysbck. The method for dumping the entire database depends on the database type.
Index Symbols 3PAR System Reporter User’s Guide, 54 3PAR System Reporter.xls, 49 3par-policy subdirectory, 125 3par-rpts subdirectory, 125 A adding a system to the list of systems to be sampled, 115 Administrator’s Guide, 25 Apache HTTP Server cgi-bin directory, 125 config.pl files, 125 Directory directives, 49 httpd.
Port Performance Reports, 21 R reading of the database, 49 Red Hat Enterprise Linux, 25 Region IO Density Report, 184 related documentation, 18 remove a storage server from the system table, 115 Remove InServ, 47 Report Information control, 92 Report Type control, 76 reports displaying results as chart or table, 92 results, 71 types of, 76 requirements, 26 S sample resolution resolutions, 67 Sample resolution control, 67 sampling a new storage server, 115 sampling components troubleshooting, 215 sampling
