Home

BMC Performance Manager for Hardware

1. next gt Herp 11 7 Write a line to a LOG file If you select the Write a line to a LOG file action you need to enter the LOG file path Where Hardware Sentry will write a line as well as the content of the line to be written Note You can use macros that will be replaced at runtime The list of available macros is described later in this chapter 11 8 Send a Basic SNMP trap Set Alert Actions Alert Action Send a basic SNMP trap Enter the SNMP trap destination SNMP Manager Host i Port 162 Community publie Enter the text to be sent with the trap Hardware problem with OBJECT_LABEL PROBLEM the text will be sent with the standard trap defined in the PATROL HIB Note You can use macro variables in the fields above See the documentation to have the complete list of available macros next gt Hep 47 If you select the Send a pop up to the PATROL Consoles action you need to enter the message that will be displayed in the pop up as well as the title of the pop up window Warning Too many pop ups could annoy the operators Note You can use macros that will be replaced at runtime The list of available macros is described later in this chapter Set Alert Actions DAR Alert Action Write a line to a LOG file Enter the LOG file path C MSHW1 LOG Enter the line that has to be written in the LOG file ASCTINE Y m d HI M2 S SSC ALARM TY
2. action you need to enter command line to be executed user and password used to run the command The command can be a program utility or a script shell and can have arguments Warning The command must be non interactive no window no user input Note You can use macros that will be replaced at runtime The list of available macros is described later in this chapter If you select the Execute a PSL command action you need to enter the PSL statement to be executed by the PATROL Agent Only a single line is permitted but it can have several PSL instructions Warning The PSL command is recommended for PATROL advanced users only Note You can use macros that will be replaced at runtime The list of available macros is described later in this chapter SNQ SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 NG SOFTWARE User Guide Configuring and Administering 11 6 Send a Pop up to the PATROL Consoles 2 Set Alert Actions Sele Alert Action Send a pop up to the PATROL Consoles Enter the pop up window title Hardware Sentry on HOSTNAHE Enter the text of the pop up that will be sent to the PATROL Consoles Hardware problem on HOSTNAME with OBJECT_LABEL NEWLINE PROBLEM Warning heavy use of pop ups is not recommended Note You can use macro variables in the fields above See the documentation to have the complete list of available macros
3. for PATROL version 1 3 01 NG Ji SO vance User Guide Configuring and Administering 11 3 Annotate the Parameter s Graph 2 Set Alert Actions Alert Action Annotate the parameter s graph Enter the text to be written in the annotation point FULLREPORT Note You can use macro variables in the fields above See the documentation to have the complete list of available macros next gt Herp 11 4 Execute an OS command 2 Set Alert Actions Alert Action Execute an OS command Enter the OS command to be executed ER Execute this command as leave empty to use the PATROL default account Usernane Password Note You can use macro variables in the fields above See the documentation to have the complete list of available macros next gt Hep 11 5 Execute a PSL command Set Alert Actions Alert Action Execute a PSL command advanced Enter the PSL command line to be executed o o ouuu Note You can use macro variables in the fields above See the documentation to have the complete list of available macros next gt nep 46 If you select the Annotate the parameter s graph action you need to enter the string that will be displayed within the annotation point Note You can use macros that will be replaced at runtime The list of available macros is described later in this chapter If you select the Execute an OS command
4. this fan should grow quickly This can lead to severe hardware damage and system crashes Recommended action Quickly check if the fan is really no more cooling the systen If so replace the fan Save As Help gt gt 6 22 2005 6 27 PM alert such as the parameter name or its value is available through the Alert Actions macros This information can be used to further customize the Alert Action triggered by Hardware Sentry and provide more details about the problem that occurs By default upon hardware failure Hardware Sentry triggers a PATROL event and annotates the parameter s graph with a comprehensive report of the problem giving details about the failure its possible consequences and the recommended action to solve the problem 27 Ae SN SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 A 2 QO m 8 a_i Key SOFT User Guide Configuring and Administering S www sentrysoftware net Section IV Configuring and Administering 1 Entering A License Key For Hardware Sentry ssessoesoesoesoescoscescossoesoesoecoessecooeo 29 2 Distributing the License over Several Computers ssssessesessesoesessesessesecseseesesossesseo 31 2 1 Distributing a configuration variable with the license Key cc ccc c cece cence n ncn n cen eenes 31 2 2 Distributing a Jile with ihe license Key vs125 04 vessesrdennsieedndeweswes ddpsearseariveoeessaapneansss 31l 3 Temp
5. Distribute this SENTRY HARDWARE license4 configuration variable to all of the hosts that run Hardware Sentry through wpconfig xpconfig or PCM The license key distributed through this mechanism will take effect within an hour 2 2 Distributing a file with the license key The other way to distribute the Hardware Sentry license key to several computers is to deploy the corresponding file through any file distribution mechanism Distribution Server Marimba ftp etc 1 On a first computer enter the license key through the graphical user interface as described in the previous section 2 Retrieve the file PATROL_HOME lib MS_HW_license4 alternatively PATROL_HOME lib MS_HW_license4 under UNIX Linux 3 Distribute the MS_HW_license4 file to the 7PATROL_HOME lib directory on all of the hosts that run Hardware Sentry 4 The license key distributed through this mechanism will take effect within an hour 31 SOFTWARE Oe www sentrysoftware net PATROL V3 5 50i Oper 1 File View Hosts Tools Options Help SN SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 User Guide Configuring and Administering ax CCE PLO El B PATROLMainMap Event Manager Gi kant Options FN Hardware Sentry 1 1 00 Parameters E EN Computer Dell PowerEdge 1600SC neters without I E Disk Controller PERC 4 SC Controller a Fans eg H E Fan 1 1 CPU 2 Expand H E Fan 1 2 Front w E Fan 1 3 Re
6. Inside Hardware Sentry fan Www sentrysoftware net 2 The HDF files connectors Each connector is an hdf file that is dedicated to one hardware information source The purpose of each hdf file is to describe how Hardware Sentry can connect to the hardware information source available on a platform and which information is available through this source For example MS_HW_Director4INT hdf will describe how to get information from the IBM Director 4 1 Agent and then monitor an IBM xSeries server A hdf file can tell the Hardware Sentry engine to do the following actions query a SNMP agent get get_next and tables execute a WBEM query execute an OS command Each hdf file uses a mix of these possible actions with some computing capabilities to make Hardware Sentry gather useful data in a given hardware information source Note The hdf files are located under the MS_HW_hdf folder under SPATROL_HOMES 1ib All hdf files are installed on all platforms Note The hdf files released by Sentry Software are encrypted and therefore cannot be updated or modified by the end user Note The whole HDF syntax will be described in a separate document and will allow users to write their own hdf files Hardware Sentry will work with custom hdf files as it works with officially released one 52 A SN SENTRY Hardware Sentry Knowledge Module for PAT
7. Options gt after a reboot j Missing device toggle ancer Note Disabling this option when some missing devices have already been detected will simply remove these devices from monitoring 42 SQ SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 a SOFT User Guide Configuring and Administering www sentrysoftware net 10 Setting Security Options Hardware Sentry Security settings Sle Security settings Use the following SNMP community for SNMP queries public Use the following account to execute external commands Username To access the Security settings dialog right repeal click the main Hardware Sentry icon computer Leave empty if you want Hardware Sentry to use the PATROL 1 1 ra RA a E S KM Commands gt Options gt Security Settings To execute external commands Hardware Sentry may use the sudo utility rather than impersonation Sudo options The settings above will be used for all the discoveries and collects of Hardware Sentry ancer weap 10 1 SNMP Community Hardware Sentry allows you to modify the SNMP community that will be used for SNMP queries In the Security Settings dialog enter the custom SNMP community in the first field Leave this field blank if you want Hardware Sentry to automatically decide which SNMP community string to use Using an alternate SNMP community can be required in secu
8. Sentry Knowledge Module for PATROL version 1 3 01 eee User Guide Configuring and Administering www sentrysoftware net 8 Pre Selecting Connectors optimizing the detection process Hardware Sentry s detection process may take some time depending on the platform Although it is only done upon the PATROL Agent startup if performance is an issue it might be wise to speed things up by pre selecting the platforms to be used to monitor Pre selecting connectors means that an administrator decides what platforms Hardware Sentry will have to assume it is running on For each connector selected Hardware Sentry will perform tests and will try to gather data using the tools provided by the connectors for that particular platform All other connectors and therefore other platforms will be ignored and no testing shall be performed for them The downside of pre selection is that if the wrong connectors are selected then no monitoring will be performed Furthermore if some hardware that does not belong to the pre selected platforms is added to the server later on it won t be detected It is therefore strongly recommended to let Hardware Sentry figure out what type of platforms can be found on the system and the pre selection option should be used by advanced users only 2 Hardware Sentry Preselect connectors 1 Preselect connectors To pre select connectors right click the main Hardware Sentry icon and select the Pre C Man
9. Sentry parameters please see the next section 37 SNQ SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 E eee User Guide Configuring and Administering www sentrysoftware net 7 Modifying Parameter Thresholds Whenever possible Hardware Sentry automatically sets thresholds for parameters These thresholds are retrieved from available information and they are generally accurate Normally there is no need to modify these thresholds but in case Hardware Sentry cannot obtain them or if prefer a more fault tolerant monitoring you can modify the thresholds of each parameter through the command menu PATROL V3 5 50i Oper 1 BAR File View Hosts Tools Options Help Br PATROLMainMap Gi kant E fal Hardware Sentry 1 1 00 TRIAL PERIOD 28 Event Manager fal Computer Dell PowerEdge 1600SC Options Disk Controller PERC 4 SC Controller Parameters 5 0 Fans sramete Fan 1 1 CPU 1 f Fan 1 2 Front Open EE Fan 1 3 Rear Expand BW IntrusionStatus agi i w Memory modules E A ht porene jee ene To do so right click on the device monitoring E fal Power supplies InfoBox a E Processors but not directly on the parameter and select FEE Temperatures Cut Bal temperature 1 1 PanerTere o Modify Thresholds E Temperature 1 2 CPUiTemp ___ Create Shortcut Voltages Properties Noa Brest Pause monitoring R
10. Sentry sends its debug output to the System Output Window of the PATROL Consoles When debugging the detection or discovery process of Hardware Sentry at the starting time of the Agent some debug information may be lost by the PATROL Console as it is not yet connected to the PATROL Agent In other cases or when you want to trace the activity of Hardware Sentry during a few minutes some debug information may be lost by the PATROL Console because its buffer is full In these cases it can be useful to send the debug output of Hardware Sentry to a specified file Note that debug file is stored on the computer where the Agent is running Warning Pay attention to the file size the debug output of Hardware Sentry could be very large when running for several days 57 XQ SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 A SOFT User Guide Troubleshooting www sentrysoftware net 2 Reinitializing Hardware Sentry In cases Hardware Sentry is not properly functioning creating missing devices you know are present or not detecting several devices as it should it is possible to reinitialize the configuration and let Hardware Sentry re start the monitoring from scratch Hardware Sentry will then run a full discovery and will detect the computer s environment set the appropriate connectors and re start the monitoring of existing devices Thresholds will be set to default so all previously manually set thresholds wi
11. User Guide Configuring and Administering www sentrysoftware net Hardware Sentry License License Information for Hardware Sentry xx Registered Licenses Once the license key has been accepted Automatic trial expires on Nov 5 2004 29 day s remaining the license dialog box Is refreshed and Term license expires on Dec 31 2666 815 day s remaining shows the new license as registered Register a new license Close Note The automatic trial period is always shown in this dialog even if it has expired This has no effect on the activation of Hardware Sentry 30 4 A 2 oe e Ld 2 Cea A aah N SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 SOFT User Guide Configuring and Administering www sentrysoftware net 2 Distributing the License over Several Computers 2 1 Distributing a configuration variable with the license key The first way to distribute the Hardware Sentry license key to several computers is to deploy the corresponding configuration variable through wpconfig Windows xpconfig Unix Linux or through PATROL Configuration Manager PCM L On a first computer enter the license key through the graphical user interface as described in the previous section Use wpconfig xpconfig or PATROL Configuration Manager to retrieve the SENTRY HARDWARE license4 configuration variable that has been set by the graphical user interface on the first computer
12. Your use of this information is subject to the terms and conditions of the applicable End User License Agreement for the product and the proprietary and restricted rights notices included in this documentation Restricted Rights Legend U S Government Restricted Rights to Computer Software UNPUBLISHED RIGHTS RESERVED UNDER THE COPYRIGHT LAWS OF THE UNITED STATES Use duplication or disclosure of any data and computer software by the U S Government is subject to restrictions as applicable set forth in FAR Section 52 227 14 DFARS 252 227 7013 DFARS 252 227 7014 DFARS 252 227 7015 and DFARS 252 227 7025 as amended from time to time Contractor Manufacturer is BMC Software Inc 2101 CityWest Blvd Houston TX 77042 2827 USA Any contract notices should be sent to this address Customer Support You can obtain technical support by using the Support page on the BMC Software Web site or by contacting Customer Support by telephone or e mail To expedite your inquiry please see Before Contacting BMC Software Support Web Site You can obtain technical support from BMC Software 24 hours a day 7 days a week at http www bmc com support_home From this Web site you can m read overviews about support services and programs that BMC Software offers m find the most current information about BMC Software products m search a database for problems similar to yours and possible solutions E order or download product documentation E
13. a high monitoring accuracy by not confusing errors that are really encountered by devices and errors due to a monitoring tool failure 26 Hardware Sentry Knowledge Module for PATROL version 1 3 01 SOFT User Guide Monitoring Gi www sentrysoftware net 6 Alert Actions Alert Actions enables the PATROL administrator to choose specific actions to be executed when a hardware failure is detected With the Alert Actions it is possible to either customize the way a hardware problem notification is performed or specify a recovery action to run when a problem occurs Alert Actions offers a large choice of actions to be executed in order to notify the operator of a problem with the hardware or to recover from a particular problem Hardware Sentry can be configured to run one several or all types of Alert Actions when an alert is triggered regarding the monitored hardware The following types of Alert Actions can be performed by Hardware Sentry trigger a PATROL event annotate the parameter s graph execute an OS command execute a PSL command send a pop up to the PATROL consoles write a line to a LOG file send a basic SNMP trap using the PATROL MIB send a custom SNMP trap Alert Actions are highly customizable One can customize the string that is sent through SNMP set the username password that is used to execute the OS Command define the content of the PATROL event sent by Har
14. and options that you used m messages received and the time and date that you received them product error messages messages from the operating system such as file system full messages from related software Hardware Sentry Knowledge Module for PATROL version 1 3 01 User Guide Copyrights and trademarks IBM RS 6000 pSeries eServer xSeries Netfinity BladeCenter and Director are trademarks or registered trademarks of International Business Machines Corporation Fujitsu Siemens Primergy and Serverview are trademarks or registered trademarks of Fujitsu Siemens Computers Corporation DELL PowerEdge PERC and OpenManage are trademarks or registered trademarks of DELL Computers Corporation HP Compaq ProLiant Integrity SuperDome and Insight Manager are trademarks or registered trademarks of Hewlett Packard Corporation NEC Express5800 and EsmPro are trademarks or registered trademarks of NEC Adaptec and Storage Manager are trademarks or registered trademarks of Adaptec Corporation LSI Logic Mylex and GAM Server are trademarks or registered trademarks of LSI Logic Corporation Intel Pentium and Itanium are trademarks or registered trademarks of Intel Corporation AMD and Opteron are trademarks or registered trademarks of Advanced Micro Devices Incorporated Sun and SPARC are trademarks or registered trademarks of Sun Microsystems Incorporated All other trademarks belong to their respective companie
15. is heavily parallelized All the hdf files are processed simultaneously 53 A SN SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 te a SOFT I ser Guide Inside Hardware Sentry ars sielie User Guide Inside Hardware Sentry Io 4 The Discovery Process The discovery process is launched just after the end of the detection process It takes the following actions Processes the Enclosure Discovery section of the detected connectors hdf that describe the computer model and create the main Hardware Sentry and computer icon class MS_HW_ENCLOSURE Most of other icons disk controllers fans temperatures etc will be created under this computer icon Launches the disk controller discovery that will processes the DiskController Discovery section of each detected hdf file and create the MS_HW_DISKCONTROLLER icons Launches the other discoveries fans temperatures voltages power supplies logical disks physical disks and other devices that will process the corresponding sections of each detected hdf file and create the corresponding icons Note With a view to optimization the discovery process is as parallelized as possible All the independent objects are processed at the same time 54 SN SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 QO m EJ jz e e a SOFT User Guide Inside Hardware Sentry laa t www sentryso
16. lt Certified Product BMC Performance Manager for Hardware by Sentry Software User Guide Supporting Hardware Sentry Knowledge Module for PATROL version 1 3 01 by Sentry Software May 8 2006 BMC SOFTWARE lt Certified MarketZone a Contacting BMC Software You can access the BMC Software Web site at http www bmc com From this Web site you can obtain information about the company its products corporate offices special events and career opportunities United States and Canada Outside United States and Canada Address BMC Software Inc 2101 CityWest Telephone 01 713 918 8800 Blvd H TX 77042 282 vd Houston TX 77042 2827 Fax 01 713 918 8000 Telephone 713 918 8800 or 800 841 2031 Fax 713 918 8000 Copyright 2006 BMC Software Inc or licensors as an unpublished work All rights reserved BMC Software the BMC Software logos and all other BMC Software product or service names are registered trademarks or trademarks of BMC Software Inc IBM is a registered trademark of International Business Machines Corporation DB2 is a registered trademark of International Business Machines Corporation Oracle is a registered trademark and the Oracle product names are registered trademarks or trademarks of Oracle Corporation All other trademarks belong to their respective companies BMC Software considers information included in this documentation to be proprietary and confidential
17. report a problem or ask a question m subscribe to receive e mail notices when new product versions are released m find worldwide BMC Software support center locations and contact information including e mail addresses fax numbers and telephone numbers Support by Telephone or E mail In the United States and Canada if you need technical support and do not have access to the Web call 800 537 1813 Outside the United States and Canada please contact your local support center for assistance To find telephone and e mail contact information for the BMC Software support center that services your location refer to the Contact Customer Support section of the Support page on the BMC Software Web site at http www bmc com support_home Before Contacting BMC Software Before you contact BMC Software have the following information available so that Customer Support can begin working on your problem immediately E product information product name product version release number license number and password trial or permanent E operating system and environment information machine type operating system type version and service pack or other maintenance level such as PUT or PTF system hardware configuration serial numbers related software database application and communication including type version and service pack or maintenance level m sequence of events leading to the problem E commands
18. see coos cece cesececceeetacdeseweseecosscesnacdecsseseecers 52 a The Deteccion Process Sinaissiacciscensatiactinsiswasivensicieessawericsi a 53 SQQ SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 SQQ SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 Aiea ihe tees ee nee d The Discovery ProC SS eaiine secs sacenc cress auc vaiewortuediavecss cue vaca caavecsswweeienacen EEIE Rna 54 S Phe Collection Proc cessivscsside cesvitacswcotetisiansecedeswsdevsertaassteiserstasseiuaswostsercctsnsaaedeen 55 Section VI Troubleshooting sunsnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn 56 I Enabling the Debug Mod icsccsvaconcdeccconticvexeusehuscuncesiesawcwescawscdseerasessbvedecewsissekenssiews 57 2 Reinitiahzing Hard Ware Senir y scsecseccttcsccsiacceccsccnseceseatecseceseasocennaseseradeeernsadeescasseve 58 3 Hardware Sentry SHOWS Nothing essessecoocooesoecoecoecoocooesoccoscoecoocoecsoccoscoecoecsoesoecoo 59 4 Unable to See Temperature Voltage or Fan ssescescossessossocsoeccossescoscoecsessesoessoeoo 60 Ss Unable to See Disk Controller osircsiseeirrserrsne iorden anions 61 O The CPU is ovrerloaded srcsesicciesscecssinccnussdesseiacwsancieboiwencitnatestevesdesssecianedacsaaisscatesues 62 SN SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 QO m EJ jz at SOFT User Guide Overview an G www sentrysoftware net en
19. that are supported by the information sources above non exhaustive list Dell PowerEdge servers under Windows and Linux Fujitsu Siemens Primergy servers under Windows and Linux Fujitsu Siemens Primergy Blade servers HP Compaq ProLiant servers under Windows and Linux HP NetServer servers under Windows only 14 SN SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 QO m j M eae User Guide Getting started fan www sentrysoftware net HP 9000 servers under HP UX HP AlphaServer servers under HP Tru64 UNIX IBM xSeries servers under Windows and Linux IBM Netfinity servers under Windows only IBM BladeCenter IBM RS 6000 pSeries and eServer p5 under AIX NEC Express5800 servers under Windows Sun Ultra and Fire servers under Solaris any PC with non RAID IDE or SCSI and SMART capable disks any PC with standard sensors supported by Motherboard Monitor As Sentry Software is continuously working on the support of new hardware information sources and new platforms that can be monitored with Hardware Sentry KM for PATROL will continue to grow Please check our web site www sentrysoftware net to find the latest updates Add ons for Hardware Sentry KM for PATROL can be obtained for free and do not need an update of the KM itself 15 Ae N SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 A 2 W Nm 8 a_i NG eee
20. 5 like memory configuration These features and configurations allow the memory modules to report statistics on failures to predict failures to hot replace a memory module upon failure etc Depending on the available information and the features provided by the motherboard and the memory modules the ErrorCount and or PredictedFailure and or Status parameters will be displayed for each discovered memory module The ErrorCount parameter reports the number of errors that have been detected by the memory module and then corrected A steadily growing value means that the memory module is not reliable and that it could encounter errors that it is unable to correct and that will then crash the system The PredictedFailure parameter is reported by the memory modules which try to predict if it is going to fail by analyzing the trend of the number of detected corrected errors thanks to the ECC technology If this parameter goes into alarm you should remove the faulty memory module and replace it with a new one The Status parameter represents the current status of the memory module An alert is triggered if the memory module reports a failure in a RAID5S like configuration or if it is missing after a computer reboot 3 3 Network Interfaces Network interfaces are devices that serve as a common interface for various other devices within a local area network LAN or as an interface to allow networked computers to connect to an outside n
21. 8 t3 4 f BE HH EE FUJITSU cconmamd S58 25255 SSE SIEME gt LEK aT E i K LA HARDWARE server RAID controller other SCSI controller The Siemens Serverview agent shows the temperatures the fans the power supplies and the voltages of the Siemens computer through SNMP The Mylex GAM Server shows the status of the physical and logical disks of the Mylex RAID Controller through SNMP The Windows WMI provider shows the status of the physical disks attached to the standard SCSI controller through WBEM Hardware Sentry detects and automatically connects to all three information sources Siemens Serverview Mylex GAM Server and Windows WMI provider Hardware Sentry gathers the useful hardware information from these sources and displays it within the PATROL framework Please keep in mind that this is only an example and that it could be applied to IBM NEC HP and other computers 10 amp SNQ SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 M eae User Guide Getting started an AG www sentrysoftware net 2 Integrating Hardware Sentry KM for PATROL Hardware Sentry KM for PATROL is a Knowledge Module KM for BMC Software PATROL Therefore it must be installed on the components of the PATROL framework the PATROL Agents the PATROL Consoles the PATROL Console Server PATROL Central Web Edition osoft Internet Explorer aici As any other KM the a
22. Create Shortcut H B Power supplies eroe Poms To check the list of the removed objects or to Figen bine ause monitorin r gt i a M Temperature MANASS ets restore a device monitoring E Temperature 1 2 CPU 1 Temp Voltages right click the Hardware Sentry computer Show connectors currently in use s 1 C O n ii select Restore device or sensor monitoring About Desktop Ku Hardware Sentry SEE Restore device or sensor monitoring If no object has been removed a message will be displayed to inform you that restoration Select a device or sensor type to restore cannot be performed Otherwise the first window of the Restore wizard will pop up all next gt ne In this window you are asked to select the type of device or sensor monitoring you would like to restore If for example you would like to restore the monitoring of a Physical Drive that you have previously removed select Physical disks from the list If you would rather restore all monitored objects that have been removed simply select the All option from the drop down box Select a type and click the Next gt button Note Hardware Sentry scans the list of removed objects in order to just display the type of objects that have actually been removed If for instance you have not removed any fans from monitoring the Fan type will not be part of the list to cho
23. DGs coos MS_HW FAN CONT HS _HW FAN MS HW DISKCONTROLLER MS_HW COMPUTER PATROL_NT yes yes yes yes yes lt komema Prompt gt DUMP KM_LIST If no MS_HW_MAIN entry is present it means that Hardware Sentry is not loaded on the agent Make sure that the MS_ HARDWARE_SENTRY1 kml file is preloaded by the PATROL Agent and loaded by the PATROL Consoles and that the MS_HW_ km are not disabled See the BMC Software documentation PATROL Agent Reference Manual to find out how to load a KM on the PATROL Agent If the MS_HW_ classes are properly loaded and no icon appears in the PATROL Console it probably means that the Hardware Sentry KM for PATROL has not been properly installed on the PATROL Agent Re install the KM and ensure that all files are properly installed typically lib files in the SPATROL_HOMES lib psl folder 59 SNQ SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 NG eae User Guide Troubleshooting fan Www sentrysoftware net 4 Unable to See Temperature Voltage or Fan The fact that the main Hardware Sentry icon is present but no temperature voltage fan information is shown may be caused by three issues Hardware Sentry did not detect any hardware information source that is able to show the temperature voltage fan data Typically the vendor specific hardware monitoring agent that comes with the server has not b
24. Guide Configuring and Administering Gi www sentrysoftware net If the manual pre selection is chosen the Next panel displays a list of connectors to choose from Hardware Sentry Preselect connectors 3 It is possible to select several connectors A GADREEE RES from the list by simply clicking on them Unselecting a connector is also performed Select one or more connectors to monitor the current platform VBEN NT Generic MS_HW_WBEMGenericNT hdf 1 1 1 1 1 IBH Director Agent 3 x on Windows HMS_HW Director3nNT hdf by clicking on its item on the list IBM Director Agent 4 x on Windows MS_HW Director4nT hdf HP NetRaid SNMP Sub Agent HMS_HW_HpNetRaidController hdf Each item contains the display name of the hp Insight Manager Compaq Drive Arrays MS_HW CpqDriveArrayNT hdf f HP TopTools Agent MS_HW_HPTopToolsnNtT hdf connector as well as its corresponding file hp Insight Manager Server Management MS HW CplgSeruNT hdf g Mylex GAM Server with SNMP support MS_HW_MylexController hdf name on the system The list is built from NEC ESMPRO MS HW NECEsmPro hdf Fujitsu Siemens Serverview MS HW ServerviewNT hdf the content of the directory Generic NT Disk Monitoring MS HW WBEMGenericDiskNT hdf Motherboard Monitor MS_HW_MBMNT hdf SPATROL HOMES 1lib MS_HW_HDF Once the selection is made click the Finish button to let Hardware Sentry re Finish Help run a detection process with the n
25. L When the parameter value reaches these thresholds the parameter triggers a warning or an alarm and this alert is sent by the PATROL Console to the operators and administrators Note If a device appears to be missing the Status parameter will be used to trigger an alert 17 go SQ SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 a User Guide Monitoring www sentrysoftware net 1 1 Fans To avoid temperatures that are too high system vendors install fans on critical devices processors power supplies etc Monitoring fans is important because they ensure a proper temperature for the system to work efficiently Depending on the available information the Speed and or SpeedPercent and or Status parameters will be displayed for each detected fan device The Speed parameter represents the speed of the corresponding fan in rotations minute An alert is triggered if the fan speed is too low for proper functioning The SpeedPercent parameter represents the speed of the corresponding fan in percentage of its maximal speed An alert is triggered if the fan speed is too low for proper cooling The Status parameter represents the current status of the fan An alert is triggered if the fan stops spinning or does not spin fast enough 1 2 Temperatures As with any electronic device chips and other components of a computer stop working when their temperature rises too high lots
26. PE OBJECT_TYPE OBJECT_DEVICEID 5 PROBLEM Note You can use macro variables in the fields above See the documentation to have the complete list of available macros If you select the Send a basic SNMP trap action you need to enter The IP address or hostname of the SNMP manager that the trap will be issued to The SNMP community string used by the SNMP manager The text that will be sent in the SNMP trap Note Upon a hardware failure Hardware Sentry will send the trap that is defined in the PATROL MIB Trap number 11 Enterprise ID 1 3 6 1 4 1 1031 1 1 2 the text is stored in the 1 3 6 1 4 1 1031 1 1 2 1 OID Note You can use macros that will be replaced at runtime The list of available macros is described later in this chapter Hardware Sentry Knowledge Module for PATROL version 1 3 01 User Guide Configuring and Administering 11 9 Send a custom SNMP trap advanced Set Alert Actions Alert Action Send a custom SNMP trap advanced Enter the SNMP trap destination meee Ee pae If you select the Send a custom SNMP trap action you need to enter the following Enterprise Object ID 1 3 6 1 4 1 1631 1 1 2 The IP address or hostname of the SNMP n manager that the trap will be issued to Enter the varBinds to be sent with the trap 7 The SNMP community string used by the OID 4 f1 3 6 1 4 1 1031 1 1 2 1 SNMP manager Value Hardware probl
27. ROL version 1 3 01 N m j Fee ee a ee eee ee es ee eee at SOFT User Guide Inside Hardware Sentry 4an eo wwe www sentrysoftware net 3 The Detection Process As Hardware Sentry KM for PATROL is preloaded by the PATROL Agent it begins its Operations upon the agent startup Once the Hardware Sentry pre discovery PSL process has activated the MS_ HW_MAIN class the PATROL Agent spawns the Hardware Sentry discovery PSL process This PSL process takes two actions the detection and the environment discovery Here is a description of this detection process Get the list of available connectors hdf files in the S PATROL_HOME 1lib MS_HW_hdf folder For each hdf file found test the detection criteria OS type NT service processes SNMP request etc Mark the hdf files as detected once all its detection criteria is successfully passed There may be several connectors detected at one time typically one connector for the temperatures voltages etc one for the RAID disk controller and one for the non RAID disk controller Launch the discovery process Note only one hdf file can describe the computer model and manufacturer as only one icon for the computer will be created The Hardware Sentry engine will ensure that only one hdf describing the computer model will be marked as detected Note to optimize the detection process and gain some time the detection process
28. TYPE Description of the problem encountered by the monitored device Zor Example The speed of this fan is critically low 1503 rpm Description of the possible consequence of the detected problem CONSEQUENCE Example The temperature of the chip component or device that was cooled by this fan should grow quickly This can lead to severe hardware damage and system crashes Recommended action to solve the problem ine OME E E RCN Example Quickly check if the fan is really no more cooling the system If so replace the fan PATROL internal ID of the instance that triggered the alert OBJECT_ID Example MS_HW_DellOpenManagehdf_11 48 Macro OBJECT_LABEL OBJECT_CLASS OBJECT_TYPE OBJECT_DEVICEID PARENT_ID PARENT_LABEL PARENT_CLASS PARENT_TYPE PARENT_DEVICEID HOSTNAME DATE TIME ASCTIME variable_name NEWLINE n FULLREPORT Hardware Sentry Knowledge Module for PATROL version 1 3 01 User Guide Configuring and Administering Description Display name of the instance that triggered the alert Example Fan 1 1 CPU 1 Class of the instance that triggered the alert Example MS_HW_FAN Type of the device that triggered the alert Example Fan Hardware Sentry internal device ID of the instance that triggered the alert Example 1 1 PATROL internal ID of the object that the faulty instance is attached to Example MS_HW
29. User Guide Monitoring Www sentrysoftware net Section III Monitoring 1 Monitoring Temperatures Fans Voltages And Power Supplies ccccscccscccsccess 17 Tiga EATS EEE A E E ae sie aloes apnea E acme oem a OI E OR E 18 lsa a LEMPET UNOS AEE S AEE TEEN TEE ENT RETA A ERE EE ESEE AETS I8 DD PONTE 0S r A E E E E T E E IS Be SV OETA EE E E APET N sare E E E EE ETE E ETETE A EEE 19 2 Monitorim SLoradE o oo ccctccseecacsccesctnsescessaceeccoescpasanne enn EEEE 20 de S A OO CIS E E EE E E E E EE EE E EEEE ER 20 Did WY SC a AE E E ee 20 PE EE E UC D A EE E EEE S E ET EE I A aE E O T ae ETT 21 3 Monitoring Processors Memory modules and Network interfaces cccceescceeses 22 Dl FPP OC CSS OF is cee woe E APE erent A E EE eee E ase E E 22 Sel MEO 0G CN oarra EErEE EEEN EE EETA EEES NETET 23 Poe NONO SCT OCC 8 T E N T EA E A 23 d Missing Device Detection cates ces ieee ccos ee chcea canner tcamece EENEN AONNE OEE E 25 5 Connecilor Monilor iDo seis ciccannen con rorarii E S 26 16 AQ SENTRY _ Hardware Sentry Knowledge Module for PATROL version 1 3 01 i Pete OL tact User Guide Monitoring 1 Monitoring Temperatures Fans Voltages And Power Supplies mh MILL_3181 a Hardware Sentry 1 1 00 W Computer HP Netserver LPr H a Disk Controller HP NetRAID 0 Hardware Sentry automatically detects the Disk Controller MegaRAID 467 Series Controllers 1 information sources available on the Dis
30. _DellOpenManagehdf_1 Display name of the object that the faulty instance is attached to Example Computer Dell PowerEdge 1600SC Class of the object that the faulty instance is attached to Example MS_HW_ENCLOSURE Type of the object that the faulty instance is attached to Example Computer enclosure Hardware Sentry internal device ID of the object that the faulty instance is attached to Example 1 Host name of the computer that the PATROL Agent is running on Example SENTRYTEST003 Current date in the YYYY MM dd format Example 2005 6 23 Current time in the HH MM SS format Example 11 14 53 Current date and time formatted as specified in the macro Example ASCTIME m d T Y will produce Jun 6 11 14 53 2005 The available formats for the ASCTIME macro are listed in the asctime description of the PSL Reference Guide Book 2 Value of the variable_name instance variable Example worstParam will give the name of the worst parameter of the instance that triggered the alert This feature is recommended for advanced users only Linefeed This is useful to produce multi line information Full hardware health report about the instance that triggered the alert Example see the output of the Show hardware health report Menu command 49 SNQ SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 NG eee User Guide Inside Hardware Sentry laan t www sentrysoftware net en www
31. anism often known as the Override parameters feature Hardware Sentry sets the alarm and warning thresholds of each parameter of each monitored object by setting a specially formatted variable under the ___tuning___ tree in the PATROL Agent configuration This is the standard by default method to set thresholds which is supported by BMC Software since version 3 4 11 of the PATROL Agent With the 3 5 release of the PATROL Agent BMC Software started to push another way to set thresholds which is likely to become the new thresholds standard in PATROL This new method is known as PATROL for Event Management formerly known as EventSpring or PATROL Configuration Manager PCM formerly known as AgentSpring Hardware Sentry fully supports PATROL for Event Management and is able to manage its thresholds through this new mechanism By default Hardware Sentry uses the Override parameters mechanism to manage its thresholds because it is the only one which does not require any additional piece of software to run on the managed computer To make Hardware Sentry uses the PATROL for Event Management mechanism for its thresholds right click on the main Hardware Sentry icon KM Commands gt Options gt Thresholds management The following dialog box appears Hardware Sentry Thresholds Management fa x Select how the alert thresholds are managed by Hardware Sentry PCM Event Management Hardware Sentry automaticall
32. ar Bi IntrusionStatus Alarm Snooze l Memory modules InfoBox w Network interfaces m Power supplies Processsors Temperatures Create Shortcut E Temperature 1 1 Planar Temp E E Temperature 1 2 CPU 1Temp Properties imdi Modify thresholds Cut Copy Remove Desktop KM Refresh parameters J2 In order to perform certain tasks maintenance for example it may be interesting to pause the monitoring of an object by Hardware Sentry When in paused state the object is still displayed in the PATROL Console but in an OFFLINE status and no collect for that particular object is executed The monitoring of the object can simply be restarted by selecting the Resume monitoring option in the menu 58 SENTRY _ Hardware Sentry Knowledge Module for PATROL version 1 3 01 oA S User Guide Configuring and Administering www sentrysoftware net 4 Terminating Monitoring of a device Information about a device may be received by Hardware Sentry through two different sources In such a rare case the monitoring of the device is performed twice and the device icon will also appear twice in the PATROL Console PATROL Y3 5 50i Oper 1 File View Hosts Tools Options Help 5 I E Be PATROLMainMap M m amp kant Event Manager Expand right click the device to remove Hardware Sentry 1 1 00 ania Sip osc De
33. ardware Sentry processes listed above are not consuming the processor time disabling Hardware Sentry may help you find out if the vendor specific hardware monitoring agent is responsible of this overload In such a case Hardware Sentry appears to be responsible for this CPU overload but only because it is querying faulty software Therefore you should check for updates of your vendor specific hardware monitoring software is Performance DEBR W Fie Action View Favorites Window Help 2 a x am e 100 90 80 Hardware Sentry discovery Hardware Sentry collect 0 000 Average 7 514 Minimum 0 000 Maximum 67 031 Duration 8 20 HERACLITE 2 Pentium Ill 533 MHz color Scale Counter instance _ Parent_ Object Compute 256 Mo SDRAM 1 000 ProcessorTime Total Processor HERACLITE 63 Notes 60208
34. cess set of commands may be more secure than Check the corresponding boxes if you want Hardware impersonating every Sentry to use the sudo utility to run these commands command executed by Hardware Sentry m A E otez you must nave re D Zusr sbin ioscan He Spi a Note This option is available on UNIX and l fusr sbin lanscan confiaured the E usr sbin lom TPE EAEE file Linux platforms only fusr bin kstat pecs Be leg to allow the PATROL E fopt IBM director CINOM bin cimcli Agent to execute the selected commands as root Enter the command line to execute the sudo utility Leave blank to use the default usr local bin sudo path 44 Editing Alert Actions To modify the Alert Actions executed by Hardware Sentry Knowledge Module for PATROL version 1 3 01 User Guide Configuring and Administering Hardware Sentry upon a hardware failure right click on the main Hardware Sentry icon gt KM Commands gt Options gt Edit Alert Actions 11 1 Select the Alert Actions to be executed Set Alert Actions for Hardware Sentry Select the Alert Actions to be executed when Hardware Sentry detects a problem with the hardware Trigger a PATROL Event Annotate the parameter s graph Execute an OS command Execute a PSL command advanced Write a line to a LOG file Send a basic SNMP trap Send a custom SNHP trap advanced BeBe E ESES Note the Alert Actions yo
35. ctive part of Hardware Sentry KM for PATROL resides on the PATROL The tree below lists the products and components you can Agent A PATROL Agent must be installed ae install Items listed as QuickStart packages contain pre each server which hardware has to be monitored packaged products and components with pre defined configurations Expand the tree to select one or more products an d a co p y th e H ar d ware S en tr y K M fo rf components or QuickStart packages Product Selection PATROL must be installed on each PATROL L a r Installation Results Pc ar Fal as Agent This version of Hardware Sentry KM for PATROL cannot monitor the hardware remotely Once Hardware Sentry KM for PATROL is integrated within the PATROL framework the Select Products and Components to Install Estimated Disk Space Needed 47 Disk Space Available 32504 MB hardware information and status of the monitored servers should be available in PATROL Please refer to the Hardware Sentry KM for PATROL Installation Guide document for further details on the installation procedure 11 Kan SN SENTRY W SOFTWARE eo www sentrysoftware net Hardware Sentry Knowledge Module for PATROL version 1 3 01 User Guide Getting started 3 The need for vendor specific hardware monitoring software Quite often the standard operating system layer is not a sufficient hardware information source and most computers req
36. d when the power supply s maximum power output is reached The Status parameter represents the current status of the power supply An alert is triggered if an error occurs with the power supply 18 Hardware Sentry Knowledge Module for PATROL version 1 3 01 User Guide Monitoring 1 4 Voltages Power supplies convert the AC line power into voltages and currents needed by the motherboard of the computer The stability of the motherboard and therefore of the overall computer strongly depends on this power converter Voltages that are too low or too high may lead to unpredictable system crashes Monitoring the value of the different voltages needed by the motherboard will help in detecting unstable system instability Depending on the available information the Voltage and or Status parameters will be displayed for each voltage sensor on the motherboard The Voltage parameter represents the voltage output in milliVolts mV An alert is triggered if the voltage goes out of the proper range The Status parameter represents the current status of the voltage An alert is triggered if the voltage output is too low for proper functioning 19 Hardware Sentry Knowledge Module for PATROL version 1 3 01 User Guide Monitoring 2 Monitoring Storage Hardware Sentry automatically detects the Disk Controller Adaptec AIC 7899 Ultrai60 m PCI SCSI Card 0 information sources available on the ge P
37. directly shows that it is missing 25 Hardware Sentry Knowledge Module for PATROL version 1 3 01 User Guide Monitoring 5 Connector Monitoring When a Hardware Sentry connector has been detected as applicable to the current platform a corresponding instance is created in the PATROL Console and its status is monitored regularly to ensure that the underlying technology is still available Example Hardware Sentry is running on a Dell server with Dell OpenManage Server Administrator Upon startup Hardware Sentry detects Dell OpenManage Server Administrator and starts using the corresponding connector to discover the server hardware configuration and monitor the discovered devices Additionally Hardware Sentry creates an icon in the PATROL Console representing the Dell OpenManage Server Administrator connector Every 2 minutes its Status parameter is updated If for some reason the Dell agent stops working an alarm is raised on the Status parameter and the devices that were discovered through to this connector are taken offline DIDEROT Hardware Sentry 1 2 00 Computer IBM RS 6000 7043 Disk Controller Wide SCSI 1 0 Controller 04 00 Network interfaces Processors Pre selected connectors Connector for IBM RS 6000 pSeries and p5 servers AIX FM Status TestReport This connector monitoring mechanism helps PATROL administrators detect hardware agent failures It also provides
38. disk and network As it does not process megabytes of data Hardware Sentry should not overload your system Hardware Sentry typically uses about 4 percent of the processor s time To diagnose if Hardware Sentry uses too much processor time check which processes are overloading your CPU Hardware Sentry is a module that runs inside the PATROL Agent Therefore you should check the PatrolAgent Exe process But keep in mind that Hardware Sentry is not the only KM that runs on your PATROL Agent Additionally Hardware Sentry may spawn the following processes Operating system Processes that Hardware sentry may be spawn CSCRIPT EXE Under Windows MS _HW_MBMClient EXE SW_Process_List Exe SW_Servicelnfo Exe prtdiag lom iostat awk grep sed Under Sun Solaris cimcli Under Linux mii tool ethtool diskinfo ioscan Under HP UX lanscan pvdisplay awk grep Isdev errpt entstat Ispv Under AIX 62 2 SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 y SOFTWARE User Guide Troubleshooting If the PatrolAgent Exe process takes all the processor time disabling Hardware Sentry in the agent configuration AgentSetup disabledKMs is a good way to find out if Hardware Sentry is the problem If the PatrolAgent Exe does not overload the CPU anymore it probably means that Hardware Sentry is responsible for this overload If the PatrolAgent Exe process and the other H
39. dware Sentry etc PATROL Central Untitled1 PATROL Central Action View Tools E Annotation Additional information regarding the VOLTAIRE_3181 MS_HW_DellOpenManagehdf_11 Speed PATROL Cental ce kl a eh Tasks D wp DIDEROT_3181 H HERACLITE_3181 gi KANT_3181 Es Application Sentry 8 3 0 Blackout KM for PATROL Hardware Sentry 1 2 00 Computer Dell Powe Disk Controller H Power supplies Processors JF Temperatures Voltages uy IntrusionStatus Detected connectors l 3 Connector for O Connector for 3 Connector for a8 Health At A Glance gH MILL_3181 H MONTESQUIEU_3181 gi PLATON_3181 H ROUSSEAU_3181 H VOLTAIRE_3181 For Help press F1 Wednesday June 22 2005 6 24 24 PM 1 500 00 Units RPM Hardware health report Wed Jun 22 18 24 24 26005 Monitored object Fan 1 1 CPU 1 Type Fan PATROL object ID MS_HW_FAN MS_HW_Dell0penManagehdf_11 Internal device ID 1 1 This object is attached to Computer Dell PowerEdge 1666SC Current value 1566 rpm Unit RPM rotation per minute Alert thresholds If Speed is in the 6 rpm 1686 rpm range Trigger an ALARH If Speed is in the 1686 rpm 1966 rpm range Trigger a WARNING S P P pm g gg Problem The speed of this fan is critically low 1566 rpm ae Consequence The temperature of the chip component or device that was cooled by EE
40. e it is shown in the PATROL Console by Hardware Sentry In addition if the processor is able to predict a future failure this information will be monitored by Hardware Sentry and shown in the PATROL Console Depending on the information available the Status and or CorrectedErrorCount and or PredictedFailure parameters will be displayed for each discovered processor CPU The Status parameter represents the current status of the processor An alert is triggered if the processor is not available for proper operation missing disable by the BIOS due to a POST error etc The PredictedFailure parameter reports the predictive failure analysis performed by the processor itself This information is based on the rate of corrected errors The CorrectedErrorCount parameter represents the number of errors that have been automatically corrected by the processor This information can be very useful to predict a failure in the near future 22 Hardware Sentry Knowledge Module for PATROL version 1 3 01 User Guide Monitoring 3 2 Memory Modules The main memory of a computer is actually as critical as the processors since almost all processor operations deal with the memory A single memory fault will lead to severe computer crash with potentially data corruption On servers the memory modules the devices where the memory data is actually stored often include auto correction features ECC and sometimes even better RAID
41. ecoecoococesoccoscsecoecoccoecoocosesoesoecoecoo 59 4 Unable to See Temperature Voltage or Fanr ccccc ccc cece ccccccncccsc ccc ccncccscccsccssccesces 60 Unable to See Disk Cont rower vibesiciicciss ceases ccusseinensscnnacteccssversseseednicisecsiessatewneanesct 61 6 Pie CPU Is OVCLlIOAGCO si scscacetssovnvecisesacearstiecwiesescesaeeteroieebeseiwewneseesewess eri ccnreneass eens 62 56 SN SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 QO m 2 jz e a SOFT User Guide Troubleshooting an www sentrysoftware net 1 Enabling the Debug Mode By default Hardware Sentry sends only the most critical information warning and error messages to the System Output Window of the PATROL Consoles Most often this information is accurate enough to ensure that Hardware Sentry is properly working If you encounter a bug and wish to report it to Sentry Software you will be asked to enable the Debug Mode and provide the debug output to the Sentry Software support team Hardware Sentry Debug mode Debug mode Debug file path optional To enable the debug mode right click the main Hardware Sentry icon KM Commands If no debug file path is provided the debug information will be sent to the System Output Window of the PATROL gt Options gt Debug Console Check the Enable debug box Warning The debug mode may slow down the KH cancer By default Hardware
42. een properly installed Hardware Sentry properly detected the vendor specific hardware monitoring agent that comes with the server but this agent is unable to provide any temperature voltage fan information Please check that all the vendor specific software is up to date the hardware agent itself the BIOS of the motherboard the BIOS of other components like management processor if any etc No temperature voltage fan sensor is provided with the computer on a few Sun servers all HP UX systems and most of the IBM RS 6000 computers 60 SNQ SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 NG eae User Guide Troubleshooting fan www sentrysoftware net oo www s 5 Unable to See Disk Controller On some computers the vendor specific hardware monitoring agent is only responsible for the monitoring of the baseboard temperatures fans voltages and power supplies In this case the disk monitoring is handled by the disk controller manufacturer Therefore you need to install additional software for your disk monitoring Please contact your server vendor to know which software must be used with your disk controller 61 Ae SNQ SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 eee User Guide Troubleshooting C www sentrysoftware net 6 The CPU is overloaded Hardware Sentry has been optimized in order to use as few system resources as possible CPU memory
43. eir sole purpose is to graphically represent how the disks are arranged in the computer Please note however that if a controller goes missing all its disk dependencies will be missing as well 2 2 Physical Disks Physical disks must be monitored to avoid loss of data unavailability and performance degradation When available the S M A R T technology will be used to predict a disk failure before it occurs Depending on the available information the PredictedFailure ErrorCount and or Status parameters will be displayed for each discovered physical disk The PredictedFailure parameter uses the S M A R T technology to predict physical disk failures An alert will be triggered if it is predicted that the Physical Disk will soon break down The ErrorCount parameter is incremented each time an error occurs on this physical disk An alert is raised from the first detected error The operator can reset the counter and clear the alert through a Menu command in the PATROL Console The Status parameter represents the current status of the physical disk An alert is triggered if the physical disk is not available for proper operation 20 SNQ SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 NG eee User Guide Monitoring fan Wwww sentrysoftware net 2 3 Logical Disks RAID or advanced disk controllers expose several physical disks as a single logical disk to the operating system The stat
44. em with OBJECT_LABEL PROBLEM All the characteristics of the trap OID 2 i 3 6 1 4 1 1031 1 1 2 2 Enterprise ID trap specific number and up Value R FULLREPORT to 4 varbinds Pima Note You can use macros that will be replaced at runtime The list of available macros is described later in this chapter Value OID 4 Value Note You can use macro variables in the fields above See the documentation to have the complete list of available macros 11 10 Macro Variables Macro variables can be used in the Alert Actions to customize a string with information related to the encountered hardware failure display name of the faulty object value of the parameter description of the problem etc Example You can configure Hardware Sentry to trigger a PATROL event containing the display name of the faulty object by entering something like Hardware problem with OBJECT_NAME in the Trigger a PATROL Event dialog box Note You can use several macros at one time The table below recapitulates the available macros in the Alert Actions Macro Description Name of the parameter that triggered the alert G aiei E Example PredictedFailure Formatted value with unit of the parameter that triggered the alert VALUE Example 67 3 C Raw value of the parameter that triggered the alert Example 67 30000 Alarm type INFORMATION WARNING or ALARM Example ALARM RAW_VALUE ALARM_
45. emove Refresh parameters Desktop KM Hardware Sentry Hodify thresholds for Voltage Core 1 Use Hardware Sentry default thresholds The first panel of the Modify thresholds Use customized thresholds f explains the purpose of this option By default Hardware Sentry retrieves and sets Here you are asked to choose between letting default thresholds for each parameter These Hardware Sentry use the default values or thresholds should be accurate and there should be no need to change them However if you would modifying them and setting your own like to modify them it is possible to do so by thresholds selecting the Use customized thresholds options If you select the first option Use Hardware Sentry default values which is the recommended one and click the Next button the next panel will display information about the parameter thresholds There is no choice to make in this panel besides confirming the use of default thresholds by clicking the Finish button Clicking on Back brings you back to the first window where you can modify your choice and select the Use customized thresholds option 38 0 SNQ SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 s SOFT User Guide Configuring and Administering an G www sentrysoftware net le This option brings you to a different panel where you are asked to enter new thresholds for the paramete
46. entry KM for PATROL is a module that allows BMC Software PATROL to monitor computer hardware disk status temperatures etc It gathers hardware information from different sources vendor specific agents standard management technologies SNMP WBEM etc and displays this information within the PATROL framework In order to work properly Hardware Sentry KM for PATROL needs certain hardware information sources to be available Depending on the platform Hardware Sentry KM for PATROL will rely on vendor specific agents and or on standard management technology such as WBEM or SNMP On startup Hardware Sentry KM for PATROL automatically detects which hardware information source is available and then uses this source to monitor the hardware of the computer Ae SN SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 E meee User Guide Getting started an sS www sentrysoftware net The big picture below represents how Hardware Sentry monitors the hardware of a Fujitsu Siemens computer This is a good example to show how Hardware Sentry works lt bmc PATROL Console Hardware Sentry KM for PATROL mc Siemens Serverview M ylex WBEM PATROL Agent Connector Connector Connector snmp 4 SNMP gt WBEM A P 4 oO FUJITSU me qa og SIEMENS MYLEX D GLOBAL ARRAY MANAGER pF 5858 S55 58 D LL 35855 35252 aR S525 28558 2ES5e A Jgd 55855 28865 SS5 5 sSeoe
47. ery process Finally when the detection and discovery processes are complete Hardware Sentry starts collecting data about the discovered hardware environment status temperatures voltages etc by querying the detected hardware information sources as described in the corresponding hdf files This is called the collection process The table below shows the synopsis of the actions performed by Hardware Sentry PRE DISCOVERY process of the MS_HW_MAIN Activation Activates the MS_HW_MAIN class class Executed once on the PATROL Agent startup Creates the MS_HW_MAIN icon License check Look for a valid license key Activates all of the Hardware Sentry classes Tests each connector in order to detect which Detection Process hardware information sources are available and can be connected to DISCOVERY process of the MS_ HW_MAIN class Executed every hour default period Discovers the hardware environment by querying the previously detected hardware information sources Discovery Process Creates the other class instances MS HW_TEMPERATURE MS_HW_PHYSICALDISK etc Polls previously detected hardware information Collector parameters of the MS_HW_MAIN class Collection Process sources to gather data about the hardware fanColl logicalDiskColl otherDeviceColl etc environment Executed every 2 minute default period 51 SNQ SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 NG eee User Guide
48. etwork It is therefore essential to make sure these devices are properly running and linked to the network For each network interface discovered the Status ErrorPercent and or LinkStatus parameters are displayed The Status parameter represents the current status of the network interface An alert is triggered when the network interface is not responding The ErrorPercent parameter gives the percentage of transmitted and received packets that were in error due to low level protocol error or physical media problems A warning is raised when this percentage goes above 10 and an alarm when above 30 A high error percentage often means that there is a serious problem with the cable or the interface The LinkStatus parameter shows the current status of the network interface to the network i e if it is plugged in By default Hardware Sentry will trigger a warning if a network interface previously connected to the network is now unplugged However it will not trigger an alert for network interfaces that have never been connected To change this setting right click the main Hardware Sentry icon and select KM Commands gt Options gt Network link monitoring This will display the dialog box shown below 23 Q SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 A SOFT User Guide Monitoring www sentrysoftware net m Hardware Sentry Network link monitoring settings BHA Hetwork link moni
49. ew connector selection Note At the end of the Pre selection wizard Hardware Sentry will run a full discovery All devices and sensors will be re detected If the connector list differs from what Hardware Sentry was using previously new devices monitored may appear in the PATROL Console and old ones may be removed If this is the case pay close attention to the PATROL System Output Window for error messages A bad selection of connectors may remove all devices and sensors and may prevent Hardware Sentry from being able to monitor the system s hardware 41 SNQ SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 SOFT User Guide Configuring and Administering Gi www sentrysoftware net 9 Enabling Disabling Missing Device Detection Hardware Sentry can detect missing devices in the system If at one point information about a device that was previously being monitored cannot be obtained anymore the device is considered missing In this case an alert is triggered and the missing status of the device is added to its label 2 Hardware Sentry Missing device detection The missing device detection works for all the lig se ect ca devices discovered by Hardware Sentry It is enabled by default and can be disabled By default Hardware Sentry monitors and detects missing through a KM Command right click oal the devices i e devices that may disappear during the monitoring or main Hardware Sentry icon gt
50. fic agents or standard technologies Hardware Sentry KM for PATROL will monitor the status of the disks RAID and non RAID disks the speed and or status of the fans the temperatures the voltage levels the status of the power supplies the processors the memory modules the network interfaces Actually Hardware Sentry KM for PATROL does not support hardware platforms but hardware information sources Currently Hardware Sentry KM for PATROL supports the following hardware information sources Adaptec Storage Manager both Standard and Browser Editions Dell OpenManage Agent under Windows and Linux Fujitsu Siemens Serverview Agent under Windows and Linux Fujitsu Siemens BX300 600 Management Module HP NetRAID SNMP sub agent under Windows HP Management Agents under Windows Linux and Tru64 HP TopTools and HP Insight Manager Agent HP UX standard monitoring tools IBM Director Agent 3 1 4 1 and 4 2 under Windows and Linux IBM BladeCenter Management Module IBM RS 6000 pSeries and eServer p5 standard tools under AIX LSI Logic Power Console Motherboard Monitor licensed by Alexander van Kaam mbm livewiredev com Mylex GAM SNMP sub agent under Windows NEC ESMPRO Agent under Windows Sun standard hardware monitoring tools under Solaris WMI under Windows S M A R T disks and network interfaces Therefore Hardware Sentry KM for PATROL is able to monitor the following platforms
51. ftware net oo www s 5 The Collection Process Once the discovery process is complete the collection process starts Every two minutes Hardware Sentry spawns several PSL processes fanColl physicalDiskColl logicalDiskColl temperatureColl etc that are responsible for the collection of information about a given device type For example fanColl will gather fan information from the different detected hardware information sources as described in the corresponding hdf Fan Collect section These PSL processes are attached to the main Hardware Sentry icon MS_HW_MAIN class and not to the corresponding classes and instances No collection is performed on objects that have been paused When a device has been marked as missing by the discovery process i e had been discovered and is no longer discovered the collection process no longer queries the hardware information source and simply sets the status of the object to alarm 55 oft 5 4r SNQ SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 I a User Guide Troubleshooting laan t www sentrysoftware net en www s Section VI Trouble shooting 1 Enabling the Debug Mod sc sccscacsccvesvouns cncesescecussuecedinscus veseawss apwatecceesvenvavaseewscesessees 57 2 Reinitializing Hardware SOU try siscsscssccosccscaseecossunsererdeanssneweseivawncssesnsesaceseesdencsGesmes 58 3 Hardware sentry SHOWS Nothing essecsecsococesoeco
52. gt KM Commands gt License w Windows InfoBox Reinitialize Cut Restore device or sensor monitoring Copy Show connectors currently in use Create Shortcut a Options F Properties pare F xMcommands ee ADOT Hardware Sentry License License Information for Hardware Sentry A dialog box appears and shows the cists license Hardware Sentry is currently Renictereu Licences working with automatic 30 day trial term yee ba license or perpetual license Automatic trial expires on Nov 5 2664 29 day s remaining To enter a new license key click on the Register a new license button Hardware Sentry Add license lease enter the license information which Choose the key type depending on the as been provided to you by your reseller license you purchased perpetual license o l or term license Product name MSHW1 If you purchased a Term License you Key type i u type Perpetual license v have to enter your Term License Expiration date J Af expiration date If you purchased a Perpetual License you TE A e should leave the expiration date field License key 9123 4567 89AB CDEF blank leave empty for perpetual license Then enter your license key in the last field and press the Register button Register Cancel 29 Q SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 Ji SOFT
53. hysical disks monitored computer and displays the g Physical Disk 0 SEAGATE ST39175LW SCSI Disk Device hardware information provided by those 4 4 Disk Controller Mylex DACS60pt1 0 sources in the PATROL Consoles in the S Logical disks corresponding container Logical Disk 0 0 RAID 1 8 5 GB If at least one hardware information source Physical disks provides useful data about the storage in the Physical Disk 0 0 0 SEAGATE 9 GB Dead computer an icon will be created for each a Physical Disk 0 0 1 SEAGATE 9 GB storage related device in the PATROL a Physical Disk 0 0 2 SEAGATE 9 GB Console disk controllers physical disks and logical disks Physical and logical disks icons are created below the disk controller icons they are attached too Each icon is labeled with a description of the device ID size vendor role etc Note If a device appears to be missing the Status parameter will trigger an alert if necessary 2 1 Disk controllers A disk controller is a card inside a computer that connects one or several physical disk drives to this computer Some intelligent disk controllers such as RAID controllers manage several physical disks as a single logical disk which is the only disk exposed to the operating system Monitoring both physical and logical disks is essential to ensure that storage is available Disk controllers do not have a parameter and no information about them is collected Th
54. ices in the computer an icon l CPU 1 1 Intel Pentium IV 2 80 GHz will be created for each device in the PATROL E Console processors memory modules and z Voltages ly IntrusionStatus network interfaces Processors memory modules and network interfaces icons are created under a computer icon Each icon is labeled with a description of the device ID size vendor role etc Note If a device appears to be missing the Status parameter will trigger an alert if necessary 3 1 Processors Processors also called CPU Central Processing Unit are obviously the most critical devices within a computer While a processor fault may often lead to a system crash without a chance for a monitoring tool to catch the error it can still be useful to monitor a server s processors In the case of a system crash due to a processor fault the system reboots automatically The reboot is either triggered by the operating system or by the motherboard itself If a processor is no longer working it is automatically disabled by the BIOS and if there is one processor left the operating system starts with one processor less Hardware Sentry monitors each processor and checks that it is present and running If a processor is missing upon a reboot Hardware Sentry will trigger an alert On some recent or high end servers processors are able to correct some operation errors by themselves like the ECC memory If this information is availabl
55. k Controller Symbios Logic 8951U PCI SCSI Adapter 0 monitored computer and displays the hardware g Fans 3 Ses information provided by those sources in the fF g Fan 0 Power Sup Fan 1 Fan 2 Chassis Fan PATROL Consoles in the corresponding H Memory modules container G4 Network interfaces B Processors If a hardware information source provides El 2 Temperatures useful data about temperatures fans voltages G fF Temperature 0 Sec Proc Temp or power supplies an icon is automatically gig Temperature created for each sensor found in the system as H 5 Temperature 1 Pri Proc Temp well as one or more parameters jF Temperature 2 Sys Bd Temp fF Temperature 3 HS Cage Temp l EQ voltages E MILL MS_HW_TEMPERATURE 1 Temperature File View Options Type Help Sek E AA We a i l U S P Temperature For each displayed sensor one or more graphs are built by polling the sensor every two minutes These graphs can be viewed by double clicking the parameter under the corresponding icon in the PATROL Console The location or the type of the sensor may appear in the icon label if this information 1s available as well State Host Name _ Last Value Units ox MILL JMS HW TEMPERATURE 1 Temperature 26 000 Celsius degrees C lt 1Total Parameters 1 0K Depending of the platform and sensors and if possible alert thresholds are automatically set by Hardware Sentry KM for PATRO
56. ll PowerEdge 1600SC PADRI 7 E EN Disk Controller PERC 4 SC Controlle mere REIS 3 E ras SE In order to avoid such redundancy the EEE Fan 1 1 CPU 1 7 p D Fen 12 frone solution is a simple remove one of the EEE Fan 1 3 Rear i s BE insite devices monitored in the PATROL EEE Memory modules G Rh interfaces InfoBox C onso 1 eC E Power supplies Cut A P r E ie Copy Create Shortcut E Temperature 1 2 CPU 1 Temp T i ee 55 M Voltages iain select the Remove option from the KM Commands Modify thresholds KM Command menu Pause monitoring BDesktop_J KM Refresh parameters The device has been removed from monitoring which means collection for this particular device will no longer be performed and the redundant icon representing the device will disappear from the PATROL Console 33 Hardware Sentry Knowledge Module for PATROL version 1 3 01 User Guide Configuring and Administering PATROL V3 5 50i Oper 1 File View Hosts Tools Options Help Event Manager ene lE options Parameters Ry PATROLMainMap amp kant Hardware Sentry 1 1 00 i o It is possible to restore device monitoring Disk Controller PE TE umn me with Hardware Sentry after it has been w Fan 1 2 Fron InfoBox MM Fan L3 Rea o removed from monitoring through the IntrusionStatus 66 29 FEE Memory modules Copy Remove option w Network ng
57. ll be lost Note After reinitializing Hardware Sentry it is not possible to go back to the previous configuration and so all changes will be lost Hardware Sentry Reinitialize Reinitialize Hardware Sentry Reinitializing Hardware Sentry will erase all configuration To Reinitialize Hardware Sentry right click and threshold information It will destroy all previously c6 Be 9 discovered objects and trigger a new platform detection as the computer icon and select Reinitialize well as a full discovery of the environment from the menu uate sabe anene be reset during the reinitialization of You can select which settings of Hardware ardware Sentry Reser the Security Seteings Sentry need to be reset Select all options if Unload connectors and perform a platform detection you think that you made some configuration Reset the debug mode Reset all thresholds errors Reset the threshold management mode to default RESPE BEE BENE SEECANgS Click on the Reinitialize button to proceed or the Cancel button to cancel This can be useful when Hardware Sentry does not act as expected after some configuration option changes or after a software hardware upgrade Reinitialize 58 Hardware Sentry Knowledge Module for PATROL version 1 3 01 SOFT User Guide Troubleshooting oe www sentrysoftware net 3 Hardware Sentry Shows Nothing Even if Hardware Sentry is unable to detect any available hardware information
58. nd preloaded If you already use PATROL for Event Management in order to modify or customize your thresholds it is strongly recommended that you use this option No threshold Hardware Sentry will not set any threshold on its monitored objects and lets you specify the thresholds through your preferred method Warning In order to avoid side effects and unpredictable behavior if you select the default or the PCM Event Management thresholds management option Hardware Sentry will automatically remove the thresholds set through the other method only for its monitored objects That is if you select the PCM Event Management option Hardware Sentry will automatically remove all the thresholds set for its monitored objects stored in the ___ tuning __ tree in the PATROL Agent configuration Note Hardware Sentry is fully compatible with the PCM Event Management thresholds management mechanism All the classes are provided with no threshold set Note the default and PCM Event Management thresholds management options still imply that Hardware Sentry automatically sets these thresholds during its discovery In other words if you use the Override parameter interface in the PATROL Console or the PATROL for Event Management interface to manually set thresholds on the Hardware Sentry parameters they will be quickly less than one hour overwritten by the discovery process For details on how to manually set thresholds on the Hardware
59. of unrecoverable errors crashes and even hardware hurt This temperature may become too high when the device is abnormally overloaded when a fan is not working properly or when the ambient temperature is too hot Monitoring the temperatures of the critical devices of your system allows you to take action before a crash occurs Depending on the available information the Temperature and or Status parameters will be displayed for each detected temperature sensor The Temperature parameter represents the current temperature reading in Celsius degrees C An alert is triggered if the temperature becomes too high The Status parameter represents the current status of the temperature An alert is triggered if the temperature gets too high 1 3 Power Supplies The power supply is the component that transforms the AC Line into electric power needed by the computer Therefore the power supply is a highly critical device of a computer that should never fail Due to this many vendors build servers with redundant power supplies Monitoring power supplies allows the operators to be alerted when a power supply fails or even in some cases when a power supply is overloaded Depending on the available information the UsedCapacity and or Status parameters will be displayed for each power supply or power unit device The UsedCapacity parameter represents the power supply s power currently in use as a percentage An alert is triggere
60. on sources are available the hardware information of the monitored computers will be displayed in the PATROL Consoles PATROL Classic Consoles and PATROL Central The operators and administrators just have to look at the PATROL Consoles to see the hardware health of their monitored servers in the same way they see the Operating system or database health Under each server icon a tree represents all the monitored hardware for this server mj KANT_3181 fx Hardware Sentry 1 1 00 Computer Dell PowerEdge 16005C E Disk Controller PERC 4 SC Controller 0 1 si Logical disks m Logical Disk 1 RAID 5 67 7 GB Degraded 4 Physical disks Physical Disk 1 SEAGATE 36 GB Online Physical Disk 2 SEAGATE 36 GB Online g Physical Disk 3 SEAGATE 36 GB Unknown status z g Fans Fan 1 1 cPu 1 g Fan 1 2 Front a Fan 1 3 Rear Memory modules Memory 1 1 DDR 256 MB Network interfaces Power supplies g Processsors CPU 1 1 Intel Pentium IV 2 80 GHz fF Temperatures iF Temperature 1 1 Planar Temp jF Temperature 1 2 CPU i Temp Voltages iy IntrusionStatus 13 SNQ SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 Kas SOFT User Guide Getting started ans www sentrysoftware net 5 What sort of hardware can be monitored with Hardware Sentry For each supported hardware information source such as vendor speci
61. orarily Suspend the Monitoring of an Object ssessessessosseecoscoscoessessosoesseeoo 32 4 Terminating Monitoring of a device essessessossoesoescoccescoecoeccescosoesooecoesoescecoeesoeoo 33 5 Restoring the Monitoring of a Device after RemoVval ssessescescessessossoesoesoesoesseeoo 34 6 Choose the way Thresholds are Managed by Hardware Sentry ccccscccscccsccesces 36 7 Modifying Parameter Thresholds sessesocossocsoceoecoesoossecsoccoscoesoosoecooccosoeesoesoecsecoo 38 8 Pre Selecting Connectors optimizing the detection Process csscccccccsccccccccccecces 40 9 Enabling Disabling Missing Device Detection cccccccccccecccscccsccecccscccscccscccsccesces 42 10 DEMING Security OPIO cccscccccccscccsssersscessrsvascsseseoteseesnetasessescscececimeresecuoscersus vases 43 10 1 SNMP COMMUNI Y yes essence eeaxeriewwias dovtss damewidbard AE EAE EE EAE E 43 10 2 Allernale User ACCOUN esirin rin e ea ET NR EUNTES 43 EER om 5610 EEES ANE EEE eT ee E EAE AEE 44 28 Hardware Sentry Knowledge Module for PATROL version 1 3 01 User Guide Configuring and Administering amp HERACLITE EEE Application Sentry 8 2 01 Options j a160 m PCI SCSI Card 0 Parameters Parameters without Icons E Open e Pause monitori E i mand Right click on the main Hardware Sentry i e PEPA ie icon
62. ose from 34 N SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 SOFT User Guide Configuring and Administering oe www sentrysoftware net Hardware Sentry Restore device or sensor monitoring The next window shows the list of devices of the selected type that can be restored Simply select one of several devices select the device s or sensor s to restore monitoring you would like to restore from Wot ae tae una eaa ccon iy the list and click the Finish button Checking the Restore all devices and sensors in the list above box will restore all devices of the selected type without selecting them one by one from the list Restore all devices and sensors Finish Herp Note If you have chosen to restore All devices monitoring in the first window this second window just displays the list of devices that will be restored for information 35 SS 8 SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 a SOFT User Guide Configuring and Administering www sentrysoftware net 6 Choose the way Thresholds are Managed by Hardware Sentry By default Hardware Sentry automatically sets the alarm and warning thresholds on the monitored parameters Depending on the computer it is running on Hardware Sentry will use different thresholds for temperature parameters voltage parameters etc To achieve this Hardware Sentry uses a standard internal PATROL mech
63. r Setting them up is easy as you enter thresholds as if you were making a sentence On the left side the drop down box is used to select the severity of the event triggered when one of the thresholds is reached none Do nothing warning or an alert The purpose of the right side of the panel is to enter the thresholds themselves Hardware Sentry Customize thresholds for Temperature Sensor 1 Parameter ID HS_ HW TEMPERATURE Sensor1 Temperature Do nothing if Temperature is not between G and p A Trigger an ALARH if Temperature is between e and 5 Trigger an ALARH if Temperature is between 7o and 299999 Hote the unit for this parameter is celcius degrees C Note If you have selected the default thresholds management option see previous section your customized thresholds are stored under the ___tuning___ tree in the PATROL Agent configuration and then can be modified through the Override parameter feature of the PATROL Console If you have selected the PCM Event Management thresholds management option your customized thresholds are stored under the AS tree in the PATROL Agent configuration and then can be modified through the PATROL for Event Management EventSpring interface or through the PATROL Configuration Manager graphical user interface If you have selected the no threshold option your customized thresholds simply will not be stored 39 SS 8 SENTRY Hardware
64. red network environments where all the SNMP agents are configured in a customized community 10 2 Alternate user account Using an alternate account different from the PATROL default account to execute external commands OS commands and WBEM queries is needed in secured systems where the PATROL Agent default account does not have all the privileges required in order to perform queries against some hardware devices This is also true for HP UX systems to access physical disks as well as IBM xSeries computers under Linux to make WBEM requests Please refer to the Hardware Sentry Installation Guide for more information specific to these platforms In the Security Settings dialog enter the username password to be used to execute eternal commands Leave these fields empty if you want Hardware Sentry to use the PATROL Agent default account Note In order to take effect these settings need the agent to be restarted 43 Hardware Sentry Knowledge Module for PATROL version 1 3 01 User Guide Configuring and Administering 10 3 Sudo Options Depending on the targeted platform Hardware Sentry may use some external system utilities to gather hardware information Sometimes the PATROL Agent default account does not have sufficient privileges to execute these commands and it is not possible in your environment to give super user rights to the PATROL Agent as described above In such case Hardware Sentry uses the sudo u
65. ring and Administering ccc cece eee een eeeeee ee eeees 28 1 Entering A License Key For Hardware Sent ry cccccccccccccccccccccccccccccccccssccssccccscces 29 2 Distributing the License Over Several Computers cccccceccccceeccccccccccceccccsecccceees 31 3 Temporarily Suspend the Monitoring Of an ODjeCt ccc ccc ccc ce ccc ce cc cceccccsscceccceeees 32 4 Terminating Monitoring Of a device cc ccc cece cece ccc ce cc ccnncceecccensceescceesccesscsesceeesceees 33 5 Restoring the Monitoring of a Device after ReMmoOval c ccc cece cece ccncccecccscccsccesces 34 6 Choose the way Thresholds are Managed by Hardware Sentry cccescccccccsccesces 36 7 Modifying Parameter Thresholds scciccnccieacsiewsn ce ctavasdnrcscotesdeiebeceseeeecevieresecnexeeues 38 8 Pre Selecting Connectors optimizing the detection Process cccsccccccccccccccccccecces 40 9 Enabling Disabling Missing Device Detection ccccccc cece cccccccceccccecccsscccsscccesccees 42 10 Seteine Security OC PLiOMs asieecsserscsine ioti eei cacwsareer nasser sccindenesvee batee cusses gules ececehcanees 43 VD Ediling AVC FC ONS ic cneynces coo vairedecnsech es ce vanesveeee EE ERN evens oes ceueonee E 45 Section V Inside Hardware Sentry ccccccccceee ee eeeeeneeeeeenneeseeenneeees 50 Le AP CHIC CCOUY sotecssearesiwrasoiianr E T ewer ieteemeniecis 51 2 The HDF files connectors a5 oc sineei dan cise dediw
66. s ce caer eee ee Contents DECHION l OVervVi W aire cieere ce eeeecd cern eecoweneeetoneresen E EAEN 7 SECTION Ul Getting Stalte iri ste ers cnt cee scorer cavennarewndeererencewevensteesssedeees 8 L Hoy doces it WORK ci nicctsaisianseressssiwasseere sat dis ERa EA RERE RE 9 2 Integrating Hardware Sentry KM for PATROL essssessessssessesoscessesessesessesessesseo 11 3 The need for vendor specific hardware monitoring softWare sesseseesesceesesseseeseo 12 4 Using Hardware Sentry KM for PATROL to monitor hardWare sssseseesesessesseo 13 5 What sort of hardware can be monitored with Hardware Sentry ssescescessesseeoo 14 SOCIO TN MOMIVOTIDG casenesec cep ee cree teeerenecuwcdae tocar siececeecadueee cet wecunsecenss 16 1 Monitoring Temperatures Fans Voltages And Power Supplies ccccscccscccsccees 17 2 WhOMI OPING StOE AGC ac sconces csstcescstescsecescezereresateccesschesecee assis cceecesceccceuaeaesseeeerese este 20 3 Monitoring Processors Memory modules and Network interfaces ccccceccceeees 22 4 Missing Device Detection siscasscccnccrcccersscecscanseewicousiscocesceneeetnscendcousviaweeseunicnredenseeces 25 Se Conneclor MONICOPING cisicncceacasedens wocatheacuaiawedesassoeconesmanceesedensoaseeteeasesesathaseuseecesce ss 26 0 Alert ACHODS giiceiericuatesanis consi aden sieweot ontuisioe sa snd nrcansiania dee saasbercsnoue mies saamotentaues nee 27 Section IV Configu
67. s Section V Inside Hardware Sentry L Architec tUr goceusticrsnaciasciadaiaananasucne EA E E EAA 51 2 The HDF files Connectors acccaccscsacseriadcecdcsicscasescrscnececcesnseeses des ceetewssenmestesesencetecess 52 3e The Deteccion Proc eS S vivessrewrtannssodiedresenaeastevs T 53 p Fhe Discovery ProCESS ceydcncccegesweneveswsenvesedsseenec NEURE EE ENEE ERS 54 S The ollection Process sesscsis sise sscsdveassvasesiecenscedueasssnniatecentetsssetecwassecssuesasosnadeaces 55 50 Hardware Sentry Knowledge Module for PATROL version 1 3 01 User Guide Inside Hardware Sentry 1 Architecture As described in the first chapter Getting Started Hardware Sentry KM for PATROL is mainly composed of a common hardware monitoring engine MS_HW_ km KM classes and ms_hw_ lib PSL libraries that run with the PATROL Agent of each monitored computer several connector files MS_HW_ hdf that are platform specific Upon startup Hardware Sentry tests each connector in order to detect which hardware information sources are available vendor specific hardware agents standards instrumentation layers etc This is called the detection process Once Hardware Sentry knows which hardware information sources are available and can be connected to it tries to discover the hardware environment by querying these selected hardware information sources as described in the corresponding hdf files This is called the discov
68. source it should create an icon labeled Hardware Sentry under the main computer PATROL icon If no Hardware Sentry related information is shown at all in the PATROL Console it means that Hardware Sentry is not running on the Agent or that an internal error occurred Execute the 7DUMP KM_LIST command in the System Output Window of the PATROL Console This should produce the following output KEk l System Output for BOURDIEU_3181 Terminal Pane i econ ae sore Last executed command cDUMP KM_LIST 28 67 2003 16 55 45 Loaded Knowledge Modules Applications 2870772003 16 55 45 8 Version Static of consoles 2870772003 16 55 28 07 2083 16 55 45 z ALL_COMPUTERS 5 28 67 2063 16 55 45 KI yes 28 67 2663 16 55 45 z MS_HW VOLTAGE CONT yes 28 67 2663 16 55 45 z MS_HW VOLTAGE yes 2870772003 16 55 45 MS_HW_TEMPERATURE_CONT yes 28 07 2003 400 ee z MS HW TEMPERATURE yes 2870772003 16 55 MS HW POWERSUPPLY_CONT s yes 2870772003 16 55 45 MS 28 67 2063 16 55 46 MS Action View i ni HW POWERSUPPLY j yes HW_OTHERDEVICE_CONT yes HW_OTHERDEVICE yes MS_HW PHYSICALDISK_CONT 3 yes HW PHYSICALDISK i yes HW LOGICALDISK_CONT i yes HW LOGICALDISK i yes 2870772003 16 55 46 MS 2870772003 16 55 28 67 2063 16 55 46 H 28 67 2063 16 55 46 MS 28 67 2063 16 55 46 MS 28 07 2063 28 07 2003 28 07 2063 28 07 2663 28 07 2063 16 16 16 16 16 rib fe le Sons S
69. tility to execute external commands as root The sudo utility helps UNIX system administrators secure their environment by authorizing some users to execute only some specified commands as another user account typically root If the case described above applies to you the sudo options may be a good workaround Here is the procedure to configure Hardware Sentry to use sudo 1 Identify which commands that are run by Hardware Sentry needs advanced privileges usr sbin diskinfo on HP UX systems opt IBM director CIMOM bin cimcli on IBM xSeries systems under Linux 2 Configure sudo to allow the PATROL Agent default account to execute the needed commands as root modify the etc sudoers file Configuring sudo requires root privileges 3 Check that the PATROL Agent default account is properly authorized to execute needed commands through the sudo utility 4 Ensure that the sudo bin directory is in the PATH variable of the PATROL environment 5 Go to the Hardware Sentry Security Settings dialog in the PATROL Console and click on the Sudo options button 6 Check the boxes corresponding to the commands that have been configured in sudo 7 Restart the PATROL Agent 2 Sudo options Sudo options The following programs may be executed by Hardware Using the sudo Sentry to gather hardware information Some of these utility for a given programs may require privileged ac
70. toring setting Select how the link status of a network interface is monitored C NEVER trigger an alert when a network interface is not linked to a network 1 e no monitoring of the link status Trigger an alert ONLY if a previously plugged in network interface is unplugged default cancer enp Three options are possible ALWAYS trigger an alert when a network interface is not linked to a network If a network card is unplugged Hardware Sentry will trigger an alert no matter what NEVER trigger an alert when a network interface s not linked to a network This basically turns off the monitoring of the network link Trigger an alert ONLY if the previously plugged in network interface is unplugged This is the default option and is explained above 24 A SNQ SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 M eae User Guide Monitoring laan t www sentrysoftware net oo www s 4 Missing Device Detection When enabled the missing device detection mechanism of Hardware Sentry alerts the operators when a device that was previously detected in the system is no longer found This mechanism is very useful to alert operators when for example a non redundant physical disk does not restart during a system reboot and therefore is no longer seen by the operating system and the monitoring software When a device is no longer discovered its Status parameter goes into alarm and its label
71. u select above will be executed each time a parameter will go into alarm These settings apply to all of the hardware components monitored by Hardware Sentry next gt nen 11 2 Trigger a PATROL Event If you have selected the Trigger a PATROL Event action you need to enter the string that will be displayed with the event A PATROL Event can be viewed from Standard PATROL Consoles Classic Console PATROL Central PATROL Enterprise Manager BMC Impact Manager Other third party products that interface with PATROL Note The PATROL event that is triggered is the 41 event from the standard PATROL catalog Note You can use macros that will be replaced at runtime The list of available macros is described later in this chapter 45 Check the boxes corresponding to the actions you would like executed upon a hardware failure By default Hardware Sentry triggers a PATROL event and annotates the parameter that raises the alert DEK Set Alert Actions Alert Action Trigger a PATROL Event Enter the text to be sent in the PATROL Event Hardware problem with OBJECT LABEL PROBLEM NEWL INE NEWL INE FULLREPORT The triggered event is the 41 event from the standard PATROL catalog Note You can use macro variables in the fields above See the documentation to have the complete list of available macros s2 SENTRY Hardware Sentry Knowledge Module
72. ually preselect connectors select connectors in the Options menu Preselecting connectors means deciding what platforms Hardware Sentry e e e e will have to assume it is running on For each connector selected The following di alog box is displayed Hardware Sentry will performed tests and will try to gather data using the tools provided by the connectors for that particular platform All other connectors will be ignored and no testing performed for them Two choices are possible in this window Let Hardware Sentry detect which connectors to use in this mode Hardware Sentry will run tests at each discovery and select the connectors that match the current system Manually pre select connectors selecting this option lets you decide which connectors Hardware Sentry should use to monitor the system Hardware Note when the panel comes up the selected option is the one that is currently in use The first option Let Hardware Sentry detect which connectors to use is the recommended one and the one used by default when Hardware Sentry is loaded on a PATROL Agent The detection process is very efficient and accurate and should not use the computer s resources extensively If this option is selected clicking on the Next button will bring up a panel with a message saying Hardware Sentry will run the detection process again 40 N SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 SOFT User
73. uire an additional vendor specific agent for Hardware Sentry KM for PATROL to work properly In most cases server vendors provide the required hardware monitoring agent for their server models Depending on the platform a single agent will monitor the temperatures fans voltages power supplies and the RAID systems or a separate agent for the environment monitoring and for the disk monitoring For example the IBM Director Agent monitors and provides information about the temperatures fans voltages power supplies and ServeRAID disks for IBM xSeries and Netfinity servers On the other hand the Siemens Serverview Agent will only monitor the sensors on the motherboard of the server temperatures voltages fans and power supplies while the Mylex GAM Server monitors the Mylex RAID controller of the server Please refer to the Hardware Sentry KM for PATROL Installation Guide document for further details on the installation procedure of the vendor specific agents supported by Hardware Sentry 12 Specific Director Serverview agent Siemens Hardware Sentry KM for PATROL Motherboard Monitor i SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 eee User Guide Getting started C www sentrysoftware net 4 Using Hardware Sentry KM for PATROL to monitor hardware Once Hardware Sentry KM for PATROL is installed and integrated to the PATROL framework and the hardware informati
74. us of a logical disk typically corresponds to the status of a RAID array on line degraded rebuilding etc For each logical disk discovered the Status parameter is displayed The Status parameter represents the current status of the logical disk An alert is triggered when the logical disk is not fully operational degraded rebuilding etc or not available at all Note For non RAID disk controllers as most of IDE controllers for example no logical disk will be displayed 21 Hardware Sentry Knowledge Module for PATROL version 1 3 01 User Guide Monitoring 3 Monitoring Processors Memory modules and Network interfaces irx PATROL Main Map Profile patrol Console Server KANT User patrol fe Tasks Hardware Sentry automatically detects the KANT_3181 KANT_ information sources available on the monitored Hardware Sentry 1 1 00 i Computer Dell PowerEdge 16005C computer and displays the hardware t a PERC 4 5C Controller 0 1 information provided by those sources in the Fans fea Memor PATROL Consoles in the corresponding emory modules E Memory 1 1 DDR 256 MB container b Network interfaces H Network interface 0 Intel R PRO 1000 MT Network c 100M1 If at least one hardware information source Network interface 1 3Com EtherLink XL 10 100 PCI F 10 Mbp provides useful data about the most critical Power supplies d h H Processsors non storage ev
75. www s Section Overview Hardware Sentry Knowledge Module for PATROL is a BMC Software PATROL module that allows administrators to monitor the hardware of their servers Hardware Sentry KM for PATROL is a single KM that is able to monitor the hardware of different server brands IBM HP Sun DELL NEC Fuyjitsu Siemens and many others Once installed on a PATROL Agent on a server Hardware Sentry KM for PATROL automatically detects the environment and starts the monitoring of the hardware status of the disks and the RAID controllers temperature of the system speed of the fans etc SNQ SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 an SOFTWARE User Guide Getting started section Il Getting started I How AGES Il WOLK wis ececceswaracrsiauieat enra sites edsasiesussot E aN 9 2 Integrating Hardware Sentry KM for PATROL 4 ccc ccc cece cece cece cece ccc cencccescceecceeees 11 3 The need for vendor specific hardware monitoring software sessessesessesessesesseo 12 4 Using Hardware Sentry KM for PATROL to monitor hardware cccccsesceees 13 5 What sort of hardware can be monitored with Hardware Sentry cccccccesccesces 14 SN SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 QO m C jz at SOFT User Guide Getting started laa t www sentrysoftware net oo www s 1 How does it work Hardware S
76. y sets the thresholds through PCH Event Management C No threshold Hardware Sentry does not set any thresholds and lets you specify them manually Remark The PCH Event Management option requires that you set up the PATROL for Event Management KH If you are currently using PATROL for Event Management to manage the thresholds within PATROL it is recommended to use this option In any other case it is recommended that you use the default option cancer You can select one of the three following options Default Hardware Sentry will manage its thresholds through the standard Override parameters mechanism Thresholds are stored in the PATROL Agent configuration under the ___tuning___ tree This option is set by default and does not require any additional software If you do not use PATROL for Event Management it is strongly recommended that you use this option 36 A SN SENTRY Hardware Sentry Knowledge Module for PATROL version 1 3 01 om Gam 5 TS ee op Ne We ee ee a tp eo wee ee at SOFT User Guide Configuring and Administering an io www sentrysoftware net of www s PCM Event Management Hardware Sentry will manage its thresholds through the PCM or Event Management mechanism Thresholds are stored in the PATROL Agent configuration under the AS tree This option requires that you set up the PATROL for Event Management KM on your PATROL Agent PATROL for Event Management has to be enabled a

BMC Performance Manager for Hardware

Contents

Download Pdf Manuals

Related Search

Related Contents