WhatsUp Gold - Performance alerting with SNMP
|WhatsUp Gold||7.x - 8.x||Windows 2003 Server
Question/Problem: How can I monitor performance using SNMP on Windows 2000, Windows XP or Windows 2003?
While the WhatsUp software includes SNMP features and components, this document relies on software provided by third-party companies. Neither the Windows SNMP server that WhatsUp will query nor the MIBs provided by SNMP Informant are supported or maintained by Ipswitch.
1.) On all systems (target machines) for which you want to monitor performance, install the Windows Simple Network Management Protocol Service. This is usually listed in the Add/Remove Programs control panel under the Management and Monitoring Tools category. It is not necessary to install the SNMP Service on the WhatsUp machine (unless you plan to monitor the WhatsUp machine itself).
2.) On all target machines, open a command prompt, enter and execute the following command:
Then, restart the target machine. Note: this should be enabled by default on XP and 2003 machines.
3.) Download the standard version of SNMP
4.) From one of the target machines, copy informant-std.mib and wtcs.mib to the WhatsUp machine. These files are usually located in C:\Program Files\SNMP Informant\standard\mibs\SMIv1. The following instructions assume that a temporary folder named C:\MIBs has been created on the WhatsUp machine and the MIB files have been copied to it.
5.) On the WhatsUp machine, close the WhatsUp application and compile the SNMP Informant MIBs. Open a command prompt, navigate to the WhatsUp installation directory, and execute the following command:
6.) You can now open the WhatsUp Console (or Net Tools) and view the MIBs you have compiled. The SNMP Informant tree is located under:
iso > org > dod > internet > private > enterprises > wtcs > informant > standard
If you cannot see the wtcs tree in Net Tools, this indicates the MIBs did not compile correctly. Double-check your work in step 5.
7.) To query a particular Object ID (OID) in Net Tools, select the SNMP tab and enter the IP Address of the target machine. Then, enter the Community name for the target machine. The default read community name for the Windows SNMP Service is public. This value is case-sensitive. Then, select the browse button next to the What field and navigate to the SNMP Informant tree and select an item. Note: Some OIDs require instance numbers and may not respond if you query them with only Get selected. If querying a particular OID does not return results, try selecting Get All Subitems and performing the query again. If none of the OIDs return any data, the SNMP service on the target machine may not be functioning correctly, traffic to the SNMP service is blocked (SNMP query traffic uses UDP port 161), or your community name may be incorrect. All Windows SNMP servers should respond to a query for *sysinfo in the What field.
Two examples for monitoring performance data via SNMP are below. These examples are provided for information only. No support for this information is provided by Ipswitch.
To monitor for free disk space on the C drive of a target machine:
1.) Open Net Tools on the WhatsUp machine and in the What field, browse to iso > org > dod > internet > private > enterprises > wtcs > informant > standard > logicalDiskTable > logicalDiskEntry > lDiskPercentFreeSpace and click OK. Notice the OID for this query is 188.8.131.52.4.1.9600.1.1.1.1.5.
2.) Enter the IP address of the target machine and the read community name in the appropriate fields and click Get All Subitems.
3.) For a computer that has the following drives:
C Drive = 17% free space
Then, the following information should be returned:
The instance number for each drive is listed after the name for the OID. The C drive usually has an instance number of 2.67.58. This means that to query for free disk space on the C drive, we should use an OID of 184.108.40.206.4.1.9600.1.1.1.1.5 and an instance number of 2.67.58. Subsequent drives should have different instance numbers, but the same OID. If you aren't sure which drive correlates to which instance number, check the free space of the target machine's drives to find a match.
4.) Now, setup a monitor in WhatsUp to query this OID and instance number on the target machine.
Select Monitors and Services from the Configure menu. Click the New button and select SNMP Monitoring from the Service type menu. Enter a name for this monitor, we'll call ours Percent free on C, and click OK.
5.) Now, enter the OID and the instance number for the C drive query. This can be entered directly into the OID field, or you can browse to it. In WhatsUp Gold, be sure to enter the complete OID and instance number (220.127.116.11.4.1.9600.1.1.1.18.104.22.168.58) into the Object ID field. In WhatsUp Professional, you can enter both into the OID field (leaving the Instance field blank), or enter 22.214.171.124.4.1.9600.1.1.1.1.5 into the OID field and 2.67.58 into the Instance field.
6.) Select Range of Values from the menu and enter 100 into the High value field. Enter the amount of free disk space at which you want to be notified into the Low value field. For example, if you want to be notified when there is less than 25% free disk space, enter 25 into the Low value field. If you wish to be notified when there is less than 75% free disk space, enter 75 into the Low value field. Save your changes.
To monitor for CPU utilization on a target machine:
1.) Open Net Tools and browse to iso > org > dod > internet > private > enterprises > wtcs > informant > standard > processorTable > cpuPercentProcessorTime. Enter the read community name and target machine's IP address, select Get All Subitems and click OK.
2.) The first CPU on the target machine should have an instance number of 1.48. This, combined with an OID of 126.96.36.199.4.1.9600.1.1.5.1.5, gives us a query of 188.8.131.52.4.1.9600.1.1.5.184.108.40.206 to see the CPU usage for the first CPU of the target machine.
3.) When creating a monitor for CPU usage, always enter 0 for the Low value and enter the value at which you wish to receive a notification for the High value. For example, if you wanted to be notified when CPU usage was above 90%, you would enter 0 in Low value and 90 in High value. If you wanted to be notified when CPU usage was above 10%, you would enter 0 in Low value and 10 in High value.
|Document #:||Revision Date:|
Return To KnowledgeBase Search Page