locked
Monitoring alerts for services not running on cluster RRS feed

  • Question

  • Hi,

    we have a Windows 2003 cluster running with SQL 2005.
    The cluster is running in active/active, the system is monitored with MOM 2005.
    On both nodes SQL has 2 instances, each node has 1 instance running and the other is for failover.
    Now MOM is alerting concerning the services that are not running on each node, but this is normal as these services are only started when the instance is failed over to the other node. How can we get rid of these alerts?

    Management pack for clusters is installed, management pack for SQL is also installed.

    Can anyone help me, I think I need to configure this in the management pack, but I'm fairly new to MOM/SCOM and they dropped me with this case.
    I already checked several hundreds of sites to find the solution but eather it hasn't been posted yet or I'm looking with the wrong search options

    thx in advance


    I found the solution to the problem
    I checked the versions from the management packs that were installed and found that it were still old versions
    I updated the management packs and cleared the MOM alerts
    Now, no alerts are generated anymore 
    • Changed type Wif Thursday, October 22, 2009 10:35 AM
    Tuesday, October 20, 2009 10:33 AM

Answers

  • mom2005 will identify the clusterresources and will monitor it from there (but you need to set your clusterresources to "managed" in the admin console).

    But this is not normal situation. i think the sql mp checks the services (this is basically the attribute on which sql rules get applied to sql servers), but only alerts when the services are set to automatic, which shouldn't be set to automatic on a clusternode. To be sure, look up the rule in the adminconsole and figure out what it checks exactly.

    Thursday, October 22, 2009 1:47 PM

All replies

  • mom2005 will identify the clusterresources and will monitor it from there (but you need to set your clusterresources to "managed" in the admin console).

    But this is not normal situation. i think the sql mp checks the services (this is basically the attribute on which sql rules get applied to sql servers), but only alerts when the services are set to automatic, which shouldn't be set to automatic on a clusternode. To be sure, look up the rule in the adminconsole and figure out what it checks exactly.

    Thursday, October 22, 2009 1:47 PM
  • "Mark as Answer", no activity for a month. Feel free to re-open this
    Anders Bengtsson | Microsoft MVP - Operations Manager | http://www.contoso.se
    Monday, November 23, 2009 10:24 PM