Forum Discussion

James_41261's avatar
James_41261
Icon for Nimbostratus rankNimbostratus
Jan 17, 2012

Automated setup of SNMP alerts

What is needed:

 

A way to generate alert emails when a node or pool member reports a down or disabled status containing information on which node and pool(s) are affected on the LTM.

 

 

What I'm trying to do:

 

I know I can setup a custom SNMP alert in user_alert.conf per node/pool that has the custom info I need preset in the body field; issue is we have A Lot of nodes and pools setup and was looking for a way to just pull node and pool info from the LTM config into a script which could then generate and write the alert entries into the user_alert.conf. This would save a lot of repetitive typing (reducing the chance of typos) and could possibly role into a cron job that would check if any new nodes/pools were added and auto create an alert entry. Looking for something that could be ran on the LTMs themselves, not dependent on another device.

 

 

 

Anyone have any scripts or processes to share?

 

management
monitoring
  • Ed_Hammond_2611's avatar
    Ed_Hammond_2611
    Icon for Nimbostratus rankNimbostratus
    Feb 06, 2012
    See if this might help:

     

    http://devcentral.f5.com/wiki/AdvDesignConfig.Acton-on-Log-using-the-alertd-deamon.ashx
  • James_41261's avatar
    James_41261
    Icon for Nimbostratus rankNimbostratus
    Feb 07, 2012
    Thanks Ed.

    I went to implementing a custom Perl script to parse details from the ltm log entry that triggers the SNMP alert. Issue I'm having now is that a "No members available for pool" alert is also triggering my "Node status Down" alert which is resulting in the alert being processed twice in 2 different fashions.

    Here is the error in the ltm log:

    Feb  5 09:09:08 tmm err tmm[5905]: 01010028:3: No members available for pool /Common/test_pool

    Here are the 2 custom alerts that seem both to be triggered:

    alert CUSTOM_GENERAL_POOL_NO_MEMBER "No members available for pool (.*?)" {
        exec command="/root/customAlertHandler.pl general"
}

alert CUSTOM_ERROR_POOL_MEMBER_STATUS_NODEDOWN "Pool (.*?) member (.*?) monitor status node down\." {
        exec command="/root/customAlertHandler.pl nodeDown"
}

    I'm unable to figure out why the Node Down alert is also being triggered by that error. Any ideas?

  • James_41261's avatar
    James_41261
    Icon for Nimbostratus rankNimbostratus
    Feb 08, 2012
    Moved my extra alert issue to a new thread:

     

    http://devcentral.f5.com/Community/GroupDetails/tabid/1082223/asg/44/aft/2161136/showtab/groupforums/Default.aspx