Forum Discussion

Joint_Commissio's avatar
Joint_Commissio
Icon for Nimbostratus rankNimbostratus
May 07, 2012

Generate an email alert when a pool is marked up by a monitor

Hello,

 

 

We currently have F5 LTM ver 9.4.8. I've been able to setup a custom SNMP alert to email me via Solarwinds Orion when a pool is down:

 

 

alert POOL_DOWN "No members available for pool (.*)" { snmptrap OID=".1.3.6.1.4.1.3375.2.4.0.856" }

 

 

Now I'm trying to write another custom alert to email me when the pool is up. I don't see any entries in the LTM logs when a pool is up other than statements that the pool member is up.

 

 

I read here http://support.f5.com/kb/en-us/solu...12531.html that when a pool is up the following is logged:

 

tmm err tmm[4779]: 01010221:3: Pool now has available members. But I don't see that in the logs when I ran a test.

 

 

How do I go about resolving this? Is it as simple as changing the log level? I'll love to write something like this:

 

 

alert POOL_UP "Pool (.*) now has available members" { snmptrap OID=".1.3.6.1.4.1.3375.2.4.0.857" }

 

 

5 Replies

  • That looks about right. You can add the custom alerts to the /config/user_alert.conf file.

     

     

    Aaron
  • Hi Hoolio,

     

     

    Thanks for your response. I'm trying to find out why the LTM never generates "Pool x now has available members" when a pool member is marked up. It does generate the "No members available for pool x" when the pool is down so my alert works. Is there a setting in the log level that I'm missing? Or is there another way to accomplish this?

     

     

    Thanks.
  • I forgot to add that I have the alert in the /config/user_alert.conf file already. I believe it's not working because the LTM is not generating the message in the alert: "Pool x now has available members"
  • I see the following in the /etc/alertd/*.h files on 10.2.3 and 11.1:

     

     

    /etc/alertd/bigip_tmm_error_maps.h:0 LOG_ERR 01010028 BIGIP_TMM_TMMERR_LAST_PMBR_DOWN "No members available for pool %s"

     

    /etc/alertd/bigip_tmm_error_maps.h:0 LOG_ERR 01010221 BIGIP_TMM_TMMERR_PMBR_BACK_UP "Pool %s now has available members"

     

     

    Do you have these entries on your version? Do you see the available members message in /var/log/ltm when a pool comes back up?

     

     

    Aaron
  • Thanks for the followup Hoolio.

     

     

    I took a look at /etc/alertd and I do have:

     

    0 LOG_ERR 01010028 BIGIP_TMM_TMMERR_LAST_PMBR_DOWN "No members available for pool %s"

     

     

    but there's no:

     

    /etc/alertd/bigip_tmm_error_maps.h: 0 LOG_ERR 01010221 BIGIP_TMM_TMMERR_PMBR_BACK_UP "Pool %s now has available members"

     

     

    The error number 01010221 is not present in the file. I guess I'm out of luck huh? I've been trying to convince management to approve an upgrade to 10.2 so that we can move away from 9.4.8 but they're dragging their feet.

     

     

    I do see the available members message in the log when a pool member (but NOT the pool) is back up:

     

    May 10 07:00:22 BMC-BIGIP01 mcpd[1802]: 01070727:5: Pool member 10.x.x.x:80 monitor status up