Forum Discussion

Brian_10532's avatar
Brian_10532
Icon for Altocumulus rankAltocumulus
Apr 11, 2017

Down due to successful monitor?

This alert makes no sense to me. Why would we ever see an alert changing a poolmember from green to red when the result of the monitor was successful?

 

Mar 12 20:14:56 10.5.32.16 Mar 12 20:14:56 denbip2000gtm02 alert gtmd[14900]: 011a4003:1: SNMP_TRAP: Pool /Common/adfs.xxxxxx.net member corlgtadfs2 (ip:port=20.87.139.151:443) state change green --> red ( Monitor /Common/adfs.xxxxxx.com from gtmd : connect: success search result true)

 

I've also seen this one...

 

Mar 12 20:21:47 10.5.32.15 Mar 12 20:21:47 denbip2000gtm01 alert gtmd[14902]: 011a4003:1: SNMP_TRAP: Pool /Common/jp1.wss.xxxxxx.net member jp1-ect_wss (ip:port=52.6.20.245:3128) state change green --> red ( Monitor /Common/wrproxy_3128 from gtmd : state: success)

 

Clearly the monitor result was a success, but why is GTM changing the state to down?

 

application delivery
BIG-IP DNS

1 Reply

  • Vijay_E's avatar
    Vijay_E
    Icon for Cirrus rankCirrus
    Apr 11, 2017

    May be the response is not "200 OK". GTM/DNS will mark the server down even if the right string is found in the response, if the status code is not "200 OK"

     

    Also, make sure you are specifying the HTTP version in your monitor.

     

    Normally, I would recommend that LTM should be checking the end servers and the GTM should just obtain the status from LTM instead of checking the servers.

     

    K13622