Forum Discussion

Federico_Battag's avatar
Federico_Battag
Icon for Nimbostratus rankNimbostratus
Nov 15, 2013

Force to StandBy button doesn't work

Hi all, I’m writing this message to ask if someone can help me about a strange behavior of my HA-Active/StandBy VIPRION 2400 (LTM) . A few days ago I upgraded from version 11.1.0 to 11.4.0 HF3. The upgrade has been slow, but it all went well.

 

The problem I have found is that if I set as Active the node which is usually in StandBy, this becomes the primary, but then if I click on "Force to StandBy" the primary doesn’t return Active. It happens that the secondary changes from Active to StandBy for 5 seconds and then return Active. The primary instead remains on StandBy. The two nodes are properly in sync. However, if I turn off the secondary (which is Active) then the primary returns Active and remains so (it's correct!). In conclusion, the mechanism works if I turn off the node, but doesn’t work if I click on "Force to standby". In the weights configuration of the two system, these are identical. I also tried to put the Active Bonus on primary and leave the secondary to zero, but nothing changes.

 

A greeting and thanks in advance FB

 

application delivery
availability
BIG-IP
deployment
ha
LTM

5 Replies

Sort By
  • Hamish's avatar
    Hamish
    Icon for Cirrocumulus rankCirrocumulus
    Nov 15, 2013

    Are you using HAGroups and also have AutoFailback enabled on the traffic-group? If so, then we've seen the same thing. (Ours also had 2 nodes claiming to be the active, although only one really was). Note that in a 6 node cluster we had the same issues when only 2 nodes were left up, but neither of them were the default device...

     

    F5 support eventually came back after duplicating the issue, and said this isn't a supported combination. We've removed the auto-failback, which has resolved the issue of nodes disagreeing who was active, but I haven't tried moving the traffic-group between nodes manually yet...

     

    They were a bit guarded about manual moving with HAGroups as well... But I haven't need to push them on that yet. It may simply have been a misunderstanding between us...

     

    H

     

  • Hamish's avatar
    Hamish
    Icon for Cirrocumulus rankCirrocumulus
    Jan 17, 2014

    I wonder if that's because something hasn't yet caught up with the concept of traffic groups versus the old active/standby methods...

     

    H

     

  • StephanManthey's avatar
    StephanManthey
    Icon for MVP rankMVP
    Dec 03, 2015

    Sorry for being a bit late with my answer.

     

    The WebUI offers multiple options to force a standby.

     

    But only one option works.

     

    Please see my post in this topic.

     

  • Cyril_M's avatar
    Cyril_M
    Icon for Altostratus rankAltostratus
    Feb 01, 2016

    I also observed this change in behavior after an upgrade, support said it's "work as design", the easiest workaround is to use "force offline" then "release offline" instead of "force to standby"

     

  • VFB's avatar
    VFB
    Icon for Cirrus rankCirrus
    Apr 22, 2016

    Go into your traffic group on both devices and and verify that "Auto Failback" isn't checked. If it is, uncheck on both devices, sync, then attempt another failover.