Forum Discussion

kridsana_52318's avatar
kridsana_52318
Icon for Nimbostratus rankNimbostratus
Nov 27, 2014

HA Connection lost after change Management IP address

Hi guy,

 

I have a problem after change mgmt IP. It's HA connection lost (result in IP conflict and downtime)

 

I have to change management IP address of BIG-IP redundant pair. But when we change it, HA connection lost and it's become active/active which cause us a downtime of application.

 

I have configsync and failover unicast IP is 2.2.2.2 (peer is 2.2.2.1) which connect directly with each other.

 

How can this occur? Is really changing mgmt IP of the box cause it HA connection lost?

 

Note. In v. 10.2.4 , we can change it just fine. Now we currently Running v.11.4.1 HF5

 

application delivery
availability
BIG-IP Access Policy Manager (APM)
design
LTM
redundancy
security
  • JG's avatar
    JG
    Icon for Cumulonimbus rankCumulonimbus
    Nov 28, 2014

    Service disruption is possible in such an operation.

     

    See ">SOL7312: Overview of the management port.

     

    To prevent the devices going into active-active, you can force-offline the standby device before operation.

     

    See ">SOL15122: Overview of the Force Offline option.

     

  • giltjr's avatar
    giltjr
    Icon for Nimbostratus rankNimbostratus
    Nov 28, 2014

    I would have to think and read about everything that needs to be updated and how quickly it must be updated in order to prevent an outage when changing the management IP address.

     

    Basically all F5's in the cluster would need to be updated at the "same" time. You would also need to make sure the DSN entries for the F5's are updated so that when the F5's do their DNS lookups they get back the new addresses.

     

    You may want to open a case with F5 support to see the best way to do this.

     

    However, do you really have an outage? I would think that one of the F5's would keep responding.

     

  • kridsana's avatar
    kridsana
    Icon for Cirrocumulus rankCirrocumulus
    Nov 28, 2014

    @giltjr So it's mean after changing management IP . Device will active/active (Downtime will occur for sure.) and then we must update a trust list (not sure if we have to create a new sync group) to make a BIG-IP sync each other again.

     

    Am I correct?

     

  • giltjr's avatar
    giltjr
    Icon for Nimbostratus rankNimbostratus
    Nov 27, 2014

    I would suggest you read how to setup HA in V11. It is no longer really a HA Pair, but a cluster. This may give you an idea of what all changes you need to make in order to get this working in V11.

     

    https://support.f5.com/kb/en-us/products/big-ip_ltm/manuals/product/bigip-device-service-clustering-admin-11-4-0/3.htmlunique_2103230074

     

    One thing to check for is your trust list on each device. From the above link you need to see what you have configured under " Device Management > Device Trust, and then either Peer List or Subordinate List"

     

    You trust the management IP address of the other devices in the cluster. So if you are changing the management IP address, you need to update the trust list.