Forum Discussion

kimhenriksen's avatar
kimhenriksen
Icon for Cirrostratus rankCirrostratus
Apr 06, 2021
Solved

vcmp ha pair resource (cpu/memory) upgrade procedure..

Hi

I am working on doing a resource upgrade (add more cpu/memory) on a vcmp ha pair. Tried to find some good info about best practices, but I´ve come up dry..

So, I am thinking of doing it just as a software upgrade.. upgrade the seconday, failing over.. letting it run for a while and then do the primary.

 

Any problems with using the force offline option during this process? Is HA functionality affected by the difference in resources on the vcmp? any tips?

  • As mentioned in the guide you will need to move the guest into the configured state before allocating resources.

     

    Do a tail on ltm log and grep for vcmp then move the guest it to the configured state.

    Look for VS_STOPPED->VS_INACTIVE, this indicates that the guest has stopped.

     

    You can then allocate the resources.

     

    There should not be any issues with regards to HA but as always take a UCS for both host and guest.

    I also do not see why you will need to move it to forced offline.

    You will not be changing any configuration of the configuration objects on the device.

     

    Once resources are allocated move the guest back into the deployed state and you should be good to go

     

    Again do a tail on ltm and grep for vcmp.

    Look for VS_STARTING->VS_RUNNING, your guest should be up and running again.

     

    Steps

     

    1. Backup host and guest UCS
    2. Tail ltm log on the host for vcmp (tail -f ltm | grep vcmp)
    3. Move guest that will be allocated more resources into configured state
    4. Look for VS_STOPPED->VS_INACTIVE in the ltm log (Shows guest is down)
    5. Assign resources
    6. Move guest to deployed
    7. Tail ltm log on the host for vcmp and look for VS_STARTING->VS_RUNNING (tail -f ltm | grep vcmp)

     

    Hope this helps

     

    *Please note that these are the steps that we followed on our Viprions*

7 Replies

    • You can use force offline options on the standby device before upgrading. There won't be any issue.
    • HA would get established once both devices are updated to the same version.
    • Import the image in Host and it should get populated on the vCMP.
    • You can follow the standard practice while updating BIGIP. You must know them already but just mentioning here.
    1. Read release notes of the version you wanted to go
    2. Take UCS backup and save it locally
    3. License reactivation if needed
    4. If you have nonprod environment, update it first.

  • This is for a software upgrade, I want to allocate more resources (cpu/memory) to a vcmp. I should have been clearer about that :)

    • SanjayP's avatar
      SanjayP
      Icon for Nacreous rankNacreous

      Apologies. I didn't read it properly. Please follow the below steps to add more cores to the vcmp guest.

       

      https://techdocs.f5.com/kb/en-us/products/big-ip_ltm/manuals/product/vcmp-administration-viprion-13-1-0/7.html

       

      force offline option is not needed as you are disabling the guest first and deploying it again.

      • kimhenriksen's avatar
        kimhenriksen
        Icon for Cirrostratus rankCirrostratus

        I´ve read that one, wondering if there are any things to do for a HA pair. But the lack of info might be the answer :)

  • As mentioned in the guide you will need to move the guest into the configured state before allocating resources.

     

    Do a tail on ltm log and grep for vcmp then move the guest it to the configured state.

    Look for VS_STOPPED->VS_INACTIVE, this indicates that the guest has stopped.

     

    You can then allocate the resources.

     

    There should not be any issues with regards to HA but as always take a UCS for both host and guest.

    I also do not see why you will need to move it to forced offline.

    You will not be changing any configuration of the configuration objects on the device.

     

    Once resources are allocated move the guest back into the deployed state and you should be good to go

     

    Again do a tail on ltm and grep for vcmp.

    Look for VS_STARTING->VS_RUNNING, your guest should be up and running again.

     

    Steps

     

    1. Backup host and guest UCS
    2. Tail ltm log on the host for vcmp (tail -f ltm | grep vcmp)
    3. Move guest that will be allocated more resources into configured state
    4. Look for VS_STOPPED->VS_INACTIVE in the ltm log (Shows guest is down)
    5. Assign resources
    6. Move guest to deployed
    7. Tail ltm log on the host for vcmp and look for VS_STARTING->VS_RUNNING (tail -f ltm | grep vcmp)

     

    Hope this helps

     

    *Please note that these are the steps that we followed on our Viprions*