I have one Basic query.
In Client network, we have Viprion (model 2400) with 2 Blade (model 2150). On each viprion , there are 5 vCMP created on both blades. We need to upgrade the IOS/Software from 11.x to latest version.
My Query is
Do we need to upgrade viprion only ? in this case all the vCMP gets upgrades?
Do we need to upgrade individual vCMP on viprion ?
Do we need to upgrade module/Blade wise ?
thanks in advance.
You need to upgrade both Viprion chesis n vcmp guest also. F5 v11.3 onwards you can directly upgrade guest n later update Viprion chesis.
I will recommend you to 1st upgrade Viprion chesis then vcmp guest because in Viprion chesis you need to re- licence Viprion before upgrade guest.
I am having similar requirement for upgrade the F5 5250V ( ACTIVE/STANDBY)
is it better to upgrade the host first then guest , what is the best practice here
any advise appreciated
It's up to you but F5 recommedation is to upgrade 1st host then guest.
Thanks you for your advice and recommendation.
Can anyone tell me , while re-activating the VIPRION (due to service date and License date issue), do I need to use the same registration key or need new key from F5 ?
Same registration key will be used. After reactivate license service date will change.
One note about reactivating vHost license (from https://support.f5.com/csp/article/K15930):
Important: Reactivating the vCMP host license will result in a temporary traffic interruption on the guest systems while the license is reactivated on the host system.
I have as well one question. In the same article there is such statement:
Before upgrading a host system, you should gracefully shut down the guests by setting the requested state to Configured.
Is that only way? What is difference between doing shutdown -h now on vGuest and changing state to Configured?
shutdown -h now
Sorry, this is such a late response, but it may be useful for others. When you use the shutdown command on a Viprion, the vCMP guest will shutdown on the blade you are logged into. If the vCMP guest spans multiple blades the others will remain up.
To use the shutdown command for the entire cluster use "#clsh shutdown -h now". The "clsh" command sends the command through the vCMP guest cluster IP, which sends it to each blade. Although, this may leave the guest in the 'Deployed' state on the vCMP host. You may still need to set it to Configured or Provisioned and back to Deployed to restart it.
The best way to gracefully shutdown a vCMP guest, as mentioned above, is to set it to the Configured state. Setting the vCMP guest to the Provisioned state also performs a graceful shutdown. The difference between the two is that, in the Configured state the host de-allocates system resources (CPU, memory, etc..), the virtual disk which remains provisioned in either state.
The following articles provide further reference on the upgrade process: