disconnect between the number of connections under the Stats and the persistence conn number

[newbie]

We're running BIG-IP 12.1.3.2 and have disabled the pool member of the pool that has 3 other members up and running. The corresponding VIP has been configured with the Default Persistence Profile " Multiple". When we look at the Stats for the connections for the pool member that had been removed from the pool, we're not seeing any connections. However, when we check the persistence for that node and the port using the show /ltm persistence persist-records node-addr 1.2.3.4 node-port 80, we're seeing multiple persistent connections:

Sys::Persistent Connections

universal c32e11e0 1.1.11.25:443 1.2.3.4:80 (slot/tmm: 1/0)

universal 1ca941b7 1.1.11.25:443 1.2.3.4:80 (slot/tmm: 1/0)

universal 9439b9eb 1.1.11.25:443 1.2.3.4:80 (slot/tmm: 1/0)

Is there a disconnect between the connection number we're seeing under the Stats for the pool member and the Persistent Connections we're seeing for that server using the show /ltm persistence persist-records command?

Thanks.

[Dario_Garrido]

Hello newbie.

I think you are matching this bug.

REF - https://cdn.f5.com/product/bugtracker/ID679854.html

I recommend you to clear those entries to check if new entries appears.

REF - https://support.f5.com/csp/article/K55632517

Could you also try to switch to another persistence method?

Regards,

Dario.

[newbie]

Thanks for your reply.

So, the persistence method that is being used for this is "Multiple", as in the following:

I had never come across this Profile before am still trying to figure out exactly how it works, so any documentation references to it would be greatly appreciated.

Thanks again...

[Dario_Garrido]

Hello Newbie.

Your output shows that you are using "universal" persist profile.

universal c32e11e0 1.1.11.25:443 1.2.3.4:80 (slot/tmm: 1/0) universal 1ca941b7 1.1.11.25:443 1.2.3.4:80 (slot/tmm: 1/0) universal 9439b9eb 1.1.11.25:443 1.2.3.4:80 (slot/tmm: 1/0)

And it seems that you are facing a bug that creates universal entries despite you were disabling the pool members.

-------

Regarding the "multiple" method that you see in the GUI, that because you have several profiles configured in the same VS. Use the next command to check the actual configuration.

tmsh list ltm virtual <VS_name>

You have a reference of how to configure "multiple" profiles here.

Regards,

Dario.