BIG-IP Pool Member/Node State: Disabled vs Forced Offline
Trying to cut through the confusion when it comes to the BIG-IP State setting of pool members and nodes. Here are the 3 options:
- Enabled (All traffic allowed)
- Disabled (Only persistent or active connections allowed)
- Forced Offline (Only active connections allowed)
Does the usage of "persistent" refer to session persistence and persistent TCP connections?
If it is referring to session persistence and we do not use "sticky" sessions, there should be no difference between Disabled and Forced Offline, correct?
If it is referring to persistent TCP connections, then Forced Offline should close any open TCP connections that aren't actively engaging in transfer, and have connected agents establish a new TCP connection to a different pool member/node on the very next HTTP request, for example.
I believe based on behavior I've witnessed it means session persistence, but wanted to make sure I'm not missing something.
As an aside, I actually would prefer it means persisted TCP connections (or give a fourth option), as it would get traffic off of my pool member/node much quicker so that I can perform maintenance/upgrades without waiting as long. Is there a different way to do this, of course in a way that wouldn't accidentally kill a TCP connection in the middle of a transfer which could cause a failure for an end user.
Note there are two great articles that both support my findings, and my confusion :)