Forum Discussion
Waseem_Shaik_97
Feb 23, 2015Nimbostratus
@ Sadorect , This is what customer is complaining that even though health monitor fails on a node..he still connecting to failed node. Alternately he would like to try different settings on Action on service down to not only allow persist connection to another node without breaking any session and also remove failed node out of equation completely.
- dragonflymrMar 04, 2015CirrostratusWhat do you exactly mean by "he still connecting to failed node"? That when node is marked down by monitor new connections are still send to this node or that existing connections are not immediately terminated? I am pretty sure that NO new connections are ever send to down node (even if there are persistence records for that node), if it is indeed happening then it has to be some kind of bug. If former is the case then it's as well normal behavior for None setting in Action On Service Down (SOL15095): The BIG-IP system takes no action on existing connections, and removes the connection table entry based on the associated profile's idle timeout value. The BIG-IP system sends a TCP Reset (RST) or ICMP Unreachable once idle timeout is reached. This is the default setting. If you need immediate reset of connection then use Reject: The BIG-IP system sends RST or ICMP messages to reset active connections and removes them from the BIG-IP connection table. I am not sure what persistence has to do with this issue. When member is down persistence records are removed (if persistence modes using them are used), even in case of persistence not using records is implemented still it will be broken for down member. So what is the problem with Reselect Tries (if that's what you mean by "reselect option"), except of course it is not really working well for any connection oriented protocols or not transparent hosts. Piotr