Posted: 2 Mar 2021 15:04 EST Last activity: 10 Mar 2021 14:16 EST
SMA shows old requestors on the node
We have a multi node environment, 2 of the nodes doesn't seem to end the requestor sessions as does the remaining. 2 nodes have multiple req. sessions accumulated from last 2 days as seen in SMA->Requestor Management with last input as default portal: @baseclass.pzHandleUnknownSession, new requestor created, Activity=FinishAssignment, Activity=@baseclass.ShowLogoffTimer, Activity=pyDeleteDocumentPg.
Overall these have about 1700 requestors out of them only about 250 are from today and remaining all are stale from last 2 days. Other 18 nodes under the same load balancer doesn't have this problem.
As seen by the load balancer, traffic is distributed evenly and no issues observed.
Restarting the JVM would help clear all of them but need to understand what went wrong on these, NodeCleaner is running and nothing found so far in tomcat or pega logs.
Any Ideas? Appreciate your time and help!!
***Edited by Moderator: Pooja Gadige to add platform capability tag***
@MarcLasserre_GCS Doesn't seem like as i went through it already and found the issue in that thread seems to be specific to a service package which was incorrectly setup, our issue was limited just requestors not terminating automatically after a http interaction from the user. All the sessions as seen in SMA were related to OOTB @baseclass.pzHandleUnknownSession, new requestor created, Activity=FinishAssignment, Activity=@baseclass.ShowLogoffTimer, Activity=pyDeleteDocumentPg..
We eventually had to restart the JVM to clear up all of them but still wondering how/why that has happened, any thoughts are appreciated.