This is a discussion on two cloneids - Websphere ; Hi, I'm solving problem with session affinity on some web application (WAS 6.0 ND with cluster of 6 appservers and memory-to-memory replication, IHS, AJAX application). There is there is strange JSESSIONID change in IHS log: 0001ZHPJjMHntWo0qlOrlcDeR7f:-34MHMC 0002ZHPJjMHntWo0qlOrlcDeR7f:-L88QKU:-34MHMC It seems that ...
Hi,
I'm solving problem with session affinity on some web application (WAS 6.0 ND with cluster of 6 appservers and memory-to-memory replication, IHS, AJAX application). There is there is strange JSESSIONID change in IHS log:
0001ZHPJjMHntWo0qlOrlcDeR7f:-34MHMC
0002ZHPJjMHntWo0qlOrlcDeR7f:-L88QKU:-34MHMC
It seems that JSESSIONID contains two cloneids. At time of change session failover to another application server in cluster. I'm note sure if that two cloneids is normal behavior in case of failover.
Unfortunately I don't know why the session failover, because nothing is logged in WebSphere plugin's log (warn loglevel) and I'm not able to simulate it in stress test. It only occurs on heavy load in production environment where we cannot turn the loglevel to trace.
Are the two cloneid in JSESSIONID normal? Has been somebody solving similar problem?
Regards,
Ludek
Hi,
We're having similar problems with two memory-to-memory replication app servers, but only with some apps on the same cluster. We've tried to change session management to "url rewriting", but now we have the same problem in the "jsessionid" parameter of the url.
Maybe you've found any workaround or solution?
Thank you in advance.
