I am working for monitoring using AppMon
Today morning Application team restarted the server.
The servers take 2 hours to restart due to long running threads.
Threads count fluctuate between 10 to 0.
They need AppMon to give threads which causing server to take long time to restart.
1.Could you please help me with the steps.
2.Could you please tell whether it is possible to get the threads happened in the morning. Should I have initiated already. .
3.I could not find ant threads dashlet in my cockpit .Did I miss to create threat dumps
4.During thread analysis any threat to AppMon tool instability or crash.Is it safe to perform thread analysis.
5.I have tried troubleshooting threads using purepaths -> high response time->method hotspots
(project team denied as it is services, we need thread names actually)
Thanks & regards
Open the Session Storage from the Cockpit to see auto recorded session for the CPU Sampling, Thread Dump and Leak Analysis Memory Snapshot.
Watch the below Performance Clinic which covered the Advanced Diagnostics Session and Thread Analysis.
Thanks for the reply.
In my session storage cockpit I could not see any dash let related to Thread dump or memory leak analysis.
I could not find thread analysis topic in the that you tube link(Sorry if I am wrong)
I am planning to do thread analysis for the suspected agent tomorrow as the restart is planned for that agent 10 AM.
Project team wants me to start thread dump analysis when they restart server.
Is it advisable to do as already the server having threads and taking lot of time to start.
So Can I start analysis during the non business hours selecting the particular time (10 AM)
Is it possible.
Thanks & regards
You might trigger thread dumps at times when the load on the system under diagnosis is low, to ensure that there are no threads leaking, or to compare the CPU usage of different threads to previously created thread dumps but I guess you can proceed as per the project team demand.
Note: While the thread information is collected, all threads are suspended to guarantee consistency of stack traces, states, and monitors.
Also have a look on the below documentation link: