Idea ID 1668588
Currently we don’t log anywhere the JVM statistics of the remote processes started by several discovery adapters. It’s difficult to tell how and what memory they consume and normally it’s a trial and error approach. The end result would be to have a better logging on remote process failures due to insufficient memory or to be able to allocate the proper memory to such processes.
We have the general jvm_statistics.log but only for the probe JVM.
In CP28 (QCCR1H119428 ) we alligned the OOTB JVM parameters to JRE8 (metaspace) but we never log the JVM statistics for the remote processes that can be started during normal discovery flows.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.