Please, see all necessary info in linked JIRA. This problem doesn't look to be JDG specific.
Dan Berindei <dberinde> made a comment on jira ISPN-2759 Are you sure TRACE is enabled in Infinispan? I'm looking at the CacheComponent code and it logs quite a bit of information at TRACE level. In fact, if TRACE was enabled, the 2nd NPE should have been in the previous line.
Tomas Sykora <tsykora> made a comment on jira ISPN-2759 Tristan's PR solves this.
Adrian Nistor <anistor> made a comment on jira ISPN-2759 Fix integrated. Thanks!
Dan Berindei <dberinde> made a comment on jira ISPN-2759 The RHQ plugin could really use some tests to smoke out this kind of problems. I've created ISPN-2761 to track that.
ER10 does NOT contain fix for this. (Changes of CacheComponent.java class in rhq-plugin module didn't go into CR3 (= neither ER10). Anyway infinispan upstream master contains this fix.
Great, ER11 is ok. I was able to stop cache, start it again and for example issue reset statistics operation too. This was something more like "smoke test", we need to test all operations (could be time consuming because of setting proper environment - for example for X-site ops. etc.) Setting this as VERIFIED, problem solved, thanks!
Assuming this bug was fixed due to VERIFIED status and removing release notes flag.