The message below occurs in our logs in batches of 7 or 8 occurrences over a few ms and separated by a gap of 4-20s. It's been happening for days. The erroring storage node is not the senior node. There is no noticeable service impact. This is a dev cluster consisting of has 8 storage, 4 proxy , 1 MBean server and 4 other process nodes. There is very little load on it. The platform is SunOs 5.10 on x86.
What does it mean?
2013-01-31 19:05:36,514 [Logger@9240084 184.108.40.206] DEBUG Coherence 2013-01-31 19:05:36.514/636281.198 Oracle Coherence GE 220.127.116.11 <D5> (thread=Invocation:Management, member=11): Error handling notification type=Platform,Domain=java.lang,subType=GarbageCollector,name=PS Scavenge,nodeId=2
The log message implies that parallel eden/surivor space collector is running on member with id =2. I don't think it implies any immediate issues but I would recommend to check you heap space graph for the node and also spit gc logs using -Xloggc:<file_location> -verbose:gc and see if the GC is tuned correctly.