Hi,
Recently we experienced that some slow query and finally lots of the
queries timeouted. We found that on some data nodes the heap usage is
nearly > 90% and GCs can not free much space.
As we already marked many fields as doc values so at this time, the
fielddata cache size is small (~ 100MB) but we found that there's lots of
segments in memory and GCs seems not able to free them.
More precisely on data nodes we have 4GB heap and nearly 3GB is occupied by
in memory segments. In the same time we checked that only a few (~5)
segments are not committed.
Our questions are
- why we have these in memory segments in memory when they are all
committed? - why when there's pressure on the heap, GC can not free them?
- what brings these segments in memory? search, aggregation and maybe
recovery process?
Thanks!
--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/05b2380d-7afe-4b43-822a-c5d1eeff2733%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.