Hi, We are running CDH 4.5, 6 nodes with hbase cluster (0.94.6). We configured major compaction to run once a week. During this time we are getting timeouts while writing to hbase (application level). When investigating it further more I have noticed that the timeout are being caused by GC "Stop the world" scenario which takes 3 minutes and not the actual Major compaction (which takes few hours). Is there any GC configuration we can set for relaxing it during Major compaction? Note that we see this GC peak just when the major compaction starts. We are running with default hbase GC settings. Yaniv
-- View this message in context: http://apache-hbase.679495.n3.nabble.com/GC-peaks-during-major-compaction-tp4062704.html Sent from the HBase User mailing list archive at Nabble.com.