Maybe you can try my action: http://apache-hbase.679495.n3.nabble.com/Region-server-not-accept-connections-intermittently-td4061107.html
Before I took the action, I almost see the RS connection problem everyday and the problem usually last about half hour. From the observation in these 2 days after the action was taken, the frequency of the problem has been reduced. The huge improvement is, even when the problem happens, the RS can recover in less than 1 minute comparing with 30-60 minutes before. I guess there should be some room to adjust the parameters more better.
于 2014/7/12 17:39, 李玉明 写道:
Can any experts look at this? Similar posts are : 1. http://mail-archives.apache.org/mod_mbox/hbase-user/201406.mbox/%3C001001cf863e$c81a60e0$584f22a0$@com%3E 2. https://issues.apache.org/jira/browse/HBASE-11306 "Client connection starvation issues under high load on Amazon EC2" 3. https://issues.apache.org/jira/browse/HBASE-11277 Regards, -- Yu Ming
