hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From James Baldassari <jbaldass...@gmail.com>
Subject ZooKeeper connect/disconnect loop in Master logs of 0.90.{3,4}
Date Thu, 29 Sep 2011 22:54:53 GMT

I'm in the process of upgrading a few CDH3 clusters from 0.90.2 to 0.90.4.
I noticed that after starting the 0.90.4 Master (also verified in 0.90.3), I
see this series of messages every 5 seconds in the Master log:

There are corresponding log entries in the ZooKeeper server log every 5
seconds: http://pastebin.com/N7VGMSFi

So the Master is connecting to ZooKeeper and then immediately disconnecting
every 5 seconds for some reason.  Is this expected?  I don't see any errors
on the master or region servers, and HBase appears to be functioning
normally.  Should I just set the log level to WARN and not worry about it,
or is this actually a problem?


  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message