It's not normal. Those happens when the heartbeats coming from the client zookeeper to the server zookeeper stop for some configurable time, if you're using the defaults in 0.90 then that's almost 2 minutes IIRC.
Unless you are using your client in extreme ways every hour, it doesn't seem likely that this would be related to GC. To rule that out you could log them. Maybe the process is getting swapped out since it's not used? J-D On Tue, Mar 8, 2011 at 7:33 AM, Barney Frank <[email protected]> wrote: > On my development environment (my laptop), I get warnings related to > zookeeper client connections timing out. The warnings are during times when > there is zero load on the client and hbase server. Should I be concerned or > is this pretty typical? My guess is that it is during a GC pause since it > happens like clockwork every hour or two so I probably will look at that. > Any suggestions would be appreciated. > > 07:40:57,267 INFO > [org.apache.hadoop.hbase.client.HConnectionManager$HConnectionImplementation] > Reconnected successfully. This disconnect could have been caused by a > network partition or a long-running GC pause, either way it's recommended > that you verify your environment. >
