Hi Ted

We are using hbase-1.0.0.
The issue seems to suddenly arise. 
It does not happen during startup of
client process.
The servers hosting hbase:meta seems normal during the issue. 
We have modified some of the hbase client properties 
apart from max and core threads
like
hbase.client.retries.number => 5
zookeeper.recovery.retry => 0
zookeeper.session.timeout => 1000
zookeeper.recovery.retry.intervalmilli => 1
hbase.rpc.timeout => 30000.
We changed this config to have fail fast during 
any hbase or hadoop deamon failure.

Thanks

Regards
Mukund Murrali

Reply via email to