I didn't know we were already at 20.20.2 ;)

It seems that your client doesn't have the address of the zookeeper
ensemble, but that part of the log is missing so I can't tell.

J-D

On Thu, Dec 2, 2010 at 12:19 PM, Peter Haidinyak <[email protected]> wrote:
> Hidey Ho,
>  I have a three node cluster up running
> Hadoop 20.20.2+737.tar  (from Cloudera)
> HBase  0.89.20100924+28 (from Cloudera)
>
> When I do a JPS everything looks ok.
>
> I am trying to connect to HBase using a Java Client.
>
>    final Configuration config = HBaseConfiguration.create();
>    m_trafficTable = new HTable(config, TRAFFIC_LOG_TABLE_NAME);
>    m_trafficTable.setAutoFlush(false);  // performance
>    m_trafficTable.setWriteBufferSize(12582912L);  // performance
>
> When the client tries to create the configuration I get the following...
>
> 2010-12-02 11:35:59,971 INFO  [Import Click Log Thread] 
> zookeeper.ZooKeeperWrapper(195): Reconnecting to zookeeper
> 2010-12-02 11:36:00,033 DEBUG [Import Click Log Thread] 
> zookeeper.ZooKeeperWrapper(201): 
> <localhost:/hbase,org.apache.hadoop.hbase.client.HConnectionManager>Connected 
> to zookeeper again
> 2010-12-02 11:36:00,035 DEBUG [Import Click Log Thread] 
> zookeeper.ZooKeeperWrapper(466): 
> <localhost:/hbase,org.apache.hadoop.hbase.client.HConnectionManager>Trying to 
> read /hbase/root-region-server
> 2010-12-02 11:36:00,171 DEBUG [Import Click Log Thread] 
> zookeeper.ZooKeeperWrapper(469): 
> <localhost:/hbase,org.apache.hadoop.hbase.client.HConnectionManager>Failed to 
> read org.apache.zookeeper.KeeperException$ConnectionLossException: 
> KeeperErrorCode = ConnectionLoss for /hbase/root-region-server
> 2010-12-02 11:36:00,172 DEBUG [Import Click Log Thread] 
> client.HConnectionManager$TableServers(1077): Sleeping 5000ms, waiting for 
> root region.
> 2
>
> This just repeats.
>
> Any ideas?
>
> Thanks
>
> -Pete
>
>

Reply via email to