Try creating the "ci" table using the accumulo shell before you run the test. The continuous ingest test is designed to run in parallel with many workers, so it doesn't try to create the table itself.
Cheers, Adam On Tue, Oct 30, 2012 at 4:53 PM, Ranjan Sen <[email protected]> wrote: > we are trying to run the continuous ingest program via the > $ACCUMULO_HOME/test/system/continuous/start-ingest.sh script. We find that > the table 'ci' (default) is not getting created. We turned on the ingest > log and find the following message: > > 30 20:38:50,746 [zookeeper.ZooSession] DEBUG: Connecting to > node0002,node0003,node0004,node0005,node0006 with timeout 30000 with auth > 30 20:38:50,924 [zookeeper.ZooCache] TRACE: zookeeper contained > /accumulo/instances/instance c64a50eb-ce2e-4a3b-9ddc-ea4421085321 > 30 20:38:50,924 [zookeeper.ZooCache] TRACE: putting > /accumulo/instances/instance c64a50eb-ce2e-4a3b-9ddc-ea4421085321 in cache > 30 20:38:51,013 [zookeeper.ZooCache] TRACE: zookeeper contained > /accumulo/c64a50eb-ce2e-4a3b-9ddc-ea4421085321 > 30 20:38:51,013 [zookeeper.ZooCache] TRACE: putting > /accumulo/c64a50eb-ce2e-4a3b-9ddc-ea4421085321 in cache > 30 20:38:51,118 [impl.ServerClient] WARN : There are no tablet servers: check > that zookeeper and accumulo are running. > 30 20:38:51,119 [impl.ServerClient] DEBUG: ClientService request failed null, > retrying ... > org.apache.thrift.transport.TTransportException: Failed to connect to a server > > > The current hdfs entry for accumulo/instance_id is > > -rw-r--r-- 3 accumulo accumulo 0 2012-10-29 21:51 > /accumulo/instance_id/f047ac35-fb66-466c-bb53-58d453900d5b > > > The ingest .out file in a datanode says (when ingest log was not turned on) > > 30 20:30:57,350 [impl.ServerClient] WARN : There are no tablet servers: check > that zookeeper and accumulo are running. > > > What is causing this problem?? > > Ranjan > > > >
