Details
Description
Right when starting CI, I got the following error message on the monitor, and potentially related DEBUG statements in the tserver log.
2014-04-17 15:04:17,940 [log.DfsLogger] ERROR: java.io.IOException: Failed to add a datanode. User may turn off this feature by setting dfs.client.block.write.replace-datanode-on-failure.policy in configu ration, where the current policy is DEFAULT. (Nodes: current=[10.20.212.138:50010, 10.20.212.116:50010], original=[10.20.212.138:50010, 10.20.212.116:50010]) 2014-04-17 15:04:17,941 [log.DfsLogger] DEBUG: DfsLogger.open() begin 2014-04-17 15:04:17,942 [fs.VolumeManagerImpl] DEBUG: Found CREATE enum CREATE 2014-04-17 15:04:17,942 [fs.VolumeManagerImpl] DEBUG: Found synch enum SYNC_BLOCK 2014-04-17 15:04:17,943 [fs.VolumeManagerImpl] DEBUG: CreateFlag set: [CREATE, SYNC_BLOCK] 2014-04-17 15:04:17,943 [fs.VolumeManagerImpl] DEBUG: creating hdfs://nameservice1/accumulo/wal/a2414.halxg.cloudera.com+10011/6a3fb666-b263-400b-b480-64cc250fbf2e with SYNCH_BLOCK flag
This error seems like it was probably completely harmless, so it should be a warn or info.