Dear Team, I have loaded 3.6GB of compressed(bz2) directly into Hive, after that I ran a simple "*select query*", namenode got crashed. There after not able to start namenode.
Envoronment: - CentOS release 5.5 (Final), Hadoop Version: 0.20.2 - Cluster size: 18 node - NameNode & SecondaryNamenode are in the same Machine It seems editlogs/fsimage got corrupted, I haven't take any backup separately, below is the exception 2011-07-14 23:37:43,378 ERROR org.apache.hadoop.hdfs.server.namenode.FSNamesystem: FSNamesystem initialization failed. java.io.FileNotFoundException: File does not exist: /opt/data/tmp/mapred/system/job_201107041958_0120/j^@^@^@^@^@^@ *Please find detailed exception in namenode's log file attached.* Earlier, I have also posted in JIRA, https://issues.apache.org/jira/browse/HADOOP-7458 , Jakob Homan directed me to post in hdfs user's list. Will there be any backup in SecondaryNamenode? Could you please assist me to recover Namenode from this issue? Thanks, Sakthivel
hadoop-hadoop-namenode-Master.log
Description: Binary data
