accumulo-notifications mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Eric Newton (JIRA)" <>
Subject [jira] [Commented] (ACCUMULO-716) Corrupt WAL file
Date Wed, 09 Jan 2013 18:56:12 GMT


Eric Newton commented on ACCUMULO-716:

I was able to reproduce this.

 * start accumulo
 * use TestIngest to put some data in
 * kill everything
 * find the last block of the last WAL file in the NameNode logs
 * find the block, and delete the last bunch of bytes
 * start accumulo

org.apache.hadoop.fs.ChecksumException: Checksum error: /blk_5930498692645763206:of:/accumulo/wal/
at 3539456
	org.apache.hadoop.fs.ChecksumException: Checksum error: /blk_5930498692645763206:of:/accumulo/wal/
at 3539456
		at org.apache.hadoop.fs.FSInputChecker.verifySum(
		at org.apache.hadoop.fs.FSInputChecker.readChecksumChunk(
		at org.apache.hadoop.fs.FSInputChecker.fill(
		at org.apache.hadoop.fs.FSInputChecker.read1(
		at org.apache.hadoop.hdfs.DFSClient$
		at org.apache.hadoop.hdfs.DFSClient$DFSInputStream.readBuffer(
		at org.apache.hadoop.hdfs.DFSClient$

I was trying to see what would happen if the disk-full occurred while trying to write out
the checksum data.
> Corrupt WAL file
> ----------------
>                 Key: ACCUMULO-716
>                 URL:
>             Project: Accumulo
>          Issue Type: Bug
>          Components: tserver
>         Environment: java version "1.6.0_33", hadoop-0.20.2-cdh3u3
>            Reporter: Josh Elser
>            Assignee: Eric Newton
> Ran wikisearch-ingest. Ended up filling up a drive used by HDFS and things failed not-so-gracefully.
Upon restart, log recovery started, appeared to finish (failed HDFS checksum on one WAL entry),
and left Accumulo in a state where no tablets were assigned.

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see:

View raw message