hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "stack (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HDFS-127) DFSClient block read failures cause open DFSInputStream to become unusable
Date Mon, 19 Oct 2009 22:44:15 GMT

    [ https://issues.apache.org/jira/browse/HDFS-127?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12767591#action_12767591
] 

stack commented on HDFS-127:
----------------------------

+1 on this patch.

Some notes:

+ This patch does not have the issue that hadoop-1911 "fixed" though it moves the failure
counter back to being a local variable.  There is no danger that it "...revert[s]- and reintroduce[s]-
HADOOP-1911?" because the patch removes the catching of exceptions inside chooseDataNode while
loop.
+ I am of the opinion that this patch should not be gated on the necessary work HDFS-656 "Clarify
error handling and retry semantics for DFS read path" (nor HDFS-378 "DFSClient should track
failures by block rather than globally").  This patch should actually help some as it narrows
scope of the failures variable.  But maybe we can do even more as part of this patch (See
below).

If a new patch is cut, the following might be considered:

+ There is no documentation of "dfs.client.max.block.acquire.failures" nor of what local counter
'failures' means.  Above its suggested that we at least doc. what a failure is ("For e.g.
it looks like 'a failure' is defined as not being able to connect/fetch from all the replicas.
Also success is defined as just being able to get 'ok' from datanode.")
+ By default, "dfs.client.max.block.acquire.failures" is 3.  If more than 3 replicas, we could
fail though good replicas out on cluster.  Should DFSClient set this.maxBlockAcquireFailures
default to max of dfs.replication/dfs.client.max.block.acquire.failures?
+ I'd put these two LOG.info messages together rather than have two LOG.info lines:

{code}
          if (nodes == null || nodes.length == 0) {
            LOG.info("No node available for block: " + blockInfo);
          }
          LOG.info("Could not obtain block " + block.getBlock()
              + " from any node. Will get new block locations from namenode and retry...");
{code}

> DFSClient block read failures cause open DFSInputStream to become unusable
> --------------------------------------------------------------------------
>
>                 Key: HDFS-127
>                 URL: https://issues.apache.org/jira/browse/HDFS-127
>             Project: Hadoop HDFS
>          Issue Type: Bug
>          Components: hdfs client
>            Reporter: Igor Bolotin
>         Attachments: 4681.patch, h127_20091016.patch, h127_20091019.patch
>
>
> We are using some Lucene indexes directly from HDFS and for quite long time we were using
Hadoop version 0.15.3.
> When tried to upgrade to Hadoop 0.19 - index searches started to fail with exceptions
like:
> 2008-11-13 16:50:20,314 WARN [Listener-4] [] DFSClient : DFS Read: java.io.IOException:
Could not obtain block: blk_5604690829708125511_15489 file=/usr/collarity/data/urls-new/part-00000/20081110-163426/_0.tis
> at org.apache.hadoop.hdfs.DFSClient$DFSInputStream.chooseDataNode(DFSClient.java:1708)
> at org.apache.hadoop.hdfs.DFSClient$DFSInputStream.blockSeekTo(DFSClient.java:1536)
> at org.apache.hadoop.hdfs.DFSClient$DFSInputStream.read(DFSClient.java:1663)
> at java.io.DataInputStream.read(DataInputStream.java:132)
> at org.apache.nutch.indexer.FsDirectory$DfsIndexInput.readInternal(FsDirectory.java:174)
> at org.apache.lucene.store.BufferedIndexInput.refill(BufferedIndexInput.java:152)
> at org.apache.lucene.store.BufferedIndexInput.readByte(BufferedIndexInput.java:38)
> at org.apache.lucene.store.IndexInput.readVInt(IndexInput.java:76)
> at org.apache.lucene.index.TermBuffer.read(TermBuffer.java:63)
> at org.apache.lucene.index.SegmentTermEnum.next(SegmentTermEnum.java:131)
> at org.apache.lucene.index.SegmentTermEnum.scanTo(SegmentTermEnum.java:162)
> at org.apache.lucene.index.TermInfosReader.scanEnum(TermInfosReader.java:223)
> at org.apache.lucene.index.TermInfosReader.get(TermInfosReader.java:217)
> at org.apache.lucene.index.SegmentTermDocs.seek(SegmentTermDocs.java:54) 
> ...
> The investigation showed that the root of this issue is that we exceeded # of xcievers
in the data nodes and that was fixed by changing configuration settings to 2k.
> However - one thing that bothered me was that even after datanodes recovered from overload
and most of client servers had been shut down - we still observed errors in the logs of running
servers.
> Further investigation showed that fix for HADOOP-1911 introduced another problem - the
DFSInputStream instance might become unusable once number of failures over lifetime of this
instance exceeds configured threshold.
> The fix for this specific issue seems to be trivial - just reset failure counter before
reading next block (patch will be attached shortly).
> This seems to be also related to HADOOP-3185, but I'm not sure I really understand necessity
of keeping track of failed block accesses in the DFS client.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message