hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Daryn Sharp (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-10301) BlockReport retransmissions may lead to storages falsely being declared zombie if storage report processing happens out of order
Date Mon, 02 May 2016 14:48:13 GMT

    [ https://issues.apache.org/jira/browse/HDFS-10301?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15266715#comment-15266715
] 

Daryn Sharp commented on HDFS-10301:
------------------------------------

Still catching up and need to review patch.  First question, how is this interleaving happening
on a frequent basis?

An interesting observation (if I interpreted the logs correctly) is processing all 4 storages
with ~14k blocks/storage appears to takes minutes to process?  Tens of seconds appear to elapse
between processing each storage.  There's some serious contention that seems indicative of
a nasty bug or suboptimal configuration exacerbating this bug.

Is the DN rpc timeout set to something very low?  Has the number of RPC handlers been greatly
increased?   Are there frequent deletes of massive trees?  Is there a lot of decomm'ing with
a low check interval?

> BlockReport retransmissions may lead to storages falsely being declared zombie if storage
report processing happens out of order
> --------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: HDFS-10301
>                 URL: https://issues.apache.org/jira/browse/HDFS-10301
>             Project: Hadoop HDFS
>          Issue Type: Bug
>          Components: namenode
>    Affects Versions: 2.6.1
>            Reporter: Konstantin Shvachko
>            Assignee: Colin Patrick McCabe
>            Priority: Critical
>         Attachments: HDFS-10301.002.patch, HDFS-10301.003.patch, HDFS-10301.01.patch,
zombieStorageLogs.rtf
>
>
> When NameNode is busy a DataNode can timeout sending a block report. Then it sends the
block report again. Then NameNode while process these two reports at the same time can interleave
processing storages from different reports. This screws up the blockReportId field, which
makes NameNode think that some storages are zombie. Replicas from zombie storages are immediately
removed, causing missing blocks.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-help@hadoop.apache.org


Mime
View raw message