You are viewing a plain text version of this content. The canonical link for it is here.
Posted to hdfs-issues@hadoop.apache.org by "Jing Zhao (Jira)" <ji...@apache.org> on 2022/03/09 21:30:00 UTC
[jira] [Commented] (HDFS-16422) Fix thread safety of EC decoding during concurrent preads
[ https://issues.apache.org/jira/browse/HDFS-16422?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17503858#comment-17503858 ]
Jing Zhao commented on HDFS-16422:
----------------------------------
Looks like we already have r/w lock protection in AbstractNativeRawDecoder and its subclasses (NativeRSRawDecoder and NativeXORRawDecoder). So does that mean the extra protection is only necessary for other decoder implementations (such as RSRawDecoder)?
HADOOP-15499 used r/w lock to replace the original object monitor (i.e. synchronized) so as to improve performance. Now looks like we're adding "synchronized" back to the APIs defined in the parent class.
I guess instead of updating the decode APIs in RawErasureDecoder, we may want to only fix the subclasses without lock protection. What do you think, [~weichiu] [~cndaimin] ?
> Fix thread safety of EC decoding during concurrent preads
> ---------------------------------------------------------
>
> Key: HDFS-16422
> URL: https://issues.apache.org/jira/browse/HDFS-16422
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: dfsclient, ec, erasure-coding
> Affects Versions: 3.3.0, 3.3.1
> Reporter: daimin
> Assignee: daimin
> Priority: Critical
> Labels: pull-request-available
> Fix For: 3.4.0, 3.2.3, 3.3.3
>
> Time Spent: 3h 40m
> Remaining Estimate: 0h
>
> Reading data on an erasure-coded file with missing replicas(internal block of block group) will cause online reconstruction: read dataUnits part of data and decode them into the target missing data. Each DFSStripedInputStream object has a RawErasureDecoder object, and when we doing pread concurrently, RawErasureDecoder.decode will be invoked concurrently too. RawErasureDecoder.decode is not thread safe, as a result of that we get wrong data from pread occasionally.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-help@hadoop.apache.org