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 "zhangwei (JIRA)" <ji...@apache.org> on 2018/01/24 07:18:00 UTC
[jira] [Updated] (HDFS-13051) dead lock occurs when rolleditlog rpc
call happen and editPendingQ is full
[ https://issues.apache.org/jira/browse/HDFS-13051?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
zhangwei updated HDFS-13051:
----------------------------
Target Version/s: 3.0.0, 2.7.5, 2.8.2, 2.8.1 (was: 2.8.1, 2.8.2, 2.7.5, 3.0.0)
Description: When we apply AsyncEditlog in our prod env, in case, deadlock will occur and crash the namenode. in the case, rollEditlog rpc call happen while editPendingQ in AsyncEditlog is full (was: When we apply AsyncEditlog in our prod env, in extremely case, deadlock will occur and crash the namenode. in the case, rollEditlog rpc call happen while AsyncEditlog asynchronized takes edit object from queue and is going to acquire AsyncEditlog object lock to executed it's logEdit method.)
Summary: dead lock occurs when rolleditlog rpc call happen and editPendingQ is full (was: dead lock occurs when using async editlog while rolleditlog rpc call happen)
> dead lock occurs when rolleditlog rpc call happen and editPendingQ is full
> --------------------------------------------------------------------------
>
> Key: HDFS-13051
> URL: https://issues.apache.org/jira/browse/HDFS-13051
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: namenode
> Affects Versions: 2.7.5
> Reporter: zhangwei
> Priority: Major
>
> When we apply AsyncEditlog in our prod env, in case, deadlock will occur and crash the namenode. in the case, rollEditlog rpc call happen while editPendingQ in AsyncEditlog is full
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-help@hadoop.apache.org