[
https://issues.apache.org/jira/browse/HDFS-2982?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13280613#comment-13280613
]
Todd Lipcon commented on HDFS-2982:
-----------------------------------
Oops, missed part of my copy-paste comment:
{code}
+ LOG.info(this + ": selecting input streams starting at " + fromTxId +
+ (inProgressOk ? " (inProgress ok) " : " (excluding inProgress) ") +
+ "from among " + elfs.size() + " candidate file(s)");
{code}
This should probably be DEBUG level. Otherwise this will show up in the "safety
check" selectInputStreams call, and potentially confuse users.
> Startup performance suffers when there are many edit log segments
> -----------------------------------------------------------------
>
> Key: HDFS-2982
> URL: https://issues.apache.org/jira/browse/HDFS-2982
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: name-node
> Affects Versions: 2.0.0
> Reporter: Todd Lipcon
> Assignee: Colin Patrick McCabe
> Priority: Critical
> Attachments: HDFS-2982.001.patch, HDFS-2982.002.patch,
> HDFS-2982.003.patch, HDFS-2982.004.patch, HDFS-2982.005.patch,
> HDFS-2982.006.patch, HDFS-2982.007.patch, HDFS-2982.008.patch
>
>
> For every one of the edit log segments, it seems like we are calling
> listFiles on the edit log directory inside of {{findMaxTransaction}}. This is
> killing performance, especially when there are many log segments and the
> directory is stored on NFS. It is taking several minutes to start up the NN
> when there are several thousand log segments present.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
