[
https://issues.apache.org/jira/browse/HDFS-10475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15459579#comment-15459579
]
Zhe Zhang commented on HDFS-10475:
----------------------------------
Thanks for confirming this Xiaoyu.
[~xkrogen] Has started some work on this direction. Now we have INFO level
logging for write lock held for over 1 seconds and read lock held for over 5
seconds (both configurable). We also have a metric for the lock queue length.
I think we can use this JIRA to discuss what other metrics to add
# A complete op -> aggregate lock time map?
# Top _n_ types of RPC calls with longest read / write lock?
# Top _n_ paths leading to longest aggregate read / write lock?
A similar idea was [mentioned |
https://issues.apache.org/jira/browse/HDFS-10713?focusedCommentId=15453607&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-15453607]
under HDFS-10713. So pinging [~jingzhao] [~liuml07] [~arpitagarwal] and
[~hanishakoneru] for opinions.
> Adding metrics for long FSNamesystem read and write locks
> ---------------------------------------------------------
>
> Key: HDFS-10475
> URL: https://issues.apache.org/jira/browse/HDFS-10475
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: namenode
> Reporter: Xiaoyu Yao
> Assignee: Erik Krogen
>
> This is a follow up of the comment on HADOOP-12916 and
> [here|https://issues.apache.org/jira/browse/HDFS-9924?focusedCommentId=15310837&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-15310837]
> add more metrics and WARN/DEBUG logs for long FSD/FSN locking operations on
> namenode similar to what we have for slow write/network WARN/metrics on
> datanode.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
