[
https://issues.apache.org/jira/browse/HDFS-16617?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Yongpan Liu updated HDFS-16617:
-------------------------------
Description:
Observer NN may have poor read performance due to various reasons (such as GC,
code, or txid lags too far, etc.).
Added latency monitoring with Active NN's txid to Observer NN to may help us
troubleshoot some of the issues.
was:
Observer NN may have poor read performance due to various reasons (serious GC,
code problems, or txid lags behind, etc.).
Added latency monitoring with Active NN's txid to Observer NN to help us
troubleshoot some of the issues.
> [SBN read] add txid lag metric for Observer NameNode
> ----------------------------------------------------
>
> Key: HDFS-16617
> URL: https://issues.apache.org/jira/browse/HDFS-16617
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: namenode
> Reporter: Yongpan Liu
> Priority: Minor
> Attachments: HDFS-16617.001.patch
>
>
> Observer NN may have poor read performance due to various reasons (such as
> GC, code, or txid lags too far, etc.).
> Added latency monitoring with Active NN's txid to Observer NN to may help us
> troubleshoot some of the issues.
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
