[
https://issues.apache.org/jira/browse/HADOOP-14035?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16038889#comment-16038889
]
Kihwal Lee commented on HADOOP-14035:
-------------------------------------
bq. Server metrics look good despite horrible client latency.
We should fix this too. If the connection is gone, the work that handlers did
will not counted in the metrics. This can be very misleading.
+1 for the patch. Also, we have been running it in production for a while now.
> Reduce fair call queue backoff's impact on clients
> --------------------------------------------------
>
> Key: HADOOP-14035
> URL: https://issues.apache.org/jira/browse/HADOOP-14035
> Project: Hadoop Common
> Issue Type: Sub-task
> Components: ipc
> Affects Versions: 2.7.0
> Reporter: Daryn Sharp
> Assignee: Daryn Sharp
> Attachments: HADOOP-14035.patch
>
>
> When fcq backoff is enabled and an abusive client overflows the call queue,
> its connection is closed, as well as subsequent good client connections.
> Disconnects are very disruptive, esp. to multi-threaded clients with multiple
> outstanding requests, or clients w/o a retry proxy (ex. datanodes).
> Until the abusive user is downgraded to a lower priority queue,
> disconnect/reconnect mayhem occurs which significantly degrades performance.
> Server metrics look good despite horrible client latency.
> The fcq should utilize selective ipc disconnects to avoid pushback
> disconnecting good clients.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: common-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: common-issues-h...@hadoop.apache.org
