[
https://issues.apache.org/jira/browse/TS-1637?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13758768#comment-13758768
]
Bin Chen commented on TS-1637:
------------------------------
mgmt_select() broadcast fd will be timeout. So checkPeers will be setting
node(timeout) down. So we should reduce tv about mgmt_select(). set the default
tv timeout 5s.
> nodes as idle/dead if we have not heard from them in awhile
> -----------------------------------------------------------
>
> Key: TS-1637
> URL: https://issues.apache.org/jira/browse/TS-1637
> Project: Traffic Server
> Issue Type: Improvement
> Components: Clustering, Management
> Affects Versions: 3.2.0
> Reporter: Bin Chen
> Assignee: Bin Chen
> Fix For: 4.2.0
>
> Attachments: TS-1637.patch, TS-1637_v2.patch
>
>
> in ClusterCom::sendReliableMessageReadTillClose(), if we can't connect peer
> node quickly(if timeout, will > 30s(default peer_timeout)), then won't send
> Shared date called by ClusterCom::sendSharedData. So peer node will marking
> me down.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
