[
https://issues.apache.org/jira/browse/TS-4317?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15224374#comment-15224374
]
Alan M. Carroll commented on TS-4317:
-------------------------------------
I'm not sure how serious it is. I would note I found this not by code
inspection but by actually having the problem occur during debugging other
efforts. It is possible it is related to the chronic issue that sometimes disks
that failed don't get added back to the cache - perhaps it's perpetually
recovery scanning instead since in the known cases there is disk I/O on the
replaced drive. On the other hand the fix is very simple and obvious.
> Cache recovery can stall, prevent disk from becoming ready.
> -----------------------------------------------------------
>
> Key: TS-4317
> URL: https://issues.apache.org/jira/browse/TS-4317
> Project: Traffic Server
> Issue Type: Bug
> Components: Cache
> Reporter: Alan M. Carroll
> Assignee: Alan M. Carroll
> Fix For: 6.2.0
>
>
> Recovery scans a stripe to find recoverable data. In a certain case it can
> loop forever, although at a limited rate which is not always noticeable. The
> disk, however, never enters a ready state.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
