[
https://issues.apache.org/jira/browse/HDDS-3096?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17142318#comment-17142318
]
Bharat Viswanadham commented on HDDS-3096:
------------------------------------------
We can revisit this in the next release.
There are few corner cases in which this can happen, with a simple fix of using
the current pipeline manager count with each pipeline report that will not
solve this issue.
> SCM does not exit Safe mode
> ---------------------------
>
> Key: HDDS-3096
> URL: https://issues.apache.org/jira/browse/HDDS-3096
> Project: Hadoop Distributed Data Store
> Issue Type: Bug
> Components: SCM
> Reporter: Bharat Viswanadham
> Assignee: Bharat Viswanadham
> Priority: Critical
> Labels: Triaged
>
> In a few scenarios, like Disks are gone, the datanode is not up or any other
> case, we may try to close pipelines.
> If we close pipelines after SCM restart, SCM will not come out of safe mode.
> This is because of the current implementation where we get the count of the
> pipeline from DB when creating a SafeMode rule object. During this, if any
> pipeline is closed/removed from DB, the Rule does not know about it, and it
> PipelineSafeMode rule is never met, this causes a situation where we never
> come out of safe mode.
>
>
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
