[
https://issues.apache.org/jira/browse/HDDS-3486?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17629567#comment-17629567
]
Siddharth Wagle commented on HDDS-3486:
---------------------------------------
What does this setting mean? Recon tracks containers from DN HB, does this
protect Recon from going out of sync? If that is the case how do we catch it? I
think we have not hammered this out properly.
At a minimum, we should track this as a metric. So an advanced user can tell if
there is a problem with Recon being able to consume the DN HBs.
A secondary consideration is doing a sequence number-based checkpoint and
periodic sync. What do you guys think? [~deveshsingh]and [~aswinshakil]
> Recon cannot track missing containers that were created and went missing
> while it is down.
> ------------------------------------------------------------------------------------------
>
> Key: HDDS-3486
> URL: https://issues.apache.org/jira/browse/HDDS-3486
> Project: Apache Ozone
> Issue Type: Bug
> Components: Ozone Recon
> Affects Versions: 1.0.0
> Reporter: Aravindan Vijayan
> Assignee: Devesh Kumar Singh
> Priority: Critical
>
> Since Recon relies on Container reports from DNs to track replicas, if a new
> container was created and went missing when Recon was down, it will not be
> able to tag it as a missing container.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
