You are viewing a plain text version of this content. The canonical link for it is here.

Posted to oak-issues@jackrabbit.apache.org by "Marcel Reutegger (JIRA)" <ji...@apache.org> on 2019/04/11 08:34:00 UTC

[jira] [Commented] (OAK-8222) oak-run: add functionality for clusternode cleanup

    [ https://issues.apache.org/jira/browse/OAK-8222?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16815200#comment-16815200 ] 

Marcel Reutegger commented on OAK-8222:
---------------------------------------

The first condition is that an active clusterNodes entry must not be deleted.

Inactive clusterNodes entries may still be used by {{DocumentNodeStore.getMinExternalRevisions()}} to get the startTime. We will have to first understand under what condition this is not needed anymore and an entry could potentially be deleted.

> oak-run: add functionality for clusternode cleanup
> --------------------------------------------------
>
>                 Key: OAK-8222
>                 URL: https://issues.apache.org/jira/browse/OAK-8222
>             Project: Jackrabbit Oak
>          Issue Type: Improvement
>          Components: documentmk, run
>            Reporter: Julian Reschke
>            Assignee: Julian Reschke
>            Priority: Minor
>
> First we need to understand under which conditions an entry in the clusternodes collection can be safely removed.
> If this is sufficient for the given use case of removing ancient, abandoned entries, we might discuss adding this direclty to {{ClusterNodeInfo}}, so that the actual operation can be made safe.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)