[
https://issues.apache.org/jira/browse/OAK-7710?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16608938#comment-16608938
]
Tomek Rękawek commented on OAK-7710:
------------------------------------
[~mreutegg], [~catholicon] - WRT to Marcel's latest comment:
> I think there's a general problem with how the CompositeNodeStore.merge()
> method interacts with the stores and the observers. The assumption is that
> the mergeLock protects a critical section, which is not the case when the
> DocumentNodeStore is used as the globalStore and external changes come in.
> Maybe the observer callbacks could be coordinated with the help of the
> CommitInfo? IIRC that's the same object passed to the merge() call.
Can we use the CommitInfo#isExternal() to differentiate between these two
cases? Eg. we can:
1. for the local changes, we can explicitly invoke the
{{dispatcher.contentChanged()}} in the {{CompositeNodeStore#merge()}} method,
2. for the external changes, we can only process events with
{{CommitInfo#isExternal() == true}} in the {{MountedNodeStoreObserver}}.
> CompositeNodeStore does not dispatch external events to observers
> -----------------------------------------------------------------
>
> Key: OAK-7710
> URL: https://issues.apache.org/jira/browse/OAK-7710
> Project: Jackrabbit Oak
> Issue Type: Bug
> Components: composite
> Reporter: Vikas Saurabh
> Assignee: Tomek Rękawek
> Priority: Major
> Fix For: 1.10
>
> Attachments: OAK-7710.patch, OAK-7710.test.patch
>
>
> Currently {{CompositeNodeStore}} only ever dispatches changes from inside its
> {{merge}} method. This then loses external events that could be read in
> background read of some underlying {{DocumentNodeStore}}.
> [^OAK-7710.test.patch] has an ignored test representing the scenario.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
