[
https://issues.apache.org/jira/browse/OAK-1332?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Jukka Zitting resolved OAK-1332.
--------------------------------
Resolution: Fixed
I added OAK-1368 to track the multiple listeners per session case.
After some fairly extensive refactoring (and some stepping on toes, sorry about
that!), I managed to significantly reduce the memory overhead of both the
queued {{Event}} instances (the amortized size of an average queued event
instance is now just around 200 bytes) and the continuation-based event
generation/queueing mechanism. I even implemented the mentioned
stop-and-continue mechanism that should solve this issue even in a worst case
scenario. Thus resolving as fixed.
As mentioned above and discussed on the mailing list, it is somewhat debatable
whether the benefits of the stop-and-continue feature really is worth the extra
complexity, so we might still decide to drop or disable it for now.
> Large number of changes to the same node can fill observation queue
> -------------------------------------------------------------------
>
> Key: OAK-1332
> URL: https://issues.apache.org/jira/browse/OAK-1332
> Project: Jackrabbit Oak
> Issue Type: Improvement
> Reporter: Michael Dürig
> Assignee: Jukka Zitting
> Priority: Minor
> Fix For: 0.16
>
>
> Changing, adding or removing a lot of (10'000s or more) direct children of
> node with on save() might cause observation queues to fill up.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
