[
https://issues.apache.org/jira/browse/SLING-8945?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17001075#comment-17001075
]
Ashish Chopra commented on SLING-8945:
--------------------------------------
Raised [0], which leverages the fact that sling-distribution-core bundle can
now (post-merge of
[#24|https://github.com/apache/sling-org-apache-sling-distribution-core/pull/24])
contain java 8 constructs. This allows putting the {{statusMap}} back to
queue-impls (where they truly belong) which
[#26|https://github.com/apache/sling-org-apache-sling-distribution-core/pull/26]
- incorrectly, in hindsight - had moved to queue-providers. A bunch of changes
at [0] are actually reverting the files to a state before
[#26|https://github.com/apache/sling-org-apache-sling-distribution-core/pull/26]
was merged.
[0] https://github.com/apache/sling-org-apache-sling-distribution-core/pull/31
> ActiveResourceQueue doesn't provide cluster-consistent processing-attempts
> view
> -------------------------------------------------------------------------------
>
> Key: SLING-8945
> URL: https://issues.apache.org/jira/browse/SLING-8945
> Project: Sling
> Issue Type: Bug
> Components: Content Distribution
> Affects Versions: Content Distribution Core 0.4.0
> Reporter: Ashish Chopra
> Assignee: Timothee Maret
> Priority: Critical
> Fix For: Content Distribution Core 0.4.2
>
> Time Spent: 10m
> Remaining Estimate: 0h
>
> [0] enables {{ActiveResourceQueue}} (introduced via SLING-8853) to capture
> processing attempts, but since those are kept in memory the
> processing-attempts view is not cluster-consistent. Specifically, if a
> {{ActiveResourceQueue}} is introspected, and its Status or its Item Status is
> requested by an instance that is _not_ processing the {{ResourceQueue}} (say,
> in response to a browser request), the statuses will be returned incorrectly
> (e.g., a blocked queue will be returned as running since the processing
> attempts will not be incremented on the instance not processing the queue).
> One way to fix this would be to persist the processing attempts alongside the
> queue-items in the resource-tree that backs the {{ResourceQueue}}.
> [0]
> https://github.com/apache/sling-org-apache-sling-distribution-core/commit/20f5575af5318c07c31e7623319cf4f1d20258c8
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
