You are viewing a plain text version of this content. The canonical link for it is here.
Posted to server-dev@james.apache.org by "Benoit Tellier (Jira)" <se...@james.apache.org> on 2020/12/11 05:57:00 UTC
[jira] [Closed] (JAMES-3289) Flux.merge bugs lead to build
instability
[ https://issues.apache.org/jira/browse/JAMES-3289?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Benoit Tellier closed JAMES-3289.
---------------------------------
Fix Version/s: 3.6.0
Resolution: Fixed
0345c32c2e7f56dcada0a378221a935a4c904ec7 fixed this
> Flux.merge bugs lead to build instability
> -----------------------------------------
>
> Key: JAMES-3289
> URL: https://issues.apache.org/jira/browse/JAMES-3289
> Project: James Server
> Issue Type: Bug
> Components: tests
> Affects Versions: 3.5.0
> Reporter: Rémi Kowalski
> Priority: Major
> Fix For: 3.6.0
>
>
> This reactor bug :
> [https://github.com/reactor/reactor-core/issues/1941]
> is the source of instability in our tests, as in the case of a first failing first publisher in a merge operation, its error would not be propagated and the result of the merge will be a successful Publisher instead of the expected failing one.
> This behavior is reproductible in :
> ConsistencyTasksIntegrationTest.shouldRecomputeMailboxCounters
> Where the merge operation in CassandraMessageIdMapper.insertIds is impacted by this bug.
> This bug should not be a real problem in real usage as it fails only when the failure is immediate. Which should not be the case with Publisher doing IOs in production usage.
> So we will add a delay in the throwing behavior in the integration tests and wait for this bug to be fixed in reactor.
> {color:#00627a} {color}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: server-dev-unsubscribe@james.apache.org
For additional commands, e-mail: server-dev-help@james.apache.org