https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=40820
--- Comment #4 from Tomás Cohen Arazi (tcohen) <[email protected]> --- (In reply to David Cook from comment #3) > (In reply to Tomás Cohen Arazi (tcohen) from comment #0) > > We added the JobsNotificationMethod preference so people can disable the use > > of RabbitMQ explicitly and not get the logs flooded with connection errors. > > > > But a patch on bug 34070 reintroduced the warn ♪┏(・o・)┛♪ > > It took me a few minutes to wrap my head around this one but I see it now. > Bug 34070 added the following: > '$error ||= "Cannot connect to the message broker, the jobs will be > processed anyway" unless $conn;' > > I didn't even know about bug 35655 until this moment. > > > e2e587273c14f092d71046f21266473445b219c5 > > I don't get anything for that commit hash? In a hurry, I just grabbed the hash from the submitted patch on bug 34070 https://bugs.koha-community.org/bugzilla3/attachment.cgi?id=182574 but you guessed right. I believe 34070 patches were wrong because they treated the 'skip if polling' behavior as an error by not acknowledging bug 35655. The code has been there for a good while and this caused a regression. > At a glance I think these patches should make sense... The patches are trivial and just wrap things around a check on JobsNotificationMethod='STOMP' which is also correct. But if we don't call ->connect() unless STOMP, then we don't need the code handling non-STOMP in ->connect. So any of the refactoring efforts should clear this for good I'd say. I might try to resurrect those patches... -- You are receiving this mail because: You are watching all bug changes. _______________________________________________ Koha-bugs mailing list [email protected] https://lists.koha-community.org/cgi-bin/mailman/listinfo/koha-bugs website : http://www.koha-community.org/ git : http://git.koha-community.org/ bugs : http://bugs.koha-community.org/
