Hello everyone, The current jenkins setup won't run builds on PRs from people who don't have write access to the repository. This means that even though Benoit fixed the cassandra-related flaky test[1] and I fixed a rabbitmq-related flaky test[2] andI then rebased it all to account for Benoit's changes, I can't go any further because I can't trigger builds to detect more flaky tests or reach a successful build :)
I need help from one of the project's committers to duplicate the PR again (as matthieu did in https://github.com/apache/james-project/pull/265) to get jenkins to build it (or to change the settings in jenkins to let it build PRs from "untrusted" users or somehow whitelist PR #265 to be built I'm not sure exactly what can be done there). Of note, the apache ci platform seems reasonably powerful when running the builds, I don't have much comparison points but a full project build (without tests) at T1C completes in about 5 minutes. I have no idea how that compares to the private linagora CI but it would be so much better than the nothing non linagora user currently enjoy ;) It is unclear from the documentation[3] whether there really are compute quotas enabled or not, I don't think there are by default and watching the activemq/maven builds run along for hours on the shared nodes I feel that james will be just fine :) The documentation [4] also mentions that apache CI boasts a free SonarQube integration with sonarcloud.io, I think it would be a nice next step after CI is enabled on the project Thanks in advance Jean As a side note I'm currently upgrading the MPT SMTP tests to use junit 5 jupiter APIS (with extensions and the like) instead of rules. [1] https://github.com/apache/james-project/pull/267 [2] https://github.com/apache/james-project/pull/264/commits/c2218b1eff07245294ee976507b9cda013a1b0b7 [3] https://cwiki.apache.org/confluence/display/INFRA/Jenkins [4] https://cwiki.apache.org/confluence/display/INFRA/SonarQube+Analysis On Thu, Nov 26, 2020 at 12:30 PM Jean Helou <jean.he...@gmail.com> wrote: > > > Likely "unstable". >> >> I will have a look at this tomorrow. >> > > Ok thanks > >>
