Thanks all for contributing to the discussion. I'll write up a JIRA before Monday and try to summarize the discussion and we can go from there. Hmmm, a LUCENE or SOLR JIRA? Does it matter?
One thing I also realized is that my use-case would be served just fine with an easy way to identify runs with awaitsfix=true set, especially if it were in the subject line. That's an easy filter to create so I can look at whichever one suits me at the moment. I did check yesterday and there are 20+ tests with the AwaitsFix annotation. An additional question is which annotation to repurpose, we can discuss that on the JIRA as well. Hoss' program won't report on tests that aren't run, so there's additional value in runs with awaitsfix=true to give it some data to chew on. Erick On Thu, Feb 22, 2018 at 6:41 AM, Yonik Seeley <ysee...@gmail.com> wrote: > On Thu, Feb 22, 2018 at 8:59 AM, Adrien Grand <jpou...@gmail.com> wrote: >> I understand your point Yonik, but the practical consequences are worse > > I think that's what we were debating though. IMO, the overall > practical consequences of simply sweeping the problem under the rug by > disabling all flakey tests is actually worse. > I agree with the issue of noise in automated reports decreasing their > relevance, but it's an independent (albeit related) issue and there > are better ways to fix that. > > But at this point, I think I'll defer to the people who have made time > to work on this problem (and adding @Ignore to flakey tests isn't > actually decreasing the number of flakey tests). > > -Yonik > > --------------------------------------------------------------------- > To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org > For additional commands, e-mail: dev-h...@lucene.apache.org > --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
