On 02/16/2017 06:11 AM, Lukas Slebodnik wrote: > On (16/02/17 09:50), William Brown wrote: >> On Wed, 2017-02-15 at 08:16 -0500, Mark Reynolds wrote: >>> On 02/15/2017 04:48 AM, Lukas Slebodnik wrote: >>>> On (15/02/17 09:05), Jenkins wrote: >>>>> See <http://vm-058-081.abc.idm.lab.eng.brq.redhat.com:8080/job/389-DS-NIGHTLY/165/> >>>>> >>>> Is there a reason why jenkins failures are ignored for a week? >>> With our VERY limited resources, and other higher priority issues going >>> on recently (like migrating to Pagure), no one has time to look into >>> these upstream failures (most of which are probably false positives due >>> to recent changes in lib389). We plan to look into them soon. >>> >>> You are also welcome to investigate some these failures if you would like :) >> And I've been buried in other tasks too. Normally it's Mark or I who >> investigate these. >> >> There is one known test case failure that I've been working to resolve >> which is this one: >> >> https://pagure.io/389-ds-base/issue/49086 >> >> Which if you look is one of the issues there in the error output: >> >> def test_ticket48272(topology_st): >> >> So I am working to fix this, but I want to fix it right, not fast. It's >> a delicate area of code. >> >> >> def test_ticket48906_dblock_default(topology_st): >> >> That test case is stupid, and my autotuning will break it. It needs to >> be cleaned out. >> >> def test_basic_dse(topology_st, import_example_ldif): >> >> No idea why this one is failing at the moment, haven't started to look >> at it. >> > I know that everyone is busy. > > But in our team we have a policy that new patches should not be pushed > if tests are not green. Your tests are obviously more stable than ours. In fact our "nightly tests" are a fairly new process. We were even hesitant to start doing this due to the inconsistency of some of the tests. Like I said before, our testing framework has also underwent some heavy construction recently. Anyway, we will try and do better. For now I will remove the reports from the public mailing list so you, and others, are not spammed with the daily failures. Once we get this sorted out I will restore those settings. Regards, Mark > > It's much better to disable flaky test rather then ignore failures > for a week > > my 2 cents > > LS > _______________________________________________ > 389-devel mailing list -- 389-devel@xxxxxxxxxxxxxxxxxxxxxxx > To unsubscribe send an email to 389-devel-leave@xxxxxxxxxxxxxxxxxxxxxxx _______________________________________________ 389-devel mailing list -- 389-devel@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to 389-devel-leave@xxxxxxxxxxxxxxxxxxxxxxx
