Pushed a fix for the StateCheckpointedITCase On Mon, Aug 24, 2015 at 12:19 PM, Maximilian Michels <m...@apache.org> wrote:
> +1 for labeling the JIRAs with "test-stability". > > On Sat, Aug 22, 2015 at 8:21 PM, Márton Balassi <balassi.mar...@gmail.com> > wrote: > > > +1 for Vasia's suggestion > > On Aug 22, 2015 8:07 PM, "Vasiliki Kalavri" <vasilikikala...@gmail.com> > > wrote: > > > > > I just came across 2 more :/ > > > I'm also in favor of tracking these with JIRA. How about > "test-stability" > > > for a label? > > > > > > -V. > > > > > > On 21 August 2015 at 12:47, Matthias J. Sax < > > mj...@informatik.hu-berlin.de > > > > > > > wrote: > > > > > > > I like the idea with the special label. Otherwise, it will be > difficult > > > > to find the correct tickets. > > > > > > > > -Matthias > > > > > > > > On 08/21/2015 12:15 PM, Till Rohrmann wrote: > > > > > I'm also in favor of JIRA, because I fear that nobody will keep the > > > wiki > > > > > page in sync. Maybe we can assign a special label for test > stability > > to > > > > > these JIRA issues. Then we can quickly find all currently instable > > test > > > > > cases. > > > > > > > > > > On Fri, Aug 21, 2015 at 11:02 AM, Robert Metzger < > > rmetz...@apache.org> > > > > > wrote: > > > > > > > > > >> I agree that we should look for a solution other than opening a > lot > > of > > > > >> small discussion threads on the mailing list. > > > > >> > > > > >> When I have a test failure, I usually search my gmail inbox to see > > > > whether > > > > >> somebody else wrote something about the error already. > > > > >> Creating a JIRA for each failing test might be a better approach. > > > > Because > > > > >> that's what bugtrackers are made for ;) (And the issues still pop > up > > > > when > > > > >> doing a gmail search) > > > > >> > > > > >> On Thu, Aug 20, 2015 at 10:16 PM, Matthias J. Sax < > > > > >> mj...@informatik.hu-berlin.de> wrote: > > > > >> > > > > >>> Thanks for the info. > > > > >>> > > > > >>> Over the weeks I lost track which errors/failing/instable tests > are > > > > know > > > > >>> an which not. Should we start a wiki page or similar to collect > > know > > > > >>> errors? If a test fails on a know error, it can just be ignored. > > This > > > > >>> would avoid "spam" on the mailing list. > > > > >>> > > > > >>> Any thoughts about this? > > > > >>> > > > > >>> -Matthias > > > > >>> > > > > >>> On 08/20/2015 10:08 PM, Robert Metzger wrote: > > > > >>>> Sachin saw the error as well, as reported here: > > > > >>>> https://issues.apache.org/jira/browse/FLINK-2468 > > > > >>>> I also see it from time to time.I have a wip branch where I > > relaxed > > > > the > > > > >>>> constraints for the test to pass a bit. > > > > >>>> > > > > >>>> On Thu, Aug 20, 2015 at 10:05 PM, Matthias J. Sax < > > > > >>>> mj...@informatik.hu-berlin.de> wrote: > > > > >>>> > > > > >>>>> Error message is: > > > > >>>>> > > > > >>>>>> Failed tests: > > > > >>>>>> > > > > >>>>> > > > > >>> > > > > >> > > > > > > > > > > StateCheckpoinedITCase>StreamFaultToleranceTestBase.runCheckpointedProgram:103->postSubmit:98 > > > > >>>>> Test inconclusive: failure occurred before first checkpoint > > > > >>>>> > > > > >>>>> See: https://travis-ci.org/mjsax/flink/jobs/76483093 > > > > >>>>> > > > > >>>>> > > > > >>>>> -Matthias > > > > >>>>> > > > > >>>>> > > > > >>>> > > > > >>> > > > > >>> > > > > >> > > > > > > > > > > > > > > > > > > >
