[Gluster-devel] Regression failure due to bad test case

[Saravanakumar Arumugam]

Hi Niels,
Thanks!

As per my understanding, core file generated due to bad test case.

<<<<<<<<<<<<<<<<<
Test Summary Report
-------------------
./tests/bugs/glusterd/bug-948686.t (Wstat: 0 Tests: 21 Failed: 2)
   Failed tests:  14-15
Files=1, Tests=21, 33 wallclock secs ( 0.02 usr  0.00 sys +  2.15 cusr  
0.68 csys =  2.85 CPU)
Result: FAIL
./tests/bugs/glusterd/bug-948686.t: bad status 1
./tests/bugs/glusterd/bug-948686.t: 1 new core files
Ignoring failure from known-bad test ./tests/bugs/glusterd/bug-948686.t
<<<<<<<<<<<<<<<<<

I think we should ignore this core file as it is due to bad test ?

The main issue here is while copying logs.
Again If I understand correctly, logs are generated due to core files.

So, ignoring core files generated due to bad test case is best choice here??

Thanks
Saravana


On 10/27/2015 03:10 PM, Niels de Vos wrote:
> On Tue, Oct 27, 2015 at 02:31:44PM +0530, Saravanakumar Arumugam wrote:
>> Hi,
>> Thanks Atin! You are correct.
>> <<<<<<<<<<
>> + '[' 1 '!=' 0 ']'
>> + set +x
>> Going to copy log tarball for processing on http://elk.cloud.gluster.org/
>> sort: write failed: standard output: Broken pipe
>> sort: write error
>> Build step 'Execute shell' marked build as failure
>> Finished: FAILURE
>> <<<<<<<<<<
>>
>> I remember facing similar issue earlier.
> Yes, and this is not an error that causes a regression test to be marked
> as failed.
>
> This run dumped a core at one point, that is what needs investigation:
>
>      [06:23:24] ./tests/performance/quick-read.t .. ok     8075 ms
>      [06:23:24]
>      All tests successful.
>      Files=1, Tests=20,  8 wallclock secs ( 0.02 usr  0.00 sys +  1.13 cusr  0.38 csys =  1.53 CPU)
>      Result: PASS
>      + RET=0
>      ++ wc -l
>      ++ ls -l /core.32471
>      + cur_count=1
>      + '[' 1 '!=' 0 ']'
>      + mkdir -p /build/install/cores
>      + mv /core.32471 /build/install/cores
>      + filename=archived_builds/build-install-20151027:03:29:01.tar
>      + rm -f /build/install/cores/liblist.txt
>      + rm -f /build/install/cores/liblist.txt.tmp
>      ++ ls /build/install/cores/core.32471
>      + CORELIST=/build/install/cores/core.32471
>
>
> Niels
>
>> Any idea how to overcome these types of failures apart from retriggering
>> regression :) ?
>>
>> Thanks,
>> Saravana
>>
>>
>> On 10/27/2015 02:13 PM, Atin Mukherjee wrote:
>>> On 10/27/2015 12:33 PM, Saravanakumar Arumugam wrote:
>>>> Hi,
>>>> I am facing regression failure http://review.gluster.org/#/c/12239
>>>>
>>>> https://build.gluster.org/job/rackspace-regression-2GB-triggered/15166/consoleFull
>>>>
>>>>
>>>> But both the test cases are marked bad??
>>>>
>>>> ./tests/bugs/glusterd/bug-948686.t
>>>>   ./tests/bugs/glusterd/bug-1238706-daemons-stop-on-peer-cleanup.t
>>>>
>>>> It should be success as both the test cases are marked as bad ??
>>> I don't think the failure was due to a bad test not passing regression.
>>> All the tests passed apart from bad test(s). Copying the log tarball to
>>> a location actually failed here. See at the bottom of the page.
>>>
>>> ~Atin
>>>> Please refer the log for details.
>>>>
>>>> Thanks,
>>>> Saravana
>>>>
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> Gluster-devel mailing list
>>>> Gluster-devel at gluster.org
>>>> http://www.gluster.org/mailman/listinfo/gluster-devel
>> _______________________________________________
>> Gluster-devel mailing list
>> Gluster-devel at gluster.org
>> http://www.gluster.org/mailman/listinfo/gluster-devel