[Gluster-devel] netbsd regression update : cdc.t

[Atin Mukherjee]

On 05/02/2015 08:52 PM, Emmanuel Dreyfus wrote:
> Atin Mukherjee <amukherj at redhat.com> wrote:
> 
>> Although I couldn't reproduce cdc.t failure but georep-setup.t failed
>> consistently and glusterd backtrace showed that it hangs on gverify.sh
> 
> That suggests the script itself blocks forever. Runnig with -x may be
> insightful.
> 
>> If you happen to see cdc.t failure again please ring a bell :)
> 
> It is 100% reproductible on nbslave78
> 
> nbslave78# cd /autobuild//glusterfs/
> nbslave78# ./run-tests.sh  -f ./tests/basic/cdc.t
In nbslave78, test case gets stuck at volume stop's
gd_syncop_mgmt_brick_op () with the following bt:

(gdb) t a a bt

Thread 6 (LWP 2):
#0  0xbb35d7d7 in _sys___nanosleep50 () from /usr/lib/libc.so.12
#1  0xbb688aa7 in __nanosleep50 () from /usr/lib/libpthread.so.1
#2  0xbb75f422 in gf_timer_proc () from
/autobuild/install/lib/libglusterfs.so.0
#3  0xbb68cbca in ?? () from /usr/lib/libpthread.so.1
#4  0xbb3acbb0 in __mknod50 () from /usr/lib/libc.so.12
#5  0xbb192000 in ?? ()
Backtrace stopped: previous frame identical to this frame (corrupt stack?)

Thread 5 (LWP 3):
#0  0xbb3ac8b7 in ____sigtimedwait50 () from /usr/lib/libc.so.12

Thread 4 (LWP 4):
#0  0xbb3c84e7 in ___lwp_park60 () from /usr/lib/libc.so.12

Thread 3 (LWP 5):
#0  0xbb3c84e7 in ___lwp_park60 () from /usr/lib/libc.so.12

Thread 2 (LWP 6):
#0  0xbb3c84e7 in ___lwp_park60 () from /usr/lib/libc.so.12

Thread 1 (LWP 1):
#0  0xbb3c84e7 in ___lwp_park60 () from /usr/lib/libc.so.12
(gdb) t 3
[Switching to thread 3 (LWP 5)]
#0  0xbb3c84e7 in ___lwp_park60 () from /usr/lib/libc.so.12
(gdb) bt
#0  0xbb3c84e7 in ___lwp_park60 () from /usr/lib/libc.so.12

Surprisingly, If I take a different commit as head in the same vm
(/home/jenkins/root/workspace/rackspace-netbsd7-regression-triggered)
its not reproduced. So my initial suspect is the delta between these two
heads.

Ccing Joseph to cross check as I see some of his patches in the delta.
> 
> ... GlusterFS Test Framework ...
> 
> 
> The following required tools are missing:
> 
>   * dbench
> 
> Running tests in file ./tests/basic/cdc.t
> [15:17:13] ./tests/basic/cdc.t .. 52/55 
> not ok 52 
> not ok 53 Got "Started" instead of "Stopped"
> volume delete: patchy: failed: Another transaction is in progress for
> patchy. Please try again after sometime.
> not ok 54 
> [15:17:13] ./tests/basic/cdc.t .. 55/55 not ok 55 
> [15:17:13] ./tests/basic/cdc.t .. Failed 4/55 subtests 
> [15:19:50]
> 
> Test Summary Report
> -------------------
> ./tests/basic/cdc.t (Wstat: 0 Tests: 55 Failed: 4)
>   Failed tests:  52-55
> Files=1, Tests=55, 157 wallclock secs ( 0.03 usr  0.05 sys +  4.01 cusr
> 11.98 csys = 16.07 CPU)
> Result: FAIL
> Failed tests ./tests/basic/cdc.t
> 

-- 
~Atin