[Gluster-devel] Regression test for 821056

[Vijay Bellur]

On 07/27/2013 09:33 AM, Vijay Bellur wrote:
> On 07/27/2013 08:50 AM, Anand Avati wrote:
>> The same test seems to finish in ~ 12s on my laptop:
>>
>> # prove --timer ./tests/bugs/bug-821056.t
>> [11:34:17] ./tests/bugs/bug-821056.t .. ok    12264 ms
>
> Well, it seems to be stemming from patch 5376 as per jenkins:
>
> http://build.gluster.org/job/regression/1769/console
>
> I consistently get this result on my laptop:
>
> [root at deepthought glusterfs.git]# prove --timer ./tests/bugs/bug-821056.t
> [09:10:58] ./tests/bugs/bug-821056.t .. ok  1220937 ms
> [09:31:19]
> All tests successful.
> Files=1, Tests=22, 1222 wallclock secs ( 0.05 usr  0.00 sys +  0.92 cusr
>   0.67 csys =  1.64 CPU)
> Result: PASS
>
> A lot of the recent regression test runs in jenkins does show
> bug-821056.t to be taking around 1040+ seconds.
>

After http://review.gluster.org/5406 got merged, the test takes < 60 
seconds now on my laptop.

[root at deepthought glusterfs.git]# prove --timer ./tests/bugs/bug-821056.t
[23:14:17] ./tests/bugs/bug-821056.t .. ok    42771 ms
[23:14:59]
All tests successful.
Files=1, Tests=22, 42 wallclock secs ( 0.03 usr  0.01 sys +  1.10 cusr 
0.64 csys =  1.78 CPU)
Result: PASS

-Vijay