[Gluster-Maintainers] Build failed in Jenkins: centos8-regression #303

[Pranith Kumar Karampuri]

On Fri, May 7, 2021 at 7:04 PM Pranith Kumar Karampuri <
pranith.karampuri at phonepe.com> wrote:

>
> On Fri, May 7, 2021 at 6:32 PM Ravishankar Narayanankutty <
> ranaraya at redhat.com> wrote:
>
>>
>> On Fri, May 7, 2021 at 10:35 AM Deepshikha Khandelwal <
>> dkhandel at redhat.com> wrote:
>>
>>> Can someone please take a look at the failing test ./tests/basic/quick-read-with-upcall.t
>>> ?
>>>
>>> On Fri, May 7, 2021 at 12:09 AM <jenkins at build.gluster.org> wrote:
>>>
>>>> See <
>>>> https://build.gluster.org/job/centos8-regression/303/display/redirect>
>>>>
>>>> 1 test(s) failed
>>>> ./tests/basic/quick-read-with-upcall.t
>>>>
>>>
>>
>>
>> According to
>> https://build.gluster.org/job/centos8-regression/303/consoleFull, the
>> test failure was:
>>
>> *21:31:39* not ok  13 [     10/      5] <  38> 'test-message0 cat /mnt/glusterfs/1/test.txt' -> 'Got "test-message1" instead of "test-message0"'
>>
>>
>> It seems https://github.com/gluster/glusterfs/issues/1190 was sent for the same issue. Pranith, in addition to enabling direct-io-mode, do we also need to set attribute-timeout and entry-timeout to zero for the mounts?
>>
>>
>
As per my understanding no. direct-io-mode will lead to a read call in the
mount and the data wouldn't be served off the cache.


>
>
>>
>> Regards,
>>
>> Ravi
>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.gluster.org/pipermail/maintainers/attachments/20210507/97fab591/attachment.html>