4:55 a.m.
--+9uAEI//NVSmpZEF
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
On 03/26, Michal Skrivanek wrote:
=20
=20
On 25 Mar 2015, at 13:58, David Caro <dcaroest(a)redhat.com> wrote:
=20
>=20
> This was done because it was requested some time ago.
>=20
> That will also remind devels that they should not merge any patches tha=
t
have
> not finished running the jenkins jobs yet.
>=20
> In a (hopefully) near future, that will not be needed as jobs will beco=
me a
> gateway to merge, meaning that the regular merge flow will
require pass=
ing the
> jenkins jobs.
=20
I would strongly object to this, until we have a positive experience of n=
o false
positives for at least couple of months.
Then it is a great idea.=20
=20
That's like expecting that every patch sent for review to pass all the
test=
s,
all the time for a month and not failing once, and I don't mean each MERGED
patch, I mean each one sent for review.
Just so you have some numbers on what you are currently asking for, some st=
ats
about builds for the last two weeks:
Number of SUCCESS builds: 50061
Number of FAILED builds: 1777
Number of UNSTABLE builds: 749
Number of ABORTED builds: 124
Number of NOTBUILT builds: 326
Total number of builds: 53037
Percentage of failed builds: 5.61%
Here, failed builds includes infra issues and test failures, we don't have
means yet to properly distinguish both.
I know this is including all the projects, not just the most 'complicated'
ones, but that means that, on small projects, where the maintainer of the
project is also writing it's own acceptance tests, the failure ratio is a l=
ot
lower than on the bigger projects, where the failures are ignored and the t=
ests
not maintained, or left for the lonely infra guy that has ~50% of it's time=
to
handle them all, doing a quick ugly calculation, that guy has 48 seconds to
review each failure and fix it (being it an infra issue or not).
So you expect that someone that has no involvement on the development of mo=
st
of the projects, single-handedly fix and triage all those builds, 48 seconds
each, and also care about all the other infra-related duties. And not only
succeed doing that, but maintain all the system from failing once during two
months.
I'm sorry but as I see it a lot has to change for that to be feasible, and I
really don't think it's fair to request it to that 1/2 guy.
=20
> In the meantime, it's just a remainder.
>=20
> On 03/25, Martin Perina wrote:
>>=20
>>=20
>> ----- Original Message -----
>>> From: "Greg Sheremeta" <gshereme(a)redhat.com>
>>> To: infra(a)ovirt.org, "Eyal Edri" <eedri(a)redhat.com>,
"David Caro" <dc=
aroest(a)redhat.com>
>>> Cc: devel(a)ovirt.org
>>> Sent: Wednesday, March 25, 2015 1:40:35 PM
>>> Subject: [ovirt-devel] gerrit jenkins bot has gotten noisier
>>>=20
>>> Hey guys,
>>>=20
>>> Sometime last week, the jenkins bot upped the verbosity of his
>>> notifications on patches. Was this by design?
>>>=20
>>> Example: https://gerrit.ovirt.org/#/c/39098/
>>>=20
>>> He now adds many "build started" messages. He used to only let
>>> us know about success / failure.
>>>=20
>>> Not a huge deal, but it's more noise to filter through -- both
>>> in gerrit and email.
>>=20
>> Agreed, especially for emails. Personally I don't care about build
>> process details, I'd like to know only final result of the build.
>> If the build failed, I could find details on jenkins.
>>=20
>> Martin
>>=20
>>>=20
>>> Thanks,
>>> Greg
>>>=20
>>> Greg Sheremeta
>>> Red Hat, Inc.
>>> Sr. Software Engineer, RHEV
>>> Cell: 919-807-1086
>>> gshereme(a)redhat.com
>>> _______________________________________________
>>> Devel mailing list
>>> Devel(a)ovirt.org
>>> http://lists.ovirt.org/mailman/listinfo/devel
>=20
> --=20
> David Caro
>=20
> Red Hat S.L.
> Continuous Integration Engineer - EMEA ENG Virtualization R&D
>=20
> Tel.: +420 532 294 605
> Email: dcaro(a)redhat.com
> Web: www.redhat.com
> RHT Global #: 82-62605
> _______________________________________________
> Devel mailing list
> Devel(a)ovirt.org
> http://lists.ovirt.org/mailman/listinfo/devel
--=20
David Caro
Red Hat S.L.
Continuous Integration Engineer - EMEA ENG Virtualization R&D
Tel.: +420 532 294 605
Email: dcaro(a)redhat.com
Web: www.redhat.com
RHT Global #: 82-62605
--+9uAEI//NVSmpZEF
Content-Type: application/pgp-signature
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQEbBAEBAgAGBQJVE9eDAAoJEEBxx+HSYmnD/vsH91dhVr+o3T/wIgABxXhkTu1c
DqTzR/qnMe4pddgvj82TiVc7pFc3FbnWZFkreetpkkXAGhxI8NdNDTsRNc7GPcnI
qc+iIw3qpOz95fdJtjcRsYe/d2l8DdZNxKWHeubetkEDDc8+MYOhrtRlSQQxfP2f
zVZPQL4ONphqove57TlyQQHDNJVuNnKHf+mR7lG0LIjhI9/3eHispQJPWhpRA4vj
FF1BswMpfR5arOGB62iCqsP6Pamt7cRoONkDx4abaDi8aPAQd5C8M47xZeoSAMkk
Kxaj9R5E/GIplXV1IboSA0mxFsJJEodEdkeDDxZBk5Ne7rBAYgNpXIT8kh1bLQ==
=PP0i
-----END PGP SIGNATURE-----
--+9uAEI//NVSmpZEF--