On Wed, Apr 12, 2017 at 3:36 PM, Michal Skrivanek
<michal.skrivanek(a)redhat.com> wrote:
On 12 Apr 2017, at 13:39, Michal Skrivanek <michal.skrivanek(a)redhat.com>
wrote:
On 12 Apr 2017, at 12:10, Barak Korren <bkorren(a)redhat.com> wrote:
On 12 April 2017 at 12:44, Michal Skrivanek <michal.skrivanek(a)redhat.com>
wrote:
>
>
> Well, Im not sure who’s handling that exactly, but vdsm-jsonrpc-java is
> getting into OST runs despite the OST failing since Saturday. That should
> not happen
There is no mechanism for blocking stuff from going into 'experimental' ATM
(Please differentiate between the 'experimental' flow that automates the
testing and the OST which is the set of testing tools that can be used in
various configurations).
'experimental' generally runs on the latest builds of all packages.
We're working on something that will allow blocking failed changes, but it
still some ways off (OVIRT-1077).
> btw, in the meantime, me and Tomas are working on narrowing down the
> original failure.
Is there a quick fix that could be put in place to get 'experimental' back
to work for everything else?
we finally managed to narrow it down and confirm the set of patches causing
the problem on Saturday (had to work around the jsonrpc-java and some other
stuff)
engine artifacts[1] passed manual OST[2] now (and sorry Ido for aborting
your job)
patches should be taken in shortly
so, they are all in, OST run, but failed on the jsonrpc-java package
Can you please manually replace it with the last working version:
http://resources.ovirt.org/repos/ovirt/experimental/master/latest.tested/...