On Tue, Jul 6, 2021 at 12:44 PM Marcin Sobczyk <msobczyk(a)redhat.com> wrote:
On 7/6/21 11:32 AM, Yedidyah Bar David wrote:
> Hi all,
>
> Recently we had several cases of OST runs that seemed to have
> successfully started their VMs, but ansible later failed to connect to
> them in the "wait_for_connection" done at start of deployment. This
> happened to both he-basic and basic, e.g. [1][2][3]. Some do pass,
> e.g. [4]. This looks like e.g. (from [1]):
>
> 20:29:58 [36m # Start vms: [32mSuccess [0m (in 0:00:02) [0m
> ...
> 20:30:08 he-basic-suite-master/test-scenarios/test_001_he_deploy.py::test_he_deploy
> started at 2021-07-05 17:30:08
> 20:30:08 -------------------------------- live log setup
> --------------------------------
> 20:30:09 2021-07-05 17:30:09+0000,180 INFO
> [ost_utils.pytest.fixtures.deployment] Waiting for SSH on the VMs
> (deployment:60)
> 20:30:09 ERROR (122s)Traceback (most recent call last):
> ...
> 20:32:15 "msg": "Failed to connect to the host via ssh: ssh:
> connect to host 192.168.200.3 port 22: No route to host",
>
> Any idea?
Yes, please see [5]. We have working images since Sunday I think,
but we need to wait until mock cache invalidates on all CI agents and
picks up the new images.
Ok, thanks.
I now pushed this trivial long-overdue patch, to allow us know which
ost-images are used:
https://gerrit.ovirt.org/c/ovirt-system-tests/+/115577
Best regards,
--
Didi