Am Di., 19. Nov. 2019 um 14:54 Uhr schrieb Gris Ge <fge(a)redhat.com>:
On Tue, Nov 19, 2019 at 02:23:24PM +0100, Till Maas wrote:
> there is a fixture to restore the Nmstate state after test runs. This
It was supposed to restore my network config after I ran the
integration test directly(not in container).
> happens unconditionally for all interfaces and causes failures when
> Nmstate does not actually support setting the reported state. I saw
> this so far with podman when there is a tap interface in the container
> that made it choke. Also it fails on other systems:
The desire state you mentioned try to set static IPv6 gateway with
DHCPv6 and Autoconf enabled which it not supported by nmstate yet.
> Do we really need it like this? Can we at least limit it to only eth1
> and eth2? AFAIU it will not remove any left-over profiles from the
> tests anyhow so eth1 and eth2 would be the only interfaces that are
> affected by this. But AFAIU, the tests fixtures clean up eth1/eth2
> also all the time already.
We need if we want to support running outside of docker.
Once test finished, we should restore the OS's network connection, so
that test report could report back to auto test system.
Currently this fixture is actually breaking running tests outside of
docker. IMHO as long as Nmstate cannot always support nmstate show |
nmstate set it is more appropriate to only support test systems that
do not need/use eth1/eth2. Or in other words: To require that they can
report back their state with other interfaces than eth1/eth2. Also if
necessary, the test system could use a custom wrapper that will
configure eth1/eth2 as required after the test exited. Therefore I
would propose to remove the fixture for now. What do you think?
Ansible RHEL Networking System Role Maintainer
Red Hat GmbH, https://www.redhat.com/de
, Registered seat: Grasbrunn,
Commercial register: Amtsgericht Muenchen, HRB 153243,
Managing Directors: Charles Cachera, Michael O'Neill, Tom Savage, Eric Shander