[Openstack] [Tempest] Holding run_cleanup for failed scenario - facilitating troubleshooting

N Vivekanandan n.vivekanandan at ericsson.com
Fri Jun 30 14:56:49 UTC 2017


Hi Tempest Team,

We were trying to troubleshoot several tempest failures in an
Openstack - Opendaylight based deployment.

Almost all the tempest scenarios tests clean up the
resources created in such tests at the end of running each
such scenario.

We were contemplating if there is a way to have a tempest scenario log
flow information from hypervisor datapaths before the scenario cleanup
kicks in (more specifically when scenarios fail) .

Is there a way to configure tempest scenarios to avoid cleanup of created resources
as part of a single scenario run?

We can have a debugger run with tempest and stop at the required point to dump information,
but not useful enough, since sometimes we won't apriori know which scenarios will fail.

--
Thanks,

Vivek
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack/attachments/20170630/3352448b/attachment.html>


More information about the Openstack mailing list