Comment 1 for bug 1859766
Revision history for this message
| Balazs Gibizer (balazs-gibizer) wrote | #1 |
© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
a60fb26
(Get the code!)
I suspect that the problem is twofold. First we have a functional test change[1] in train and master that was not backported to older branches. This change removes a global state hack from the fake virt driver. Second we most probably backported tests from master (and train) to stain and rocky that does not properly set up the global state as on master it is not need, but it is needed on stain and back.
At the moment I didn't find what is the offending backport. But from the failed test log I see that a test case starts up a compute service on host1[2] and then later the nova code tries to access a compute service on host2 [3] that was never started.
[1] https:/
[2] https:/
[3] https:/