Comment 1 for bug 1999671
Revision history for this message
| Alex Kavanagh (ajkavanagh) wrote Re: [jammy][yoga] Tempest times out reaching neutron-api | #1 |
© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
a60fb26
(Get the code!)
I'm not able to triage just yet, but I have a feeling that this might be a bug related to synchronisation between the OVN and neutron backends. e.g. OVN maintains its own (internal) database of what's connected and this is synced to the neutron database (on mysql). It may be that the syncing has missed out or been delayed (e.g. the failed port look-up was about a minute behind the port being assigned). So this could also be resource contention if the neutron servers (or the machines that they are being run on) are heavily loaded; this is fairly typical in the SQA lab.
If the above is true, then it's more a neutron/OVN bug than a charm bug, and we may need to reallocate it. In order to discount/rule-out the resource contention (and thus a race-hazard type issue), it would be good to capture the load stats for the machine in question - which may not currently be recorded.