Comment 14 for bug 1786346
Revision history for this message
| Sahid Orentino (sahid-ferdjaoui) wrote | #14 |
© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
73416ba
(Get the code!)
The bug is reported on Pike, why that has been reverted from master without to have clearly identified the root cause? Crying everywhere it's "racy" without any explanation does not really make sense, sorry.
How the patch on compute manager is going to address the issue? It's not.
The specific work which has been done on libvirt driver for linux bridge to slow the migration progress is basically because the TAP device is created by libvirt itself when calling migrateToURI which the neutron agent is going to discover and then correctly setup L2 network. That was basically the whole purpose :/
The issue has been noticed on Pike, it would have been interesting to get more details, basically the neutron logs, since we all know that this patch came with some fix in Neutron, perhaps one of them has been badly backported. I mean at least a bit more of investigations would have been appreciated.