neutron should rely on nova doing the job instead of trying to 'fix' it.
'Fixing' it introduces race conditions between lb agent and nova VIF
driver. Particularly, lb agent can scan for new tap devices in the
middle of nova plumbing qbr-tap setup, and attempt to do it on its own.
So if agent is more lucky to plug the tap device into the bridge, nova
may fail to do the same, getting the following error:
libvirtError: Unable to add bridge brqxxx-xx port tapxxx-xx: Device or
resource busy
This also requires a change in how the port admin_state_up is implemented
by setting the tap device's link state instead of moving it in or out
of the bridge.
Conflicts:
neutron/common/constants.py
neutron/plugins/ml2/drivers/linuxbridge/agent/linuxbridge_neutron_agent.py
neutron/tests/unit/plugins/ml2/drivers/linuxbridge/agent/test_linuxbridge_neutron_agent.py
Co-Authored-By: Sean M. Collins <sean@coreitpro.com>
Co-Authored-By: Darragh O'Reilly <darragh.oreilly@hp.com>
Co-Authored-By: Andreas Scheuring <andreas.scheuring@de.ibm.com>
Closes-Bug: #1312016
(cherry picked from commit f42ea67995)
===
Also squashed in the following follow up fix:
lb: Correct String formatting to get rid of logged ValueError
The following error is caused by a missing String formatting in the
linuxbridge agent:
"ValueError: unsupported format character 'a' (0x61) at index 90
Logged from file linuxbridge_neutron_agent.py, line 447"
In addition a duplicated word in the log text has been fixed.
Change-Id: I587f1165fc7084dc9c4806149b65652f6e27b14e
(cherry picked from commit 1f86d8687b)
===
Also squashed in:
Only ensure admin state on ports that exist
The linux bridge agent was calling ensure_port_admin state
unconditionally on ports in treat_devices_added_or_updated.
This would cause it to throw an error on interfaces that
didn't exist so it would restart the entire processing loop.
If another port was being updated in the same loop before this
one, that port would experience a port status life-cycle of
DOWN->BUILD->ACTIVE->BUILD->ACTIVE
^ <--- Exception in unrelated port causes cycle
to start over again.
This causes the bug below because the first active transition will
cause Nova to boot the VM. At this point tempest tests expect the
ports that belong to the VM to be in the ACTIVE state so it filters
Neutron port list calls with "status=ACTIVE". Therefore tempest would
not get any ports back and assume there was some kind of error with
the port and bail.
This patch just makes sure the admin state call is skipped if the port
doesn't exist and it includes a basic unit test to prevent a regression.
Conflicts:
neutron/plugins/ml2/drivers/linuxbridge/agent/linuxbridge_neutron_agent.py
Closes-Bug: #1523638
Change-Id: I5330c6111cbb20bf45aec9ade7e30d34e8dd16ca
(cherry picked from commit 96c67e22f9)
===
Change-Id: I02971103407b4ec11a65218e9ef7e2708915d938
4cb9062319