yahoo-eng-team team mailing list archive

[Oleg Bondarev <obondarev@xxxxxxxxxxxx>]

Public bug reported:

Currently If 'dvr' l3 agent on a compute node goes down, routers are
removed from it; then when agent is up again routers are not added back
- so VMs on that compute node lose all routing.

Scheduling/unscheduling of DVR routers to the l3 agent in 'dvr' mode
running on a compute node is done according to dvr serviced ports
created/deleted on that compute node. It doesn't make sense to
reschedule router from l3 agent on compute node even if it's down - no
other l3 agent can handle VMs running on that compute node.

** Affects: neutron
     Importance: Undecided
     Assignee: Oleg Bondarev (obondarev)
         Status: New


** Tags: l3-dvr-backlog

** Description changed:

  Currently If 'dvr' l3 agent on a compute node goes down, routers are
  removed from it; then when agent is up again routers are not added back
  - so VMs on that compute node lose all routing.
  
- Scheduling/unscheduling of DVR routers to the l3 agent in 'dvr' mode running on a compute node is done according
- to dvr serviced ports created/deleted on that compute node. It doesn't make sense to reschedule router from 
- l3 agent on compute node even if it's down - no other l3 agent can handle VMs running on that compute node.
+ Scheduling/unscheduling of DVR routers to the l3 agent in 'dvr' mode
+ running on a compute node is done according to dvr serviced ports
+ created/deleted on that compute node. It doesn't make sense to
+ reschedule router from l3 agent on compute node even if it's down - no
+ other l3 agent can handle VMs running on that compute node.

-- 
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to neutron.
https://bugs.launchpad.net/bugs/1470889

Title:
  DVR: should not remove routers form down l3 agents on compute nodes

Status in OpenStack Neutron (virtual network service):
  New

Bug description:
  Currently If 'dvr' l3 agent on a compute node goes down, routers are
  removed from it; then when agent is up again routers are not added
  back - so VMs on that compute node lose all routing.

  Scheduling/unscheduling of DVR routers to the l3 agent in 'dvr' mode
  running on a compute node is done according to dvr serviced ports
  created/deleted on that compute node. It doesn't make sense to
  reschedule router from l3 agent on compute node even if it's down - no
  other l3 agent can handle VMs running on that compute node.

To manage notifications about this bug go to:
https://bugs.launchpad.net/neutron/+bug/1470889/+subscriptions