Loopback interface inside 'vrouter' namespace is down after deploy on non-primary controllers
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Fuel for OpenStack |
Fix Released
|
High
|
Oleksiy Molchanov |
Bug Description
Scenario:
- Create HA cluster (Ubuntu/
- add 3 controllers + ceph-osd
- add 2 compute + ceph-osd
I guess ceph-osd roles are not necessary in this case, but I haven't tested another scenario yet.
- deploy cluster
Actual results: 'lo' interfaces are down inside 'vrouter' namespace on node-2 and node-3.
Cluster roles: http://
Interfaces in 'vrouter' namespace on primary controller: http://
Interfaces in 'vrouter' namespace on other controllers: http://
Services that was started inside the 'vrouter' namespace (for example, dnsmasq and ntpd) are inaccessible from the router:
root@node-3:~# ip netns exec vrouter ntpq -pn
127.0.0.1: timed out, nothing received
***Request timed out
description: | updated |
Changed in fuel: | |
status: | New → Triaged |
importance: | Undecided → High |
assignee: | Fuel Library Team (fuel-library) → Oleksiy Molchanov (omolchanov) |
Changed in fuel: | |
status: | Triaged → In Progress |
tags: | added: on-verification |
After reboot the controller (node-3) , 'lo' interface started successfully:
http:// paste.openstack .org/show/ 199369/