[lxc-users] Nova LXD - Timeout waiting for vif plugging callback

[Joshua Schaeffer]

I'm trying to get OpenStack setup with nova-compute-lxd and having problems with the networking portions. Wondering if anybody has run into this problem before. All my servers run Ubuntu 16.04. On my compute node I have nova-compute-lxd 15.0.2 installed and lxd version 3.0.1 installed (not sure this one matters):

    root at bllcloudcmp02:~# dpkg -l | grep lxd
    ii  lxd                               3.0.1-0ubuntu1~16.04.4                     amd64        Container hypervisor based on LXC - daemon
    ii  lxd-client                        3.0.1-0ubuntu1~16.04.4                     amd64        Container hypervisor based on LXC - client
    ii  nova-compute-lxd                  15.0.2-0ubuntu1~cloud0                     all          Openstack Compute - LXD container hypervisor support
    ii  python-nova-lxd                   15.0.2-0ubuntu1~cloud0                     all          OpenStack Compute Python libraries - LXD driver
    ii  python-pylxd                      2.2.4-0ubuntu0.17.04.1~cloud0              all          Python library for interacting with LXD REST API

Everything from OpenStack appears to be operational:

    root at bllcloudctl02:~# openstack compute service list
    +----+------------------+---------------+----------+---------+-------+----------------------------+
    | ID | Binary           | Host          | Zone     | Status  | State | Updated At                 |
    +----+------------------+---------------+----------+---------+-------+----------------------------+
    |  3 | nova-conductor   | bllcloudapi02 | internal | enabled | up    | 2018-08-14T00:52:31.000000 |
    |  5 | nova-scheduler   | bllcloudapi02 | internal | enabled | up    | 2018-08-14T00:52:32.000000 |
    |  6 | nova-consoleauth | bllcloudapi02 | internal | enabled | up    | 2018-08-14T00:52:34.000000 |
    |  7 | nova-compute     | bllcloudcmp02 | nova     | enabled | up    | 2018-08-14T00:52:31.000000 |
    +----+------------------+---------------+----------+---------+-------+----------------------------+
    root at bllcloudctl02:~# openstack network agent list
    +--------------------------------------+--------------------+---------------+-------------------+-------+-------+---------------------------+
    | ID                                   | Agent Type         | Host          | Availability Zone | Alive | State | Binary                    |
    +--------------------------------------+--------------------+---------------+-------------------+-------+-------+---------------------------+
    | 2e0e53b1-7c99-4acb-ac69-79c3bd9aed12 | DHCP agent         | bllcloudnet02 | nova              | True  | UP    | neutron-dhcp-agent        |
    | 86f9449f-1d85-4b80-9a4e-5c35b198f695 | Metadata agent     | bllcloudnet02 | None              | True  | UP    | neutron-metadata-agent    |
    | c8a6fc41-9f93-4ca8-a222-83b12baada8f | Linux bridge agent | bllcloudnet02 | None              | True  | UP    | neutron-linuxbridge-agent |
    | cc4073ce-664a-4046-aac1-8c0f1bfeee94 | Linux bridge agent | bllcloudcmp02 | None              | True  | UP    | neutron-linuxbridge-agent |
    | dca89565-7d65-49d6-8d26-60c7fc075d86 | L3 agent           | bllcloudnet02 | nova              | True  | UP    | neutron-l3-agent          |
    +--------------------------------------+--------------------+---------------+-------------------+-------+-------+---------------------------+

When I try to start a server I get the following in the nova log (some text removed for readability):

    INFO nova.scheduler.client.report [...] Submitted allocation for instance
    INFO os_vif [...] Successfully plugged vif VIFBridge(active=False,address=fa:16:3e:5b:d1:28,bridge_name='brqd03aa728-2e',has_traffic_filtering=True,id=7758ba79-b165-4883-95d8-83d39316f4bb,network=Network(d03aa728-2e48-4f99-8ba3-017559cd0890),plugin='linux_bridge',port_profile=<?>,preserve_on_delete=False,vif_name='tap7758ba79-b1')
    WARNING nova.virt.lxd.driver [...] Timeout waiting for vif plugging callback for instance instance-00000015

This ultimately results in an error and the container not being created because I have the vif_plugging_is_fatal set to true (the default). I haven't been able to figure out why the container never gets a response after the vif interface is created. If I set vif_plugging_is_fatal to false the container gets created, but I can't ping the IP address that gets assigned to it. Any help would be appreciated.

Thanks,
Joshua Schaeffer
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linuxcontainers.org/pipermail/lxc-users/attachments/20180813/83fd5a54/attachment.html>