container bridge script should report an error if it doesn't succeed

Bug #1660937 reported by John A Meinel
10
This bug affects 2 people
Affects Status Importance Assigned to Milestone
Canonical Juju
Invalid
Medium
Unassigned

Bug Description

Right now, the script we use to rewrite /etc/network/interfaces for bridging doesn't pay attention to included subdirectories (/etc/network/interfaces.d/*). On many clouds, cloud-init creates 'eth0' in /e/n/i/50cloud.cfg

Which means that the Controller may decide "I want you to bridge eth0", and trigger the bridge script to run, but it won't actually be able to create br-eth0, but doesn't actually report a failure. Which means we continue, and then fail late when we try to use br-eth0 but it doesn't exist.

This is not a critical issue today, because for clouds that do use /e/n/i/50cloud.cfg we don't support getting IP addresses from the underlying cloud anyway. If/when we do, we'll need to fix this.

We'd like to move away from a Python script anyway, so maybe we can just do it there.

Curtis Hovey (sinzui)
Changed in juju:
milestone: 2.2-beta1 → 2.2-beta2
Changed in juju:
milestone: 2.2-beta2 → 2.2-beta3
Changed in juju:
milestone: 2.2-beta3 → 2.2-beta4
Changed in juju:
milestone: 2.2-beta4 → 2.2-rc1
Tim Penhey (thumper)
Changed in juju:
milestone: 2.2-rc1 → none
importance: High → Medium
Revision history for this message
Joseph Phillips (manadart) wrote :

We don't use e/n/i for new deployments, and certainly the Netplan bridger works with config in the xxx.d form.

Changed in juju:
status: Triaged → Invalid
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.