This RFE is a result of testing / iterating on the CI squad's 12th sprint:
https://trello.com/c/JEGLSVh6/51-reproduce-ci-jobs-with-libvirt
Multiple members of the CI squad hit this issue, which isn't immediately apparent. Since the reproducer script runs on the local machine, it uses SSH to connect to 127.0.0.2 (own host). As this is a non-obvious "first run experience" issue, it's a good candidate to resolve. There are 2 approaches proposed thus far:
- add own key to authorized_keys
- ansible_connection=local (avoiding SSH for local host alltogether).
There's already a patch posted for the latter option...
---
https://review.openstack.org/#/c/566149
Presently the user's own key needs to be added to authorized_users.
Since the --libvirt option is for localhost only, use the
'local' connection plugin. This removes a common first point
of frustration for users and will avoid ssh entirely for the virthost
connection
Change-Id: I7cafe0e4315cc6877e0e50ebfcc8e0f8307582d1
---
Change abandoned by Emilien Macchi (<email address hidden>) on branch: master
Review: https:/
Reason: This patch has no activity for a while, per policy, we abandon it. Feel free to re-open if you think you're working on it.