restore with bootstrap does not update controller instance id
Bug #1695115 reported by
Ian Booth
This bug affects 1 person
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Canonical Juju |
Incomplete
|
Low
|
Eric Claude Jones |
Bug Description
When doing a juju restore with the -b flag to bootstrap a new controller, the model database is not updated to reflect the fact that the controller machine will have a new instance id. This means that the instance poller continually tries to poll the old instance and errors out, filling the logs.
Changed in juju: | |
milestone: | 2.3-beta1 → 2.3-beta2 |
Changed in juju: | |
milestone: | 2.3-beta2 → none |
Changed in juju: | |
assignee: | nobody → Eric Claude Jones (ecjones) |
tags: | added: restore-backup |
tags: | removed: restore-backup |
To post a comment you must log in.
Hi Ian,
I cannot replicate this issue. I have tried this on both MAAS 2.3 and Joyent providers. I am seeing the database update the instanceData collection and models collection appropriately. I am not seeing the log fill with messages.
https:/ /paste. ubuntu. com/p/cQ8GDKmQN H/
Please provide with further detail.