Changing Glance backend with pending nodes issue
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Fuel for OpenStack |
Fix Committed
|
High
|
Julia Aranovich |
Bug Description
Summary:
After the user added nodes to be deployed, and before deploying, if the user changes the storage configuration, the partition layout will be rendered unusable, even using the "load defaults" button (cinder LVM to Ceph).
Steps to reproduce:
- Create a new environment with Fuel UI, and configure the glance backend to default.
- Add new nodes to this new environment.
- (Optional, tested both ways) Change the partition layout.
- Go to settings, change the Glance backend to Ceph RBD backend for images.
- Access again the node storage layout.
Expected behaviour:
- Disk storage layout should be default as if we added a new node.
Real behaviour:
- From Swift backend to Ceph backend, the layout UI screen isn't usable and shows an error dialog (screenshot attached).
- From Ceph backend to Swift backend, using the "load defaults" switches back to the default partition settings, but it will not be properly saved.
Workaround:
- Deleting and adding back again the nodes will fix the issue.
Changed in fuel: | |
assignee: | nobody → Fuel Python Team (fuel-python) |
importance: | Undecided → High |
milestone: | none → 6.1 |
Changed in fuel: | |
status: | New → Confirmed |
tags: | added: tricky |
Changed in fuel: | |
assignee: | Fuel UI Team (fuel-ui) → Julia Aranovich (jkirnosova) |
status: | Confirmed → In Progress |
We do not support this use case. Have you done it with UI or with console client?