-
Bug
-
Resolution: Not a Bug
-
Critical
-
None
-
4.17, 4.17.z, 4.18, 4.18.z
-
None
-
False
-
-
Customer Escalated
Description of problem:
Replacing an already installed host on an installed cluster using SaaS assisted-service fails.
Procedure:
- Install a cluster using assisted service in console.redhat.com
- When the cluster is installed, remove the node manually from the installed cluster by
- Cordoning, draining, and deleting the node
$ oc adm cordon <node_name> $ oc adm delete <node_name>
- Generate a new discovery ISO on console.redhat.com
- Boot the host that was removed from the cluster with the new discovery ISO
Actual results:
Host doesn't get re-registered and stays installed into the original cluster on console.redhat.com
assisted-installer-agent shows the following logs
Mar 27 17:04:34 next_step_runne[4861]: time="27-03-2025 17:04:34" level=info msg="Result for domain-resolution already exists in assisted service" file="step_processor.go:64" request_id=585d3465-af20-4441-8a8d-c63587bc6078
Expected results:
Host should be rediscovered
Additional Information
Original Issue Description
Below procedure on assisted cluster fails to remove the node from assisted service inventory.
8----8<----8<----8<----8<----8<----8<----8
1. Cordon the node
2. Drain the node
3. Delete the node using oc adm delete
After this visit console.redhat.com and follow the cluster.
1. Click on Add host option.
2. Add the needed details there.
3. Generate ISO and boot using it.
8----8<----8<----8<----8<----8<----8<----8
This will result to error as on the node.
"Result for inventory already exists in assisted service"
Version-Release number of selected component (if applicable):
Any release
How reproducible:
As per description.
Actual results:
Readding node fails.
Expected results:
Shouldn't fail.
Additional info:
This seems to be block as the API option I tried doesn't work even.
- depends on
-
RFE-7364 Add documentation for "Re-adding" node for assisted service. ( Disater recovery)
-
- Backlog
-
- links to