You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
🐛 Machine Controller should try to retrieve node on delete
Consider this scenario:
* Machine is created
* InfraMachine is created
* MachineHealthCheck is set to 10 minutes
* [10 minutes pass]
* Machine is marked as needing remediation
* Machine and InfraMachine goes into deletion flow
* Node finally joins the cluster, say 10 minutes + few seconds
* InfraMachine is still waiting for VM to be deleted
* Machine keeps retrying to delete, but `nodeRef` was never set
* Machine, InfraMachine go away (finalizer removed)
* Node object sticks around in the cluster
This changset allows the Machine controller to look into the cluster
during deletion flow if the Node eventually got created before the
infrastructure provider was able to fully delete the machine.
Signed-off-by: Vince Prignano <[email protected]>
0 commit comments