Node Scheduling can't be removed

[Toonix]

When you set a node Scheduling, you can't remove it.

Automatically pick nodes for each pod matching scheduling rules:
Can't be removed form a deployment

Useful	Info
Versions	Rancher v2.0.0 UI: v2.0.41
Access	local admin
Route	containers.run

[tfiduccia]

Version - 2.0 master 5/16
Now when I switch to a specific node back to Automatically pick nodes, it never finishes the edit. The new pod is stuck in pending and the old pod stays active.

[sangeethah]

Tested with v2.0.3-rc1

This issue is still reproducible.

When I remove the node scheduling rule as part of upgrade ,I dont see pods getting recreated.
Also when I view the workload details, the node scheduling rules are still present.

[sangeethah]

Tested with v2.0.3-rc2

Create a workload with node scheduling enabled ( Run all pods for this workload on a specific node or Automatically pick nodes for each pod matching scheduling rules ).
Edit workload and remove the node scheduling rule.
Workload upgrade succeeds and pods get recreated as expected.
Node scheduling rules get removed in the workload after upgrade as expected.