When running "kubectl apply -f cmk-uninstall-all-daemonset.yaml" the uninstall worked on one of my two nodes, but failed on the other. I'm now left with pod/cmk-uninstall-all-nodes-bm8lp in CrashLoopBackoff, daemonset.apps/cmk-uninstall-all-nodes still running, and /etc/cmk still present.

The final logs for the failed pod were as follows:

WARNING:root:"cmk-nodereport" for node "controller-0" does not exist.
INFO:root:"cmk-nodereport" for node "controller-0" removed.
INFO:root:Removing "cmk-reconcilereport" from Kubernetes API server for node "controller-0".
INFO:root:Converted "controller-0" to "controller-0" for TPR/CRD name
WARNING:root:"cmk-reconcilereport" for node "controller-0" does not exist.
INFO:root:"cmk-reconcilereport" for node "controller-0" removed.
INFO:root:Removing node taint.
INFO:root:Patching node controller-0:
[
{
"op": "replace",
"path": "/spec/taints",
"value": []
}
]
INFO:root:Removed node taint with key"cmk".
INFO:root:Removing node ERs
INFO:root:Patching node status controller-0:
[
{
"op": "remove",
"path": "/status/capacity/cmk.intel.com~1exclusive-cores"
}
]
ERROR:root:Aborting uninstall: Exception when removing ER: (422)
Reason: Unprocessable Entity
HTTP response headers: HTTPHeaderDict({'Content-Type': 'application/json', 'Content-Length': '187', 'Date': 'Wed, 15 May 2019 17:44:02 GMT'})
HTTP response body: {"kind":"Status","apiVersion":"v1","metadata":{},"status":"Failure","message":"the server rejected our request due to an error in our request","reason":"Invalid","details":{},"code":422}