openshift/node-problem-detector-operator

An operator to run Node Problem Detector on OpenShift

node-problem-detector-operator

An operator to run Node Problem Detector on OpenShift

To deploy the operator:

oc create -f deploy/crd.yaml
oc create -f deploy/ns.yaml
oc create -f deploy/sa.yaml
oc create -f deploy/rbac.yaml
oc create -f deploy/operator.yaml
oc create -f deploy/cr.yaml

To uninstall the operator and Node Problem Detector:

oc delete -f deploy/cr.yaml
oc delete -f deploy/operator.yaml
oc delete -f deploy/rbac.yaml
oc adm policy remove-scc-from-user -n openshift-node-problem-detector privileged -z node-problem-detector
oc delete -f deploy/sa.yaml
oc delete -f deploy/ns.yaml
oc delete -f deploy/crd.yaml

node-problem-detector-operator's Issues

image not pullable

latest version of operator uses an invalid image:

Back-off pulling image "registry.svc.ci.openshift.org/openshift/origin-v4.0:node-problem-detector-operator"

docker pull registry.svc.ci.openshift.org/openshift/origin-v4.0:node-problem-detector-operator
Error response from daemon: received unexpected HTTP status: 503 Service Unavailable

missing rbac

After installing operator 0.0.1 it logs:

node-problem-detector-2200398493\" is forbidden: user \"system:serviceaccount:openshift-node-problem-detector:node-problem-detector-operator\" (groups=[\"system:serviceaccounts\" \"system:serviceaccounts:openshift-node-problem-detector\" \"system:authenticated\"]) is attempting to grant RBAC permissions not currently held:\n{APIGroups:[\"events.k8s.io\"], Resources:[\"events\"], Verbs:[\"create\" \"patch\" \"update\"]}"

and it is not able to create a ds for the actual NPDs

Future Release Branches Frozen For Merging | branch:release-4.17 branch:release-4.18

The following branches are being fast-forwarded from the current development branch (master) as placeholders for future releases. No merging is allowed into these release branches until they are unfrozen for production release.

• release-4.17
• release-4.18

For more information, see the branching documentation.

Future Release Branches Frozen For Merging | branch:release-4.5 branch:release-4.4

The following branches are being fast-forwarded from the current development branch (master) as placeholders for future releases. No merging is allowed into these release branches until they are unfrozen for production release.

• release-4.5
• release-4.4

Contact the Test Platform or Automated Release teams for more information.

kubelet-health.sh is not working inside the container

I am unable to get KubeletProblem status in API server, the status is always showing Unknown.
I tried to run the script manually on host and it is working fine.

#!/usr/bin/env bash

set -eou pipefail

data=$(curl \
  -s \
  http://127.0.0.1:10250/healthz
)

if [[ "$data" != "ok" ]]; then
  exit 20
fi

exit 0

origin@master1:/home/origin>./kubelet-health.sh
origin@master1:/home/origin>echo $?
20
origin@master1:/home/origin>v

But it is not working inside the container

origin@bastion1:/home/origin>oc get po -o wide
NAME                                              READY     STATUS    RESTARTS   AGE       IP            NODE       NOMINATED NODE
node-problem-detector-48jdh                       1/1       Running   0          19m       10.128.0.28   master1    <none>
node-problem-detector-9fccv                       1/1       Running   0          20m       10.128.0.34   compute1   <none>
node-problem-detector-9nlgv                       1/1       Running   0          19m       10.128.0.30   master3    <none>
node-problem-detector-c2dvv                       1/1       Running   0          20m       10.128.0.35   compute2   <none>
node-problem-detector-nzgzn                       1/1       Running   0          19m       10.128.0.31   infra1     <none>
node-problem-detector-operator-56df5c5cb4-4rm48   1/1       Running   0          6h        10.1.6.49     compute2   <none>
node-problem-detector-x8v5v                       1/1       Running   0          20m       10.128.0.32   infra2     <none>
node-problem-detector-zhkcx                       1/1       Running   0          20m       10.128.0.29   master2    <none>
node-problem-detector-zr5nf                       1/1       Running   0          20m       10.128.0.33   infra3     <none>
origin@bastion1:/home/origin>

origin@bastion1:/home/origin>oc exec -it node-problem-detector-48jdh -- /bin/bash
root@master1:/# cd /etc/npd-plugins/
root@master1:/etc/npd-plugins# ./kubelet-health.sh
./kubelet-health.sh: line 8: warning: command substitution: ignored null byte in input
root@master1:/etc/npd-plugins#

image pull error when installed from operator hub

I just installed this operator from operato hub and I'm getting this error:

Failed to pull image "openshift/ose-node-problem-detector:v4.0": rpc error: code = Unknown desc = Error reading manifest v4.0 in docker.io/openshift/ose-node-problem-detector: errors: denied: requested access to the resource is denied unauthorized: authentication required

on the daemon set image.