Description of problem: Issue with https://access.redhat.com/documentation/en-us/openshift_container_platform/3.9/html/installation_and_configuration/install-config-cluster-metrics#openshift-prometheus-undeploy Rerunning this uninstallation playbook goes past this issue and it uninstalls the Prometheus. It happens repeatedly. ~~~ TASK [openshift_prometheus : Apply the node exporter template file] ************************************************************************************************************************** Wednesday 27 June 2018 12:41:58 +0000 (0:00:02.410) 0:02:42.225 ******** fatal: [master0.xyz.com]: FAILED! => {"changed": true, "cmd": "oc process -f \"/tmp/prometheus-ansible-MtDs2l/node-exporter-template.yaml\" --param IMAGE=\"registry.access.redhat.com/openshift3/prometheus-node-exporter:v3.9.30\" --param MEMORY_REQUESTS=\"30Mi\" --param CPU_REQUESTS=\"100m\" --param MEMORY_LIMITS=\"50Mi\" --param CPU_LIMITS=\"200m\" --config=/tmp/prometheus-ansible-MtDs2l/admin.kubeconfig -n \"openshift-metrics\" | oc apply --config=/tmp/prometheus-ansible-MtDs2l/admin.kubeconfig -f - -n \"openshift-metrics\"", "delta": "0:00:00.907052", "end": "2018-06-27 12:42:00.736659", "failed": true, "msg": "non-zero return code", "rc": 1, "start": "2018-06-27 12:41:59.829607", "stderr": "error: unable to process template\n processedtemplates.template.openshift.io \"prometheus-node-exporter\" is forbidden: unable to create new content in namespace openshift-metrics because it is being terminated.\nerror: no objects passed to apply", "stderr_lines": ["error: unable to process template", " processedtemplates.template.openshift.io \"prometheus-node-exporter\" is forbidden: unable to create new content in namespace openshift-metrics because it is being terminated.", "error: no objects passed to apply"], "stdout": "", "stdout_lines": []} ~~~ Version-Release number of the following components: openshift-ansible: 3.9.32-1 prometheus-node-exporter:v3.9.30
This is a bug, looks like it affects at least 3.9 and later. PR for 3.10: https://github.com/openshift/openshift-ansible/pull/10225 I'll backport to 3.9 once it's merged to 3.10.
Tested with the fix, issue is fixed, but the current latest openshift-ansible version is openshift-ansible-3.9.45-1 which does not contain the fix, need new openshift-ansible package to verify, change it to MODIFIED
The fix should be available now as of 3.9.46
Tested with openshift-ansible-3.9.50-1, there is not prometheus uninstallation error.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2018:3748