Uploaded image for project: 'OpenShift Logging'
  1. OpenShift Logging
  2. LOG-3763

Error logs deleting elasticsearch CR when not installed EO

XMLWordPrintable

    • False
    • False
    • NEW
    • VERIFIED
    • Hide
      Before this update, if CLO was installed without Elasticsearch Operator, the CLO pod would continuously display an error message regarding the deletion of Elasticsearch. With this update, CLO will perform additional checks before displaying any error messages, resulting in the absence of error messages related to Elasticsearch deletion.
      Show
      Before this update, if CLO was installed without Elasticsearch Operator, the CLO pod would continuously display an error message regarding the deletion of Elasticsearch. With this update, CLO will perform additional checks before displaying any error messages, resulting in the absence of error messages related to Elasticsearch deletion.
    • Log Collection - Sprint 233
    • Moderate

      Description of problem:

      When installed only the CLO since not needed the EO because all the logs are logforwarded, the CLO pod returns continually the message error:

      $ oc logs $(oc get pod -l name=cluster-logging-operator -o name) |grep "Failure deleting elasticsearch elasticsearch CR" |head -1
{"_ts":"2023-03-13T09:13:10.794416893Z","_level":"0","_component":"cluster-logging-operator","_message":"Error removing component","_error":{"msg":"Failure deleting elasticsearch elasticsearch CR for \"instance\": no matches for kind \"Elasticsearch\" in version \"logging.openshift.io/v1\""}}

$ oc logs $(oc get pod -l name=cluster-logging-operator -o name)  |grep -c "Failure deleting elasticsearch elasticsearch CR" 
174

      Version-Release number of selected component (if applicable):

      $ oc get csv
NAME                     DISPLAY                     VERSION   REPLACES                 PHASE
cluster-logging.v5.6.3   Red Hat OpenShift Logging   5.6.3     cluster-logging.v5.6.2   Succeeded

      How reproducible:

      Always

      Steps to Reproduce:

      1. Deploy CLO using channel stable-5.6 without deploying EO since the internal Elasticsearch won't be used

      $ oc get csv
NAME                     DISPLAY                     VERSION   REPLACES                 PHASE
cluster-logging.v5.6.3   Red Hat OpenShift Logging   5.6.3     cluster-logging.v5.6.2   Succeeded

      2. Create the CLO CR defining only the collector

      $ cat clo-instance-only-collector.yaml 
apiVersion: "logging.openshift.io/v1"
kind: "ClusterLogging"
metadata:
  name: "instance"
  namespace: "openshift-logging"
spec:
  managementState: "Managed"
  collection:
    logs:
      type: "fluentd"
      fluentd:
        resources: {}

      3. Check the logs of the CLO operator pod where visible always a error message trying to delete the Elasticsearch CR

      $ oc logs $(oc get pod -l name=cluster-logging-operator -o name) |grep "Failure deleting elasticsearch elasticsearch CR" |head -1
{"_ts":"2023-03-13T09:13:10.794416893Z","_level":"0","_component":"cluster-logging-operator","_message":"Error removing component","_error":{"msg":"Failure deleting elasticsearch elasticsearch CR for \"instance\": no matches for kind \"Elasticsearch\" in version \"logging.openshift.io/v1\""}}

$ oc logs $(oc get pod -l name=cluster-logging-operator -o name)  |grep -c "Failure deleting elasticsearch elasticsearch CR" 
174

      {}Workaround{}:

      Create the Elasticsearch CR until a fix is available:

      $ oc create -f  https://raw.githubusercontent.com/openshift/elasticsearch-operator/release-5.6/bundle/manifests/logging.openshift.io_elasticsearches.yaml

      Review that not more "Failure deleting elasticsearch elasticsearch CR" visible in the clusterLogging Operator pod after applying the workaround:

      oc logs $(oc get pod -l name=cluster-logging-operator -o name) |grep "Failure deleting elasticsearch elasticsearch CR" |head -1

      Actual results:

      ClusterLogging operator pod returning continually the error deleting the Elasticsearch CR

      Expected results:

      No errors visible in the clusterLogging Operator deleting the Elasticsearch CR since it's not expected to be installed because the logs are only forwarded outside and not used the internal Elasticsearch

              vparfono Vitalii Parfonov
              rhn-support-ocasalsa Oscar Casal Sanchez
              Kabir Bharti Kabir Bharti
              Votes:
              0 Vote for this issue
              Watchers:
              3 Start watching this issue

                Created:
                Updated:
                Resolved: