Loading...

XML

Word

Printable

Details

Type: Bug
Resolution: Done-Errata
Priority: Normal
Fix Version/s: Logging 5.7.8
Affects Version/s: Logging 5.6.z, Logging 5.7.z
Component/s: Log Collection
Labels:
- devel_ack+

Blocked:
False
Blocked Reason:
None
Ready:
False
Docs QE Status:
NEW
QE Status:
VERIFIED
Release Note Text:

Hide
Before these changes, CLO did not inform the user about an invalid configuration when configuring JSON log parser without setting 'structuredTypeKey' or 'structuredTypeName'. With this update, this situation will be checked, and an invalid condition will be set for CLO, alerting the user about the configuration issue.

Show
Before these changes, CLO did not inform the user about an invalid configuration when configuring JSON log parser without setting 'structuredTypeKey' or 'structuredTypeName'. With this update, this situation will be checked, and an invalid condition will be set for CLO, alerting the user about the configuration issue.
Release Note Type:
Bug Fix

Sprint:
Log Collection - Sprint 243, Log Collection - Sprint 244
Severity:
Moderate

SFDC Cases Links:
SFDC Cases Counter:

Description

Description of problem:

When in the clusterLogForwarder is set parse: json using the default outputRefs and without setting outputDefaults.elasticsearch.structuredTypeKey or outputDefaults.elasticsearch.structuredTypeName, then, any change done in the clusterlogging CR is not applied and no an error is visible in the clusterLogForwarder. It's only visible in the clusterLogging pod something like:

$ clo=$(oc get pod -l name=cluster-logging-operator -n openshift-logging -o name)
$ oc logs $clo -n openshift-logging  |grep -i "structuredTypeKey or structuredTypeName must be defined for Elasticsearch output"  |tail -1 
{"_ts":"2023-08-14T14:27:27.865948363Z","_level":"0","_component":"cluster-logging-operator","_message":"Error reconciling clusterlogging instance","_error":{"msg":"structuredTypeKey or structuredTypeName must be defined for Elasticsearch output named \"default\" when JSON parsing is enabled on pipeline \"container-logs\" that references it"}}

This could lead that the change applied is correct since no error in the clusterLogForwarder visible:

$ oc get clusterlogforwarder instance -o yaml -n openshift-logging
apiVersion: logging.openshift.io/v1
kind: ClusterLogForwarder
metadata:
  creationTimestamp: "2023-08-14T14:13:39Z"
  generation: 2
  name: instance
  namespace: openshift-logging
  resourceVersion: "335236"
  uid: 066a36a2-b104-43a5-a4d1-7c8fe39184c1
spec:
  pipelines:
  - inputRefs:
    - application
    - infrastructure
    - audit
    name: container-logs
    outputRefs:
    - default
    parse: json
status:
  conditions:
  - lastTransitionTime: "2023-08-14T14:27:01Z"
    status: "True"
    type: Ready
  outputs:
    default:
    - lastTransitionTime: "2023-08-14T14:27:01Z"
      status: "True"
      type: Ready
  pipelines:
    container-logs:
    - lastTransitionTime: "2023-08-14T14:27:01Z"
      status: "True"
      type: Ready

It should be expected that in the the CLF status is showed an error indicating that the pipeline or configuration created is not correct as it shows in other situations or even not allowing to apply a wrong configuration.

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

$ oc get csv
NAME                                       DISPLAY                                          VERSION                   REPLACES                        PHASE
cluster-logging.v5.7.4                     Red Hat OpenShift Logging                        5.7.4                     cluster-logging.v5.7.3          Succeeded
elasticsearch-operator.v5.7.4              OpenShift Elasticsearch Operator                 5.7.4                     elasticsearch-operator.v5.7.3   Succeeded

How reproducible:

Always

Steps to Reproduce:

Create or edit the clusterLogforwarder using a configuration similar to [1]
Review the logs from the clusterlogging operator pod and check that any posterior changes applied to the clusterLogging instance won't be applied since the clusterLogforwarder configuration is wrong and showed the error structuredTypeKey or structuredTypeName must be defined for Elasticsearch output named
Verify that in the clusterlogforwader instance status not errors are visible [3]
Apply any change to the clusterLogging instance and observe is not applied

Actual results:

The clusterLogForwarder is not showing any error and also allowing to create a wrong configuration.

Expected results:

The clusterlogforwarder shows an error indicating that the configuration is wrong. Added to this, for improving the user experience, really, applying this configuration should be given an error and not allowing it

Additional info:

[1]

$ cat clf.yaml 
apiVersion: logging.openshift.io/v1
kind: ClusterLogForwarder
metadata:
  name: instance 
  namespace: openshift-logging 
spec:
  pipelines: 
   - name: container-logs
     inputRefs:
     - application
     - infrastructure
     - audit
     outputRefs:
     - default
     parse: json
 $ oc create -f clf.yaml

[2]

$ oc logs $clo -n openshift-logging  |grep -i "structuredTypeKey or structuredTypeName must be defined for Elasticsearch output"  |tail -1  {"_ts":"2023-08-14T14:27:27.865948363Z","_level":"0","_component":"cluster-logging-operator","_message":"Error reconciling clusterlogging instance","_error":{"msg":"structuredTypeKey or structuredTypeName must be defined for Elasticsearch output named \"default\" when JSON parsing is enabled on pipeline \"container-logs\" that references it"}}

[3]

$ oc get clusterlogforwarder instance  -o yaml 
apiVersion: logging.openshift.io/v1
kind: ClusterLogForwarder
metadata:
  creationTimestamp: "2023-08-14T14:32:26Z"
  generation: 1
  name: instance
  namespace: openshift-logging
  resourceVersion: "340727"
  uid: d6b48f9c-270f-46ed-b6ab-1a10bac5439b
spec:
  pipelines:
  - inputRefs:
    - application
    - infrastructure
    - audit
    name: container-logs
    outputRefs:
    - default
    parse: json
status:
  conditions:
  - lastTransitionTime: "2023-08-14T14:38:05Z"
    status: "True"
    type: Ready
  outputs:
    default:
    - lastTransitionTime: "2023-08-14T14:38:05Z"
      status: "True"
      type: Ready
  pipelines:
    container-logs:
    - lastTransitionTime: "2023-08-14T14:38:05Z"
      status: "True"
      type: Ready

[4]

$ oc get clusterlogging instance -o yaml 
apiVersion: "logging.openshift.io/v1"
kind: "ClusterLogging"
metadata:
  name: "instance"
  namespace: "openshift-logging"
spec:   collection:
    logs:
      type: "fluentd"
      fluentd:
        resources: {}
        nodeSelector: 
          test: failure