Description of problem: Aggregated logging diagnostics always run successfully without any existing errors report. Issue reproduced while running most test cases here: goo.gl/FM33AM. Version-Release number of selected component (if applicable): openshift v1.4.0-alpha.0+e22007f kubernetes v1.4.0+776c994 etcd 3.1.0-alpha.1 docker.io/openshift/origin-logging-curator latest a3b368d5a230 3 weeks ago 226.2 MB docker.io/openshift/origin-logging-deployment latest a2164db0cbfa 3 weeks ago 701.4 MB docker.io/openshift/origin-logging-elasticsearch latest e0c2ce98d1cd 3 weeks ago 439.6 MB docker.io/openshift/origin-logging-kibana latest 5afef55d878f 3 weeks ago 452.7 MB docker.io/openshift/origin-logging-auth-proxy latest 223796812cdb 3 months ago 711.6 MB How reproducible: always Steps to Reproduce: 1.Launch instance on EC2 from image devenv-rhel7_5144 2.Start openshift serivce on the instance 3.Diagnose Aggregated logging from master machine without logging components deployed Actual results: # oadm diagnostics AggregatedLogging [Note] Determining if client configuration exists for client/cluster diagnostics Info: Successfully read a client config file at '/etc/origin/master/admin.kubeconfig' Info: Using context for cluster-admin access: 'default/{master-dns}:8443/system:admin' [Note] Running diagnostic: AggregatedLogging Description: Check aggregated logging integration for proper configuration [Note] Summary of diagnostics execution (version v1.4.0-alpha.0+e22007f): [Note] Completed with no errors or warnings seen. Expected results: Should report the missing of logging components as listed in https://gist.github.com/jcantrill/d608072970f70a8737e566af59b84cb9 Additional info:
The diagnostic requires you to have configured the LoggingPublicUrl in the master config in order to run. Otherwise it doesn't know where to find the logging deployment, and just silently completes. I think a quick message to this effect would be a good idea.
I have a pull request in to clarify this: https://github.com/openshift/origin/pull/11295
Commit pushed to master at https://github.com/openshift/origin https://github.com/openshift/origin/commit/905f0c7ee369ce1a5f6e676358e2c4464e27158d diagnostics: clarify when logging not configured bug 1382903 https://bugzilla.redhat.com/show_bug.cgi?id=1382903 When the LoggingPublicUrl isn't configured in master config, the diagnostic no longer silently succeeds, but instead is skipped with a message about why.
Verified on origin, it's fixed: # openshift version openshift v1.5.0-alpha.0+76a5795-71 kubernetes v1.4.0+776c994 etcd 3.1.0-rc.0 # oadm diagnostics AggregatedLogging [Note] Determining if client configuration exists for client/cluster diagnostics Info: Successfully read a client config file at '/etc/origin/master/admin.kubeconfig' Info: Using context for cluster-admin access: 'default/ec2-54-208-172-89-compute-1-amazonaws-com:8443/system:admin' [Note] Skipping diagnostic: AggregatedLogging Description: Check aggregated logging integration for proper configuration Because: No LoggingPublicURL is defined in the master configuration [Note] Summary of diagnostics execution (version v1.5.0-alpha.0+76a5795-71): [Note] Completed with no errors or warnings seen.
Verified on OCP 3.5, it's fixed: # openshift version openshift v3.5.0.14+20b49d0 kubernetes v1.5.2+43a9be4 etcd 3.1.0 # oadm diagnostics AggregatedLogging [Note] Determining if client configuration exists for client/cluster diagnostics Info: Successfully read a client config file at '/root/.kube/config' Info: Using context for cluster-admin access: 'default/ip-172-18-5-92-ec2-internal:8443/system:admin' [Note] Skipping diagnostic: AggregatedLogging Description: Check aggregated logging integration for proper configuration Because: No LoggingPublicURL is defined in the master configuration [Note] Summary of diagnostics execution (version v3.5.0.14+20b49d0): [Note] Completed with no errors or warnings seen.