Red Hat Bugzilla – Bug 1523047
[3.9] exception output while using view archive link in pod log for default project
Last modified: 2018-08-09 18:14:26 EDT
OCP env set up work is blocked by https://bugzilla.redhat.com/show_bug.cgi?id=1525357 Will verify it after BZ # 1525357 is fixed
Moving this back to 'new'. The issue here is there are certain projects where the logs go to the 'operations' index which is the case here. The webconsole does not take that into account.
Commits pushed to master at https://github.com/openshift/openshift-ansible https://github.com/openshift/openshift-ansible/commit/d0fed2b6d24ddbedbc33acffd1c9decf65a54e66 bug 1523047. Annotate ops projects with an .operation prefix https://github.com/openshift/openshift-ansible/commit/32b1287b26c7a3e047d51fd63b5ab3a08fa4c0f1 Merge pull request #6765 from jcantrill/bz1523047_annotate_logging_prefix Automatic merge from submit-queue. bug 1523047. Annotate ops projects with an .operation prefix
Commits pushed to master at https://github.com/openshift/origin-web-console https://github.com/openshift/origin-web-console/commit/4115a78cabf3c5cc05ef645e4e8aa29256158447 bug 1523047. fix log archive link for operation namespaces https://github.com/openshift/origin-web-console/commit/5dcf858df5b48e96ec84d30117df719ba1d77f50 Merge pull request #2678 from jcantrill/1523047_kibana_link_for_ops Automatic merge from submit-queue. bug 1523047. fix log archive link for operation namespaces
Tested with ose-web-console/images/v3.9.0-0.22.0.0, covered non-ops and ops enabled cluster, when click "view archive" link in pod log, it can navigate to kibana UI without error throw out in kibana UI. See the attached pictures.
Created attachment 1384682 [details] click "view archive" link in default project -- non-ops cluster
Created attachment 1384683 [details] click "view archive" link in default project -- ops enabled cluster
Verified with fluentd 3.7 images, other component used 3.9 images since defect https://bugzilla.redhat.com/show_bug.cgi?id=1531157
Backport to 3.7 targeted to https://bugzilla.redhat.com/show_bug.cgi?id=1547348
Issue is reproduced, see the attached picture, click "View Archive" link from default project, the warning shows again in Kibana UI, the index is .all, should use .operations.* index in Kibana UI Warning Discover: "project.default.47990a88-21d8-11e8-a453-fa163e5fe7e9.*" is not a configured pattern. Using the default index pattern: ".all" All logging component version is v3.9.3-1
Created attachment 1405662 [details] Issue is reproduced, waring shows again
(In reply to Junqi Zhao from comment #16) > Issue is reproduced, see the attached picture, click "View Archive" link > from default project, the warning shows again in Kibana UI, the index is > .all, should use .operations.* index in Kibana UI > > > Warning > Discover: "project.default.47990a88-21d8-11e8-a453-fa163e5fe7e9.*" is not a > configured pattern. Using the default index pattern: ".all" > > All logging component version is v3.9.3-1 env: # openshift version openshift v3.9.3 kubernetes v1.9.1+a0ce1bc657 etcd 3.2.16 deployed logging without ops cluster enabled
Please provide the project annotations: `oc get project default -o yaml'. The annotations provide hints to the webconsole to properly build the url. If the new image was deployed without running ansible, you may not have executed the code to annotate the project. Additionally, please provide the version of the webconsole image so we can confirm it also has the code that consumes the annotation.
backport to 3.6 Bug 1547688 - [3.6] exception output while using view archive link in pod log for default project
Can you advise if this logging deployment additionally uses an operations instance of the logging stack too? It looks like we still may have an error where the annotation is not applied if the ops cluster is not deployed [1] [1] https://github.com/openshift/openshift-ansible/blob/release-3.9/roles/openshift_logging/tasks/annotate_ops_projects.yaml#L15
The merge of https://github.com/openshift/openshift-ansible/pull/7522 allows properly annotating projects for the prefix regardless if the 'ops' EFK stack is deployed. Needs a new ansible release to resolve once merged. Manual work around is to annotation the default project with: openshift.io/logging.data.prefix: ".operations"
logging component images version: v3.9.38 Tested with enabled ops cluster, it will navigate to kibana-ops UI without error Tested with disabled ops cluster, it will navigate to kibana UI without error the index pattern is .operations.* in both scenarios
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-2018:2335