Description of problem: During upgrade, the EO waits for the cluster to go 'green' each time, however it can get away with waiting for 'yellow'. Also per the elasticsearch documentation we should be setting the allocation to be "primaries" not "none". How reproducible: Always
Move to verified. 1). Make the ES to Yellow status (It is easy as https://bugzilla.redhat.com/show_bug.cgi?id=1838153) 1.2) Install 4.4 Cluster Logging with nodeCount=1 and ZeroRedundancy 1.2) Upgrade CLO to 4.5 1.3) The ES status #oc get csv NAME DISPLAY VERSION REPLACES PHASE clusterlogging.4.5.0-202005221517 Cluster Logging 4.5.0-202005221517 clusterlogging.4.4.0-202005221357 Succeeded elasticsearch-operator.4.4.0-202005220258 Elasticsearch Operator 4.4.0-202005220258 Succeeded #oc exec -c elasticsearch elasticsearch-cdm-a9h73w86-1-758c54d858-b9zb6 -- es_cluster_health { "cluster_name" : "elasticsearch", "status" : "yellow", "timed_out" : false, "number_of_nodes" : 1, "number_of_data_nodes" : 1, "active_primary_shards" : 16, "active_shards" : 16, "relocating_shards" : 0, "initializing_shards" : 0, "unassigned_shards" : 10, "delayed_unassigned_shards" : 0, "number_of_pending_tasks" : 1, "number_of_in_flight_fetch" : 0, "task_max_waiting_in_queue_millis" : 0, "active_shards_percent_as_number" : 61.53846153846154 } 2) Upgrade EO to 4.5. 3) Check the ES pods. You can see the ES had been upgraded to 4.5. (ose-elasticsearch-proxy). (Note: the ES couldn't be running for https://bugzilla.redhat.com/show_bug.cgi?id=1838929. I think that doesn't blocked this bug.) $ oc get pods NAME READY STATUS RESTARTS AGE cluster-logging-operator-558d8f8f7-4w6nm 1/1 Running 0 11m curator-1590223200-pplnv 0/1 Completed 0 17m curator-1590223800-rtd2f 0/1 Error 0 7m18s elasticsearch-cdm-a9h73w86-1-64fdf5c8f5-pvbxv 1/2 CrashLoopBackOff 5 5m23s $oc get pods elasticsearch-cdm-a9h73w86-1-64fdf5c8f5-pvbxv -o yaml |grep 'image:' image: quay.io/openshift-qe-optional-operators/ose-logging-elasticsearch6@sha256:1d2c67ad5a6bbebfc4d44c6e943b3c1727cb33731f67c35e69d4436ff8b46774 image: quay.io/openshift-qe-optional-operators/ose-elasticsearch-proxy@sha256:cc93bc0d0e7a5c92f6380fde91b6bade54994b1d03949441d49a719fbfd55e23
As the workaround to BZ1838929. After I used registry.svc.ci.openshift.org/origin/4.5:logging-elasticsearch6 instead of downstream image. The ES pod can be running.
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-2020:2409