Description of problem: I'm unable to deploy a CouchbaseCluster with the couchbase operator and get the below error: The field status.members is invalid Version-Release number of selected component (if applicable): openshift-install v0.16.1 built from commit e3fceacc975953f56cb09931e6be015a36eb6075 OpenShift cluster version is 4.0.0-0.9 registry.connect.redhat.com/couchbase/operator:1.1.0-1 How reproducible: Steps to Reproduce: 1. Install OCP 4.0 cluster on AWS. 2. Login the cluster as kubeadmin user and install the couchbase operator. 3. Create an application project (i.e. kevin-sandbox). 4. From the Developer Catalog, create a Couchbase Cluster and use the default values. 5. Here is the YAML file populated with default values: apiVersion: couchbase.com/v1 kind: CouchbaseCluster metadata: name: cb-example namespace: kevin-sandbox spec: authSecret: cb-example-auth baseImage: registry.connect.redhat.com/couchbase/server buckets: - conflictResolution: seqno enableFlush: true evictionPolicy: fullEviction ioPriority: high memoryQuota: 128 name: default replicas: 1 type: couchbase cluster: analyticsServiceMemoryQuota: 1024 autoFailoverMaxCount: 3 autoFailoverOnDataDiskIssues: true autoFailoverOnDataDiskIssuesTimePeriod: 120 autoFailoverServerGroup: false autoFailoverTimeout: 120 clusterName: cb-example dataServiceMemoryQuota: 256 eventingServiceMemoryQuota: 256 indexServiceMemoryQuota: 256 indexStorageSetting: memory_optimized searchServiceMemoryQuota: 256 servers: - name: all_services services: - data - index - query - search - eventing - analytics size: 3 version: 5.5.2-1 6. Click through the details for the new cb-example CouchbaseCluster that was created. Actual results: I received 'The field status.members is invalid', along with no Couchbase pod (the two are related as the existence of a Couchbase pod would have inherently defined status.members). Unfortunately, I don't get much other detail as 'oc get events' is empty and 'oc logs -f couchbase-operator-ff85f5954-nvtq7' from the couchbase-operators project did not provide any relevant info either. Expected results: The CouchbaseCluster pod should be up and running. Additional info:
Matt from Couchbase here. Spotted this. One question, if we wanted to repro in house, is it accurate to say it'd be with OpenShift 4.0 which is AWS only at this stage. Any guidance on repro? Also, we looked at this github console #922. Our testing wasn't so much through Operator Hub, but just plain CLI 'oc'. Does it work that way?
Matt, This morning, I took the YAML above and attempted to install via CLI: 'oc create -f couchbase-cluster.yml' and encountered the same issue. Here's the output of 'oc get couchbasecluster cb-example -o yaml' which appears to be missing the entire status section, including status.members which it's complaining about. apiVersion: couchbase.com/v1 kind: CouchbaseCluster metadata: creationTimestamp: 2019-04-26T14:46:21Z generation: 1 name: cb-example namespace: kevin-sandbox resourceVersion: "19319" selfLink: /apis/couchbase.com/v1/namespaces/kevin-sandbox/couchbaseclusters/cb-example uid: 0e8b2036-6832-11e9-b0a8-027be685655c spec: authSecret: cb-example-auth baseImage: registry.connect.redhat.com/couchbase/server buckets: - conflictResolution: seqno enableFlush: true evictionPolicy: fullEviction ioPriority: high memoryQuota: 128 name: default replicas: 1 type: couchbase cluster: analyticsServiceMemoryQuota: 1024 autoFailoverMaxCount: 3 autoFailoverOnDataDiskIssues: true autoFailoverOnDataDiskIssuesTimePeriod: 120 autoFailoverServerGroup: false autoFailoverTimeout: 120 clusterName: cb-example dataServiceMemoryQuota: 256 eventingServiceMemoryQuota: 256 indexServiceMemoryQuota: 256 indexStorageSetting: memory_optimized searchServiceMemoryQuota: 256 servers: - name: all_services services: - data - index - query - search - eventing - analytics size: 3 version: 5.5.2-1
Kevin, Matt Thanks for your information! There is a similar bug 1699794 for etcd operator. I think this is the operator itself issue. Anyway, left it in this component is OK since we haven't a proprietary component for the couchbase operator bugs yet. Since the couchbase operator from the "Certified Operators" source, I increased the processing priority.
Jian, Yes, I came across that bug 1669794 you filed before opening mine. I tried a few other operators from OperatorHub including etcd, and they seem to work just fine for me, hence I thought to report this separately.
Tony, can you take a look at this?
Moving out to 4.1.z. This does not need to block the OCP 4.1.0 release.
*** Bug 1685394 has been marked as a duplicate of this bug. ***
Closing this bug as this is a community or ISV certified operator that has issues. If this is a community operator, please use Github to open an issue and track there. If this is a certified operator, please reach out to Tony Cambell directly so that he is turn can speak to the vendor about fixing the issue.
Unfortunately I am unable to verify the error experienced by Kevin (original poster) is related to the UI warning/error. The "The field status.members is invalid" message is related to the descriptors component of the OpenShift console (https://github.com/openshift/console/tree/master/frontend/public/components/operator-lifecycle-manager/descriptors). Essentially it is a custom UI component designed to pull data from the status portion of the couchbase cluster resource and display it for quick insight into the state of the "members" of the cluster. It appears the operator followed the documentation for displaying the status of members but potentially it is modifying the status in a way that the descriptors component does not expect therefore rendering the path (status.members) in a way that cannot be read. I am looking further into it as it is an eye sore but it is not an issue that is causing the operator to not install. Environment: OpenShift 4.1.11 Kubernetes v1.13.4+df9cebc Installed via OpenShift Installer deployed on AWS Couchbase Operator 1.1.0
The needinfo request[s] on this closed bug have been removed as they have been unresolved for 1000 days