Description of problem: console shows empty Filesystem usage in project Utilization charts Version-Release number of selected component (if applicable): 4.9.0-0.nightly-2021-10-18-141510 How reproducible: Always Steps to Reproduce: 1. cluster admin user goes to /k8s/cluster/projects/openshift-apiserver page and check project Utilization charts 2. 3. Actual results: 1. Filesystem data is Not available, it shows 'No datapoints found' The query console using is: query=sum(pod:container_fs_usage_bytes:sum{container="",pod!="",namespace='openshift-apiserver'}) BY (namespace) Expected results: 1. Filesystem data should be returned Additional info: Cluster Filesystem usage can be returned correctly and it is using sum( ( max by (device, instance) (node_filesystem_size_bytes{device=~"/.*"}) - max by (device, instance) (node_filesystem_free_bytes{device=~"/.*"}) ) * on(instance) group_left(role) ( label_replace(max by (node) (kube_node_role{role=~".+"}), "instance", "$1", "node", "(.*)") ) )
pod Filesystem query also returns 'No datapoints found', it is using query: pod:container_fs_usage_bytes:sum{pod='console-5ff588fc68-2gkhn',namespace='openshift-console'}
@monitoring team can you please advise what queries Console UI should now use ? Current queries are: namespace filesystem usage sum(pod:container_fs_usage_bytes:sum{container="",pod!="",namespace='openshift-apiserver'}) BY (namespace) top pod filesystem consumers in namespace topk(25, sort_desc(sum(pod:container_fs_usage_bytes:sum{container="",pod!="",namespace='openshift-apiserver'}) BY (pod, namespace))) top project consumers topk(25, sort_desc( sum by (namespace) ( ( sum(avg_over_time(pod:container_fs_usage_bytes:sum{container="", pod!=""}[5m])) BY (namespace, pod) * on(pod,namespace) group_left(node) (node_namespace_pod:kube_pod_info:) ) * on(node) group_left(role) (max by (node) (kube_node_role{role=~"<%= nodeType %>"})) ) )) pod filesystem usage pod:container_fs_usage_bytes:sum{pod='apiserver-56c4b75c74-4tkz9',namespace='openshift-apiserver'} other resource filesystem usage sum(pod:container_fs_usage_bytes:sum * on(pod) group_left(workload, workload_type) namespace_workload_pod:kube_pod_owner:relabel{workload='foo', workload_type='bar'}) by (pod)
Hi @rawagner - we are going to reinstate the `container_fs_usage_bytes` metric. Already merged to 4.10 and I have the back port underway for 4.9.z
checked with 4.10.0-0.nightly-2021-10-31-133814, could see the project and pod filesystem usuage on UI, see the pricture
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 (Moderate: OpenShift Container Platform 4.10.3 security update), 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/RHSA-2022:0056