Description of problem: Hi, I've configured Elasticsearch log Curator to delete 1 day older indexed data of user defined project(eg: myproj-qe) and observed curator could not delete "myproj-qe" project data which is older than 1 day. Could you please check is below steps are correct as per doc.? and should it create indices as format "projectName-timestamp" in ES and later can be rollover by Curator by age? reference RH links:- https://access.redhat.com/documentation/en-us/openshift_container_platform/4.5/html/release_notes/ocp-4-5-release-notes https://docs.openshift.com/container-platform/4.6/logging/config/cluster-logging-curator.html Version-Release number of selected component (if applicable): #oc version Client Version: 4.7.0-fc.2 Server Version: 4.7.0-fc.2 Kubernetes Version: v1.20.0+394a5a3 How reproducible: Steps to Reproduce: 1. Setup cluster logging instance 2. create new project("myproj-qe") and run some sample network workload(iperf3) to produce logs as application log for ES app index. # oc get pods -n myproj-qe NAME READY STATUS RESTARTS AGE network-stress-client-5v9gn 1/1 Running 0 19h network-stress-server-q4mmd 1/1 Running 0 19h 3. edit Curator configuration file- oc edit configmap/curator. and add below in config.yaml: section myproj-qe: delete: days: 1 4.checked the Kibana application log after 1 day, it still showing data of "myproj-qe" project in "app" index, which is older than 1 day. Actual results: Expected results: Additional info: NOTES: we have enabled the debug log of curator, and observed, the "filtertype: age" is calculated based on "Source"(eg. indices name) and "timestring" associated with the indices name(eg. curator expecting indices format- myproj-%Y.%m.%d), But in OCP the index name generated by ES are generic in nature (e.g., infra-000001 or app-000003 or audit-000021) and Logs from all application namespaces are stored in a given app-###### index name. Also we could not see any index format( eg. myproj-qe-### ) in ES. OCP ES index format : -------------------- # oc exec elasticsearch-cdm-ofr1aa67-1-bf99cb4dd-r56p9 -- indices health status index uuid pri rep docs.count docs.deleted store.size pri.store.size green open app-000003 xekH_V4NSGSjHV0n91Tx6Q 1 0 14108 0 5 5 green open audit-000021 uXocn7PITF6kukm3TKvxfQ 1 0 567615 0 453 453 green open infra-000021 AFtdNCaYRjyW_777AfG7eQ 1 0 190334 0 119 119 some lines of Curator DEBUG logs(log details attached ) =============================== ... + cat /opt/app-root/src/actions.yaml --- actions: 1: action: delete_indices description: auto-generated options: ignore_empty_list: true continue_if_exception: false timeout_override: '300' filters: - filtertype: pattern kind: regex value: '^project\.myproj\-qe\..*$' exclude: false - filtertype: age source: name direction: older timestring: '%Y.%m.%d' unit: days unit_count: 1 ...... 2021-01-22 09:00:05,535 INFO curator.cli run:148 Preparing Action ID: 1, "delete_indices" ... 2021-01-22 09:00:06,055 DEBUG curator.cli process_action:93 Running "DELETE_INDICES" 2021-01-22 09:00:06,055 DEBUG curator.indexlist __get_indices:65 Getting all indices 2021-01-22 09:00:06,075 DEBUG curator.utils get_indices:648 Detected Elasticsearch version 6.8.1 2021-01-22 09:00:06,076 DEBUG curator.utils get_indices:650 All indices: ['infra-000190', 'infra-000197', 'app-000039', 'app-000030', 'infra-000205', 'infra-000193', 'infra-000192', 'audit-000261', 'app-000038', 'infra-000195', '.kibana_-377444158_kubeadmin', 'app-000024', 'app-000020', 'infra-000188', 'app-000027', 'audit-000251', 'infra-000203', 'audit-000260', 'app-000023', 'infra-000206', '.security', 'app-000040', 'app-000018', 'infra-000200', 'audit-000256', 'infra-000191', 'app-000036', 'infra-000204', 'app-000017', 'infra-000198', 'audit-000258', 'infra-000187', 'audit-000257', 'infra-000189', 'app-000019', 'app-000028', 'app-000034', 'infra-000196', 'app-000037', 'audit-000254', 'audit-000250', 'app-000029', 'app-000025', 'infra-000194', '.kibana_1', 'audit-000262', 'audit-000263', 'app-000032', 'app-000031', 'app-000021', 'audit-000259', 'audit-000255', 'infra-000201', 'infra-000202', 'audit-000253', 'app-000026', 'app-000022', 'app-000035', 'audit-000252', 'app-000033', 'infra-000199'] ... 2021-01-22 09:00:05,972 DEBUG curator.indexlist filter_by_age:490 Filtering indices by age 2021-01-22 09:00:05,972 DEBUG curator.indexlist _get_name_based_ages:280 Getting ages of indices by "name" 2021-01-22 09:00:05,972 DEBUG curator.indexlist empty_list_check:226 Checking for empty list 2021-01-22 09:00:05,972 INFO curator.cli run:202 Skipping action "delete_indices" due to empty list: <class 'curator.exceptions.NoIndices'> 2021-01-22 09:00:05,972 INFO curator.cli run:223 Action ID: 1, "delete_indices" completed. 2021-01-22 09:00:05,972 DEBUG curator.cli run:128 action_disabled = False 2021-01-22 09:00:05,972 DEBUG curator.cli run:132 continue_if_exception = False 2021-01-22 09:00:05,972 DEBUG curator.cli run:134 timeout_override = 300 2021-01-22 09:00:05,972 DEBUG curator.cli run:136 ignore_empty_list = True 2021-01-22 09:00:05,972 DEBUG curator.cli run:138 allow_ilm_indices = False 2021-01-22 09:00:05,973 INFO curator.cli run:148 Preparing Action ID: 2, "delete_indices"
Created attachment 1751131 [details] Curator DEBUG logs oc set env cronjob/curator CURATOR_LOG_LEVEL=DEBUG CURATOR_SCRIPT_LOG_LEVEL=DEBUG
Created attachment 1751132 [details] ES index format as - app,infra,audit elasticsearch index format sample. in 4.7
Created attachment 1751133 [details] Notes from RH docs about index in ocp version for "user defined project"
Curator has been deprecated in favor of the new indexmanagement jobs and will be removed all-together in 4.7/OpenShift Logging 5.0