Unified Manager cluster discovery stuck due to unresponsive Zapi calls
Applies to
- ActiveIQ Unified Manager(AIQUM) 9.x
- All OS Platforms
- ZAPI acquisition
Issue
- Cluster discovery stuck for one cluster in AIQUM showing Health Poll status in Progress
- Initial Discovery never finishes for a newly added cluster
au.log/log_netappfoundation_<cluster_IP>_<DS_ID>_<Date_and_time>_xxx_sample.logunder recordings shows[netappfoundation] <New_CLUSTER IP> - while executing ZAPIs on datasource: <CLUSTER IP> IP: <CLUSTER IP> for ZAPI: <ZAPI>, java.io.EOFExceptionfor the cluster attempting initial discovery- Extensive execution time and returned records for different ZAPI calls in foundation poll recordings for existing clusters under
log_netappfoundation_xxxx_iterator_response_time.txt - Some of the known ZAPI calls exhibiting the behavior are:
job-schedule-get-iterjob-schedule-cron-get-iterquota-report-iterqtree-list-itersnapmirror-history-get-iter- Example of error:
datasource, zapi, timestamp, max-records, iterations, total-clock-time, total-zapi-time, total-invocations, total-records, returned-records-listxx.xx.xx.xx<Cluster_IP>, snapmirror-history-get-iter, Fri Dec 03 06:58:17 TRT 2021, 1000, 8216, 10052702, 9745756, 8216, 8215035, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000,....[truncated]