Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

Agenda Topics:

Problem statement:

Timeout issue in SO during the VNF resource instantiation 

This is found in the basic_vm, basic_cnf flows (rarely in the basic_network too) and is random in nature.

Possible cause:

Maria DB has been shared wit other components today. This could have induced issues around the inventory access (critical section / semaphores...)

There could be an issue around the timing of the db entry vs the actual call back.

This issue is not found in the CSIT.

Possible Way outs:

  1. The shared DB scaleout to one pod instead of the Gallera being on the mariaDB cluster 3 pods. To corner to problem. This could be tested on a snapshot version.
    1. https://docs.camunda.org/manual/latest/user-guide/process-engine/database/mariadb-galera-configuration/  configuration of mariadb needed for testing the fix. (Sylvain)
  2. Deploy the CSIT on the oom (currently its on the docker), can help us test the actual load of SO (to a greater extent). (Byung, Irshad).
  3. Minor improvements Like the aaf user onap@so.user printed in the logs taking some CPU cycles. (Seshu / Mukesh / Ramesh / Irshad  to examine this further).


H release:

Jira Legacy
serverSystem Jira
columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
maximumIssues20
jqlQueryproject = "Service Orchestrator" AND status != Closed AND ( priority = High OR priority = Highest ) AND status != Done AND issuetype = Bug ORDER BY priority DESC
serverId4733707d-2057-3a0f-ae5e-4fd8aff50176

...