[SERVER-70215] fsm_workload_helpers/kill_session.js Can Fail with Unexpected Error Codes Created: 04/Oct/22  Updated: 12/Dec/23

Status: Backlog
Project: Core Server
Component/s: Sharding
Affects Version/s: None
Fix Version/s: None

Type: Bug Priority: Major - P3
Reporter: Brett Nawrocki Assignee: Backlog - Cluster Scalability
Resolution: Unresolved Votes: 0
Labels: cs-subteam1, sharding-nyc-subteam1
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Depends
Assigned Teams:
Cluster Scalability
Operating System: ALL
Participants:
Linked BF Score: 3
Story Points: 2

 Description   

fsm_workload_helpers/kill_session.js's killSession function begins by refreshing the logical session cache, expecting that this command may fail during certain circumstances. However, it is also possible that this command will fail with NoProgressMade (and possibly other errors) if the writes to the session collection are killed by another session killer during the refresh (see this comment on BF-25968).

killSession should either consider additional error codes to be acceptable, or perhaps retry at least once in case of an error.



 Comments   
Comment by Max Hirschhorn [ 24/Oct/22 ]

One thought would be to set TestData.disableImplicitSessions to avoid the refreshLogicalSessionCacheNow and $listSessions aggregation from including an lsid in their request. This way they wouldn't be candidates to get interrupted by the killSession() function itself.

Generated at Thu Feb 08 06:15:35 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.