[SERVER-46880] Disable multiversion and multiversion_auth tests in evergreen Created: 16/Mar/20  Updated: 06/Dec/22  Resolved: 16/Mar/20

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

Type: Task Priority: Major - P3
Reporter: Vamsi Boyapati Assignee: Backlog - Storage Execution Team
Resolution: Won't Do Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Assigned Teams:
Storage Execution
Participants:

 Description   

Currently we are working on the solution for upgrade/downgrade for durable history. We need to control what state the database is in when going from 4.4 to 4.2 (it needs to have had a clean shutdown) and between 4.4 <-> 4.0 ( in this case the database ends up in corruption). There are more cases than we anticipated. While we are working on this solution, we would like to efficiently use our time in triaging as well, so it is better we could cut down the noise from the known failures.

In this direction we would like to disable multiversion & multiversion_auth tests from evergreen.



 Comments   
Comment by Eric Milkie [ 16/Mar/20 ]

I don't think we should disable the tests – instead, we'll put a freeze on the 4.4 branch itself until you can get them fixed. We're only backporting test fixes now anyway, so there is nothing that can't wait until later this week to go in.

Comment by Luke Chen [ 16/Mar/20 ]

The set of multiversion tasks failed in the wiredtiger drop commit build:

  • multiversion
  • multiversion_auth
  • replica_sets_multiversion
  • replica_sets_jscore_multiversion_passthrough
  • rollback_multiversion_fuzzer
  • sharding_jscore_multiversion_passthrough
  • sharding_multiversion
  • concurrency_sharded_replication_multiversion
  • jstestfuzz_replication_multiversion
Comment by Alexander Gorrod [ 16/Mar/20 ]

milkie Do you think this is the right approach? If so do you think someone on your team could make this change for us?

We need to get the scenarios passing for the rc0 - so the changes should be in within the next week, I'm not sure if the test noise from a failing test for that long justifies disabling it or who is in a good position to make that call.

Generated at Thu Feb 08 05:12:42 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.