[SERVER-48680] Collection drop may fail in migration_fails_if_exists_in_rangedeletions.js test due to concurrent chunk migration Created: 09/Jun/20  Updated: 29/Oct/23  Resolved: 30/Jun/20

Status: Closed
Project: Core Server
Component/s: Sharding
Affects Version/s: None
Fix Version/s: 4.7.0

Type: Bug Priority: Minor - P4
Reporter: Max Hirschhorn Assignee: Matthew Saltz (Inactive)
Resolution: Fixed Votes: 0
Labels: sharding-wfbf-day
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Depends
Backwards Compatibility: Fully Compatible
Operating System: ALL
Sprint: Sharding 2020-06-29
Participants:
Linked BF Score: 25

 Description   

The migration_fails_if_exists_in_rangedeletions.js test causes a chunk migration to block forever due to the presence of an overlapping range to clean up and the range deleter being paused using a failpoint on the recipient shard. Disabling the suspendRangeDeletion failpoint allows the chunk migration to proceed. If the ongoing chunk migration doesn't complete quickly enough, then the drop command to clean up the sharded collection before running the next test scenario will fail with a LockBusy error response.

Re-running the moveChunk command which had its maxTimeMS expire after the suspendRangeDeletion failpoint was disabled would ensure the chunk migration is no longer in progress before the drop command is run.



 Comments   
Comment by Githook User [ 29/Jun/20 ]

Author:

{'name': 'Matthew Saltz', 'email': 'matthew.saltz@mongodb.com', 'username': 'saltzm'}

Message: SERVER-48680 Retry moveChunk request after failed attempt
Branch: master
https://github.com/mongodb/mongo/commit/4eacc5768f3967882fe5edeeb7ec2739bb95a1a8

Generated at Thu Feb 08 05:17:48 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.