[SERVER-70909] Avoid range deletions while draining a shard that is being removed Created: 27/Oct/22  Updated: 26/Oct/23

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

Type: Task Priority: Major - P3
Reporter: Tommaso Tocci Assignee: Backlog - Catalog and Routing
Resolution: Unresolved Votes: 0
Labels: oldshardingemea, range-deleter-improvements, sharding-emea-pm-iteration-planning
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Depends
depends on SERVER-77359 Allow enabling/disabling the range de... Backlog
Assigned Teams:
Catalog and Routing
Participants:
Case:

 Description   

When removeShard is called the balancer starts migrating all the data for sharded collections out of that shard. The chunks are migrated through the moveRange command and  when migrations complete the range deleter kick in on the shard being removed to delete the donated ranges.
The range deletions procedure can consume a non-negligible amount of resource on the node and it can affect significantly the throughput of concurrent outgoing migrations.

Considering that the shard being removed will be discarded as soon as the draining process complete, there is no reason to actually perform range deletions and garbage collect the orphaned ranges.


Generated at Thu Feb 08 06:17:29 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.