[SERVER-74094] Support multiple namespaces in bulkWrite sub-batching logic Created: 16/Feb/23  Updated: 17/Feb/23  Resolved: 17/Feb/23

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

Type: Task Priority: Major - P3
Reporter: Lingzhi Deng Assignee: Lingzhi Deng
Resolution: Duplicate Votes: 0
Labels: milestone-2
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Depends
depends on SERVER-72787 Implement sub-batching logic for bulk... Closed
Duplicate
duplicates SERVER-74095 Support multiple namespaces in bulkWr... Closed
Participants:

 Description   

SERVER-72787 added sub-batching logic for bulkWrite and reused most of the targeting logic from BatchWriteOp. But one thing we missed was that the target function group targeted batches based on ShardEndpoint (which is a struct with shardId and shard/db version). But since bulkWrite support batched writes against multiple namespaces, targeted writes for different namespaces could have different ShardEndpoints (based on EndpointComp) even if they target the same shard. Therefore, we would end up mistakenly batching writes against different namespaces into separate batches if even they target the same shard.

This ticket is to fix that and allow sub-batching writes of multiple namespaces as long as they target the same shard (regardless of the endpoint comparisons).



 Comments   
Comment by Lingzhi Deng [ 17/Feb/23 ]

Looks like somehow JIRA created two identical tickets

Generated at Thu Feb 08 06:26:29 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.