[SERVER-34424] a changeStream on a database will incorrectly notify that unsharded collections in the database have been dropped after movePrimary Created: 11/Apr/18  Updated: 31/Jul/23

Status: Open
Project: Core Server
Component/s: Sharding
Affects Version/s: 3.7.3
Fix Version/s: None

Type: Bug Priority: Major - P3
Reporter: Esha Maharishi (Inactive) Assignee: Backlog - Query Execution
Resolution: Unresolved Votes: 0
Labels: movePrimary, sharding-common-backlog
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Related
related to SERVER-72099 Writes done as part of movePrimary sh... Backlog
is related to SERVER-34462 changeStreams will incorrectly notify... Open
Assigned Teams:
Query Execution
Operating System: ALL
Sprint: Sharding EMEA 2023-04-17
Participants:

 Description   

One fix for this would be to mark "dropCollections" done on all the unsharded collections in the database at the end of movePrimary as "fromMigrate", so that they are ignored by changeStreams.

Also, this problem would go away if unsharded collections were treated as sharded collections with one chunk.



 Comments   
Comment by Esha Maharishi (Inactive) [ 01/Jul/19 ]

andre.spiegel, just curious, are they trying to movePrimary because they are removing a shard?

Comment by Andy Schwerin [ 20/Apr/18 ]

Moving to sharding team Needs Triage. Either of the two solutions in the description make sense, and are projects in the sharding catalog. This is not urgent, given the infrequency of movePrimary.

Comment by Andy Schwerin [ 13/Apr/18 ]

This is probably yacceptable until we implement changes to allow whole-database changestreams to survive individual collection drops.

Generated at Thu Feb 08 04:36:38 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.