[SERVER-77066] Acquisitions should be the ones establishing a snapshot unless we are reading atPointInTime or in a multiDoctransaction Created: 12/May/23  Updated: 13/Jun/23  Resolved: 13/Jun/23

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

Type: Improvement Priority: Major - P3
Reporter: Daniel Gomez Ferro Assignee: Daniel Gomez Ferro
Resolution: Won't Fix Votes: 0
Labels: PM-2144-Milestone-0
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Assigned Teams:
Storage Execution
Sprint: Execution EMEA Team 2023-06-26
Participants:

 Description   

We need to perform the shardingPlacementCheck before opening the snapshot to guarantee consistency with the sharding state, this can only be disregarded if we are reading atPointInTime.

"Recursive" acquisitions can only be done on a namespace that isNamespaceAlwaysUnsharded()



 Comments   
Comment by Daniel Gomez Ferro [ 13/Jun/23 ]

This is a problem as shown in SERVER-77506 but we can't enforce it because non-snapshot multidoc transactions rely on this.

Generated at Thu Feb 08 06:34:26 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.