[SERVER-30817] Make full validate block until a new majority-confirmed checkpoint is created Created: 24/Aug/17  Updated: 06/Dec/22  Resolved: 04/Feb/19

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

Type: Bug Priority: Major - P3
Reporter: Geert Bosch Assignee: Backlog - Storage Execution Team
Resolution: Duplicate Votes: 0
Labels: neweng
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Duplicate
duplicates SERVER-29494 WT validate should wait for explicit ... Closed
Assigned Teams:
Storage Execution
Operating System: ALL
Participants:

 Description   

When doing verifyTable on a table with writes since the last checkpoint, WiredTiger will implicitly create a new checkpoint. However, with recovery to a timestamp no implicit checkpoints are allowed. So, the validate command needs to be changed to do the following:

1. Lock the collection to validate in MODE_S
2. Signal the checkpointing thread that a new checkpoint should be created
3. Block until the new checkpoint is created

This ensures that when we call verifyTable, no dirty data exists and no implicit checkpoint will be created.



 Comments   
Comment by Daniel Gottlieb (Inactive) [ 02/Sep/17 ]

geert.bosch Is this ticket looking for the same functionality as SERVER-29494?

Generated at Thu Feb 08 04:25:06 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.