[SERVER-15857] Enhance repairCursor command for mmapv1 Created: 24/Apr/14  Updated: 06/Dec/22  Resolved: 14/Sep/18

Status: Closed
Project: Core Server
Component/s: MMAPv1, Storage
Affects Version/s: None
Fix Version/s: None

Type: Improvement Priority: Minor - P4
Reporter: Ian Whalen (Inactive) Assignee: Backlog - Storage Execution Team
Resolution: Won't Fix Votes: 0
Labels: mongodump, repair
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Assigned Teams:
Storage Execution
Participants:

 Description   

This is a ticket for multiple proposed extensions to mongodump --repair:

  • Currently, it traverses the list of extents front to back and then back to front. Within each extent, the same back and forth iteration should happen for the records for the case in which the intra-extent record chain is broken.
  • De-duplication should be done across the two stages, otherwise we are returning possibly 2x the number of records.
  • Possibly search for the magic extent number to find unreachable extents.


 Comments   
Comment by Michael O'Brien [ 29/Oct/14 ]

Going to move this back to the SERVER project. Rationale: mongodump in 2.7.8+ works by calling the repairCursor command against a mongod, so this change would need to be implemented in server code rather than tools code - additionally, the implemenation of repair referenced in the description is specific to the mmapv1 storage engine, and different storage engines may need to implement repairCursor differently.

Generated at Thu Feb 08 03:39:12 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.