[SERVER-85636] Poor FTDC compression due to apparent schema change observed on 7.3.0-alpha1 Created: 23/Jan/24  Updated: 24/Jan/24  Resolved: 24/Jan/24

Status: Closed
Project: Core Server
Component/s: None
Affects Version/s: None
Fix Version/s: 8.0.0-rc0, 7.3.0-rc1

Type: Bug Priority: Major - P3
Reporter: John Morales Assignee: Henrik Edin
Resolution: Fixed Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Backports
Problem/Incident
is caused by SERVER-80290 Change WiredTigerUtil::exportTableToB... Closed
Related
related to SERVER-85654 Create test to prevent FTDC compressi... Open
Assigned Teams:
Storage Execution
Backwards Compatibility: Fully Compatible
Operating System: ALL
Backport Requested:
v7.3
Sprint: Execution Team 2024-02-05
Participants:

 Description   

Environment

  • MongoDB 7.3.0-alpha1
  • Single node replica set
  • macOS

Happened to observe that a new 10MB FTDC file was being rolled every 8 minutes instead of the more typical 4-8 hours.

Based on some internal discussion this is thought to be caused by frequent schema changes brought on by a separate memory optimization.

This is apparently trivially reproducible on linux. Example logs from a linux repro:

"full-time diagnostic data capture schema change: field name change","attr":{"from":"block-manager","to":"cursor"}}
"full-time diagnostic data capture schema change: field name change","attr":{"from":"cursor","to":"block-cache"}}



 Comments   
Comment by Henrik Edin [ 24/Jan/24 ]

Fixed by reverting SERVER-80290

Generated at Thu Feb 08 06:58:14 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.