[SERVER-61467] Record temporary disk usage by queries, index builds, etc. Created: 13/Nov/21  Updated: 18/May/23

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

Type: Improvement Priority: Major - P3
Reporter: Bruce Lucas (Inactive) Assignee: Backlog - Query Execution
Resolution: Unresolved Votes: 2
Labels: needs-triage
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Depends
Related
related to SERVER-63675 Track usedDisk in serverStatus Investigating
Assigned Teams:
Query Execution
Participants:
Case:

 Description   

When diagnosing issues related to i/o capacity it may be necessary to account for all i/o generated by mongod. Currently we monitor in FTDC disk usage for WT data files and WT journal, and can easily compute mongod log usage.

However we don't record i/o volume (bytes written and read) for temporary files used by queries for sorting, index builds, etc. Can we record this in FTDC? In order to be useful this would need to be recorded as a running total as the query proceeds (as opposed to simply bumping a counter at the end of the operation).



 Comments   
Comment by Steven Vannelli [ 18/Oct/22 ]

Reopening this ticket since the PM-2247 has been reopened.

Comment by Steve La (Inactive) [ 25/Feb/22 ]

Given this is more a new feature request, I recommend this goes through query-product-triage so we can prioritize this with every other things we may want to do post 7.0
If we do this, I would suggest to include it in our roadmap, stream-view ( and potentially turn it into a project ) given cross team impact
joe.sack FYI

Generated at Thu Feb 08 05:52:29 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.