[SERVER-63675] Track usedDisk in serverStatus Created: 15/Feb/22  Updated: 10/Mar/23

Status: Investigating
Project: Core Server
Component/s: None
Affects Version/s: 5.2.0
Fix Version/s: None

Type: Improvement Priority: Major - P3
Reporter: Kevin Arhelger Assignee: Chris Harris
Resolution: Unresolved Votes: 0
Labels: diagnostics
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Related
is related to SERVER-61467 Record temporary disk usage by querie... Open
Participants:
Case:

 Description   

This request is for two query related counters in Server Status. These names are just suggestions.

usedDisk: Incremented when queries or aggregations wrote data to temporary files.
failedDisk: Incremented when queries or aggregations cannot complete due to reaching the sorting or aggregation memory limit. (Queries without allowDiskUse:true exceeding 100MB sorting limit for example).



 Comments   
Comment by Kevin Arhelger [ 15/Feb/22 ]

bruce.lucas,

I would find it helpful to have a breakout of read and write operations and bytes written / read as these operations progress. That should provide a more complete picture of IO activity for a given mongodb instance.

Per operation io statistics for tmp would also be helpful in the slow query message and profiler.

Comment by Bruce Lucas (Inactive) [ 15/Feb/22 ]

I wonder if it would be also be helpful to have a field that tracks the amount of disk used. Ideally for such a metric to be useful I think it would need to track the usage as the query is running, so that it doesn't show up as just a large bump at the end of a long-running query, making it hard to correlate with system disk usage metrics.

Generated at Thu Feb 08 05:58:22 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.