[SERVER-50806] repeated finds on nested data without index causes accumulating return times Created: 09/Sep/20  Updated: 27/Oct/23  Resolved: 10/Sep/20

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

Type: Bug Priority: Major - P3
Reporter: Philipp Kunz Assignee: Dmitry Agranat
Resolution: Works as Designed Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Operating System: ALL
Participants:

 Description   

When repteadly querying for nested data like this

    db.inventory.find( { "size.uom": "in" } )

the return always adds a dew seconds until basically becoming unusable.



 Comments   
Comment by Dmitry Agranat [ 10/Sep/20 ]

Thank you phil@lossless.com for letting us know that the issue was in the ODM, I will go ahead and close this case.

Comment by Philipp Kunz [ 10/Sep/20 ]

This can be closed.

Comment by Philipp Kunz [ 10/Sep/20 ]

Sorry for the confusion, somehow the pipeline is not public, although settings say it should be.

Anyway I found the issue as actually being the result of the way the decorator worked. I fixed the problem in the ODM.

Comment by Daniel Pasette (Inactive) [ 09/Sep/20 ]

Hi Philipp
That link is giving me a 404. Can you check it?

Comment by Philipp Kunz [ 09/Sep/20 ]

You can see the odd behaviour here: https://gitlab.com/pushrocks/smartdata/-/jobs/728181329

Test 3 builds a simple collection of 2000 Cars

Test 4 queries a top-level non indexed property, showing consistent return times

and Test 5 queries a nested non indexed property exposing the odd behaviour.

Generated at Thu Feb 08 05:23:38 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.