Record bucket 'Add To Bucket' actions should be batched or serialized.

Bug #1913458 reported by Bill Erickson
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Fix Released

Bug Description

Evergreen 3.6

Along the lines of bug #1896285, the "Add To Bucket" action is blasting servers with "" API calls when the contents to be added are a large list.

Calls should be serialized (at minimum) or batched (ideal) to resolve the issue and avoid another area in the code where too many parallel requests are sent.

On a related note, suggestions for a new LP tag to group these?

Revision history for this message
Bill Erickson (berick) wrote :

Code pushed:;a=shortlog;h=refs/heads/user/berick/lp1913458-add-to-bucket-batching

This addresses both Add To Bucket and Remove From Bucket actions. They are now batched/streaming API calls that require clients send only identifiers instead of fieldmapper objects for an added bit of efficiency.

Changed in evergreen:
milestone: none → 3.6.2
assignee: Bill Erickson (berick) → nobody
tags: added: pullrequest
tags: added: parallel-requests
Changed in evergreen:
milestone: 3.6.2 → 3.6.3
Revision history for this message
Jeff Davis (jdavis-sitka) wrote :

Works for me. Working branch user/jeffdavis/lp1913458-add-to-bucket-batching-signoff has a signoff.

tags: added: signedoff
Changed in evergreen:
assignee: nobody → Jane Sandberg (sandbej)
Revision history for this message
Jane Sandberg (sandbergja) wrote :

Works for me too. Thanks, Bill and Jeff. Pushed to master and rel_3_6.

Changed in evergreen:
importance: Undecided → Medium
status: New → Fix Committed
assignee: Jane Sandberg (sandbej) → nobody
Changed in evergreen:
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers