New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 592667 link

Starred by 1 user

Issue metadata

Status: Duplicate
Merged: issue 620731
Owner: ----
Closed: Jan 2017
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Feature



Sign in to add a comment

When recipe-roller creates a new roll, it should close all previous pending ones

Project Member Reported by aga...@chromium.org, Mar 7 2016

Issue description

What steps will reproduce the problem?
1. Land a non-trivial recipe change
2. Let recipe-roller generate a review in another repo
3. Ignore that review for a while
4. Land another change
5. Let recipe-roller generate another review

What is the expected output? What do you see instead?
Recipe-roller should close the previous review, now that it has been replaced by the more recent one. Instead, you just get a ton of nearly-identical emails in a row and you have to inspect them carefully to figure out which one should actually be reviewed.

Please use labels and text to provide additional information.

 
Labels: Infra-Recipes
There are some known issues around this kind of behavior. Any CL links you could provide which show this more explicitly?
In the future we should effectively be landing one change at a time, right? So each should be queued, landing over the previous one after downstream is caught up.
I believe this is a different mode.

The recipe roller should close the old issues, and post a comment about moving to the new CL. That's not really possible with the current design, though, because it doesn't have any idea of continuity, but only the raw diff, I think. But this would be very nice to have for the user.

Comment 5 by aga...@chromium.org, Apr 26 2016

Components: Infra>Platform>Recipes
Labels: -Infra-Recipes

Comment 6 by benhenry@google.com, Nov 23 2016

Labels: -Type-Bug Type-Feature
Cc: phajdan.jr@chromium.org
Mergedinto: 620731
Status: Duplicate (was: Available)
I believe this is fixed with https://chromium-review.googlesource.com/c/371918/ .

We could do some further refinements like mentioned here, such as posting comments on old CLs if needed.

Sign in to add a comment