Skip to content

feat(beetmover): Bug 1943594 - Support partner build numbers in push-to-releases - #1550

Open
hneiva wants to merge 1 commit into
masterfrom
hneiva/bug-1943594-partner-build-numbers
Open

hneiva wants to merge 1 commit into
masterfrom
hneiva/bug-1943594-partner-build-numbers

Conversation

@hneiva

@hneiva hneiva commented Oct 27, 2026

Copy link
Copy Markdown
Contributor

push-to-releases only copied partner repacks from the v1 partner build directory, so off-cycle partner respins (e.g. v202501231439) were silently excluded.

Add an optional partner_build_numbers payload field mapping "partner/subpartner" to an integer >= 1. Partners missing from the map, or payloads without the field, keep using partner build number 1. Both the S3 and GCS push paths log the build number used for each partner and warn when a requested partner has no files under its candidates prefix.

@hneiva
hneiva requested a review from a team as a code owner October 27, 2026 21:33
@hneiva

hneiva commented Oct 28, 2026

Copy link
Copy Markdown
Contributor Author

Example payload:

partner_build_numbers is optional. It maps a "partner/subpartner" from partners to the partner build number to push from. Partners not in the map, or payloads without the field, use 1.

"partners": ["smi/smi-001", "acer/acer-002"],
"partner_build_numbers": {
    "smi/smi-001": 202501231439
}

With version: "134.0" and build_number: 2:

Partner Copied from (candidates/134.0-candidates/build2/partner-repacks/…) Copied to
smi/smi-001 smi/smi-001/v202501231439/ releases/partners/smi/smi-001/134.0/
acer/acer-002 acer/acer-002/v1/ (default) releases/partners/acer/acer-002/134.0/

@bhearsum bhearsum left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Before this is reviewed, it would be good to understand how the Gecko and/or Ship It side is going to work. This does look like it might work correctly for a manual ship_firefox run as part of off cycle repacks, because we can stick whatever we want in the parameters.

This doesn't look like it would help in the case where we have the following sequence of operations:

  1. promote_firefox runs, creating initial partner repacks
  2. an off cycle partner repack is created
  3. ship_firefox is run via Ship It

In that situation, Ship It has no knowledge of the off cycle partner repack, and thus will not be able to put the appropriate information in the parameters.

It's also unclear to me if this will remove the need for the horrible hack of running and then cancelling a bunch of stuff (per https://bugzilla.mozilla.org/show_bug.cgi?id=1943594#c3) when running this just for an off cycle partner repack.

I'm not opposed to initially landing a fix just for the manual-ship_firefox-for-offcycle case, but I would prefer that we at least try to architect the full solution before implementing piecemeal fixes that might cause issues or need to be reverted to fix related issues.

…to-releases

push-to-releases only copied partner repacks from the v1 partner build
directory, so off-cycle partner respins (e.g. v202501231439) were silently
excluded.

Add an optional `partner_build_numbers` payload field mapping
"partner/subpartner" to an integer >= 1. Partners missing from the map, or
payloads without the field, keep using partner build number 1. Both the S3
and GCS push paths log the build number used for each partner and warn when
a requested partner has no files under its candidates prefix.
@hneiva
hneiva force-pushed the hneiva/bug-1943594-partner-build-numbers branch from 915ce2b to 046346d Compare November 6, 2026 17:51
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants