842741 - wrong release config bumped during Thunderbird 17.0.3

Status: RESOLVED FIXED

(Release Engineering :: Release Automation: Other, defect)

Description

[[e-mail] (:bhearsum)]

Related to bug 815302, we bumped comm-esr17's release config rather than comm-release's.

Comment 1

[Rail Aliiev [:rail]]

Since 17.0.3 and 17.0.3esr use the same branch,
http://hg.mozilla.org/build/tools/file/6efdaa82df4d/buildfarm/release/release-runner.py#l321 bumps the same files. The only difference coming from ship it is the name of release.

We can add a work around to detect "esr" and only then use esr config, but it sounds error prone.

We can also add another field to ship-it to override release config file (and forget to change it later :) ).

Or we can use comm-release repo for releases. :)

Comment 2

[[e-mail] (:bhearsum)]

> We can add a work around to detect "esr" and only then use esr config, but
> it sounds error prone.

I don't think this would work....esr already works fine. It's non-esr releases building from esr that have issues.

(In reply to Rail Aliiev [:rail] from comment #1)
> We can also add another field to ship-it to override release config file
> (and forget to change it later :) ).

I don't think this is a good idea. It would let you do really dumb things (like use mozilla-beta release configs for a final release) that I don't know how to guard against.

Comment 3

[[e-mail] (:bhearsum)]

Attachment: horrible hack

I wracked my brain trying to find a better fix for this, but the simple fact is that our in the rapid release world, our release automation assumes a 1:1 relationship between branches and releases. I wish we'd remembered this before implementing the Thunderbird-releases-out-of-esr17 :(.

We should take some time to think about whether or not this is a scenario that's worthwhile to support in the long term. If it is, we should make some higher level changes to the automation (like not putting branch names in config file names).

In the meantime, this should get us back into a working state, and also pain me greatly.

Comment 4

[Rail Aliiev [:rail]]

Comment on attachment 717181 [details] [diff] [review]
horrible hack

This exactly what I thought :)
For esr24 we shouldn't use the same repo