1129550 - Stop generating aus2 snippets

Status: RESOLVED FIXED

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

Comment 1

[Rail Aliiev [:rail]]

Attachment: buildbotcustom-no-snippets.diff

Extracted from win64 related staging work

Comment 2

[Rail Aliiev [:rail]]

Comment on attachment 8559285 [details] [diff] [review]
buildbotcustom-no-snippets.diff

nm this, it's better to get rid of the code

Comment 3

[Nick Thomas [:nthomas] (UTC+12)]

See also bug 933426.

Comment 4

[[e-mail] (:bhearsum)]

Attachment: switch update server url + switch off snippet uploading everywhere

Comment 5

[Rail Aliiev [:rail]]

Comment on attachment 8559957 [details] [diff] [review]
switch update server url + switch off snippet uploading everywhere

I don't think this will work. We have a bunch of places in factory.py using this variable.

Don't worry about it, this variable will be gone with the patch I'm testing

Comment 6

[[e-mail] (:bhearsum)]

(In reply to Rail Aliiev [:rail] from comment #5)
> Comment on attachment 8559957 [details] [diff] [review]
> switch update server url + switch off snippet uploading everywhere
> 
> I don't think this will work. We have a bunch of places in factory.py using
> this variable.
> 
> Don't worry about it, this variable will be gone with the patch I'm testing

I don't think this is true for releases. It's used a lot in nightlies, but only in ReleaseUpdatesFactory for releases AFAICT.

Comment 7

[[e-mail] (:bhearsum)]

Attachment: don't upload snippets + remove dead code

I verified this with checkconfig, it has some risk, but not a ton...

Comment 8

[Rail Aliiev [:rail]]

Comment on attachment 8559957 [details] [diff] [review]
switch update server url + switch off snippet uploading everywhere

these 2 patches will work together

Comment 9

[[e-mail] (:bhearsum)]

Comment on attachment 8560520 [details] [diff] [review]
don't upload snippets + remove dead code

Oops, wrong patch.

Comment 10

[[e-mail] (:bhearsum)]

Attachment: more comprehensive buildbotcustom patch

Comment 11

[Rail Aliiev [:rail]]

Comment on attachment 8560522 [details] [diff] [review]
more comprehensive buildbotcustom patch

Review of attachment 8560522 [details] [diff] [review]:
-----------------------------------------------------------------

looks good to me.

Comment 12

[[e-mail] (:bhearsum)]

Should be all done here. No more snippets for future releases!

Comment 13

[Rail Aliiev [:rail]]

I'm going to keep this open to delete moaar code! Not today though :)

Comment 14

[Rail Aliiev [:rail]]

Attachment: no_aus-buildbotcustom-3.diff

* passed test-masters.sh
* the list of builders is the same
* dump_masters.sh shows the following deletions for a build master: https://gist.github.com/rail/36005eda2408e130771b

Comment 15

[Rail Aliiev [:rail]]

Attachment: no_aus-buildbot-configs-1.diff

Comment 16

[Rail Aliiev [:rail]]

Attachment: no_aus-tools.diff

Comment 17

[Rail Aliiev [:rail]]

Attachment: no_aus-mozharness.diff

Comment 18

[[e-mail] (:bhearsum)]

Comment on attachment 8561573 [details] [diff] [review]
no_aus-buildbotcustom-3.diff

Review of attachment 8561573 [details] [diff] [review]:
-----------------------------------------------------------------

I'm r+'ing these patches based on visual inspection + dump master output. I'd really really like to see this tested in staging though - these are pretty risky changes, and they're only validated by nightly builds.

Comment 19

[Rail Aliiev [:rail]]

Attachment: no_aus-buildbotcustom-android.diff

This is a follow up for attachment 8561573 [details] [diff] [review] to fix the android issue. It turned out that the script I deleted sets some properties in http://hg.mozilla.org/build/tools/file/a7c2695e5ed7/scripts/android/android_snippet.py#l249 calculated by http://hg.mozilla.org/build/tools/file/a7c2695e5ed7/scripts/android/android_snippet.py#l155

I replaced them with in-line and renamed the method to match the other one in factory.py.

Android builds look fine now, waiting for windows...

Comment 20

[Jordan Lund (:jlund)]

Comment on attachment 8561578 [details] [diff] [review]
no_aus-mozharness.diff

Review of attachment 8561578 [details] [diff] [review]:
-----------------------------------------------------------------

\o/\o/ long live balrog

Comment 21

[Rail Aliiev [:rail]]

Comment on attachment 8562448 [details] [diff] [review]
no_aus-buildbotcustom-android.diff

See comment 19, windows worked fine, both win32 (mozharness based) and win64 (MBF based). I additionally ran second win32 build to make sure it works fine with partials. L10N jobs worked fine too.

The only build type I had troubles to test was Android nightly L10N. For some reason it was trying to upload files to stage.m.o.

Comment 22

[Rail Aliiev [:rail]]

If you want to take a look at the staging results, they are still at http://dev-master1.srv.releng.scl3.mozilla.com:8033/builders

Comment 23

[Rail Aliiev [:rail]]

Comment on attachment 8561578 [details] [diff] [review]
no_aus-mozharness.diff

https://hg.mozilla.org/build/mozharness/rev/44e2e5f6a0ad

Comment 24

[Rail Aliiev [:rail]]

Comment on attachment 8561575 [details] [diff] [review]
no_aus-buildbot-configs-1.diff

https://hg.mozilla.org/build/buildbot-configs/rev/69ffb581b00c

Comment 25

[Rail Aliiev [:rail]]

Comment on attachment 8561573 [details] [diff] [review]
no_aus-buildbotcustom-3.diff

https://hg.mozilla.org/build/buildbotcustom/rev/0ebc46469595

Comment 26

[Rail Aliiev [:rail]]

Comment on attachment 8562448 [details] [diff] [review]
no_aus-buildbotcustom-android.diff

combined with the first patch 
https://hg.mozilla.org/build/buildbotcustom/rev/0ebc46469595

Comment 27

[Rail Aliiev [:rail]]

The tools patch will be landed after a reconfig (it deletes some files used in production)

Comment 28

[Rail Aliiev [:rail]]

Comment on attachment 8561577 [details] [diff] [review]
no_aus-tools.diff

https://hg.mozilla.org/build/tools/rev/a0ac6a4b2646

Comment 29

[Rail Aliiev [:rail]]

No more tears^W snippets!