+++ This bug was initially created as a clone of Bug #1224861 +++ Since Z3c L port will rely on that manifest, I think we should make sure this is included in the b2g bumper config. Should this be landed in mozilla-central? In mozharness? Both?

I can do the patches but I need to know agains which repo :)

Any integration branch is ok (b2g-i, m-i). You can land it with DONTBUILD in the commit message - there is no reason to build things with that change.

Oh, I think I misread the question. Bumper lives in https://hg.mozilla.org/build/mozharness/

(In reply to Rail Aliiev [:rail] from comment #3) > Oh, I think I misread the question. Bumper lives in > https://hg.mozilla.org/build/mozharness/ Sure, but there is copy in testing/mozharness. Which one is the canonical one I should update?

Copy in mozilla-central I meant

testing/mozharness is the main repo and https://hg.mozilla.org/build/mozharness/ is a fork for a couple of projects that are not tracking changes in the main repo. Bumper and vcssync are 2 of those afaik. Hal, maybe we should remove the outdated bumper and vcssync configs in testing/mozharness to avoid this confusion?

I have hacked a b2g-manifest PR to use that: https://travis-ci.org/mozilla-b2g/b2g-manifest/builds/91360384 Do we need to ensure all repos are mirrors properly before landing that?

11:32:56 INFO - needs sync? git://github.com/mozilla//android_external_busybox -> https://git.mozilla.org/b2g/android_external_busybox 11:32:56 INFO - needs sync? git://github.com/mozilla//device-sony-amami -> https://git.mozilla.org/b2g/device-sony-amami 11:32:56 INFO - needs sync? git://github.com/mozilla//device-sony-flamingo -> https://git.mozilla.org/b2g/device-sony-flamingo 11:32:56 INFO - needs sync? git://codeaurora.org//platform/external/libxml2 -> https://git.mozilla.org/external/caf/platform/external/libxml2 11:32:56 INFO - needs sync? git://github.com/mozilla//device-sony-tianchi -> https://git.mozilla.org/b2g/device-sony-tianchi 11:32:56 INFO - needs sync? git://github.com/mozilla//device-sony-honami -> https://git.mozilla.org/b2g/device-sony-honami 11:32:56 INFO - needs sync? git://github.com/mozilla//device-sony-sirius -> https://git.mozilla.org/b2g/device-sony-sirius 11:32:56 INFO - needs sync? git://github.com/mozilla//device-qcom-sepolicy -> https://git.mozilla.org/b2g/device-qcom-sepolicy 11:32:56 INFO - needs sync? git://github.com/mozilla//device-sony-eagle -> https://git.mozilla.org/b2g/device-sony-eagle 11:32:56 INFO - needs sync? git://github.com/mozilla//macaddrsetup -> https://git.mozilla.org/b2g/macaddrsetup 11:32:56 INFO - needs sync? git://github.com/mozilla//device-sony-yukon -> https://git.mozilla.org/b2g/device-sony-yukon 11:32:56 INFO - needs sync? git://github.com/mozilla//platform_hardware_broadcom_libbt -> https://git.mozilla.org/b2g/platform_hardware_broadcom_libbt 11:32:56 INFO - needs sync? git://github.com/mozilla//device-sony-rhine -> https://git.mozilla.org/b2g/device-sony-rhine 11:32:56 INFO - needs sync? git://github.com/mozilla//device-sony-scorpion -> https://git.mozilla.org/b2g/device-sony-scorpion

(In reply to Alexandre LISSY :gerard-majax from comment #8) > I have hacked a b2g-manifest PR to use that: > https://travis-ci.org/mozilla-b2g/b2g-manifest/builds/91360384 > > Do we need to ensure all repos are mirrors properly before landing that? I think so. Let's ask Hal, maybe there is something else (legal approval or something?).

Comment on attachment 8687922 [details] [diff] [review] Add Z3c L manifest to b2g bumper 302 Hal

Comment on attachment 8687922 [details] [diff] [review] Add Z3c L manifest to b2g bumper Review of attachment 8687922 [details] [diff] [review]: ----------------------------------------------------------------- lgtm

(In reply to Rail Aliiev [:rail], on PTO Nov 21 - Mozlandia from comment #6) > testing/mozharness is the main repo and > https://hg.mozilla.org/build/mozharness/ is a fork for a couple of projects > that are not tracking changes in the main repo. Bumper and vcssync are 2 of > those afaik. > > Hal, maybe we should remove the outdated bumper and vcssync configs in > testing/mozharness to avoid this confusion? I opened bug 1225364 to track this

(In reply to Rail Aliiev [:rail], on PTO Nov 21 - Mozlandia from comment #10) > (In reply to Alexandre LISSY :gerard-majax from comment #8) > > I have hacked a b2g-manifest PR to use that: > > https://travis-ci.org/mozilla-b2g/b2g-manifest/builds/91360384 > > > > Do we need to ensure all repos are mirrors properly before landing that? > > I think so. Let's ask Hal, maybe there is something else (legal approval or > something?). Yes -- if those repos aren't mirrored already, the build will break and kittens will cry. https://wiki.mozilla.org/ReleaseEngineering/RepositoryCreationRequest#Actions has the process

Filed bug 1225441 for this

Dependency fixed, we should be good :)

Catlee, do you know in which repo this should land, is this a checkin for the sheriffs ?

It should land both in build/mozharness, and in gecko/testing/mozharness

hm callek is this landing to build/mozharness something you guys do ? I guess this has to be coordinated or ?

I can land on build/mozharness and merge the hg branch to production anytime requested.

So what is blocking this from landing now?

I am not familiar with b2g_bumper and what blockers we have for deploying. afaik, we can land whenever assuming the code looks good and is expected

please land it at your earliest convenience

Who should land this?

Comment on attachment 8687922 [details] [diff] [review] Add Z3c L manifest to b2g bumper landed and merged to production branch: https://hg.mozilla.org/build/mozharness/rev/5cf030da6858 I am assuming it will automatically get picked up and deployed by bumper.

(In reply to Alexandre LISSY :gerard-majax from comment #16) > Filed bug 1225441 for this While closed, that bug was incomplete -- there are more repos being used in the manifest than requested. b2g bumper is failing on at least: 11:55:39 WARNING - https://git.mozilla.org/b2g/device-sony-seagull:refs/heads/sony-aosp-l - got output: fatal: repository 'https://git.mozilla.org/b2g/device-sony-seagull/' not found Backing out this change, as it blocks all b2g-inbound activity

https://hg.mozilla.org/build/mozharness/rev/ae8b447ce6af9a7bdd795703193e0f0d27d82b4b bug 1224903 bustage Backed out changeset 3078d9b64d4e; r=bustage

Comment on attachment 8687922 [details] [diff] [review] Add Z3c L manifest to b2g bumper reverted due to bustage - see comment 27

(In reply to Hal Wine [:hwine] (use NI) from comment #27) > (In reply to Alexandre LISSY :gerard-majax from comment #16) > > Filed bug 1225441 for this > > While closed, that bug was incomplete -- there are more repos being used in > the manifest than requested. b2g bumper is failing on at least: > 11:55:39 WARNING - > https://git.mozilla.org/b2g/device-sony-seagull:refs/heads/sony-aosp-l - got > output: fatal: repository 'https://git.mozilla.org/b2g/device-sony-seagull/' > not found > > Backing out this change, as it blocks all b2g-inbound activity No, that was good at the time it was requested. Things have moved and we landed new devices in the meantime, sorry.

Can we reland this asap once the missing repo has been mirrored?

when you are ready, feel free to ping me and I can re-land. available in irc too via #releng

https://hg.mozilla.org/build/mozharness/rev/70ea02a6e9df9bdd6858161dbb092df54eebf66e bug 1224903 - reland now that bug 1225441 fixed; r=me

Comment on attachment 8687922 [details] [diff] [review] Add Z3c L manifest to b2g bumper relanding after bug 1225441 re-fixed

hwine, jlund, catlee mentioned that this need to land on gecko/testing/mozharness too, is this done and can i remove the checkin needed flag now ? :)

(In reply to Carsten Book [:Tomcat] from comment #35) > hwine, jlund, catlee mentioned that this need to land on > gecko/testing/mozharness too, is this done and can i remove the checkin > needed flag now ? :) I believe catlee was incorrect in comment 19. b2g_bumper configs should not be used by intree processes :catlee can you confim - you are the b2g_bumper guru. If we do need it, I need to document that.

We don't need it. I thought we were trying to keep repos in sync though? If not, we should delete the in-tree versions.

(In reply to Chris AtLee [:catlee] from comment #37) > We don't need it. No more checkins needed > I thought we were trying to keep repos in sync though? If > not, we should delete the in-tree versions. That is covered in bug 1225364 and bug 1239003 Done with bug

> Done with bug \o/ and with checkin-needed :)