ReleaseEngineering/ThunderbirdTryServer: Difference between revisions

Jump to navigation Jump to search
→‎Tips and Tricks: add examples of basic try runs to start with
(Remove buildbot content and clarify a few things)
(→‎Tips and Tricks: add examples of basic try runs to start with)
Line 29: Line 29:


* Pushing a try build with try-comm-central will always pick the latest m-c changeset.
* Pushing a try build with try-comm-central will always pick the latest m-c changeset.
* A good example that should cover most cases is: <code>../mach try -b o -p macosx64,linux64 -u all</code>
* For UI changes do: <code>../mach try -b o -p macosx64,linux64 -u mozmill</code>


=== Pushing mozilla-central patches ===
=== Pushing mozilla-central patches ===
== For TaskCluster builds ==
== For TaskCluster builds ==
The process described above of applying M-C patches to C-C try pushes does not work for TaskCluster builds. For those, modify the <code>payload</code> section in the top-level <code>.taskcluster.yml</code> and point <code>GECKO_HEAD_REPOSITORY</code> to <code> https://hg.mozilla.org/try</code> and also modify <code>GECKO_HEAD_REF</code> to point to the changeset you previously pushed to M-C's try with <code>mach try empty</code>.
The process described above of applying M-C patches to C-C try pushes does not work for TaskCluster builds. For those, modify the <code>payload</code> section in the top-level <code>.taskcluster.yml</code> and point <code>GECKO_HEAD_REPOSITORY</code> to <code> https://hg.mozilla.org/try</code> and also modify <code>GECKO_HEAD_REF</code> to point to the changeset you previously pushed to M-C's try with <code>mach try empty</code>.
52

edits

Navigation menu