We have a suite of tests that run when when a PR Is opened against our main branch, but they don't run against the major-release branch. But we want to test those changes too! If anything it's more important, because they're breaking changes.
That said, we will still catch test failures. It will just be as part of the release process, which is not ideal as (a) it delays a release and (b) the context of the specific PR will be lost, we may have to ping the submitter, etc.
We have a suite of tests that run when when a PR Is opened against our main branch, but they don't run against the major-release branch. But we want to test those changes too! If anything it's more important, because they're breaking changes.
That said, we will still catch test failures. It will just be as part of the release process, which is not ideal as (a) it delays a release and (b) the context of the specific PR will be lost, we may have to ping the submitter, etc.
See here for the headache this caused.