Repository navigation
fix(server): warn on dangling provider refs at startup - #4661
TaylorMutch merged 3 commits into
Conversation
Signed-off-by: Eric Curtin <eric.curtin@docker.com>
|
@johntmyers @elezar PTAL, and |
Signed-off-by: Eric Curtin <eric.curtin@docker.com>
|
Merged main to fix conflicts. @johntmyers @elezar PTAL, and |
|
Is ok |
|
/ok to test 30c1f3b |
|
Label |
|
E2E failures look unrelated to this change: k3s-ha pod gating timeout, ssh_host_identity container exit, and workspace operator |
EmilienM
left a comment
There was a problem hiding this comment.
Approved. This does what #4657 asks: a missing provider gets one warning per sandbox, and store errors and the #2962 invalid-composition guard stay fatal. I left three small comments, none blocking: test ordering, the sequential lookups, and the UpdateConfig path that shares this helper. Heads up that my approval doesn't count toward required reviews, so you'll still need a maintainer's.
Signed-off-by: Eric Curtin <eric.curtin@docker.com>
|
Addressed review feedback. @johntmyers @elezar PTAL, and |
|
/ok to test 8c3231e |
Summary
A deleted provider no longer blocks gateway startup. The preflight warns and keeps validating.
Related Issue
Closes #4657
Changes
UpdateConfiguses the same check, so it also tolerates a dangling reference.Testing
Ran
cargo clippyand theopenshell-servergrpc::unit tests, including new dangling reference cases for startup and global policy delete.Checklist