ci: Renovate commits too frequently, and too often breaks linting on main branch #1517
Labels
No labels
Blocked
Bug
Cherry-picking
Database
Dependencies
Dependencies/Renovate
Difficulty
Easy
Difficulty
Hard
Difficulty
Medium
Documentation
Enhancement
Good first issue
Help wanted
Inherited
Matrix/Administration
Matrix/Appservices
Matrix/Auth
Matrix/Client
Matrix/Core
Matrix/E2EE
Matrix/Federation
Matrix/Hydra
Matrix/MSC
Matrix/Media
Matrix/T&S
Meta
Meta/CI
Meta/Packaging
Priority
Blocking
Priority
High
Priority
Low
Security
Status
Confirmed
Status
Duplicate
Status
Invalid
Status
Needs Investigation
Support
To-Merge
Wont fix
old/ci/cd
old/rust
No milestone
No project
No assignees
3 participants
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
continuwuation/continuwuity#1517
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Given our release cadence of about once a month or so, I think it would be preferable to "batch up" all low-impact Renovate/automated package security update PRs into a weekly or twice a month merge, one which ideally is tested (perhaps manually) for linter compliance.
Also there are just too many PRs open in general by Renovate.
I wanna look at what people have open, but every time I have to click the "labels" menu and hide "Dependencies/Renovate" to filter the spam.
Someone please find a solution to this distracting issue!! Thank you
git:to gitea:Renovatecommits too frequently, and too often breaks linting onmainbranchRenovatecommits too frequently, and too often breaks linting onmainbranchThe renovate config already implements batching for patch-level dependency updates. I'd say you could batch minor-level updates too, as is already done for Actions, but this appears to have been a deliberate decision by the maintainers, presumably to inspect minor-level and above updates to dependencies individually.
If not, however, this would be trivial to resolve. The solution can be as simple as adding
minortomatchUpdateTypesfor more groups.The part I find weird personally is that some PRs use
chore(deps):and others usefix(deps):cc @Jade, do you have any objections to batching all non-major cargo dependency updates, or am I correct in assuming the intention is to allow you to inspect minor releases?
CI failures on renovate commits to main are because they are usually merged in quick sequence, causing earlier runs to be cancelled.
If you inspect the currently open PRs, you will notice that most of them have errors. They need manual work, while the ones we have merged do not. This is why most updates are not bundled.
The label is there so you can filter out these PRs.
Also, afaik renovate is not rust-semver aware by default so bundling all minor updates would bundle breaking changes in zerover crates.
That was what I noticed too, it counts minor releases as minor releases even for projects in a 0.y.z state, which is a problem. You could get around that, but half of the Rust ecosystem is 0.y.z anyway. Are there non-zerover crates that have required manual intervention for minor releases? If not, we could batch minor updates for packages above version 1.0.0, I suppose.
It is also worth noting that Renovate makes up a whole third of the merged PR volume, so cutting that down wouldn't hurt.
gitea:to ci:Renovatecommits too frequently, and too often breaks linting onmainbranchRenovatecommits too frequently, and too often breaks linting onmainbranch