forgejo/services/packages/container/cleanup.go
Mathieu Fenniak ac487e93d9 fix: reduce runtime of container cleanup by relying on mass digest cleanup (#10297)
The package cleanup routine checks every container version for whether it is referenced by a multi-platform manifest, which appears to be a performance problem indicated by CPU profiling collected in #9358 on SQLite systems.  This PR removes that check completely, which isn't necessary since #4698 added a much more performant mass-cleanup of these dangling platform versions.

May fix #9358 completely, but it leaves fundamental scalability concerns with SQLite due to long-running transactions.  The transactions will be shorter with this change.  Requires end-user testing to confirm if sufficiently fixed.

## Checklist

The [contributor guide](https://forgejo.org/docs/next/contributor/) contains information that will be helpful to first time contributors. There also are a few [conditions for merging Pull Requests in Forgejo repositories](https://codeberg.org/forgejo/governance/src/branch/main/PullRequestsAgreement.md). You are also welcome to join the [Forgejo development chatroom](https://matrix.to/#/#forgejo-development:matrix.org).

### Tests

- I added test coverage for Go changes...
  - [x] in their respective `*_test.go` for unit tests.
  - [x] in the `tests/integration` directory if it involves interactions with a live Forgejo server.
- I added test coverage for JavaScript changes...
  - [ ] in `web_src/js/*.test.js` if it can be unit tested.
  - [ ] in `tests/e2e/*.test.e2e.js` if it requires interactions with a live Forgejo server (see also the [developer guide for JavaScript testing](https://codeberg.org/forgejo/forgejo/src/branch/forgejo/tests/e2e/README.md#end-to-end-tests)).

### Documentation

- [ ] I created a pull request [to the documentation](https://codeberg.org/forgejo/docs) to explain to Forgejo users how to use this change.
- [x] I did not document these changes and I do not expect someone else to do it.

### Release notes

- [ ] I do not want this change to show in the release notes.
- [x] I want the title to show in the release notes with a link to this pull request.
- [ ] I want the content of the `release-notes/<pull request number>.md` to be be used for the release notes instead of the title.

<!--start release-notes-assistant-->

## Release notes
<!--URL:https://codeberg.org/forgejo/forgejo-->
- Bug fixes
  - [PR](https://codeberg.org/forgejo/forgejo/pulls/10297): <!--number 10297 --><!--line 0 --><!--description cmVkdWNlIHJ1bnRpbWUgb2YgY29udGFpbmVyIGNsZWFudXAgYnkgcmVseWluZyBvbiBtYXNzIGRpZ2VzdCBjbGVhbnVw-->reduce runtime of container cleanup by relying on mass digest cleanup<!--description-->
<!--end release-notes-assistant-->

Reviewed-on: https://codeberg.org/forgejo/forgejo/pulls/10297
Reviewed-by: Michael Kriese <michael.kriese@gmx.de>
Reviewed-by: Earl Warren <earl-warren@noreply.codeberg.org>
Co-authored-by: Mathieu Fenniak <mathieu@fenniak.net>
Co-committed-by: Mathieu Fenniak <mathieu@fenniak.net>
2025-12-05 15:45:47 +01:00

98 lines
2.5 KiB
Go

// Copyright 2022 The Gitea Authors. All rights reserved.
// SPDX-License-Identifier: MIT
package container
import (
"context"
"time"
packages_model "forgejo.org/models/packages"
container_model "forgejo.org/models/packages/container"
"forgejo.org/modules/optional"
packages_service "forgejo.org/services/packages"
digest "github.com/opencontainers/go-digest"
)
// Cleanup removes expired container data
func Cleanup(ctx context.Context, olderThan time.Duration) error {
if err := cleanupExpiredBlobUploads(ctx, olderThan); err != nil {
return err
}
if err := CleanupSHA256(ctx, olderThan); err != nil {
return err
}
return cleanupExpiredUploadedBlobs(ctx, olderThan)
}
// cleanupExpiredBlobUploads removes expired blob uploads
func cleanupExpiredBlobUploads(ctx context.Context, olderThan time.Duration) error {
pbus, err := packages_model.FindExpiredBlobUploads(ctx, olderThan)
if err != nil {
return err
}
for _, pbu := range pbus {
if err := RemoveBlobUploadByID(ctx, pbu.ID); err != nil {
return err
}
}
return nil
}
// cleanupExpiredUploadedBlobs removes expired uploaded blobs not referenced by a manifest
func cleanupExpiredUploadedBlobs(ctx context.Context, olderThan time.Duration) error {
pfs, err := container_model.SearchExpiredUploadedBlobs(ctx, olderThan)
if err != nil {
return err
}
for _, pf := range pfs {
if err := packages_service.DeletePackageFile(ctx, pf); err != nil {
return err
}
}
pvs, _, err := packages_model.SearchVersions(ctx, &packages_model.PackageSearchOptions{
Type: packages_model.TypeContainer,
Version: packages_model.SearchValue{
ExactMatch: true,
Value: container_model.UploadVersion,
},
IsInternal: optional.Some(true),
HasFiles: optional.Some(false),
})
if err != nil {
return err
}
for _, pv := range pvs {
if err := packages_model.DeleteAllProperties(ctx, packages_model.PropertyTypeVersion, pv.ID); err != nil {
return err
}
if err := packages_model.DeleteVersionByID(ctx, pv.ID); err != nil {
return err
}
}
return nil
}
func ShouldBeSkipped(pv *packages_model.PackageVersion) bool {
// Always skip the "latest" tag
if pv.LowerVersion == "latest" {
return true
}
// Check if the version is a digest (or untagged)
if digest.Digest(pv.LowerVersion).Validate() == nil {
// Don't apply PackageCleanupRules to image versions that aren't tags; rely on `CleanupSHA256` to do mass
// cleanup of them once they're dangling due to not being referenced.
return true
}
return false
}