mirror of
https://codeberg.org/forgejo/forgejo.git
synced 2026-05-15 15:30:26 +00:00
In the case you hit some API error (Github ratelimit was often a problem) or the instance restarted in the middle of your migration, you would be left with data on the disk and/or database. Upon retrying the migration the migration code would (rightfully) fail because it's trying to migrate stuff that already exists.
This was hit so often on Codeberg it was better to force people to delete and start whole migration process again: 28ee60c91f
Delete the repository data before retrying to solve this.
Reviewed-on: https://codeberg.org/forgejo/forgejo/pulls/12370
Reviewed-by: Mathieu Fenniak <mfenniak@noreply.codeberg.org>
201 lines
6.4 KiB
Go
201 lines
6.4 KiB
Go
// Copyright 2020 The Gitea Authors. All rights reserved.
|
|
// SPDX-License-Identifier: MIT
|
|
|
|
package repository
|
|
|
|
import (
|
|
"context"
|
|
"fmt"
|
|
"strings"
|
|
"time"
|
|
|
|
"forgejo.org/models/db"
|
|
repo_model "forgejo.org/models/repo"
|
|
system_model "forgejo.org/models/system"
|
|
"forgejo.org/modules/git"
|
|
"forgejo.org/modules/log"
|
|
repo_module "forgejo.org/modules/repository"
|
|
"forgejo.org/modules/util"
|
|
|
|
"xorm.io/builder"
|
|
)
|
|
|
|
// GitFsckRepos calls 'git fsck' to check repository health.
|
|
func GitFsckRepos(ctx context.Context, timeout time.Duration, args git.TrustedCmdArgs) error {
|
|
log.Trace("Doing: GitFsck")
|
|
|
|
if err := db.Iterate(
|
|
ctx,
|
|
builder.Expr("id>0 AND is_fsck_enabled=?", true),
|
|
func(ctx context.Context, repo *repo_model.Repository) error {
|
|
select {
|
|
case <-ctx.Done():
|
|
return db.ErrCancelledf("before fsck of %s", repo.FullName())
|
|
default:
|
|
}
|
|
return GitFsckRepo(ctx, repo, timeout, args)
|
|
},
|
|
); err != nil {
|
|
log.Trace("Error: GitFsck: %v", err)
|
|
return err
|
|
}
|
|
|
|
log.Trace("Finished: GitFsck")
|
|
return nil
|
|
}
|
|
|
|
// GitFsckRepo calls 'git fsck' to check an individual repository's health.
|
|
func GitFsckRepo(ctx context.Context, repo *repo_model.Repository, timeout time.Duration, args git.TrustedCmdArgs) error {
|
|
log.Trace("Running health check on repository %-v", repo)
|
|
repoPath := repo.RepoPath()
|
|
if err := git.Fsck(ctx, repoPath, timeout, args); err != nil {
|
|
log.Warn("Failed to health check repository (%-v): %v", repo, err)
|
|
if err = system_model.CreateRepositoryNotice("Failed to health check repository (%s): %v", repo.FullName(), err); err != nil {
|
|
log.Error("CreateRepositoryNotice: %v", err)
|
|
}
|
|
}
|
|
return nil
|
|
}
|
|
|
|
// GitGcRepos calls 'git gc' to remove unnecessary files and optimize the local repository
|
|
func GitGcRepos(ctx context.Context, timeout time.Duration, args git.TrustedCmdArgs) error {
|
|
log.Trace("Doing: GitGcRepos")
|
|
|
|
if err := db.Iterate(
|
|
ctx,
|
|
builder.Gt{"id": 0},
|
|
func(ctx context.Context, repo *repo_model.Repository) error {
|
|
select {
|
|
case <-ctx.Done():
|
|
return db.ErrCancelledf("before GC of %s", repo.FullName())
|
|
default:
|
|
}
|
|
// we can ignore the error here because it will be logged in GitGCRepo
|
|
_ = GitGcRepo(ctx, repo, timeout, args)
|
|
return nil
|
|
},
|
|
); err != nil {
|
|
return err
|
|
}
|
|
|
|
log.Trace("Finished: GitGcRepos")
|
|
return nil
|
|
}
|
|
|
|
// GitGcRepo calls 'git gc' to remove unnecessary files and optimize the local repository
|
|
func GitGcRepo(ctx context.Context, repo *repo_model.Repository, timeout time.Duration, args git.TrustedCmdArgs) error {
|
|
log.Trace("Running git gc on %-v", repo)
|
|
command := git.NewCommand(ctx, "gc").AddArguments(args...).
|
|
SetDescription(fmt.Sprintf("Repository Garbage Collection: %s", repo.FullName()))
|
|
var stdout string
|
|
var err error
|
|
stdout, _, err = command.RunStdString(&git.RunOpts{Timeout: timeout, Dir: repo.RepoPath()})
|
|
if err != nil {
|
|
log.Error("Repository garbage collection failed for %-v. Stdout: %s\nError: %v", repo, stdout, err)
|
|
desc := fmt.Sprintf("Repository garbage collection failed for %s. Stdout: %s\nError: %v", repo.RepoPath(), stdout, err)
|
|
if err := system_model.CreateRepositoryNotice(desc); err != nil {
|
|
log.Error("CreateRepositoryNotice: %v", err)
|
|
}
|
|
return fmt.Errorf("Repository garbage collection failed in repo: %s: Error: %w", repo.FullName(), err)
|
|
}
|
|
|
|
// Now update the size of the repository
|
|
if err := repo_module.UpdateRepoSize(ctx, repo); err != nil {
|
|
log.Error("Updating size as part of garbage collection failed for %-v. Stdout: %s\nError: %v", repo, stdout, err)
|
|
desc := fmt.Sprintf("Updating size as part of garbage collection failed for %s. Stdout: %s\nError: %v", repo.RepoPath(), stdout, err)
|
|
if err := system_model.CreateRepositoryNotice(desc); err != nil {
|
|
log.Error("CreateRepositoryNotice: %v", err)
|
|
}
|
|
return fmt.Errorf("Updating size as part of garbage collection failed in repo: %s: Error: %w", repo.FullName(), err)
|
|
}
|
|
|
|
return nil
|
|
}
|
|
|
|
func gatherMissingRepoRecords(ctx context.Context) (repo_model.RepositoryList, error) {
|
|
repos := make([]*repo_model.Repository, 0, 10)
|
|
if err := db.Iterate(
|
|
ctx,
|
|
builder.Gt{"id": 0},
|
|
func(ctx context.Context, repo *repo_model.Repository) error {
|
|
select {
|
|
case <-ctx.Done():
|
|
return db.ErrCancelledf("during gathering missing repo records before checking %s", repo.FullName())
|
|
default:
|
|
}
|
|
isDir, err := util.IsDir(repo.RepoPath())
|
|
if err != nil {
|
|
return fmt.Errorf("Unable to check dir for %s. %w", repo.FullName(), err)
|
|
}
|
|
if !isDir {
|
|
repos = append(repos, repo)
|
|
}
|
|
return nil
|
|
},
|
|
); err != nil {
|
|
if strings.HasPrefix(err.Error(), "Aborted gathering missing repo") {
|
|
return nil, err
|
|
}
|
|
if err2 := system_model.CreateRepositoryNotice("gatherMissingRepoRecords: %v", err); err2 != nil {
|
|
log.Error("CreateRepositoryNotice: %v", err2)
|
|
}
|
|
return nil, err
|
|
}
|
|
return repos, nil
|
|
}
|
|
|
|
// DeleteMissingRepositories deletes all repository records that lost Git files.
|
|
func DeleteMissingRepositories(ctx context.Context) error {
|
|
repos, err := gatherMissingRepoRecords(ctx)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
|
|
if len(repos) == 0 {
|
|
return nil
|
|
}
|
|
|
|
for _, repo := range repos {
|
|
select {
|
|
case <-ctx.Done():
|
|
return db.ErrCancelledf("during DeleteMissingRepositories before %s", repo.FullName())
|
|
default:
|
|
}
|
|
log.Trace("Deleting %d/%d...", repo.OwnerID, repo.ID)
|
|
if err := DeleteRepositoryDirectly(ctx, repo.ID, DeleteRepositoryOpts{}); err != nil {
|
|
log.Error("Failed to DeleteRepository %-v: Error: %v", repo, err)
|
|
if err2 := system_model.CreateRepositoryNotice("Failed to DeleteRepository %s [%d]: Error: %v", repo.FullName(), repo.ID, err); err2 != nil {
|
|
log.Error("CreateRepositoryNotice: %v", err)
|
|
}
|
|
}
|
|
}
|
|
return nil
|
|
}
|
|
|
|
// ReinitMissingRepositories reinitializes all repository records that lost Git files.
|
|
func ReinitMissingRepositories(ctx context.Context) error {
|
|
repos, err := gatherMissingRepoRecords(ctx)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
|
|
if len(repos) == 0 {
|
|
return nil
|
|
}
|
|
|
|
for _, repo := range repos {
|
|
select {
|
|
case <-ctx.Done():
|
|
return db.ErrCancelledf("during ReinitMissingRepositories before %s", repo.FullName())
|
|
default:
|
|
}
|
|
log.Trace("Initializing %d/%d...", repo.OwnerID, repo.ID)
|
|
if err := git.InitRepository(ctx, repo.RepoPath(), true, repo.ObjectFormatName); err != nil {
|
|
log.Error("Unable (re)initialize repository %d at %s. Error: %v", repo.ID, repo.RepoPath(), err)
|
|
if err2 := system_model.CreateRepositoryNotice("InitRepository [%d]: %v", repo.ID, err); err2 != nil {
|
|
log.Error("CreateRepositoryNotice: %v", err2)
|
|
}
|
|
}
|
|
}
|
|
return nil
|
|
}
|