gitea/act_runner
1
0
Fork 0
mirror of https://gitea.com/gitea/act_runner synced 2026-09-21 19:37:07 +02:00
Code Activity

perf: cut redundant work out of job setup and teardown (#1218)

Browse Source 
Implement speedups to job start and shutdown.

- Create the job container while services are still becoming healthy, and poll their health at a flat one second instead of a 2s to 32s doubling backoff
- Pull each service image once instead of twice, and fetch a warm action cache once instead of twice
- Report the job result before reclaiming its volumes, and reap volumes stranded by a runner that died mid-job

| Step | Scenario | Before | After |
| --- | --- | --- | --- |
| Complete job | Large workspace volume | 4.2s | 0.4s |
| Set up job | One service, 2s health interval | 7.08s | 3.26s |
| Set up job | Two cached actions from github.com | 1.81s | 1.34s |
| Set up job | Two cached actions from gitea.com | 2.42s | 1.95s |
| Set up job | Two actions, cold action cache | 6.62s | unchanged |
| Set up job | Minimal job, no services or actions | 0.62s | unchanged |

Assisted-by: Claude Code:Opus 5
Reviewed-on: https://gitea.com/gitea/runner/pulls/1218
Reviewed-by: bircni <bircni@icloud.com>
Co-authored-by: silverwind <me@silverwind.io>
This commit is contained in:
silverwind
2026-09-10 19:00:06 +00:00
committed by bircni
parent 498282caaa
commit 54978255f5
11 changed files with 591 additions and 254 deletions
Download Patch File Download Diff File Expand all files Collapse all files
internal/app/run/runner.go
+31 -9
View File
@@ -146,6 +146,8 @@ func (r *Runner) Close() error {
// removeOrphanNetworks is a variable so tests can substitute one that needs no Docker daemon.
var removeOrphanNetworks = container.RemoveOrphanNetworks
var removeOrphanJobVolumes = container.RemoveOrphanJobVolumes
// OnIdle performs lightweight maintenance during polling idle windows.
// It runs synchronously on the poller goroutine; shouldRunIdleCleanup
// throttles invocations to runner.idle_cleanup_interval so the impact on
@@ -167,21 +169,20 @@ func (r *Runner) OnIdle(ctx context.Context) {
if hostRoot := filepath.FromSlash(r.cfg.Host.WorkdirParent); hostRoot != "" {
r.cleanupStaleDirs(ctx, hostRoot, isHostScratchDir)
}
r.cleanupOrphanNetworks(ctx)
r.cleanupOrphanDockerResources(ctx)
}
// cleanupOrphanNetworks reclaims the per-job networks of jobs this runner did not live to
// tear down. A labelled network with no containers on it is finished with, and as for the
// directories above, a task beginning during the pass is safe because the cutoff keeps a
// network it has created but not yet attached a container to out of scope.
func (r *Runner) cleanupOrphanNetworks(ctx context.Context) {
if r.uuid == "" || !r.requiresDocker() {
func (r *Runner) cleanupOrphanDockerResources(ctx context.Context) {
if r.uuid == "" || (!r.requiresDocker() && !dockerReachable(ctx)) {
return
}
cutoff := r.now().Add(-r.cfg.Runner.WorkdirCleanupAge)
if err := removeOrphanNetworks(ctx, r.uuid, cutoff); err != nil {
log.Warnf("failed to clean up networks left behind by earlier jobs: %v", err)
}
if err := removeOrphanJobVolumes(ctx, r.uuid, cutoff); err != nil {
log.Warnf("failed to clean up volumes left behind by earlier jobs: %v", err)
}
}
func (r *Runner) shouldRunIdleCleanup() bool {
@@ -287,6 +288,7 @@ func (r *Runner) Run(ctx context.Context, task *runnerv1.Task) error {
defer r.runningTasks.Delete(task.Id)
r.runningCount.Add(1)
defer r.runningCount.Add(-1)
start := time.Now()
@@ -294,15 +296,25 @@ func (r *Runner) Run(ctx context.Context, task *runnerv1.Task) error {
defer cancel()
// A proxy URL may carry credentials, and every job is given it; keep them out of the log.
reporter := report.NewReporter(ctx, cancel, r.client, task, r.cfg, proxyPasswords()...)
var volumeCleanup []common.Executor
var volumeCleanupMu sync.Mutex
if r.cfg.Runner.PostTaskScript == "" {
ctx = runner.WithJobVolumeCleanup(ctx, func(cleanup common.Executor) {
volumeCleanupMu.Lock()
defer volumeCleanupMu.Unlock()
volumeCleanup = append(volumeCleanup, cleanup)
})
}
var runErr error
defer func() {
r.runningCount.Add(-1)
lastWords := ""
if runErr != nil {
lastWords = runErr.Error()
}
_ = reporter.Close(lastWords)
if err := cleanupJobVolumes(ctx, volumeCleanup); err != nil {
log.Warnf("task %d volume cleanup after reporting: %v", task.Id, err)
}
metrics.JobDuration.Observe(time.Since(start).Seconds())
metrics.JobsTotal.WithLabelValues(metrics.ResultToStatusLabel(reporter.Result())).Inc()
@@ -313,6 +325,16 @@ func (r *Runner) Run(ctx context.Context, task *runnerv1.Task) error {
return nil
}
func cleanupJobVolumes(ctx context.Context, cleanups []common.Executor) error {
ctx, cancel := context.WithTimeout(context.WithoutCancel(ctx), time.Minute)
defer cancel()
var errs []error
for _, cleanup := range cleanups {
errs = append(errs, cleanup(ctx))
}
return errors.Join(errs...)
}
func (r *Runner) cloneEnvs() map[string]string {
// Reserve space for the per-task keys injected by run():
// ACTIONS_ID_TOKEN_REQUEST_URL, ACTIONS_ID_TOKEN_REQUEST_TOKEN, ACTIONS_RUNTIME_TOKEN,