mirror of
https://gitea.com/gitea/act_runner.git
synced 2026-04-24 04:40:22 +08:00
## Summary Many teams self-host Gitea + Act Runner at scale. The current runner design causes excessive HTTP requests to the Gitea server, leading to high server load. This PR addresses three root causes: aggressive fixed-interval polling, per-task status reporting every 1 second regardless of activity, and unoptimized HTTP client configuration. ## Problem The original architecture has these issues: **1. Fixed 1-second reporting interval (RunDaemon)** - Every running task calls ReportLog + ReportState every 1 second (2 HTTP requests/sec/task) - These requests are sent even when there are no new log rows or state changes - With 200 runners × 3 tasks each = **1,200 req/sec just for status reporting** **2. Fixed 2-second polling interval (no backoff)** - Idle runners poll FetchTask every 2 seconds forever, even when no jobs are queued - No exponential backoff or jitter — all runners can synchronize after network recovery (thundering herd) - 200 idle runners = **100 req/sec doing nothing useful** **3. HTTP client not tuned** - Uses http.DefaultClient with MaxIdleConnsPerHost=2, causing frequent TCP/TLS reconnects - Creates two separate http.Client instances (one for Ping, one for Runner service) instead of sharing **Total: ~1,300 req/sec for 200 runners with 3 tasks each** ## Solution ### Adaptive Event-Driven Log Reporting Replace the recursive `time.AfterFunc(1s)` pattern in RunDaemon with a goroutine-based select event loop using three trigger mechanisms: | Trigger | Default | Purpose | |---------|---------|---------| | `log_report_max_latency` | 3s | Guarantee even a single log line is delivered within this time | | `log_report_interval` | 5s | Periodic sweep — steady-state cadence | | `log_report_batch_size` | 100 rows | Immediate flush during bursty output (e.g., npm install) | **Key design**: `log_report_max_latency` (3s) must be less than `log_report_interval` (5s) so the max-latency timer fires before the periodic ticker for single-line scenarios. State reporting is decoupled to its own `state_report_interval` (default 5s), with immediate flush on step transitions (start/stop) via a stateNotify channel for responsive frontend UX. Additionally: - Skip ReportLog when `len(rows) == 0` (no pending log rows) - Skip ReportState when `stateChanged == false && len(outputs) == 0` (nothing changed) - Move expensive `proto.Clone` after the early-return check to avoid deep copies on no-op paths ### Polling Backoff with Jitter Replace fixed `rate.Limiter` with adaptive exponential backoff: - Track `consecutiveEmpty` and `consecutiveErrors` counters - Interval doubles with each empty/error response: `base × 2^(n-1)`, capped at `fetch_interval_max` (default 60s) - Add ±20% random jitter to prevent thundering herd - Fetch first, sleep after ��� preserves burst=1 behavior for immediate first fetch on startup and after task completion ### HTTP Client Tuning - Configure custom `http.Transport` with `MaxIdleConnsPerHost=10` (was 2) - Share a single `http.Client` between PingService and RunnerService - Add `IdleConnTimeout=90s` for clean connection lifecycle ## Load Reduction For 200 runners × 3 tasks (70% with active log output): | Component | Before | After | Reduction | |-----------|--------|-------|-----------| | Polling (idle) | 100 req/s | ~3.4 req/s | 97% | | Log reporting | 420 req/s | ~84 req/s | 80% | | State reporting | 126 req/s | ~25 req/s | 80% | | **Total** | **~1,300 req/s** | **~113 req/s** | **~91%** | ## Frontend UX Impact | Scenario | Before | After | Notes | |----------|--------|-------|-------| | Continuous output (npm install) | ~1s | ~5s | Periodic ticker sweep | | Single line then silence | ~1s | ≤3s | maxLatencyTimer guarantee | | Bursty output (100+ lines) | ~1s | <1s | Batch size immediate flush | | Step start/stop | ~1s | <1s | stateNotify immediate flush | | Job completion | ~1s | ~1s | Close() retry unchanged | ## New Configuration Options All have safe defaults — existing config files need no changes: ```yaml runner: fetch_interval_max: 60s # Max backoff interval when idle log_report_interval: 5s # Periodic log flush interval log_report_max_latency: 3s # Max time a log row waits (must be < log_report_interval) log_report_batch_size: 100 # Immediate flush threshold state_report_interval: 5s # State flush interval (step transitions are always immediate) ``` Config validation warns on invalid combinations: - `fetch_interval_max < fetch_interval` → auto-corrected - `log_report_max_latency >= log_report_interval` → warning (timer would be redundant) ## Test Plan - [x] `go build ./...` passes - [x] `go test ./...` passes (all existing + 3 new tests) - [x] `golangci-lint run` — 0 issues - [x] TestReporter_MaxLatencyTimer — verifies single log line flushed by maxLatencyTimer before logTicker - [x] TestReporter_BatchSizeFlush — verifies batch size threshold triggers immediate flush - [x] TestReporter_StateNotifyFlush — verifies step transition triggers immediate state flush - [x] TestReporter_EphemeralRunnerDeletion — verifies Close/RunDaemon race safety - [x] TestReporter_RunDaemonClose_Race — verifies concurrent Close safety Reviewed-on: https://gitea.com/gitea/act_runner/pulls/819 Reviewed-by: Nicolas <173651+bircni@noreply.gitea.com> Co-authored-by: Bo-Yi Wu <appleboy.tw@gmail.com> Co-committed-by: Bo-Yi Wu <appleboy.tw@gmail.com>
275 lines
8.7 KiB
Go
275 lines
8.7 KiB
Go
// Copyright 2022 The Gitea Authors. All rights reserved.
|
|
// SPDX-License-Identifier: MIT
|
|
|
|
package run
|
|
|
|
import (
|
|
"context"
|
|
"encoding/json"
|
|
"fmt"
|
|
"maps"
|
|
"os"
|
|
"path/filepath"
|
|
"strings"
|
|
"sync"
|
|
"time"
|
|
|
|
runnerv1 "code.gitea.io/actions-proto-go/runner/v1"
|
|
"connectrpc.com/connect"
|
|
"github.com/docker/docker/api/types/container"
|
|
"github.com/nektos/act/pkg/artifactcache"
|
|
"github.com/nektos/act/pkg/common"
|
|
"github.com/nektos/act/pkg/model"
|
|
"github.com/nektos/act/pkg/runner"
|
|
log "github.com/sirupsen/logrus"
|
|
|
|
"gitea.com/gitea/act_runner/internal/pkg/client"
|
|
"gitea.com/gitea/act_runner/internal/pkg/config"
|
|
"gitea.com/gitea/act_runner/internal/pkg/labels"
|
|
"gitea.com/gitea/act_runner/internal/pkg/report"
|
|
"gitea.com/gitea/act_runner/internal/pkg/ver"
|
|
)
|
|
|
|
// Runner runs the pipeline.
|
|
type Runner struct {
|
|
name string
|
|
|
|
cfg *config.Config
|
|
|
|
client client.Client
|
|
labels labels.Labels
|
|
envs map[string]string
|
|
|
|
runningTasks sync.Map
|
|
}
|
|
|
|
func NewRunner(cfg *config.Config, reg *config.Registration, cli client.Client) *Runner {
|
|
ls := labels.Labels{}
|
|
for _, v := range reg.Labels {
|
|
if l, err := labels.Parse(v); err == nil {
|
|
ls = append(ls, l)
|
|
}
|
|
}
|
|
envs := make(map[string]string, len(cfg.Runner.Envs))
|
|
maps.Copy(envs, cfg.Runner.Envs)
|
|
if cfg.Cache.Enabled == nil || *cfg.Cache.Enabled {
|
|
if cfg.Cache.ExternalServer != "" {
|
|
envs["ACTIONS_CACHE_URL"] = cfg.Cache.ExternalServer
|
|
} else {
|
|
cacheHandler, err := artifactcache.StartHandler(
|
|
cfg.Cache.Dir,
|
|
cfg.Cache.Host,
|
|
cfg.Cache.Port,
|
|
log.StandardLogger().WithField("module", "cache_request"),
|
|
)
|
|
if err != nil {
|
|
log.Errorf("cannot init cache server, it will be disabled: %v", err)
|
|
// go on
|
|
} else {
|
|
envs["ACTIONS_CACHE_URL"] = cacheHandler.ExternalURL() + "/"
|
|
}
|
|
}
|
|
}
|
|
|
|
// set artifact gitea api
|
|
artifactGiteaAPI := strings.TrimSuffix(cli.Address(), "/") + "/api/actions_pipeline/"
|
|
envs["ACTIONS_RUNTIME_URL"] = artifactGiteaAPI
|
|
envs["ACTIONS_RESULTS_URL"] = strings.TrimSuffix(cli.Address(), "/")
|
|
|
|
// Set specific environments to distinguish between Gitea and GitHub
|
|
envs["GITEA_ACTIONS"] = "true"
|
|
envs["GITEA_ACTIONS_RUNNER_VERSION"] = ver.Version()
|
|
|
|
return &Runner{
|
|
name: reg.Name,
|
|
cfg: cfg,
|
|
client: cli,
|
|
labels: ls,
|
|
envs: envs,
|
|
}
|
|
}
|
|
|
|
func (r *Runner) Run(ctx context.Context, task *runnerv1.Task) error {
|
|
if _, ok := r.runningTasks.Load(task.Id); ok {
|
|
return fmt.Errorf("task %d is already running", task.Id)
|
|
}
|
|
r.runningTasks.Store(task.Id, struct{}{})
|
|
defer r.runningTasks.Delete(task.Id)
|
|
|
|
ctx, cancel := context.WithTimeout(ctx, r.cfg.Runner.Timeout)
|
|
defer cancel()
|
|
reporter := report.NewReporter(ctx, cancel, r.client, task, r.cfg)
|
|
var runErr error
|
|
defer func() {
|
|
lastWords := ""
|
|
if runErr != nil {
|
|
lastWords = runErr.Error()
|
|
}
|
|
_ = reporter.Close(lastWords)
|
|
}()
|
|
reporter.RunDaemon()
|
|
runErr = r.run(ctx, task, reporter)
|
|
|
|
return nil
|
|
}
|
|
|
|
// getDefaultActionsURL
|
|
// when DEFAULT_ACTIONS_URL == "https://github.com" and GithubMirror is not blank,
|
|
// it should be set to GithubMirror first.
|
|
func (r *Runner) getDefaultActionsURL(task *runnerv1.Task) string {
|
|
giteaDefaultActionsURL := task.Context.Fields["gitea_default_actions_url"].GetStringValue()
|
|
if giteaDefaultActionsURL == "https://github.com" && r.cfg.Runner.GithubMirror != "" {
|
|
return r.cfg.Runner.GithubMirror
|
|
}
|
|
return giteaDefaultActionsURL
|
|
}
|
|
|
|
func (r *Runner) run(ctx context.Context, task *runnerv1.Task, reporter *report.Reporter) (err error) {
|
|
defer func() {
|
|
if r := recover(); r != nil {
|
|
err = fmt.Errorf("panic: %v", r)
|
|
}
|
|
}()
|
|
|
|
reporter.Logf("%s(version:%s) received task %v of job %v, be triggered by event: %s", r.name, ver.Version(), task.Id, task.Context.Fields["job"].GetStringValue(), task.Context.Fields["event_name"].GetStringValue())
|
|
|
|
workflow, jobID, err := generateWorkflow(task)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
|
|
plan, err := model.CombineWorkflowPlanner(workflow).PlanJob(jobID)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
job := workflow.GetJob(jobID)
|
|
reporter.ResetSteps(len(job.Steps))
|
|
|
|
taskContext := task.Context.Fields
|
|
|
|
log.Infof("task %v repo is %v %v %v", task.Id, taskContext["repository"].GetStringValue(),
|
|
r.getDefaultActionsURL(task),
|
|
r.client.Address())
|
|
|
|
preset := &model.GithubContext{
|
|
Event: taskContext["event"].GetStructValue().AsMap(),
|
|
RunID: taskContext["run_id"].GetStringValue(),
|
|
RunNumber: taskContext["run_number"].GetStringValue(),
|
|
RunAttempt: taskContext["run_attempt"].GetStringValue(),
|
|
Actor: taskContext["actor"].GetStringValue(),
|
|
Repository: taskContext["repository"].GetStringValue(),
|
|
EventName: taskContext["event_name"].GetStringValue(),
|
|
Sha: taskContext["sha"].GetStringValue(),
|
|
Ref: taskContext["ref"].GetStringValue(),
|
|
RefName: taskContext["ref_name"].GetStringValue(),
|
|
RefType: taskContext["ref_type"].GetStringValue(),
|
|
HeadRef: taskContext["head_ref"].GetStringValue(),
|
|
BaseRef: taskContext["base_ref"].GetStringValue(),
|
|
Token: taskContext["token"].GetStringValue(),
|
|
RepositoryOwner: taskContext["repository_owner"].GetStringValue(),
|
|
RetentionDays: taskContext["retention_days"].GetStringValue(),
|
|
}
|
|
if t := task.Secrets["GITEA_TOKEN"]; t != "" {
|
|
preset.Token = t
|
|
} else if t := task.Secrets["GITHUB_TOKEN"]; t != "" {
|
|
preset.Token = t
|
|
}
|
|
|
|
if actionsIDTokenRequestURL := taskContext["actions_id_token_request_url"].GetStringValue(); actionsIDTokenRequestURL != "" {
|
|
r.envs["ACTIONS_ID_TOKEN_REQUEST_URL"] = actionsIDTokenRequestURL
|
|
r.envs["ACTIONS_ID_TOKEN_REQUEST_TOKEN"] = taskContext["actions_id_token_request_token"].GetStringValue()
|
|
task.Secrets["ACTIONS_ID_TOKEN_REQUEST_TOKEN"] = r.envs["ACTIONS_ID_TOKEN_REQUEST_TOKEN"]
|
|
}
|
|
|
|
giteaRuntimeToken := taskContext["gitea_runtime_token"].GetStringValue()
|
|
if giteaRuntimeToken == "" {
|
|
// use task token to action api token for previous Gitea Server Versions
|
|
giteaRuntimeToken = preset.Token
|
|
}
|
|
r.envs["ACTIONS_RUNTIME_TOKEN"] = giteaRuntimeToken
|
|
|
|
eventJSON, err := json.Marshal(preset.Event)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
|
|
maxLifetime := 3 * time.Hour
|
|
if deadline, ok := ctx.Deadline(); ok {
|
|
maxLifetime = time.Until(deadline)
|
|
}
|
|
|
|
workdirParent := strings.TrimLeft(r.cfg.Container.WorkdirParent, "/")
|
|
if r.cfg.Container.BindWorkdir {
|
|
// Append the task ID to isolate concurrent jobs from the same repo.
|
|
workdirParent = fmt.Sprintf("%s/%d", workdirParent, task.Id)
|
|
}
|
|
workdir := filepath.FromSlash(fmt.Sprintf("/%s/%s", workdirParent, preset.Repository))
|
|
|
|
runnerConfig := &runner.Config{
|
|
// On Linux, Workdir will be like "/<parent_directory>/<owner>/<repo>"
|
|
// On Windows, Workdir will be like "\<parent_directory>\<owner>\<repo>"
|
|
Workdir: workdir,
|
|
BindWorkdir: r.cfg.Container.BindWorkdir,
|
|
ActionCacheDir: filepath.FromSlash(r.cfg.Host.WorkdirParent),
|
|
|
|
ReuseContainers: false,
|
|
ForcePull: r.cfg.Container.ForcePull,
|
|
ForceRebuild: r.cfg.Container.ForceRebuild,
|
|
LogOutput: true,
|
|
JSONLogger: false,
|
|
Env: r.envs,
|
|
Secrets: task.Secrets,
|
|
GitHubInstance: strings.TrimSuffix(r.client.Address(), "/"),
|
|
AutoRemove: true,
|
|
NoSkipCheckout: true,
|
|
PresetGitHubContext: preset,
|
|
EventJSON: string(eventJSON),
|
|
ContainerNamePrefix: fmt.Sprintf("GITEA-ACTIONS-TASK-%d", task.Id),
|
|
ContainerMaxLifetime: maxLifetime,
|
|
ContainerNetworkMode: container.NetworkMode(r.cfg.Container.Network),
|
|
ContainerOptions: r.cfg.Container.Options,
|
|
ContainerDaemonSocket: r.cfg.Container.DockerHost,
|
|
Privileged: r.cfg.Container.Privileged,
|
|
DefaultActionInstance: r.getDefaultActionsURL(task),
|
|
PlatformPicker: r.labels.PickPlatform,
|
|
Vars: task.Vars,
|
|
ValidVolumes: r.cfg.Container.ValidVolumes,
|
|
InsecureSkipTLS: r.cfg.Runner.Insecure,
|
|
}
|
|
|
|
rr, err := runner.New(runnerConfig)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
executor := rr.NewPlanExecutor(plan)
|
|
|
|
reporter.Logf("workflow prepared")
|
|
|
|
// add logger recorders
|
|
ctx = common.WithLoggerHook(ctx, reporter)
|
|
|
|
if !log.IsLevelEnabled(log.DebugLevel) {
|
|
ctx = runner.WithJobLoggerFactory(ctx, NullLogger{})
|
|
}
|
|
|
|
execErr := executor(ctx)
|
|
reporter.SetOutputs(job.Outputs)
|
|
|
|
if r.cfg.Container.BindWorkdir {
|
|
// Remove the entire task-specific directory (e.g. /workspace/<task_id>).
|
|
taskDir := filepath.FromSlash("/" + workdirParent)
|
|
if err := os.RemoveAll(taskDir); err != nil {
|
|
log.Warnf("failed to clean up workspace %s: %v", taskDir, err)
|
|
}
|
|
}
|
|
|
|
return execErr
|
|
}
|
|
|
|
func (r *Runner) Declare(ctx context.Context, labels []string) (*connect.Response[runnerv1.DeclareResponse], error) {
|
|
return r.client.Declare(ctx, connect.NewRequest(&runnerv1.DeclareRequest{
|
|
Version: ver.Version(),
|
|
Labels: labels,
|
|
}))
|
|
}
|