mirror of
https://gitea.com/gitea/act_runner.git
synced 2026-07-22 20:47:45 +00:00
fix: stop host-mode jobs from leaking processes on Windows (#1080)
## Problem On host-mode Windows runners, a job can leave processes running after it finishes, and those leftovers hold file handles that block deletion of the workspace. Today a step's tree is only torn down when the step is *cancelled* (`process.Killer`). A step that completes leaves whatever it spawned alive, and the existing workspace scan in `terminateRunningProcesses` misses two shapes of leftover: orphans whose parent already exited (no tree to walk, and their executable often lives outside the workspace), and processes that merely *run in* the workspace but reference no path from it — `Win32_Process` exposes no working directory, so the scan cannot match them. ## Solution Two additions in `internal/pkg/process`, both no-ops outside Windows: - **`process.Group`** — a job-scoped Windows Job Object created with `JOB_OBJECT_LIMIT_KILL_ON_JOB_CLOSE`. Step processes are assigned to it before they get their own `Killer`, so the step's job nests inside the job's: cancellation still kills exactly the step's tree, while `Remove` closing the group makes the kernel terminate everything still assigned, whatever its parentage. The kernel also drops the handle when the runner exits, so a crashed runner cannot strand processes. - **`process.KillProcessesWithCWDUnder`** — a best-effort net for processes that never joined the job (started via a service or scheduled task). It reads each process's working directory from its PEB and terminates those under a workspace dir. Processes it cannot open are skipped. --------- Co-authored-by: silverwind <me@silverwind.io> Reviewed-on: https://gitea.com/gitea/runner/pulls/1080 Reviewed-by: silverwind <2021+silverwind@noreply.gitea.com>
This commit is contained in:
91
internal/pkg/process/leftover_windows_test.go
Normal file
91
internal/pkg/process/leftover_windows_test.go
Normal file
@@ -0,0 +1,91 @@
|
||||
// Copyright 2026 The Gitea Authors. All rights reserved.
|
||||
// SPDX-License-Identifier: MIT
|
||||
|
||||
package process
|
||||
|
||||
import (
|
||||
"context"
|
||||
"os"
|
||||
"os/exec"
|
||||
"path/filepath"
|
||||
"syscall"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/stretchr/testify/require"
|
||||
"golang.org/x/sys/windows"
|
||||
)
|
||||
|
||||
// startSleeperIn launches a long-running child whose cwd is dir but whose
|
||||
// executable lives outside it — what the path scan cannot see.
|
||||
func startSleeperIn(t *testing.T, dir string) *exec.Cmd {
|
||||
t.Helper()
|
||||
ping := filepath.Join(os.Getenv("SystemRoot"), "System32", "ping.exe")
|
||||
cmd := exec.Command(ping, "-n", "60", "127.0.0.1")
|
||||
cmd.Dir = dir
|
||||
cmd.SysProcAttr = &syscall.SysProcAttr{HideWindow: true}
|
||||
require.NoError(t, cmd.Start())
|
||||
t.Cleanup(func() {
|
||||
_ = cmd.Process.Kill()
|
||||
_, _ = cmd.Process.Wait()
|
||||
})
|
||||
return cmd
|
||||
}
|
||||
|
||||
// processCWD opens pid and reads its working directory. Test-only; the reaper
|
||||
// shares one handle across read and kill via processCWDHandle.
|
||||
func processCWD(pid uint32) (string, error) {
|
||||
h, err := windows.OpenProcess(windows.PROCESS_QUERY_INFORMATION|windows.PROCESS_VM_READ, false, pid)
|
||||
if err != nil {
|
||||
return "", err
|
||||
}
|
||||
defer func() { _ = windows.CloseHandle(h) }()
|
||||
return processCWDHandle(h)
|
||||
}
|
||||
|
||||
// awaitCWD waits until pid's PEB reports a readable cwd (not populated the
|
||||
// instant the process starts) and returns it.
|
||||
func awaitCWD(t *testing.T, pid int) string {
|
||||
t.Helper()
|
||||
var cwd string
|
||||
require.Eventually(t, func() bool {
|
||||
var err error
|
||||
cwd, err = processCWD(uint32(pid))
|
||||
return err == nil && cwd != ""
|
||||
}, 5*time.Second, 50*time.Millisecond, "process cwd should become readable")
|
||||
return cwd
|
||||
}
|
||||
|
||||
func TestProcessCWDReadsWorkingDirectory(t *testing.T) {
|
||||
dir := t.TempDir()
|
||||
cmd := startSleeperIn(t, dir)
|
||||
cwd := awaitCWD(t, cmd.Process.Pid)
|
||||
require.True(t, dirContainsPath(dir, cwd), "cwd %q should be within %q", cwd, dir)
|
||||
}
|
||||
|
||||
func TestKillProcessesWithCWDUnderTerminatesMatch(t *testing.T) {
|
||||
dir := t.TempDir()
|
||||
// Executable outside dir and arguments carry no workspace path, so only the
|
||||
// working directory links this process to dir.
|
||||
cmd := startSleeperIn(t, dir)
|
||||
awaitCWD(t, cmd.Process.Pid)
|
||||
|
||||
killed, err := KillProcessesWithCWDUnder(context.Background(), []string{dir})
|
||||
require.NoError(t, err)
|
||||
require.GreaterOrEqual(t, killed, 1)
|
||||
|
||||
require.Eventually(t, func() bool {
|
||||
return !processAlive(cmd.Process.Pid)
|
||||
}, 5*time.Second, 50*time.Millisecond, "matched process should exit")
|
||||
}
|
||||
|
||||
func TestKillProcessesWithCWDUnderSparesUnrelated(t *testing.T) {
|
||||
workspace := t.TempDir()
|
||||
elsewhere := t.TempDir()
|
||||
cmd := startSleeperIn(t, elsewhere)
|
||||
awaitCWD(t, cmd.Process.Pid)
|
||||
|
||||
_, err := KillProcessesWithCWDUnder(context.Background(), []string{workspace})
|
||||
require.NoError(t, err)
|
||||
require.True(t, processAlive(cmd.Process.Pid), "process outside owned dirs must be spared")
|
||||
}
|
||||
Reference in New Issue
Block a user