fix(campaign): signal a timed-out run so it reaps its sidecar

CommandContext kills outright, so a run stopped by --run-timeout never ran its
own shutdown and left a sidecar holding a port and a quarter gigabyte,
reparented to init and deaf to SIGTERM. The timeout exists for unattended
hosts, which is exactly where nobody is watching to reap what it leaves.

Claude-Session: https://claude.ai/code/session_01A5KmftdEJ49A9z5mF5ESrX
This commit is contained in:
pj committed 2026-08-14 17:27:13 +05:30
1 parent 9a2bbbf769
commit dd17f831ef
2 files changed
+42

No files matched your search

@@ -2,6 +2,7 @@ package main
import (
"bytes"
"context"
"encoding/json"
"io"
"os"
@@ -9,6 +10,7 @@ import (
"slices"
"strings"
"testing"
"time"
)
// stubSanderling answers `version`, writes a run directory shaped like the one
@@ -142,3 +144,28 @@ func TestRun_EndToEndAgainstStubBinary(t *testing.T) {
t.Errorf("progress output: %q", stdout.String())
}
}
func TestExecuteCommand_RunTimeoutSignalsSoTheRunReapsItsChildren(t *testing.T) {
directory := t.TempDir()
marker := filepath.Join(directory, "child-reaped")
script := filepath.Join(directory, "wedged")
body := "#!/bin/sh\n" +
"sleep 300 &\n" +
"child=$!\n" +
"trap 'kill $child; echo reaped > " + marker + "; exit 143' TERM\n" +
"wait $child\n"
if err := os.WriteFile(script, []byte(body), 0o755); err != nil {
t.Fatal(err)
}
ctx, cancel := context.WithTimeout(context.Background(), time.Second)
defer cancel()
if _, err := executeCommand(ctx, script, nil, io.Discard); err != nil {
t.Fatalf("execute: %v", err)
}
if _, err := os.Stat(marker); err != nil {
t.Fatal("the run timeout killed the run outright, so it never reaped its own children: " +
"an unattended sweep leaks one sidecar per wedged run")
}
}