mirror of
https://github.com/status-im/consul.git
synced 2025-01-25 05:00:32 +00:00
2b377dc624
This turns out to have a lot more subtelty than we accounted for. The test suite is especially prone to races now we can only poll the child and many extra levels of indirectoin are needed to correctly run daemon process without it becoming a Zombie. I ran this test suite in a loop with parallel enabled to verify for races (-race doesn't find any as they are logical inter-process ones not actual data races). I made it through ~50 runs before hitting an error due to timing which is much better than before. I want to go back and see if we can do better though. Just getting this up.
71 lines
1.5 KiB
Go
71 lines
1.5 KiB
Go
package proxy
|
|
|
|
import (
|
|
"os"
|
|
"path/filepath"
|
|
"testing"
|
|
"time"
|
|
|
|
"github.com/hashicorp/consul/testutil/retry"
|
|
"github.com/stretchr/testify/require"
|
|
)
|
|
|
|
func TestExternalWait(t *testing.T) {
|
|
t.Parallel()
|
|
|
|
require := require.New(t)
|
|
td, closer := testTempDir(t)
|
|
defer closer()
|
|
path := filepath.Join(td, "file")
|
|
|
|
cmd := helperProcess("restart", path)
|
|
require.NoError(cmd.Start())
|
|
exitCh := make(chan struct{})
|
|
// Launch waiter to make sure this process isn't zombified when it exits part
|
|
// way through the test.
|
|
go func() {
|
|
cmd.Process.Wait()
|
|
close(exitCh)
|
|
}()
|
|
defer cmd.Process.Kill()
|
|
|
|
// Create waiter
|
|
pollInterval := 1 * time.Millisecond
|
|
waitCh, closer := externalWait(cmd.Process.Pid, pollInterval)
|
|
defer closer()
|
|
|
|
// Wait for the file to exist so we don't rely on timing to not race with
|
|
// process startup.
|
|
retry.Run(t, func(r *retry.R) {
|
|
_, err := os.Stat(path)
|
|
if err == nil {
|
|
return
|
|
}
|
|
|
|
r.Fatalf("error: %s", err)
|
|
})
|
|
|
|
// waitCh should not be closed until process quits. We'll wait a bit to verify
|
|
// we weren't just too quick to see a process exit
|
|
select {
|
|
case <-waitCh:
|
|
t.Fatal("waitCh should not be closed yet")
|
|
default:
|
|
}
|
|
|
|
// Delete the file
|
|
require.NoError(os.Remove(path))
|
|
|
|
// Wait for the child to actually exit cleanly
|
|
<-exitCh
|
|
|
|
// Now we _should_ see waitCh close (need to wait at least a whole poll
|
|
// interval)
|
|
select {
|
|
case <-waitCh:
|
|
// OK
|
|
case <-time.After(10 * pollInterval):
|
|
t.Fatal("waitCh should be closed")
|
|
}
|
|
}
|