func (r *remote) handleConnectionChange() {
var transientFailureCount = 0
ticker := time.NewTicker(500 * time.Millisecond)
defer ticker.Stop()
healthClient := grpc_health_v1.NewHealthClient(r.rpcConn)
for {
<-ticker.C
ctx, cancel := context.WithTimeout(context.Background(), containerdHealthCheckTimeout)
_, err := healthClient.Check(ctx, &grpc_health_v1.HealthCheckRequest{})
cancel()
if err == nil {
continue
}
logrus.Debugf("libcontainerd: containerd health check returned error: %v", err)
if r.daemonPid != -1 {
if strings.Contains(err.Error(), "is closing") {
// Well, we asked for it to stop, just return
return
}
// all other errors are transient
// Reset state to be notified of next failure
transientFailureCount++
if transientFailureCount >= maxConnectionRetryCount {
transientFailureCount = 0
if system.IsProcessAlive(r.daemonPid) {
system.KillProcess(r.daemonPid)
}
<-r.daemonWaitCh
if err := r.runContainerdDaemon(); err != nil { //FIXME: Handle error
logrus.Errorf("libcontainerd: error restarting containerd: %v", err)
}
continue
}
}
}
}
func (r *remote) runContainerdDaemon() error {
pidFilename := filepath.Join(r.stateDir, containerdPidFilename)
f, err := os.OpenFile(pidFilename, os.O_RDWR|os.O_CREATE, 0600)
if err != nil {
return err
}
defer f.Close()
// File exist, check if the daemon is alive
b := make([]byte, 8)
n, err := f.Read(b)
if err != nil && err != io.EOF {
return err
}
if n > 0 {
pid, err := strconv.ParseUint(string(b[:n]), 10, 64)
if err != nil {
return err
}
if system.IsProcessAlive(int(pid)) {
logrus.Infof("libcontainerd: previous instance of containerd still alive (%d)", pid)
r.daemonPid = int(pid)
return nil
}
}
// rewind the file
_, err = f.Seek(0, os.SEEK_SET)
if err != nil {
return err
}
// Truncate it
err = f.Truncate(0)
if err != nil {
return err
}
// Start a new instance
args := []string{
"-l", fmt.Sprintf("unix://%s", r.rpcAddr),
"--metrics-interval=0",
"--start-timeout", "2m",
"--state-dir", filepath.Join(r.stateDir, containerdStateDir),
}
if goruntime.GOOS == "solaris" {
args = append(args, "--shim", "containerd-shim", "--runtime", "runc")
} else {
args = append(args, "--shim", "docker-containerd-shim")
if r.runtime != "" {
args = append(args, "--runtime")
args = append(args, r.runtime)
}
}
if r.debugLog {
args = append(args, "--debug")
}
if len(r.runtimeArgs) > 0 {
for _, v := range r.runtimeArgs {
args = append(args, "--runtime-args")
args = append(args, v)
}
logrus.Debugf("libcontainerd: runContainerdDaemon: runtimeArgs: %s", args)
}
cmd := exec.Command(containerdBinary, args...)
// redirect containerd logs to docker logs
cmd.Stdout = os.Stdout
cmd.Stderr = os.Stderr
cmd.SysProcAttr = setSysProcAttr(true)
cmd.Env = nil
// clear the NOTIFY_SOCKET from the env when starting containerd
for _, e := range os.Environ() {
if !strings.HasPrefix(e, "NOTIFY_SOCKET") {
cmd.Env = append(cmd.Env, e)
}
}
if err := cmd.Start(); err != nil {
return err
}
logrus.Infof("libcontainerd: new containerd process, pid: %d", cmd.Process.Pid)
if err := setOOMScore(cmd.Process.Pid, r.oomScore); err != nil {
system.KillProcess(cmd.Process.Pid)
return err
}
if _, err := f.WriteString(fmt.Sprintf("%d", cmd.Process.Pid)); err != nil {
system.KillProcess(cmd.Process.Pid)
return err
}
r.daemonWaitCh = make(chan struct{})
go func() {
cmd.Wait()
close(r.daemonWaitCh)
}() // Reap our child when needed
r.daemonPid = cmd.Process.Pid
return nil
}