[client] Add health check flag to status command and expose daemon status in output (#5650)

Author: lixmal

client/Dockerfile

@@ -17,8 +17,8 @@ ENV \
     NETBIRD_BIN="/usr/local/bin/netbird" \
     NB_LOG_FILE="console,/var/log/netbird/client.log" \
     NB_DAEMON_ADDR="unix:///var/run/netbird.sock" \
-    NB_ENTRYPOINT_SERVICE_TIMEOUT="5" \
-    NB_ENTRYPOINT_LOGIN_TIMEOUT="5"
+    NB_ENTRYPOINT_SERVICE_TIMEOUT="30" \
+    NB_ENTRYPOINT_LOGIN_TIMEOUT="30"
 
 ENTRYPOINT [ "/usr/local/bin/netbird-entrypoint.sh" ]
 

client/Dockerfile-rootless

@@ -23,8 +23,8 @@ ENV \
     NB_DAEMON_ADDR="unix:///var/lib/netbird/netbird.sock" \
     NB_LOG_FILE="console,/var/lib/netbird/client.log" \
     NB_DISABLE_DNS="true" \
-    NB_ENTRYPOINT_SERVICE_TIMEOUT="5" \
-    NB_ENTRYPOINT_LOGIN_TIMEOUT="1"
+    NB_ENTRYPOINT_SERVICE_TIMEOUT="30" \
+    NB_ENTRYPOINT_LOGIN_TIMEOUT="30"
 
 ENTRYPOINT [ "/usr/local/bin/netbird-entrypoint.sh" ]
 

client/cmd/status.go

@@ -28,6 +28,7 @@ var (
 	ipsFilterMap         map[string]struct{}
 	prefixNamesFilterMap map[string]struct{}
 	connectionTypeFilter string
+	checkFlag            string
 )
 
 var statusCmd = &cobra.Command{
@@ -49,13 +50,18 @@ func init() {
 	statusCmd.PersistentFlags().StringSliceVar(&prefixNamesFilter, "filter-by-names", []string{}, "filters the detailed output by a list of one or more peer FQDN or hostnames, e.g., --filter-by-names peer-a,peer-b.netbird.cloud")
 	statusCmd.PersistentFlags().StringVar(&statusFilter, "filter-by-status", "", "filters the detailed output by connection status(idle|connecting|connected), e.g., --filter-by-status connected")
 	statusCmd.PersistentFlags().StringVar(&connectionTypeFilter, "filter-by-connection-type", "", "filters the detailed output by connection type (P2P|Relayed), e.g., --filter-by-connection-type P2P")
+	statusCmd.PersistentFlags().StringVar(&checkFlag, "check", "", "run a health check and exit with code 0 on success, 1 on failure (live|ready|startup)")
 }
 
 func statusFunc(cmd *cobra.Command, args []string) error {
 	SetFlagsFromEnvVars(rootCmd)
 
 	cmd.SetOut(cmd.OutOrStdout())
 
+	if checkFlag != "" {
+		return runHealthCheck(cmd)
+	}
+
 	err := parseFilters()
 	if err != nil {
 		return err
@@ -68,15 +74,17 @@ func statusFunc(cmd *cobra.Command, args []string) error {
 
 	ctx := internal.CtxInitState(cmd.Context())
 
-	resp, err := getStatus(ctx, false)
+	resp, err := getStatus(ctx, true, false)
 	if err != nil {
 		return err
 	}
 
 	status := resp.GetStatus()
 
-	if status == string(internal.StatusNeedsLogin) || status == string(internal.StatusLoginFailed) ||
-		status == string(internal.StatusSessionExpired) {
+	needsAuth := status == string(internal.StatusNeedsLogin) || status == string(internal.StatusLoginFailed) ||
+		status == string(internal.StatusSessionExpired)
+
+	if needsAuth && !jsonFlag && !yamlFlag {
 		cmd.Printf("Daemon status: %s\n\n"+
 			"Run UP command to log in with SSO (interactive login):\n\n"+
 			" netbird up \n\n"+
@@ -99,7 +107,17 @@ func statusFunc(cmd *cobra.Command, args []string) error {
 		profName = activeProf.Name
 	}
 
-	var outputInformationHolder = nbstatus.ConvertToStatusOutputOverview(resp.GetFullStatus(), anonymizeFlag, resp.GetDaemonVersion(), statusFilter, prefixNamesFilter, prefixNamesFilterMap, ipsFilterMap, connectionTypeFilter, profName)
+	var outputInformationHolder = nbstatus.ConvertToStatusOutputOverview(resp.GetFullStatus(), nbstatus.ConvertOptions{
+		Anonymize:            anonymizeFlag,
+		DaemonVersion:        resp.GetDaemonVersion(),
+		DaemonStatus:         nbstatus.ParseDaemonStatus(status),
+		StatusFilter:         statusFilter,
+		PrefixNamesFilter:    prefixNamesFilter,
+		PrefixNamesFilterMap: prefixNamesFilterMap,
+		IPsFilter:            ipsFilterMap,
+		ConnectionTypeFilter: connectionTypeFilter,
+		ProfileName:          profName,
+	})
 	var statusOutputString string
 	switch {
 	case detailFlag:
@@ -121,7 +139,7 @@ func statusFunc(cmd *cobra.Command, args []string) error {
 	return nil
 }
 
-func getStatus(ctx context.Context, shouldRunProbes bool) (*proto.StatusResponse, error) {
+func getStatus(ctx context.Context, fullPeerStatus bool, shouldRunProbes bool) (*proto.StatusResponse, error) {
 	conn, err := DialClientGRPCServer(ctx, daemonAddr)
 	if err != nil {
 		//nolint
@@ -131,7 +149,7 @@ func getStatus(ctx context.Context, shouldRunProbes bool) (*proto.StatusResponse
 	}
 	defer conn.Close()
 
-	resp, err := proto.NewDaemonServiceClient(conn).Status(ctx, &proto.StatusRequest{GetFullPeerStatus: true, ShouldRunProbes: shouldRunProbes})
+	resp, err := proto.NewDaemonServiceClient(conn).Status(ctx, &proto.StatusRequest{GetFullPeerStatus: fullPeerStatus, ShouldRunProbes: shouldRunProbes})
 	if err != nil {
 		return nil, fmt.Errorf("status failed: %v", status.Convert(err).Message())
 	}
@@ -185,6 +203,83 @@ func enableDetailFlagWhenFilterFlag() {
 	}
 }
 
+func runHealthCheck(cmd *cobra.Command) error {
+	check := strings.ToLower(checkFlag)
+	switch check {
+	case "live", "ready", "startup":
+	default:
+		return fmt.Errorf("unknown check %q, must be one of: live, ready, startup", checkFlag)
+	}
+
+	if err := util.InitLog(logLevel, util.LogConsole); err != nil {
+		return fmt.Errorf("init log: %w", err)
+	}
+
+	ctx := internal.CtxInitState(cmd.Context())
+
+	isStartup := check == "startup"
+	resp, err := getStatus(ctx, isStartup, isStartup)
+	if err != nil {
+		return err
+	}
+
+	switch check {
+	case "live":
+		return nil
+	case "ready":
+		return checkReadiness(resp)
+	case "startup":
+		return checkStartup(resp)
+	default:
+		return nil
+	}
+}
+
+func checkReadiness(resp *proto.StatusResponse) error {
+	daemonStatus := internal.StatusType(resp.GetStatus())
+	switch daemonStatus {
+	case internal.StatusIdle, internal.StatusConnecting, internal.StatusConnected:
+		return nil
+	case internal.StatusNeedsLogin, internal.StatusLoginFailed, internal.StatusSessionExpired:
+		return fmt.Errorf("readiness check: daemon status is %s", daemonStatus)
+	default:
+		return fmt.Errorf("readiness check: unexpected daemon status %q", daemonStatus)
+	}
+}
+
+func checkStartup(resp *proto.StatusResponse) error {
+	fullStatus := resp.GetFullStatus()
+	if fullStatus == nil {
+		return fmt.Errorf("startup check: no full status available")
+	}
+
+	if !fullStatus.GetManagementState().GetConnected() {
+		return fmt.Errorf("startup check: management not connected")
+	}
+
+	if !fullStatus.GetSignalState().GetConnected() {
+		return fmt.Errorf("startup check: signal not connected")
+	}
+
+	var relayCount, relaysConnected int
+	for _, r := range fullStatus.GetRelays() {
+		uri := r.GetURI()
+		if !strings.HasPrefix(uri, "rel://") && !strings.HasPrefix(uri, "rels://") {
+			continue
+		}
+		relayCount++
+		if r.GetAvailable() {
+			relaysConnected++
+		}
+	}
+
+	if relayCount > 0 && relaysConnected == 0 {
+		return fmt.Errorf("startup check: no relay servers available (0/%d connected)", relayCount)
+	}
+
+	return nil
+}
+
 func parseInterfaceIP(interfaceIP string) string {
 	ip, _, err := net.ParseCIDR(interfaceIP)
 	if err != nil {

client/internal/debug/debug.go

@@ -31,7 +31,6 @@ import (
 	nbstatus "github.com/netbirdio/netbird/client/status"
 	mgmProto "github.com/netbirdio/netbird/shared/management/proto"
 	"github.com/netbirdio/netbird/util"
-	"github.com/netbirdio/netbird/version"
 )
 
 const readmeContent = `Netbird debug bundle
@@ -418,7 +417,10 @@ func (g *BundleGenerator) addStatus() error {
 		fullStatus := g.statusRecorder.GetFullStatus()
 		protoFullStatus := nbstatus.ToProtoFullStatus(fullStatus)
 		protoFullStatus.Events = g.statusRecorder.GetEventHistory()
-		overview := nbstatus.ConvertToStatusOutputOverview(protoFullStatus, g.anonymize, version.NetbirdVersion(), "", nil, nil, nil, "", profName)
+		overview := nbstatus.ConvertToStatusOutputOverview(protoFullStatus, nbstatus.ConvertOptions{
+			Anonymize:   g.anonymize,
+			ProfileName: profName,
+		})
 		statusOutput := overview.FullDetailSummary()
 
 		statusReader := strings.NewReader(statusOutput)

client/netbird-entrypoint.sh

@@ -1,12 +1,11 @@
 #!/usr/bin/env bash
 set -eEuo pipefail
 
-: ${NB_ENTRYPOINT_SERVICE_TIMEOUT:="5"}
-: ${NB_ENTRYPOINT_LOGIN_TIMEOUT:="5"}
+: ${NB_ENTRYPOINT_SERVICE_TIMEOUT:="30"}
+: ${NB_ENTRYPOINT_LOGIN_TIMEOUT:="30"}
 NETBIRD_BIN="${NETBIRD_BIN:-"netbird"}"
 export NB_LOG_FILE="${NB_LOG_FILE:-"console,/var/log/netbird/client.log"}"
 service_pids=()
-log_file_path=""
 
 _log() {
   # mimic Go logger's output for easier parsing
@@ -33,60 +32,50 @@ on_exit() {
   fi
 }
 
-wait_for_message() {
-  local timeout="${1}" message="${2}"
-  if test "${timeout}" -eq 0; then
-    info "not waiting for log line ${message@Q} due to zero timeout."
-  elif test -n "${log_file_path}"; then
-    info "waiting for log line ${message@Q} for ${timeout} seconds..."
-    grep -E -q "${message}" <(timeout "${timeout}" tail -F "${log_file_path}" 2>/dev/null)
-  else
-    info "log file unsupported, sleeping for ${timeout} seconds..."
-    sleep "${timeout}"
-  fi
-}
-
-locate_log_file() {
-  local log_files_string="${1}"
-
-  while read -r log_file; do
-    case "${log_file}" in
-    console | syslog) ;;
-    *)
-      log_file_path="${log_file}"
-      return
-      ;;
-    esac
-  done < <(sed 's#,#\n#g' <<<"${log_files_string}")
-
-  warn "log files parsing for ${log_files_string@Q} is not supported by debug bundles"
-  warn "please consider removing the \$NB_LOG_FILE or setting it to real file, before gathering debug bundles."
-}
-
 wait_for_daemon_startup() {
   local timeout="${1}"
+  if [[ "${timeout}" -eq 0 ]]; then
+    info "not waiting for daemon startup due to zero timeout."
+    return
+  fi
 
-  if test -n "${log_file_path}"; then
-    if ! wait_for_message "${timeout}" "started daemon server"; then
-      warn "log line containing 'started daemon server' not found after ${timeout} seconds"
-      warn "daemon failed to start, exiting..."
-      exit 1
+  local deadline=$((SECONDS + timeout))
+  while [[ "${SECONDS}" -lt "${deadline}" ]]; do
+    if "${NETBIRD_BIN}" status --check live 2>/dev/null; then
+      return
     fi
-  else
-    warn "daemon service startup not discovered, sleeping ${timeout} instead"
-    sleep "${timeout}"
-  fi
+    sleep 1
+  done
+
+  warn "daemon did not become responsive after ${timeout} seconds, exiting..."
+  exit 1
 }
 
 login_if_needed() {
   local timeout="${1}"
 
-  if test -n "${log_file_path}" && wait_for_message "${timeout}" 'peer has been successfully registered|management connection state READY'; then
+  if "${NETBIRD_BIN}" status --check ready 2>/dev/null; then
     info "already logged in, skipping 'netbird up'..."
-  else
+    return
+  fi
+
+  if [[ "${timeout}" -eq 0 ]]; then
     info "logging in..."
     "${NETBIRD_BIN}" up
+    return
   fi
+
+  local deadline=$((SECONDS + timeout))
+  while [[ "${SECONDS}" -lt "${deadline}" ]]; do
+    if "${NETBIRD_BIN}" status --check ready 2>/dev/null; then
+      info "already logged in, skipping 'netbird up'..."
+      return
+    fi
+    sleep 1
+  done
+
+  info "logging in..."
+  "${NETBIRD_BIN}" up
 }
 
 main() {
@@ -95,7 +84,6 @@ main() {
   service_pids+=("$!")
   info "registered new service process 'netbird service run', currently running: ${service_pids[@]@Q}"
 
-  locate_log_file "${NB_LOG_FILE}"
   wait_for_daemon_startup "${NB_ENTRYPOINT_SERVICE_TIMEOUT}"
   login_if_needed "${NB_ENTRYPOINT_LOGIN_TIMEOUT}"