feat: health check endpoint (#971)

* chore: move Nostr relay readiness methods to service.go * feat: basic health checks implemented * chore: rename alby info variable * feat: add basic health check ui, remove unnecessary alarms * fix: health indicator layout on mobile --------- Co-authored-by: Roland Bewick <[email protected]>
getAlby · Jan 17, 2025 · ff9ef56 · ff9ef56
1 parent fae02f3
commit ff9ef56
Show file tree

Hide file tree

Showing 18 changed files with 253 additions and 46 deletions.
diff --git a/api/api.go b/api/api.go
@@ -1024,6 +1024,50 @@ func (api *api) GetLogOutput(ctx context.Context, logType string, getLogRequest
 	return &GetLogOutputResponse{Log: string(logData)}, nil
 }
 
+func (api *api) Health(ctx context.Context) (*HealthResponse, error) {
+	var alarms []HealthAlarm
+
+	albyInfo, err := api.albyOAuthSvc.GetInfo(ctx)
+	if err != nil {
+		return nil, err
+	}
+	if !albyInfo.Healthy {
+		alarms = append(alarms, NewHealthAlarm(HealthAlarmKindAlbyService, albyInfo.Incidents))
+	}
+
+	isNostrRelayReady := api.svc.IsRelayReady()
+	if !isNostrRelayReady {
+		alarms = append(alarms, NewHealthAlarm(HealthAlarmKindNostrRelayOffline, nil))
+	}
+
+	lnClient := api.svc.GetLNClient()
+
+	if lnClient != nil {
+		nodeStatus, err := lnClient.GetNodeStatus(ctx)
+		if err != nil {
+			return nil, err
+		}
+		if nodeStatus == nil || !nodeStatus.IsReady {
+			alarms = append(alarms, NewHealthAlarm(HealthAlarmKindNodeNotReady, nodeStatus))
+		}
+
+		channels, err := lnClient.ListChannels(ctx)
+		if err != nil {
+			return nil, err
+		}
+
+		offlineChannels := slices.DeleteFunc(channels, func(channel lnclient.Channel) bool {
+			return channel.Active
+		})
+
+		if len(offlineChannels) > 0 {
+			alarms = append(alarms, NewHealthAlarm(HealthAlarmKindChannelsOffline, nil))
+		}
+	}
+
+	return &HealthResponse{Alarms: alarms}, nil
+}
+
 func (api *api) parseExpiresAt(expiresAtString string) (*time.Time, error) {
 	var expiresAt *time.Time
 	if expiresAtString != "" {

diff --git a/api/models.go b/api/models.go
@@ -56,6 +56,7 @@ type API interface {
 	RestoreBackup(unlockPassword string, r io.Reader) error
 	MigrateNodeStorage(ctx context.Context, to string) error
 	GetWalletCapabilities(ctx context.Context) (*WalletCapabilitiesResponse, error)
+	Health(ctx context.Context) (*HealthResponse, error)
 }
 
 type App struct {
@@ -366,3 +367,28 @@ type Channel struct {
 type MigrateNodeStorageRequest struct {
 	To string `json:"to"`
 }
+
+type HealthAlarmKind string
+
+const (
+	HealthAlarmKindAlbyService       HealthAlarmKind = "alby_service"
+	HealthAlarmKindNodeNotReady                      = "node_not_ready"
+	HealthAlarmKindChannelsOffline                   = "channels_offline"
+	HealthAlarmKindNostrRelayOffline                 = "nostr_relay_offline"
+)
+
+type HealthAlarm struct {
+	Kind       HealthAlarmKind `json:"kind"`
+	RawDetails any             `json:"rawDetails,omitempty"`
+}
+
+func NewHealthAlarm(kind HealthAlarmKind, rawDetails any) HealthAlarm {
+	return HealthAlarm{
+		Kind:       kind,
+		RawDetails: rawDetails,
+	}
+}
+
+type HealthResponse struct {
+	Alarms []HealthAlarm `json:"alarms,omitempty"`
+}
diff --git a/frontend/src/components/layouts/AppLayout.tsx b/frontend/src/components/layouts/AppLayout.tsx
@@ -48,12 +48,15 @@ import {
 } from "src/components/ui/tooltip";
 import { useAlbyMe } from "src/hooks/useAlbyMe";
 
+import clsx from "clsx";
 import { useAlbyInfo } from "src/hooks/useAlbyInfo";
+import { useHealthCheck } from "src/hooks/useHealthCheck";
 import { useInfo } from "src/hooks/useInfo";
 import { useNotifyReceivedPayments } from "src/hooks/useNotifyReceivedPayments";
 import { useRemoveSuccessfulChannelOrder } from "src/hooks/useRemoveSuccessfulChannelOrder";
 import { deleteAuthToken } from "src/lib/auth";
 import { cn } from "src/lib/utils";
+import { HealthAlarm } from "src/types";
 import { isHttpMode } from "src/utils/isHttpMode";
 import { openLink } from "src/utils/openLink";
 import ExternalLink from "../ExternalLink";
@@ -230,6 +233,7 @@ export default function AppLayout() {
                       <AlbyHubLogo className="text-foreground" />
                     </Link>
                     <AppVersion />
+                    <HealthIndicator />
                   </div>
                   <MainMenuContent />
                 </nav>
@@ -285,9 +289,9 @@ export default function AppLayout() {
                       <Link to="/">
                         <AlbyHubLogo className="text-foreground" />
                       </Link>
-                      {/* align shield with x icon */}
-                      <div className="mr-2">
+                      <div className="mr-1 flex gap-2 items-center justify-center">
                         <AppVersion />
+                        <HealthIndicator />
                       </div>
                     </div>
                     <MainMenuContent />
@@ -369,6 +373,65 @@ function AppVersion() {
   );
 }
 
+function HealthIndicator() {
+  const { data: health } = useHealthCheck();
+  if (!health) {
+    return null;
+  }
+
+  const ok = !health.alarms?.length;
+
+  function getAlarmTitle(alarm: HealthAlarm) {
+    // TODO: could show extra data from alarm.rawDetails
+    // for some alarm types
+    switch (alarm.kind) {
+      case "alby_service":
+        return "One or more Alby Services are offline";
+      case "channels_offline":
+        return "One or more channels are offline";
+      case "node_not_ready":
+        return "Node is not ready";
+      case "nostr_relay_offline":
+        return "Could not connect to relay";
+      default:
+        return "Unknown error";
+    }
+  }
+
+  return (
+    <TooltipProvider>
+      <Tooltip>
+        <TooltipTrigger>
+          <span className="text-xs flex items-center text-muted-foreground">
+            <div
+              className={clsx(
+                "w-2 h-2 rounded-full",
+                ok ? "bg-green-300" : "bg-destructive"
+              )}
+            />
+          </span>
+        </TooltipTrigger>
+        <TooltipContent>
+          {ok ? (
+            <p>Alby Hub is running</p>
+          ) : (
+            <div>
+              <p className="font-semibold">
+                {health.alarms.length} issues were found
+              </p>
+              <ul className="mt-2 max-w-xs whitespace-pre-wrap list-disc list-inside">
+                {health.alarms.map((alarm) => (
+                  <li key={alarm.kind}>{getAlarmTitle(alarm)}</li>
+                ))}
+              </ul>
+            </div>
+          )}
+        </TooltipContent>
+      </Tooltip>
+    </TooltipProvider>
+  );
+}
+
 const MenuItem = ({
   to,
   children,

diff --git a/frontend/src/hooks/useHealthCheck.ts b/frontend/src/hooks/useHealthCheck.ts
@@ -0,0 +1,16 @@
+import useSWR, { SWRConfiguration } from "swr";
+
+import { HealthResponse } from "src/types";
+import { swrFetcher } from "src/utils/swr";
+
+const pollConfiguration: SWRConfiguration = {
+  refreshInterval: 30000,
+};
+
+export function useHealthCheck(poll = true) {
+  return useSWR<HealthResponse>(
+    "/api/health",
+    swrFetcher,
+    poll ? pollConfiguration : undefined
+  );
+}
diff --git a/frontend/src/types.ts b/frontend/src/types.ts
@@ -157,6 +157,21 @@ export interface InfoResponse {
   autoUnlockPasswordEnabled: boolean;
 }
 
+export type HealthAlarmKind =
+  | "alby_service"
+  | "node_not_ready"
+  | "channels_offline"
+  | "nostr_relay_offline";
+
+export type HealthAlarm = {
+  kind: HealthAlarmKind;
+  rawDetails: unknown;
+};
+
+export type HealthResponse = {
+  alarms: HealthAlarm[];
+};
+
 export type Network = "bitcoin" | "testnet" | "signet";
 
 export type AppMetadata = { app_store_app_id?: string } & Record<

diff --git a/http/http_service.go b/http/http_service.go
@@ -153,6 +153,7 @@ func (httpSvc *HttpService) RegisterSharedRoutes(e *echo.Echo) {
 	restrictedGroup.POST("/api/send-payment-probes", httpSvc.sendPaymentProbesHandler)
 	restrictedGroup.POST("/api/send-spontaneous-payment-probes", httpSvc.sendSpontaneousPaymentProbesHandler)
 	restrictedGroup.GET("/api/log/:type", httpSvc.getLogOutputHandler)
+	restrictedGroup.GET("/api/health", httpSvc.healthHandler)
 
 	httpSvc.albyHttpSvc.RegisterSharedRoutes(restrictedGroup, e)
 }
@@ -1072,3 +1073,14 @@ func (httpSvc *HttpService) restoreBackupHandler(c echo.Context) error {
 
 	return c.NoContent(http.StatusNoContent)
 }
+
+func (httpSvc *HttpService) healthHandler(c echo.Context) error {
+	healthResponse, err := httpSvc.api.Health(c.Request().Context())
+	if err != nil {
+		return c.JSON(http.StatusInternalServerError, ErrorResponse{
+			Message: fmt.Sprintf("Failed to check node health: %v", err),
+		})
+	}
+
+	return c.JSON(http.StatusOK, healthResponse)
+}
diff --git a/lnclient/breez/breez.go b/lnclient/breez/breez.go
@@ -426,7 +426,9 @@ func (bs *BreezService) GetLogOutput(ctx context.Context, maxLen int) ([]byte, e
 }
 
 func (bs *BreezService) GetNodeStatus(ctx context.Context) (nodeStatus *lnclient.NodeStatus, err error) {
-	return nil, nil
+	return &lnclient.NodeStatus{
+		IsReady: true,
+	}, nil
 }
 
 func (bs *BreezService) SignMessage(ctx context.Context, message string) (string, error) {

diff --git a/lnclient/cashu/cashu.go b/lnclient/cashu/cashu.go
@@ -258,7 +258,9 @@ func (cs *CashuService) GetNetworkGraph(ctx context.Context, nodeIds []string) (
 func (cs *CashuService) UpdateLastWalletSyncRequest() {}
 
 func (cs *CashuService) GetNodeStatus(ctx context.Context) (nodeStatus *lnclient.NodeStatus, err error) {
-	return nil, nil
+	return &lnclient.NodeStatus{
+		IsReady: true,
+	}, nil
 }
 
 func (cs *CashuService) SendPaymentProbes(ctx context.Context, invoice string) error {

diff --git a/lnclient/greenlight/greenlight.go b/lnclient/greenlight/greenlight.go
@@ -659,7 +659,9 @@ func (gs *GreenlightService) GetStorageDir() (string, error) {
 }
 
 func (gs *GreenlightService) GetNodeStatus(ctx context.Context) (nodeStatus *lnclient.NodeStatus, err error) {
-	return nil, nil
+	return &lnclient.NodeStatus{
+		IsReady: true,
+	}, nil
 }
 
 func (gs *GreenlightService) GetNetworkGraph(ctx context.Context, nodeIds []string) (lnclient.NetworkGraphResponse, error) {

diff --git a/lnclient/ldk/ldk.go b/lnclient/ldk/ldk.go
@@ -255,35 +255,38 @@ func NewLDKService(ctx context.Context, cfg config.Config, eventPublisher events
 	}).Info("LDK node synced successfully")
 
 	if ls.network == "bitcoin" {
-		// try to connect to some peers to retrieve P2P gossip data. TODO: Remove once LDK can correctly do gossip with CLN and Eclair nodes
-		// see https://github.com/lightningdevkit/rust-lightning/issues/3075
-		peers := []string{
-			"031b301307574bbe9b9ac7b79cbe1700e31e544513eae0b5d7497483083f99e581@45.79.192.236:9735",   // Olympus
-			"0364913d18a19c671bb36dd04d6ad5be0fe8f2894314c36a9db3f03c2d414907e1@192.243.215.102:9735", // LQwD
-			"035e4ff418fc8b5554c5d9eea66396c227bd429a3251c8cbc711002ba215bfc226@170.75.163.209:9735",  // WoS
-			"02fcc5bfc48e83f06c04483a2985e1c390cb0f35058baa875ad2053858b8e80dbd@35.239.148.251:9735",  // Blink
-			"027100442c3b79f606f80f322d98d499eefcb060599efc5d4ecb00209c2cb54190@3.230.33.224:9735",    // c=
-			"038a9e56512ec98da2b5789761f7af8f280baf98a09282360cd6ff1381b5e889bf@64.23.162.51:9735",    // Megalith LSP
-		}
-		logger.Logger.Info("Connecting to some peers to retrieve P2P gossip data")
-		for _, peer := range peers {
-			parts := strings.FieldsFunc(peer, func(r rune) bool { return r == '@' || r == ':' })
-			port, err := strconv.ParseUint(parts[2], 10, 16)
-			if err != nil {
-				logger.Logger.WithError(err).Error("Failed to parse port number")
-				continue
+		go func() {
+			// try to connect to some peers in the background to retrieve P2P gossip data.
+			// TODO: Remove once LDK can correctly do gossip with CLN and Eclair nodes
+			// see https://github.com/lightningdevkit/rust-lightning/issues/3075
+			peers := []string{
+				"031b301307574bbe9b9ac7b79cbe1700e31e544513eae0b5d7497483083f99e581@45.79.192.236:9735",   // Olympus
+				"0364913d18a19c671bb36dd04d6ad5be0fe8f2894314c36a9db3f03c2d414907e1@192.243.215.102:9735", // LQwD
+				"035e4ff418fc8b5554c5d9eea66396c227bd429a3251c8cbc711002ba215bfc226@170.75.163.209:9735",  // WoS
+				"02fcc5bfc48e83f06c04483a2985e1c390cb0f35058baa875ad2053858b8e80dbd@35.239.148.251:9735",  // Blink
+				// "027100442c3b79f606f80f322d98d499eefcb060599efc5d4ecb00209c2cb54190@3.230.33.224:9735",    // c=
+				"038a9e56512ec98da2b5789761f7af8f280baf98a09282360cd6ff1381b5e889bf@64.23.162.51:9735", // Megalith LSP
 			}
-			err = ls.ConnectPeer(ctx, &lnclient.ConnectPeerRequest{
-				Pubkey:  parts[0],
-				Address: parts[1],
-				Port:    uint16(port),
-			})
-			if err != nil {
-				logger.Logger.WithFields(logrus.Fields{
-					"peer": peer,
-				}).WithError(err).Error("Failed to connect to peer")
+			logger.Logger.Info("Connecting to some peers to retrieve P2P gossip data")
+			for _, peer := range peers {
+				parts := strings.FieldsFunc(peer, func(r rune) bool { return r == '@' || r == ':' })
+				port, err := strconv.ParseUint(parts[2], 10, 16)
+				if err != nil {
+					logger.Logger.WithError(err).Error("Failed to parse port number")
+					continue
+				}
+				err = ls.ConnectPeer(ctx, &lnclient.ConnectPeerRequest{
+					Pubkey:  parts[0],
+					Address: parts[1],
+					Port:    uint16(port),
+				})
+				if err != nil {
+					logger.Logger.WithFields(logrus.Fields{
+						"peer": peer,
+					}).WithError(err).Error("Failed to connect to peer")
+				}
 			}
-		}
+		}()
 	}
 
 	// setup background sync
@@ -1644,8 +1647,10 @@ func deleteOldLDKLogs(ldkLogDir string) {
 }
 
 func (ls *LDKService) GetNodeStatus(ctx context.Context) (nodeStatus *lnclient.NodeStatus, err error) {
+	status := ls.node.Status()
 	return &lnclient.NodeStatus{
-		InternalNodeStatus: ls.node.Status(),
+		IsReady:            status.IsRunning && status.IsListening,
+		InternalNodeStatus: status,
 	}, nil
 }
 

diff --git a/lnclient/lnd/lnd.go b/lnclient/lnd/lnd.go
@@ -1060,7 +1060,7 @@ func lndPaymentToTransaction(payment *lnrpc.Payment) (*lnclient.Transaction, err
 		DescriptionHash: descriptionHash,
 		ExpiresAt:       expiresAt,
 		SettledAt:       settledAt,
-		//TODO: Metadata:  (e.g. keysend),
+		// TODO: Metadata:  (e.g. keysend),
 	}, nil
 }
 
@@ -1131,6 +1131,7 @@ func (svc *LNDService) GetNodeStatus(ctx context.Context) (nodeStatus *lnclient.
 	}
 
 	return &lnclient.NodeStatus{
+		IsReady: true, // Assuming that, if GetNodeInfo() succeeds, the node is online and accessible.
 		InternalNodeStatus: map[string]interface{}{
 			"info":         info,
 			"config":       debugInfo.Config,

diff --git a/lnclient/models.go b/lnclient/models.go
@@ -100,6 +100,7 @@ type Channel struct {
 }
 
 type NodeStatus struct {
+	IsReady            bool        `json:"isReady"`
 	InternalNodeStatus interface{} `json:"internalNodeStatus"`
 }
 

diff --git a/lnclient/phoenixd/phoenixd.go b/lnclient/phoenixd/phoenixd.go
@@ -507,7 +507,9 @@ func (svc *PhoenixService) GetLogOutput(ctx context.Context, maxLen int) ([]byte
 }
 
 func (svc *PhoenixService) GetNodeStatus(ctx context.Context) (nodeStatus *lnclient.NodeStatus, err error) {
-	return nil, nil
+	return &lnclient.NodeStatus{
+		IsReady: true,
+	}, nil
 }
 
 func (svc *PhoenixService) GetStorageDir() (string, error) {
-Original file line number
+Diff line change
@@ Expand Up / @@ -100,6 +100,7 @@ type Channel struct { @@
     }
     type NodeStatus struct {
+    	IsReady            bool        `json:"isReady"`
     	InternalNodeStatus interface{} `json:"internalNodeStatus"`
     }
@@ Expand Down @@