refactor: removed uncessary panics in zetaclientd process (#2210)

* removed uncessary panics in zetaclientd process

* added changelog entry

* fix unit test and renamed  as

* throw and handle error from MonitorCore()

* resolved conflict in ObserveInTx()

* return error from WaitForZetacoreToCreateBlocks() and SetAccountNumber() and handle them in start.go

---------

Co-authored-by: Lucas Bertrand <[email protected]>

Author: ws4charlie

changelog.md

@@ -19,13 +19,14 @@
 ### Refactor
 
 * [2094](https://github.com/zeta-chain/node/pull/2094) - upgrade go-tss to use cosmos v0.47
-* [2110](https://github.com/zeta-chain/node/pull/2110) - move non-query rate limiter logic to zetaclient side and code refactor.
+* [2110](https://github.com/zeta-chain/node/pull/2110) - move non-query rate limiter logic to zetaclient side and code refactor
 * [2032](https://github.com/zeta-chain/node/pull/2032) - improve some general structure of the ZetaClient codebase
 * [2097](https://github.com/zeta-chain/node/pull/2097) - refactor lightclient verification flags to account for individual chains
 * [2071](https://github.com/zeta-chain/node/pull/2071) - Modify chains struct to add all chain related information
 * [2118](https://github.com/zeta-chain/node/pull/2118) - consolidate inbound and outbound naming
 * [2124](https://github.com/zeta-chain/node/pull/2124) - removed unused variables and method
-* [2150](https://github.com/zeta-chain/node/pull/2150) - created `chains` `zetacore` `orchestrator` packages in zetaclient and reorganized source files accordingly.
+* [2150](https://github.com/zeta-chain/node/pull/2150) - created `chains` `zetacore` `orchestrator` packages in zetaclient and reorganized source files accordingly
+* [2210](https://github.com/zeta-chain/node/pull/2210) - removed uncessary panics in the zetaclientd process
 * [2205](https://github.com/zeta-chain/node/pull/2205) - remove deprecated variables pre-v17
 
 ### Tests

cmd/zetaclientd-supervisor/main.go

@@ -16,10 +16,13 @@ import (
 )
 
 func main() {
+	// load zetaclient config
 	cfg, err := config.Load(app.DefaultNodeHome)
 	if err != nil {
-		panic(fmt.Errorf("failed to load config: %w", err))
+		fmt.Println("failed to load config: ", err)
+		os.Exit(1)
 	}
+
 	// log outputs must be serialized since we are writing log messages in this process and
 	// also directly from the zetaclient process
 	serializedStdout := &serializedWriter{upstream: os.Stdout}
@@ -38,13 +41,15 @@ func main() {
 
 	hotkeyPassword, tssPassword, err := promptPasswords()
 	if err != nil {
-		panic(fmt.Errorf("unable to get passwords: %w", err))
+		logger.Error().Err(err).Msg("unable to get passwords")
+		os.Exit(1)
 	}
 
 	_, enableAutoDownload := os.LookupEnv("ZETACLIENTD_SUPERVISOR_ENABLE_AUTO_DOWNLOAD")
 	supervisor, err := newZetaclientdSupervisor(cfg.ZetaCoreURL, logger, enableAutoDownload)
 	if err != nil {
-		panic(fmt.Errorf("unable to get supervisor: %w", err))
+		logger.Error().Err(err).Msg("unable to get supervisor")
+		os.Exit(1)
 	}
 	supervisor.Start(ctx)
 

cmd/zetaclientd/debug.go

@@ -24,7 +24,6 @@ import (
 	"github.com/zeta-chain/zetacore/zetaclient/config"
 	clientcontext "github.com/zeta-chain/zetacore/zetaclient/context"
 	"github.com/zeta-chain/zetacore/zetaclient/keys"
-	"github.com/zeta-chain/zetacore/zetaclient/metrics"
 	"github.com/zeta-chain/zetacore/zetaclient/zetacore"
 )
 
@@ -62,22 +61,14 @@ func DebugCmd() *cobra.Command {
 			var ballotIdentifier string
 			chainLogger := zerolog.New(io.Discard).Level(zerolog.Disabled)
 
-			telemetryServer := metrics.NewTelemetryServer()
-			go func() {
-				err := telemetryServer.Start()
-				if err != nil {
-					panic("telemetryServer error")
-				}
-			}()
-
+			// create a new zetacore client
 			client, err := zetacore.NewClient(
 				&keys.Keys{OperatorAddress: sdk.MustAccAddressFromBech32(sample.AccAddress())},
 				debugArgs.zetaNode,
 				"",
 				debugArgs.zetaChainID,
 				false,
-				telemetryServer)
-
+				nil)
 			if err != nil {
 				return err
 			}
@@ -89,14 +80,13 @@ func DebugCmd() *cobra.Command {
 			if err != nil {
 				return err
 			}
-
 			chain := chains.GetChainFromChainID(chainID)
 			if chain == nil {
 				return fmt.Errorf("invalid chain id")
 			}
 
+			// get ballot identifier according to the chain type
 			if chains.IsEVMChain(chain.ChainId) {
-
 				evmObserver := evmobserver.Observer{
 					Mu: &sync.Mutex{},
 				}
@@ -204,6 +194,7 @@ func DebugCmd() *cobra.Command {
 			}
 			fmt.Println("BallotIdentifier : ", ballotIdentifier)
 
+			// query ballot
 			ballot, err := client.GetBallot(ballotIdentifier)
 			if err != nil {
 				return err

cmd/zetaclientd/p2p_diagnostics.go

@@ -151,68 +151,70 @@ func RunDiagnostics(startLogger zerolog.Logger, peers p2p.AddrList, hotkeyPk cry
 	// every 1min, print out the p2p diagnostic
 	ticker := time.NewTicker(time.Duration(cfg.P2PDiagnosticTicker) * time.Second)
 	round := 0
-	for {
-		select {
-		case <-ticker.C:
-			round++
-			// Now, look for others who have announced
-			// This is like your friend telling you the location to meet you.
-			startLogger.Info().Msgf("Searching for other peers...")
-			peerChan, err := routingDiscovery.FindPeers(context.Background(), "ZetaZetaOpenTheDoor")
+
+	for range ticker.C {
+		round++
+		// Now, look for others who have announced
+		// This is like your friend telling you the location to meet you.
+		startLogger.Info().Msgf("Searching for other peers...")
+		peerChan, err := routingDiscovery.FindPeers(context.Background(), "ZetaZetaOpenTheDoor")
+		if err != nil {
+			return err
+		}
+
+		peerCount := 0
+		okPingPongCount := 0
+		for peer := range peerChan {
+			peerCount++
+			if peer.ID == host.ID() {
+				startLogger.Info().Msgf("Found myself #(%d): %s", peerCount, peer)
+				continue
+			}
+			startLogger.Info().Msgf("Found peer #(%d): %s; pinging the peer...", peerCount, peer)
+			stream, err := host.NewStream(context.Background(), peer.ID, protocol.ID(ProtocolID))
 			if err != nil {
-				panic(err)
+				startLogger.Error().Err(err).Msgf("fail to create stream to peer %s", peer)
+				continue
 			}
 
-			peerCount := 0
-			okPingPongCount := 0
-			for peer := range peerChan {
-				peerCount++
-				if peer.ID == host.ID() {
-					startLogger.Info().Msgf("Found myself #(%d): %s", peerCount, peer)
-					continue
-				}
-				startLogger.Info().Msgf("Found peer #(%d): %s; pinging the peer...", peerCount, peer)
-				stream, err := host.NewStream(context.Background(), peer.ID, protocol.ID(ProtocolID))
-				if err != nil {
-					startLogger.Error().Err(err).Msgf("fail to create stream to peer %s", peer)
-					continue
-				}
-				message := fmt.Sprintf("round %d %s => %s", round, host.ID().String()[len(host.ID().String())-5:], peer.ID.String()[len(peer.ID.String())-5:])
-				_, err = stream.Write([]byte(message))
-				if err != nil {
-					startLogger.Error().Err(err).Msgf("fail to write to stream to peer %s", peer)
-					err = stream.Close()
-					if err != nil {
-						startLogger.Warn().Err(err).Msgf("fail to close stream to peer %s", peer)
-					}
-					continue
-				}
-				//startLogger.Debug().Msgf("wrote %d bytes", nw)
-				buf := make([]byte, 1024)
-				nr, err := stream.Read(buf)
+			// write a message to the stream
+			message := fmt.Sprintf("round %d %s => %s", round, host.ID().String()[len(host.ID().String())-5:], peer.ID.String()[len(peer.ID.String())-5:])
+			_, err = stream.Write([]byte(message))
+			if err != nil {
+				startLogger.Error().Err(err).Msgf("fail to write to stream to peer %s", peer)
+				err = stream.Close()
 				if err != nil {
-					startLogger.Error().Err(err).Msgf("fail to read from stream to peer %s", peer)
-					err = stream.Close()
-					if err != nil {
-						startLogger.Warn().Err(err).Msgf("fail to close stream to peer %s", peer)
-					}
-					continue
+					startLogger.Warn().Err(err).Msgf("fail to close stream to peer %s", peer)
 				}
-				//startLogger.Debug().Msgf("read %d bytes", nr)
-				startLogger.Debug().Msgf("echoed message: %s", string(buf[:nr]))
+				continue
+			}
+
+			// read the echoed message
+			buf := make([]byte, 1024)
+			nr, err := stream.Read(buf)
+			if err != nil {
+				startLogger.Error().Err(err).Msgf("fail to read from stream to peer %s", peer)
 				err = stream.Close()
 				if err != nil {
 					startLogger.Warn().Err(err).Msgf("fail to close stream to peer %s", peer)
 				}
+				continue
+			}
+			startLogger.Debug().Msgf("echoed message: %s", string(buf[:nr]))
+			err = stream.Close()
+			if err != nil {
+				startLogger.Warn().Err(err).Msgf("fail to close stream to peer %s", peer)
+			}
 
-				if string(buf[:nr]) != message {
-					startLogger.Error().Msgf("ping-pong failed with peer #(%d): %s; want %s got %s", peerCount, peer, message, string(buf[:nr]))
-					continue
-				}
-				startLogger.Info().Msgf("ping-pong success with peer #(%d): %s;", peerCount, peer)
-				okPingPongCount++
+			// check if the message is echoed correctly
+			if string(buf[:nr]) != message {
+				startLogger.Error().Msgf("ping-pong failed with peer #(%d): %s; want %s got %s", peerCount, peer, message, string(buf[:nr]))
+				continue
 			}
-			startLogger.Info().Msgf("Expect %d peers in total; successful pings (%d/%d)", peerCount, okPingPongCount, peerCount-1)
+			startLogger.Info().Msgf("ping-pong success with peer #(%d): %s;", peerCount, peer)
+			okPingPongCount++
 		}
+		startLogger.Info().Msgf("Expect %d peers in total; successful pings (%d/%d)", peerCount, okPingPongCount, peerCount-1)
 	}
+	return nil
 }

cmd/zetaclientd/start.go

@@ -91,24 +91,38 @@ func start(_ *cobra.Command, _ []string) error {
 		}
 	}()
 
-	// CreateZetaCoreClient:  zetacore client is used for all communication to zetacore , which this client connects to.
+	// CreateZetacoreClient:  zetacore client is used for all communication to zetacore , which this client connects to.
 	// Zetacore accumulates votes , and provides a centralized source of truth for all clients
-	zetacoreClient, err := CreateZetaCoreClient(cfg, telemetryServer, hotkeyPass)
+	zetacoreClient, err := CreateZetacoreClient(cfg, telemetryServer, hotkeyPass)
 	if err != nil {
-		panic(err)
+		startLogger.Error().Err(err).Msg("CreateZetacoreClient error")
+		return err
+	}
+
+	// Wait until zetacore is ready to create blocks
+	err = zetacoreClient.WaitForZetacoreToCreateBlocks()
+	if err != nil {
+		startLogger.Error().Err(err).Msg("WaitForZetacoreToCreateBlocks error")
+		return err
 	}
-	zetacoreClient.WaitForCoreToCreateBlocks()
 	startLogger.Info().Msgf("Zetacore client is ready")
-	zetacoreClient.SetAccountNumber(authz.ZetaClientGranteeKey)
+
+	// Set grantee account number and sequence number
+	err = zetacoreClient.SetAccountNumber(authz.ZetaClientGranteeKey)
+	if err != nil {
+		startLogger.Error().Err(err).Msg("SetAccountNumber error")
+		return err
+	}
 
 	// cross-check chainid
 	res, err := zetacoreClient.GetNodeInfo()
 	if err != nil {
-		panic(err)
+		startLogger.Error().Err(err).Msg("GetNodeInfo error")
+		return err
 	}
 
 	if strings.Compare(res.GetDefaultNodeInfo().Network, cfg.ChainID) != 0 {
-		startLogger.Warn().Msgf("chain id mismatch, zeta-core chain id %s, zeta client chain id %s; reset zeta client chain id", res.GetDefaultNodeInfo().Network, cfg.ChainID)
+		startLogger.Warn().Msgf("chain id mismatch, zetacore chain id %s, zetaclient configured chain id %s; reset zetaclient chain id", res.GetDefaultNodeInfo().Network, cfg.ChainID)
 		cfg.ChainID = res.GetDefaultNodeInfo().Network
 		err := zetacoreClient.UpdateChainID(cfg.ChainID)
 		if err != nil {
@@ -118,7 +132,12 @@ func start(_ *cobra.Command, _ []string) error {
 
 	// CreateAuthzSigner : which is used to sign all authz messages . All votes broadcast to zetacore are wrapped in authz exec .
 	// This is to ensure that the user does not need to keep their operator key online , and can use a cold key to sign votes
-	CreateAuthzSigner(zetacoreClient.GetKeys().GetOperatorAddress().String(), zetacoreClient.GetKeys().GetAddress())
+	signerAddress, err := zetacoreClient.GetKeys().GetAddress()
+	if err != nil {
+		startLogger.Error().Err(err).Msg("error getting signer address")
+		return err
+	}
+	CreateAuthzSigner(zetacoreClient.GetKeys().GetOperatorAddress().String(), signerAddress)
 	startLogger.Debug().Msgf("CreateAuthzSigner is ready")
 
 	// Initialize core parameters from zetacore
@@ -265,8 +284,12 @@ func start(_ *cobra.Command, _ []string) error {
 	}
 
 	// Orchestrator wraps the zetacore client and adds the observers and signer maps to it . This is the high level object used for CCTX interactions
-	mo1 := orchestrator.NewOrchestrator(zetacoreClient, signerMap, observerMap, masterLogger, telemetryServer)
-	mo1.MonitorCore(appContext)
+	orchestrator := orchestrator.NewOrchestrator(zetacoreClient, signerMap, observerMap, masterLogger, telemetryServer)
+	err = orchestrator.MonitorCore(appContext)
+	if err != nil {
+		startLogger.Error().Err(err).Msg("Orchestrator failed to start")
+		return err
+	}
 
 	// start zeta supply checker
 	// TODO: enable

cmd/zetaclientd/utils.go

@@ -21,7 +21,7 @@ func CreateAuthzSigner(granter string, grantee sdk.AccAddress) {
 	authz.SetupAuthZSignerList(granter, grantee)
 }
 
-func CreateZetaCoreClient(cfg config.Config, telemetry *metrics.TelemetryServer, hotkeyPassword string) (*zetacore.Client, error) {
+func CreateZetacoreClient(cfg config.Config, telemetry *metrics.TelemetryServer, hotkeyPassword string) (*zetacore.Client, error) {
 	hotKey := cfg.AuthzHotkey
 	if cfg.HsmMode {
 		hotKey = cfg.HsmHotKey