fix: validate gridNet in NewClient and add logging for state management

Author: SalmaElsoly

backend/internal/core/workflows/deployer_activities.go

@@ -65,6 +65,7 @@ func ensureClient(state ewf.State) {
 
 func DeployNetworkStep() ewf.StepFn {
 	return func(ctx context.Context, state ewf.State) error {
+		log := logger.ForOperation("deployer_activities", "deploy_network")
 		ensureClient(state)
 
 		config, err := getConfig(state)
@@ -91,7 +92,9 @@ func DeployNetworkStep() ewf.StepFn {
 		statemanager.StoreCluster(state, cluster)
 		err = kubeClient.DeployNetwork(ctx, &cluster)
 		// Save GridClient state after network deployment
-		statemanager.SaveGridClientState(state, kubeClient)
+		if err := statemanager.SaveGridClientState(state, kubeClient); err != nil {
+			log.Warn().Err(err).Msg("failed to save GridClient state after network deployment")
+		}
 		statemanager.StoreCluster(state, cluster)
 		if err != nil {
 			nodeIDs := make([]uint32, 0, len(cluster.Nodes))
@@ -114,6 +117,7 @@ func DeployNetworkStep() ewf.StepFn {
 
 func UpdateNetworkStep() ewf.StepFn {
 	return func(ctx context.Context, state ewf.State) error {
+		log := logger.ForOperation("deployer_activities", "update_network")
 		ensureClient(state)
 
 		config, err := getConfig(state)
@@ -144,7 +148,9 @@ func UpdateNetworkStep() ewf.StepFn {
 		}
 
 		// Save GridClient state after network update
-		statemanager.SaveGridClientState(state, kubeClient)
+		if err := statemanager.SaveGridClientState(state, kubeClient); err != nil {
+			log.Warn().Err(err).Msg("failed to save GridClient state after network update")
+		}
 		statemanager.StoreCluster(state, cluster)
 		state["node"] = node
 		return nil
@@ -153,6 +159,7 @@ func UpdateNetworkStep() ewf.StepFn {
 
 func AddNodeStep() ewf.StepFn {
 	return func(ctx context.Context, state ewf.State) error {
+		log := logger.ForOperation("deployer_activities", "add_node")
 		ensureClient(state)
 
 		config, err := getConfig(state)
@@ -184,7 +191,9 @@ func AddNodeStep() ewf.StepFn {
 		}
 
 		// Save GridClient state after node deployment
-		statemanager.SaveGridClientState(state, kubeClient)
+		if err := statemanager.SaveGridClientState(state, kubeClient); err != nil {
+			log.Warn().Err(err).Msg("failed to save GridClient state after node deployment")
+		}
 		statemanager.StoreCluster(state, cluster)
 
 		// Store the deployed node in state for verification step
@@ -239,7 +248,9 @@ func DeployLeaderNodeStep() ewf.StepFn {
 
 		log.Debug().Str("node_name", leaderNode.Name).Msg("Leader node deployed successfully")
 
-		statemanager.SaveGridClientState(state, kubeClient)
+		if err := statemanager.SaveGridClientState(state, kubeClient); err != nil {
+			log.Warn().Err(err).Msg("failed to save GridClient state after leader node deployment")
+		}
 		statemanager.StoreCluster(state, cluster)
 		return nil
 	}
@@ -291,7 +302,9 @@ func BatchDeployAllNodesStep(metrics *metricsLib.Metrics) ewf.StepFn {
 
 		batchErr := kubeClient.BatchDeployNodes(ctx, &cluster, nodesToDeploy, config.SSHPublicKey)
 
-		statemanager.SaveGridClientState(state, kubeClient)
+		if err := statemanager.SaveGridClientState(state, kubeClient); err != nil {
+			log.Warn().Err(err).Msg("failed to save GridClient state after batch node deployment")
+		}
 		statemanager.StoreCluster(state, cluster)
 
 		if batchErr != nil {
@@ -524,6 +537,7 @@ func DeleteAllUserClustersStep(clusterRepo models.ClusterRepository, metrics *me
 
 func RemoveDeploymentNodeStep() ewf.StepFn {
 	return func(ctx context.Context, state ewf.State) error {
+		log := logger.ForOperation("deployer_activities", "remove_deployment_node")
 		ensureClient(state)
 
 		config, err := getConfig(state)
@@ -553,7 +567,10 @@ func RemoveDeploymentNodeStep() ewf.StepFn {
 		}
 
 		// Save GridClient state after node removal
-		statemanager.SaveGridClientState(state, kubeClient)
+		if err := statemanager.SaveGridClientState(state, kubeClient); err != nil {
+			log.Warn().Err(err).Msg("failed to save GridClient state after node removal")
+		}
+
 		statemanager.StoreCluster(state, existingCluster)
 		return nil
 	}
@@ -563,7 +580,9 @@ func closeClient(ctx context.Context, wf *ewf.Workflow, err error) {
 	log := logger.ForOperation("deployer_activities", "close_client").With().Str("workflow_name", wf.Name).Logger()
 	if kubeClient, ok := wf.State["kubeclient"].(*kubedeployer.Client); ok {
 		// Save final GridClient state before closing
-		statemanager.SaveGridClientState(wf.State, kubeClient)
+		if err := statemanager.SaveGridClientState(wf.State, kubeClient); err != nil {
+			log.Warn().Err(err).Msg("failed to save GridClient state before closing client")
+		}
 
 		kubeClient.Close()
 		delete(wf.State, "kubeclient")

backend/internal/deployment/kubedeployer/client.go

@@ -1,6 +1,7 @@
 package kubedeployer
 
 import (
+	"fmt"
 	"kubecloud/internal/infrastructure/gridclient"
 
 	sdktrace "go.opentelemetry.io/otel/sdk/trace"
@@ -12,10 +13,11 @@ type Client struct {
 
 // NewClient creates a new Client instance
 func NewClient(mnemonic, gridNet string, debug bool, tp *sdktrace.TracerProvider) (*Client, error) {
-	var opts []gridclient.ClientOpts
-	if gridNet != "" {
-		opts = append(opts, gridclient.WithNetwork(gridNet))
+	if gridNet == "" {
+		return nil, fmt.Errorf("gridNet is required and cannot be empty")
 	}
+	var opts []gridclient.ClientOpts
+	opts = append(opts, gridclient.WithNetwork(gridNet))
 	if debug {
 		opts = append(opts, gridclient.WithDebug())
 	}

backend/internal/deployment/statemanager/client_manager.go

@@ -85,7 +85,9 @@ func GetKubeClient(state ewf.State, config ClientConfig) (*kubedeployer.Client,
 
 	// Store the new client in state for reuse
 	state["kubeclient"] = kubeClient
-	SaveGridClientState(state, kubeClient)
+	if err := SaveGridClientState(state, kubeClient); err != nil {
+		log.Warn().Err(err).Msg("failed to save GridClient state after creating kubeclient")
+	}
 
 	log.Debug().Msg("Created and stored fresh kubeclient")
 	return kubeClient, nil
@@ -114,7 +116,9 @@ func CloseClient(state ewf.State, kubeClient *kubedeployer.Client) error {
 		return nil
 	}
 
-	SaveGridClientState(state, kubeClient)
+	if err := SaveGridClientState(state, kubeClient); err != nil {
+		log.Warn().Err(err).Msg("failed to save GridClient state before closing client")
+	}
 	kubeClient.Close()
 	delete(state, "kubeclient")
 

backend/internal/deployment/statemanager/gridclient_state.go

@@ -1,25 +1,31 @@
 package statemanager
 
 import (
+	"fmt"
 	"kubecloud/internal/deployment/kubedeployer"
+	"kubecloud/internal/infrastructure/logger"
 
 	"github.com/xmonader/ewf"
 )
 
 const gridClientStateKey = "gridclient_state"
 
 // SaveGridClientState saves the GridClient state to the workflow state
-func SaveGridClientState(workflowState ewf.State, kubeClient *kubedeployer.Client) {
+func SaveGridClientState(workflowState ewf.State, kubeClient *kubedeployer.Client) error {
+	log := logger.ForOperation("statemanager", "save_gridclient_state")
+
 	if kubeClient == nil || kubeClient.GridClient == nil {
-		return
+		return fmt.Errorf("kubeClient or its GridClient is nil")
 	}
 
 	stateData, err := kubeClient.GridClient.GetState()
 	if err != nil {
-		return
+		return fmt.Errorf("failed to get GridClient state: %w", err)
 	}
 
 	workflowState[gridClientStateKey] = stateData
+	log.Debug().Msg("GridClient state saved successfully")
+	return nil
 }
 
 // RestoreGridClientState restores the GridClient state from the workflow state

backend/internal/infrastructure/gridclient/grid_client.go

@@ -11,6 +11,7 @@ import (
 	"kubecloud/internal/infrastructure/logger"
 	"math"
 	"net/http"
+	"sync"
 	"time"
 
 	"github.com/cosmos/go-bip39"
@@ -79,6 +80,7 @@ type GridClient interface {
 type gridClient struct {
 	gridClient     *deployer.TFPluginClient
 	systemMnemonic string
+	mu             sync.RWMutex
 }
 
 var _ GridClient = (*gridClient)(nil)

backend/internal/infrastructure/gridclient/grid_client_state.go

@@ -15,6 +15,7 @@ type gridClientState struct {
 
 // GetState returns the current GridClient state as JSON bytes
 func (s *gridClient) GetState() ([]byte, error) {
+
 	if s.gridClient.State == nil {
 		return nil, fmt.Errorf("gridclient state is nil")
 	}
@@ -24,6 +25,7 @@ func (s *gridClient) GetState() ([]byte, error) {
 		NetworkSubnets:         make(map[string]map[uint32]string),
 	}
 
+	s.mu.RLock()
 	// Save CurrentNodeDeployments
 	for nodeID, contractIDs := range s.gridClient.State.CurrentNodeDeployments {
 		gridState.CurrentNodeDeployments[nodeID] = []uint64(contractIDs)
@@ -34,15 +36,13 @@ func (s *gridClient) GetState() ([]byte, error) {
 		gridState.NetworkSubnets[networkName] = network.Subnets
 	}
 
+	s.mu.RUnlock()
+
 	return json.Marshal(gridState)
 }
 
 // RestoreState restores the GridClient state from JSON bytes
 func (s *gridClient) RestoreState(stateData []byte) error {
-	if s.gridClient.State == nil {
-		return fmt.Errorf("gridclient state is nil")
-	}
-
 	if len(stateData) == 0 {
 		return nil // Nothing to restore
 	}
@@ -52,6 +52,11 @@ func (s *gridClient) RestoreState(stateData []byte) error {
 		return fmt.Errorf("failed to unmarshal state: %w", err)
 	}
 
+	if s.gridClient.State == nil {
+		return fmt.Errorf("gridclient state is nil")
+	}
+
+	s.mu.Lock()
 	// Restore CurrentNodeDeployments
 	s.gridClient.State.CurrentNodeDeployments = make(map[uint32]state.ContractIDs)
 	for nodeID, contractIDs := range savedState.CurrentNodeDeployments {
@@ -65,6 +70,7 @@ func (s *gridClient) RestoreState(stateData []byte) error {
 			Subnets: subnets,
 		}
 	}
+	s.mu.Unlock()
 
 	return nil
 }