Merge pull request #108 from mfreeman451/99-feat-sparkgraphs-in-grid-…

…mode

99 feat sparkgraphs in grid mode

README.md

@@ -30,45 +30,45 @@ ServiceRadar can be installed via direct downloads from GitHub releases.
 Install these components on your monitored host:
 ```bash
 # Download and install core components
-curl -LO https://github.com/mfreeman451/serviceradar/releases/download/1.0.3/serviceradar-agent_1.0.5.deb \
-     -O https://github.com/mfreeman451/serviceradar/releases/download/1.0.3/serviceradar-poller_1.0.5.deb
+curl -LO https://github.com/mfreeman451/serviceradar/releases/download/1.0.3/serviceradar-agent_1.0.6.deb \
+     -O https://github.com/mfreeman451/serviceradar/releases/download/1.0.3/serviceradar-poller_1.0.6.deb
 
-sudo dpkg -i serviceradar-agent_1.0.5.deb serviceradar-poller_1.0.5.deb
+sudo dpkg -i serviceradar-agent_1.0.6.deb serviceradar-poller_1.0.6.deb
 ```
 
 On a separate machine (recommended) or the same host:
 ```bash
 # Download and install cloud service
-curl -LO https://github.com/mfreeman451/serviceradar/releases/download/1.0.3/serviceradar-cloud_1.0.5.deb
-sudo dpkg -i serviceradar-cloud_1.0.5.deb
+curl -LO https://github.com/mfreeman451/serviceradar/releases/download/1.0.3/serviceradar-cloud_1.0.6.deb
+sudo dpkg -i serviceradar-cloud_1.0.6.deb
 ```
 
 #### Optional: Dusk Node Monitoring
 If you're running a [Dusk](https://dusk.network/) node and want specialized monitoring:
 ```bash
-curl -LO https://github.com/mfreeman451/serviceradar/releases/download/1.0.5/serviceradar-dusk-checker_1.0.5.deb
-sudo dpkg -i serviceradar-dusk-checker_1.0.5.deb
+curl -LO https://github.com/mfreeman451/serviceradar/releases/download/1.0.6/serviceradar-dusk-checker_1.0.6.deb
+sudo dpkg -i serviceradar-dusk-checker_1.0.6.deb
 ```
 
 #### Distributed Setup
 For larger deployments where components run on different hosts:
 
 1. On monitored hosts:
 ```bash
-curl -LO https://github.com/mfreeman451/serviceradar/releases/download/1.0.5/serviceradar-agent_1.0.5.deb
-sudo dpkg -i serviceradar-agent_1.0.5.deb
+curl -LO https://github.com/mfreeman451/serviceradar/releases/download/1.0.6/serviceradar-agent_1.0.6.deb
+sudo dpkg -i serviceradar-agent_1.0.6.deb
 ```
 
 2. On monitoring host:
 ```bash
-curl -LO https://github.com/mfreeman451/serviceradar/releases/download/1.0.3/serviceradar-poller_1.0.5.deb
-sudo dpkg -i serviceradar-poller_1.0.5.deb
+curl -LO https://github.com/mfreeman451/serviceradar/releases/download/1.0.3/serviceradar-poller_1.0.6.deb
+sudo dpkg -i serviceradar-poller_1.0.6.deb
 ```
 
 3. On cloud host:
 ```bash
-curl -LO https://github.com/mfreeman451/serviceradar/releases/download/1.0.3/serviceradar-cloud_1.0.5.deb
-sudo dpkg -i serviceradar-cloud_1.0.5.deb
+curl -LO https://github.com/mfreeman451/serviceradar/releases/download/1.0.3/serviceradar-cloud_1.0.6.deb
+sudo dpkg -i serviceradar-cloud_1.0.6.deb
 ```
 
 ## Architecture
@@ -170,19 +170,19 @@ cd serviceradar
 
 1. **Agent Installation** (on monitored hosts):
 ```bash
-sudo dpkg -i serviceradar-dusk-checker_1.0.5.deb  # For Dusk nodes
+sudo dpkg -i serviceradar-dusk-checker_1.0.6.deb  # For Dusk nodes
 # or
-sudo dpkg -i serviceradar-agent_1.0.5.deb  # For other hosts
+sudo dpkg -i serviceradar-agent_1.0.6.deb  # For other hosts
 ```
 
 2. **Poller Installation** (on any host in your network):
 ```bash
-sudo dpkg -i serviceradar-poller_1.0.5.deb
+sudo dpkg -i serviceradar-poller_1.0.6.deb
 ```
 
 3. **Cloud Installation** (on a reliable host):
 ```bash
-sudo dpkg -i serviceradar-cloud_1.0.5.deb
+sudo dpkg -i serviceradar-cloud_1.0.6.deb
 ```
 
 ## Configuration
@@ -258,13 +258,17 @@ Default location: `/etc/serviceradar/poller.json`
         {
           "service_type": "port",
           "service_name": "SSH",
-          "port": 22
+          "details": "127.0.0.1:22"
         },
         {
           "service_type": "grpc",
           "service_name": "dusk",
           "details": "127.0.0.1:50052"
         },
+        {
+          "service_type": "icmp",
+          "service_name": "ping"
+        },
         {
           "service_type": "sweep",
           "service_name": "network_sweep",
@@ -276,7 +280,7 @@ Default location: `/etc/serviceradar/poller.json`
   "cloud_address": "changeme:50052",
   "listen_addr": ":50053",
   "poll_interval": "30s",
-  "poller_id": "home-poller-1",
+  "poller_id": "dusk",
   "service_name": "PollerService",
   "service_type": "grpc"
 }
@@ -292,6 +296,11 @@ Default location: `/etc/serviceradar/cloud.json`
   "grpc_addr": ":50052",
   "alert_threshold": "5m",
   "known_pollers": ["home-poller-1"],
+  "metrics": {
+    "enabled": true,
+    "retention": 100,
+    "max_nodes": 10000
+  },
   "webhooks": [
     {
       "enabled": false,

buildAll.sh

@@ -1,6 +1,6 @@
 #!/bin/bash
 
-VERSION=${VERSION:-1.0.5}
+VERSION=${VERSION:-1.0.6}
 
 
 ./setup-deb-poller.sh

cmd/cloud/main.go

@@ -29,6 +29,8 @@ func run() error {
 		return err
 	}
 
+	log.Printf("cfg: %v", cfg)
+
 	// Create root context for lifecycle management
 	ctx := context.Background()
 
@@ -39,7 +41,7 @@ func run() error {
 	}
 
 	// Create and configure API server
-	apiServer := api.NewAPIServer()
+	apiServer := api.NewAPIServer(server.GetMetricsManager())
 	server.SetAPIServer(apiServer)
 
 	// Start HTTP API server in background

pkg/agent/checker.go

@@ -8,7 +8,13 @@ import (
 	"net"
 	"os/exec"
 	"regexp"
+	"strconv"
 	"strings"
+	"time"
+)
+
+const (
+	partsForPortDetails = 2
 )
 
 var (
@@ -55,21 +61,54 @@ type PortChecker struct {
 	Port int
 }
 
+func NewPortChecker(details string) (*PortChecker, error) {
+	if details == "" {
+		return nil, errDetailsRequired
+	}
+
+	// Split the details into host and port
+	parts := strings.Split(details, ":")
+	if len(parts) != partsForPortDetails {
+		return nil, errInvalidDetailsFormat
+	}
+
+	host := parts[0]
+
+	port, err := strconv.Atoi(parts[1])
+	if err != nil {
+		return nil, fmt.Errorf("%w: %d", errInvalidPort, port)
+	}
+
+	if port <= 0 || port > 65535 {
+		return nil, fmt.Errorf("%w: %d", errInvalidPort, port)
+	}
+
+	return &PortChecker{
+		Host: host,
+		Port: port,
+	}, nil
+}
+
+// Check validates if a port is accessible.
 func (p *PortChecker) Check(ctx context.Context) (isAccessible bool, statusMsg string) {
 	var d net.Dialer
 
 	addr := fmt.Sprintf("%s:%d", p.Host, p.Port)
 
+	start := time.Now()
+
 	conn, err := d.DialContext(ctx, "tcp", addr)
 	if err != nil {
-		return false, fmt.Sprintf("Port %d is not accessible: %v", p.Port, err)
+		return false, fmt.Sprintf(`{"error": "Port %d is not accessible: %v"}`, p.Port, err)
 	}
 
+	responseTime := time.Since(start).Nanoseconds()
+
 	if err = conn.Close(); err != nil {
 		log.Printf("Error closing connection: %v", err)
-
-		return false, "Error closing connection"
+		return false, `{"error": "Error closing connection"}`
 	}
 
-	return true, fmt.Sprintf("Port %d is accessible", p.Port)
+	// Return raw data
+	return true, fmt.Sprintf(`{"host": "%q", "port": %d, "response_time": %d}`, p.Host, p.Port, responseTime)
 }

pkg/agent/errors.go

@@ -3,8 +3,8 @@ package agent
 import "errors"
 
 var (
-	errGrpcAddressRequired = errors.New("address is required for gRPC checker")
-	errUnknownCheckerType  = errors.New("unknown checker type")
-	errGrpcMissingConfig   = errors.New("no configuration or address provided for gRPC checker")
-	errShutdown            = errors.New("error while shutting down")
+	errShutdown             = errors.New("error while shutting down")
+	errInvalidPort          = errors.New("invalid port")
+	errDetailsRequired      = errors.New("details field is required for port checks")
+	errInvalidDetailsFormat = errors.New("invalid details format: expected 'host:port'")
 )

pkg/agent/external_checker.go

@@ -3,8 +3,10 @@ package agent
 
 import (
 	"context"
+	"encoding/json"
 	"fmt"
 	"log"
+	"time"
 
 	grpcpkg "github.com/mfreeman451/serviceradar/pkg/grpc"
 	"github.com/mfreeman451/serviceradar/proto"
@@ -76,31 +78,38 @@ func (e *ExternalChecker) Check(ctx context.Context) (isAccessible bool, statusM
 	healthy, err := e.client.CheckHealth(ctx, "")
 	if err != nil {
 		log.Printf("External checker %s: Health check failed: %v", e.serviceName, err)
-
-		return false, fmt.Sprintf("Failed to check %s: %v", e.serviceName, err)
+		return false, fmt.Sprintf("Health check failed: %v", err)
 	}
 
 	if !healthy {
 		log.Printf("External checker %s: Service reported unhealthy", e.serviceName)
-
-		return false, fmt.Sprintf("%s is not healthy", e.serviceName)
+		return false, "Service reported unhealthy"
 	}
 
 	// Then get block details through AgentService
 	client := proto.NewAgentServiceClient(e.client.GetConnection())
 
+	start := time.Now()
 	status, err := client.GetStatus(ctx, &proto.StatusRequest{
 		ServiceName: e.serviceName,
 		ServiceType: e.serviceType,
 	})
+
 	if err != nil {
 		log.Printf("External checker %s: Failed to get details: %v", e.serviceName, err)
 
-		return true, fmt.Sprintf("%s is healthy but details unavailable", e.serviceName)
+		return true, "Service healthy but details unavailable"
 	}
 
-	log.Printf("External checker %s: Received status message: %s", e.serviceName, status.Message)
+	responseTime := time.Since(start).Nanoseconds()
+
+	// Parse status.Message into response structure
+	var details map[string]interface{}
+	if err := json.Unmarshal([]byte(status.Message), &details); err != nil {
+		return true, fmt.Sprintf(`{"response_time": %d, "error": "invalid details format"}`, responseTime)
+	}
 
+	// Pass the details directly in the status message
 	return true, status.Message
 }
 

pkg/agent/icmp_checker.go

@@ -0,0 +1,58 @@
+// Package agent pkg/agent/icmp_checker.go
+package agent
+
+import (
+	"context"
+	"fmt"
+	"time"
+
+	"github.com/mfreeman451/serviceradar/pkg/models"
+	"github.com/mfreeman451/serviceradar/pkg/scan"
+)
+
+type ICMPChecker struct {
+	Host  string
+	Count int
+}
+
+const (
+	defaultICMPCount  = 3
+	defaultConcurrent = 1
+)
+
+func (p *ICMPChecker) Check(ctx context.Context) (isDown bool, results string) {
+	scanner := scan.NewCombinedScanner(defaultICMPCount*time.Second, defaultConcurrent, p.Count)
+
+	target := models.Target{
+		Host: p.Host,
+		Mode: models.ModeICMP,
+	}
+
+	resultChan, err := scanner.Scan(ctx, []models.Target{target})
+	if err != nil {
+		return false, fmt.Sprintf("ICMP check failed: %v", err)
+	}
+
+	var totalResponseTime time.Duration
+
+	var successfulPings int
+
+	for result := range resultChan {
+		if result.Error != nil {
+			return false, result.Error.Error()
+		}
+
+		if result.Available {
+			totalResponseTime += result.RespTime
+			successfulPings++
+		}
+	}
+
+	if successfulPings == 0 {
+		return false, "No successful ICMP replies"
+	}
+
+	avgResponseTime := totalResponseTime / time.Duration(successfulPings)
+
+	return true, fmt.Sprintf("%d", avgResponseTime)
+}