feat: show error reason when monitors go DOWN

Propagate check failure reasons through the entire stack:
- Checker captures specific errors (DNS, timeout, HTTP status, SSL, etc.)
- Engine tracks LastError, StatusChangedAt, LastSuccessAt per monitor
- State transitions persisted to new state_changes table
- Detail panel shows error reason, HTTP code, state duration, last
  success time, and last 5 state change events
- Monitor table shows inline error preview for DOWN services
- Alert messages include error reason
- Probe nodes forward error reasons to leader

15 files changed across models, checker, engine, store, TUI, and probes.

internal/cluster/cluster_test.go

@@ -67,6 +67,8 @@ func (m *mockStore) DeleteMaintenanceWindow(int) error                   { retur
 func (m *mockStore) IsMonitorInMaintenance(int) (bool, error)               { return false, nil }
 func (m *mockStore) GetPreference(string) (string, error)                   { return "", nil }
 func (m *mockStore) SetPreference(string, string) error                     { return nil }
+func (m *mockStore) SaveStateChange(int, string, string, string) error      { return nil }
+func (m *mockStore) GetStateChanges(int, int) ([]models.StateChange, error) { return nil, nil }
 func (m *mockStore) Close() error                                           { return nil }
 
 // --- Cluster Start Tests ---

internal/cluster/probe.go

@@ -130,6 +130,7 @@ type probeResultItem struct {
 	SiteID      int    `json:"site_id"`
 	LatencyNs   int64  `json:"latency_ns"`
 	IsUp        bool   `json:"is_up"`
+	ErrorReason string `json:"error_reason,omitempty"`
 }
 
 func probeExecuteChecks(ctx context.Context, sites []models.Site, strict, insecure *http.Client, allowPrivate bool) []probeResultItem {
@@ -157,6 +158,7 @@ loop:
 				SiteID:      s.ID,
 				LatencyNs:   cr.LatencyNs,
 				IsUp:        cr.Status == "UP",
+				ErrorReason: cr.ErrorReason,
 			})
 			mu.Unlock()
 		}(site)

internal/metrics/prometheus_test.go

@@ -2,13 +2,14 @@ package metrics
 
 import (
 	"context"
-	"gitea.lerkolabs.com/lerko/uptop/internal/models"
-	"gitea.lerkolabs.com/lerko/uptop/internal/monitor"
 	"net/http"
 	"net/http/httptest"
 	"strings"
 	"testing"
 	"time"
+
+	"gitea.lerkolabs.com/lerko/uptop/internal/models"
+	"gitea.lerkolabs.com/lerko/uptop/internal/monitor"
 )
 
 type mockStore struct {
@@ -64,6 +65,8 @@ func (m *mockStore) DeleteMaintenanceWindow(int) error                   { retur
 func (m *mockStore) IsMonitorInMaintenance(int) (bool, error)               { return false, nil }
 func (m *mockStore) GetPreference(string) (string, error)                   { return "", nil }
 func (m *mockStore) SetPreference(string, string) error                     { return nil }
+func (m *mockStore) SaveStateChange(int, string, string, string) error      { return nil }
+func (m *mockStore) GetStateChanges(int, int) ([]models.StateChange, error) { return nil, nil }
 func (m *mockStore) Close() error                                           { return nil }
 
 func TestMetricsHandler(t *testing.T) {

internal/models/models.go

@@ -35,6 +35,18 @@ type Site struct {
 	HasSSL          bool
 	LastCheck       time.Time
 	SentSSLWarning  bool
+	LastError       string
+	StatusChangedAt time.Time
+	LastSuccessAt   time.Time
+}
+
+type StateChange struct {
+	ID          int
+	SiteID      int
+	FromStatus  string
+	ToStatus    string
+	ErrorReason string
+	ChangedAt   time.Time
 }
 
 type AlertConfig struct {

internal/monitor/aggregator.go

@@ -15,6 +15,7 @@ type NodeResult struct {
 	IsUp        bool
 	LatencyNs   int64
 	CheckedAt   time.Time
+	ErrorReason string
 }
 
 func AggregateStatus(results []NodeResult, strategy AggregationStrategy) (isUp bool, avgLatencyNs int64) {

internal/monitor/checker.go

@@ -2,6 +2,7 @@ package monitor
 
 import (
 	"context"
+	"fmt"
 	"net"
 	"net/http"
 	"strconv"
@@ -21,6 +22,7 @@ type CheckResult struct {
 	LatencyNs   int64
 	HasSSL      bool
 	CertExpiry  time.Time
+	ErrorReason string
 }
 
 func RunCheck(site models.Site, strict, insecure *http.Client, globalInsecure bool, allowPrivate ...bool) CheckResult {
@@ -35,7 +37,7 @@ func RunCheck(site models.Site, strict, insecure *http.Client, globalInsecure bo
 			if ips, err := net.LookupIP(host); err == nil {
 				for _, ip := range ips {
 					if isPrivateIP(ip) {
-						return CheckResult{SiteID: site.ID, Status: "DOWN"}
+						return CheckResult{SiteID: site.ID, Status: "DOWN", ErrorReason: "target resolves to private IP"}
 					}
 				}
 			}
@@ -52,7 +54,7 @@ func RunCheck(site models.Site, strict, insecure *http.Client, globalInsecure bo
 	case "dns":
 		return runDNSCheck(site)
 	default:
-		return CheckResult{SiteID: site.ID, Status: "DOWN"}
+		return CheckResult{SiteID: site.ID, Status: "DOWN", ErrorReason: "unsupported monitor type: " + site.Type}
 	}
 }
 
@@ -68,7 +70,7 @@ func runHTTPCheck(site models.Site, strict, insecure *http.Client, globalInsecur
 
 	req, err := http.NewRequestWithContext(ctx, method, site.URL, nil)
 	if err != nil {
-		return CheckResult{SiteID: site.ID, Status: "DOWN"}
+		return CheckResult{SiteID: site.ID, Status: "DOWN", ErrorReason: "invalid request: " + err.Error()}
 	}
 
 	client := strict
@@ -88,6 +90,7 @@ func runHTTPCheck(site models.Site, strict, insecure *http.Client, globalInsecur
 
 	if err != nil {
 		result.Status = "DOWN"
+		result.ErrorReason = truncateError(err.Error(), 256)
 		return result
 	}
 	defer resp.Body.Close()
@@ -95,6 +98,11 @@ func runHTTPCheck(site models.Site, strict, insecure *http.Client, globalInsecur
 	result.StatusCode = resp.StatusCode
 	if !isCodeAccepted(resp.StatusCode, site.AcceptedCodes) {
 		result.Status = "DOWN"
+		expected := site.AcceptedCodes
+		if expected == "" {
+			expected = "200-299"
+		}
+		result.ErrorReason = fmt.Sprintf("HTTP %d (expected %s)", resp.StatusCode, expected)
 	}
 
 	if site.CheckSSL && resp.TLS != nil && len(resp.TLS.PeerCertificates) > 0 {
@@ -103,6 +111,7 @@ func runHTTPCheck(site models.Site, strict, insecure *http.Client, globalInsecur
 		result.CertExpiry = cert.NotAfter
 		if time.Now().After(cert.NotAfter) {
 			result.Status = "SSL EXP"
+			result.ErrorReason = "SSL certificate expired"
 		}
 	}
 
@@ -117,7 +126,7 @@ func runPingCheck(site models.Site) CheckResult {
 
 	pinger, err := probing.NewPinger(host)
 	if err != nil {
-		return CheckResult{SiteID: site.ID, Status: "DOWN"}
+		return CheckResult{SiteID: site.ID, Status: "DOWN", ErrorReason: "ping setup: " + err.Error()}
 	}
 	pinger.Count = 1
 	pinger.Timeout = siteTimeout(site)
@@ -127,8 +136,11 @@ func runPingCheck(site models.Site) CheckResult {
 	err = pinger.Run()
 	latency := time.Since(start)
 
-	if err != nil || pinger.Statistics().PacketsRecv == 0 {
+	if err != nil {
-		return CheckResult{SiteID: site.ID, Status: "DOWN", LatencyNs: latency.Nanoseconds()}
+		return CheckResult{SiteID: site.ID, Status: "DOWN", LatencyNs: latency.Nanoseconds(), ErrorReason: "ping failed: " + err.Error()}
+	}
+	if pinger.Statistics().PacketsRecv == 0 {
+		return CheckResult{SiteID: site.ID, Status: "DOWN", LatencyNs: latency.Nanoseconds(), ErrorReason: "no ICMP response"}
 	}
 
 	stats := pinger.Statistics()
@@ -148,7 +160,7 @@ func runPortCheck(site models.Site) CheckResult {
 	latency := time.Since(start)
 
 	if err != nil {
-		return CheckResult{SiteID: site.ID, Status: "DOWN", LatencyNs: latency.Nanoseconds()}
+		return CheckResult{SiteID: site.ID, Status: "DOWN", LatencyNs: latency.Nanoseconds(), ErrorReason: truncateError(err.Error(), 256)}
 	}
 	_ = conn.Close()
 	return CheckResult{SiteID: site.ID, Status: "UP", LatencyNs: latency.Nanoseconds()}
@@ -199,10 +211,10 @@ func runDNSCheck(site models.Site) CheckResult {
 	latency := time.Since(start)
 
 	if err != nil {
-		return CheckResult{SiteID: site.ID, Status: "DOWN", LatencyNs: latency.Nanoseconds()}
+		return CheckResult{SiteID: site.ID, Status: "DOWN", LatencyNs: latency.Nanoseconds(), ErrorReason: "DNS query failed: " + err.Error()}
 	}
 	if r.Rcode != dns.RcodeSuccess {
-		return CheckResult{SiteID: site.ID, Status: "DOWN", StatusCode: r.Rcode, LatencyNs: latency.Nanoseconds()}
+		return CheckResult{SiteID: site.ID, Status: "DOWN", StatusCode: r.Rcode, LatencyNs: latency.Nanoseconds(), ErrorReason: "DNS RCODE: " + dns.RcodeToString[r.Rcode]}
 	}
 	return CheckResult{SiteID: site.ID, Status: "UP", LatencyNs: latency.Nanoseconds()}
 }
@@ -235,3 +247,10 @@ func isCodeAccepted(code int, accepted string) bool {
 	}
 	return false
 }
+
+func truncateError(s string, max int) string {
+	if len(s) <= max {
+		return s
+	}
+	return s[:max-3] + "..."
+}

internal/monitor/monitor.go

@@ -283,6 +283,9 @@ func (e *Engine) UpdateSiteConfig(site models.Site) {
 		site.LastCheck = existing.LastCheck
 		site.SentSSLWarning = existing.SentSSLWarning
 		site.FailureCount = existing.FailureCount
+		site.LastError = existing.LastError
+		site.StatusChangedAt = existing.StatusChangedAt
+		site.LastSuccessAt = existing.LastSuccessAt
 		e.liveState[site.ID] = site
 		e.addToTokenIndex(site)
 	}
@@ -393,33 +396,45 @@ func (e *Engine) checkByID(id int) {
 		updatedSite.CertExpiry = result.CertExpiry
 		updatedSite.Latency = time.Duration(result.LatencyNs)
 		updatedSite.LastCheck = time.Now()
-		e.handleStatusChange(updatedSite, result.Status, result.StatusCode, time.Duration(result.LatencyNs))
+		e.handleStatusChange(updatedSite, result.Status, result.StatusCode, time.Duration(result.LatencyNs), result.ErrorReason)
 	}
 }
 
 func (e *Engine) checkPush(site models.Site) {
 	deadline := site.LastCheck.Add(time.Duration(site.Interval) * time.Second).Add(pushGracePeriod)
 	if time.Now().After(deadline) {
-		e.handleStatusChange(site, "DOWN", 0, 0)
+		e.handleStatusChange(site, "DOWN", 0, 0, "heartbeat missed")
 	} else if site.Status != "UP" {
-		e.handleStatusChange(site, "UP", 200, 0)
+		e.handleStatusChange(site, "UP", 200, 0, "")
 	}
 }
 
-func (e *Engine) handleStatusChange(site models.Site, rawStatus string, code int, latency time.Duration) {
+func (e *Engine) handleStatusChange(site models.Site, rawStatus string, code int, latency time.Duration, errorReason string) {
 	if !e.IsActive() {
 		return
 	}
 
 	newState := site
 	newState.StatusCode = code
+	newState.LastError = errorReason
+
+	if rawStatus == "UP" {
+		newState.LastSuccessAt = time.Now()
+		newState.LastError = ""
+	} else {
+		newState.LastSuccessAt = site.LastSuccessAt
+	}
 
 	if site.Status == "UP" && rawStatus != "UP" {
 		newState.FailureCount++
 		if newState.FailureCount > site.MaxRetries {
 			newState.Status = rawStatus
 			newState.FailureCount = site.MaxRetries + 1
+			if errorReason != "" {
+				e.AddLog(fmt.Sprintf("Monitor '%s' confirmed DOWN: %s", site.Name, errorReason))
+			} else {
 				e.AddLog(fmt.Sprintf("Monitor '%s' confirmed DOWN", site.Name))
+			}
 		} else {
 			e.AddLog(fmt.Sprintf("Monitor '%s' failed check %d/%d", site.Name, newState.FailureCount, site.MaxRetries))
 		}
@@ -431,6 +446,14 @@ func (e *Engine) handleStatusChange(site models.Site, rawStatus string, code int
 		newState.FailureCount = site.MaxRetries + 1
 	}
 
+	if newState.Status != site.Status && site.Status != "PENDING" {
+		newState.StatusChangedAt = time.Now()
+	} else if site.StatusChangedAt.IsZero() && newState.Status != "PENDING" {
+		newState.StatusChangedAt = time.Now()
+	} else {
+		newState.StatusChangedAt = site.StatusChangedAt
+	}
+
 	inMaint := e.isInMaintenance(site.ID)
 
 	if site.Type == "http" && site.CheckSSL && site.HasSSL {
@@ -455,12 +478,19 @@ func (e *Engine) handleStatusChange(site models.Site, rawStatus string, code int
 
 	e.recordCheck(site.ID, latency, rawStatus == "UP")
 
+	if newState.Status != site.Status && site.Status != "PENDING" {
+		go func() { _ = e.db.SaveStateChange(site.ID, site.Status, newState.Status, errorReason) }()
+	}
+
 	isBroken := func(s string) bool { return s == "DOWN" || s == "SSL EXP" }
 	if !isBroken(site.Status) && isBroken(newState.Status) && newState.Status != "PENDING" {
 		if inMaint {
 			e.AddLog(fmt.Sprintf("Monitor '%s' is DOWN (alerts suppressed — maintenance)", site.Name))
 		} else {
 			msg := fmt.Sprintf("Monitor '%s' is DOWN (%s)", site.Name, rawStatus)
+			if errorReason != "" {
+				msg = fmt.Sprintf("Monitor '%s' is DOWN: %s", site.Name, errorReason)
+			}
 			if site.Type == "push" {
 				msg = fmt.Sprintf("Push Monitor '%s' missed heartbeat.", site.Name)
 			}
@@ -554,7 +584,7 @@ func (e *Engine) SetAggStrategy(strategy AggregationStrategy) {
 	e.aggStrategy = strategy
 }
 
-func (e *Engine) IngestProbeResult(nodeID string, siteID int, latencyNs int64, isUp bool) {
+func (e *Engine) IngestProbeResult(nodeID string, siteID int, latencyNs int64, isUp bool, errorReason string) {
 	e.probeResultsMu.Lock()
 	if e.probeResults[siteID] == nil {
 		e.probeResults[siteID] = make(map[string]NodeResult)
@@ -564,6 +594,7 @@ func (e *Engine) IngestProbeResult(nodeID string, siteID int, latencyNs int64, i
 		IsUp:        isUp,
 		LatencyNs:   latencyNs,
 		CheckedAt:   time.Now(),
+		ErrorReason: errorReason,
 	}
 	results := make([]NodeResult, 0, len(e.probeResults[siteID]))
 	for _, r := range e.probeResults[siteID] {
@@ -588,7 +619,7 @@ func (e *Engine) IngestProbeResult(nodeID string, siteID int, latencyNs int64, i
 	updatedSite := site
 	updatedSite.Latency = time.Duration(avgLatency)
 	updatedSite.LastCheck = time.Now()
-	e.handleStatusChange(updatedSite, rawStatus, 0, time.Duration(avgLatency))
+	e.handleStatusChange(updatedSite, rawStatus, 0, time.Duration(avgLatency), errorReason)
 }
 
 func (e *Engine) GetProbeResults(siteID int) map[string]NodeResult {
@@ -601,3 +632,11 @@ func (e *Engine) GetProbeResults(siteID int) map[string]NodeResult {
 	}
 	return cp
 }
+
+func (e *Engine) GetStateChanges(siteID int, limit int) []models.StateChange {
+	changes, err := e.db.GetStateChanges(siteID, limit)
+	if err != nil {
+		return nil
+	}
+	return changes
+}

internal/monitor/monitor_test.go

@@ -2,10 +2,11 @@ package monitor
 
 import (
 	"fmt"
-	"gitea.lerkolabs.com/lerko/uptop/internal/models"
 	"sync"
 	"testing"
 	"time"
+
+	"gitea.lerkolabs.com/lerko/uptop/internal/models"
 )
 
 // --- Mock Store ---
@@ -73,6 +74,8 @@ func (m *mockStore) EndMaintenanceWindow(int) error                      { retur
 func (m *mockStore) DeleteMaintenanceWindow(int) error                      { return nil }
 func (m *mockStore) GetPreference(string) (string, error)                   { return "", nil }
 func (m *mockStore) SetPreference(string, string) error                     { return nil }
+func (m *mockStore) SaveStateChange(int, string, string, string) error      { return nil }
+func (m *mockStore) GetStateChanges(int, int) ([]models.StateChange, error) { return nil, nil }
 func (m *mockStore) Close() error                                           { return nil }
 
 func (m *mockStore) GetAllAlerts() ([]models.AlertConfig, error) {
@@ -174,7 +177,7 @@ func TestHandleStatusChange_PendingToUp(t *testing.T) {
 	site := models.Site{ID: 1, Name: "test", Status: "PENDING", MaxRetries: 3, AlertID: 1}
 	injectSite(e, site)
 
-	e.handleStatusChange(site, "UP", 200, 10*time.Millisecond)
+	e.handleStatusChange(site, "UP", 200, 10*time.Millisecond, "")
 
 	s, _ := getSite(e, 1)
 	if s.Status != "UP" {
@@ -195,7 +198,7 @@ func TestHandleStatusChange_UpIncrementFailure(t *testing.T) {
 	site := models.Site{ID: 1, Name: "test", Status: "UP", MaxRetries: 3, FailureCount: 0}
 	injectSite(e, site)
 
-	e.handleStatusChange(site, "DOWN", 500, 0)
+	e.handleStatusChange(site, "DOWN", 500, 0, "test error")
 
 	s, _ := getSite(e, 1)
 	if s.Status != "UP" {
@@ -213,7 +216,7 @@ func TestHandleStatusChange_UpToDown_ExceedsRetries(t *testing.T) {
 	site := models.Site{ID: 1, Name: "test", Status: "UP", MaxRetries: 2, FailureCount: 2, AlertID: 1}
 	injectSite(e, site)
 
-	e.handleStatusChange(site, "DOWN", 500, 0)
+	e.handleStatusChange(site, "DOWN", 500, 0, "test error")
 
 	s, _ := getSite(e, 1)
 	if s.Status != "DOWN" {
@@ -236,7 +239,7 @@ func TestHandleStatusChange_UpToDown_ZeroRetries(t *testing.T) {
 	site := models.Site{ID: 1, Name: "test", Status: "UP", MaxRetries: 0, FailureCount: 0, AlertID: 1}
 	injectSite(e, site)
 
-	e.handleStatusChange(site, "DOWN", 0, 0)
+	e.handleStatusChange(site, "DOWN", 0, 0, "test error")
 
 	s, _ := getSite(e, 1)
 	if s.Status != "DOWN" {
@@ -255,7 +258,7 @@ func TestHandleStatusChange_DownToUp_Recovery(t *testing.T) {
 	site := models.Site{ID: 1, Name: "test", Status: "DOWN", FailureCount: 4, AlertID: 1}
 	injectSite(e, site)
 
-	e.handleStatusChange(site, "UP", 200, 5*time.Millisecond)
+	e.handleStatusChange(site, "UP", 200, 5*time.Millisecond, "")
 
 	s, _ := getSite(e, 1)
 	if s.Status != "UP" {
@@ -276,7 +279,7 @@ func TestHandleStatusChange_DownStaysDown(t *testing.T) {
 	site := models.Site{ID: 1, Name: "test", Status: "DOWN", MaxRetries: 2, FailureCount: 3}
 	injectSite(e, site)
 
-	e.handleStatusChange(site, "DOWN", 0, 0)
+	e.handleStatusChange(site, "DOWN", 0, 0, "test error")
 
 	s, _ := getSite(e, 1)
 	if s.Status != "DOWN" {
@@ -295,7 +298,7 @@ func TestHandleStatusChange_SSLExpired(t *testing.T) {
 	site := models.Site{ID: 1, Name: "test", Status: "UP", MaxRetries: 0, AlertID: 1}
 	injectSite(e, site)
 
-	e.handleStatusChange(site, "SSL EXP", 0, 0)
+	e.handleStatusChange(site, "SSL EXP", 0, 0, "SSL certificate expired")
 
 	s, _ := getSite(e, 1)
 	if s.Status != "SSL EXP" {
@@ -315,7 +318,7 @@ func TestHandleStatusChange_AlertSuppressedMaintenance(t *testing.T) {
 	site := models.Site{ID: 1, Name: "test", Status: "UP", MaxRetries: 0, AlertID: 1}
 	injectSite(e, site)
 
-	e.handleStatusChange(site, "DOWN", 0, 0)
+	e.handleStatusChange(site, "DOWN", 0, 0, "test error")
 
 	s, _ := getSite(e, 1)
 	if s.Status != "DOWN" {
@@ -346,7 +349,7 @@ func TestHandleStatusChange_RecoverySuppressedMaintenance(t *testing.T) {
 	site := models.Site{ID: 1, Name: "test", Status: "DOWN", AlertID: 1}
 	injectSite(e, site)
 
-	e.handleStatusChange(site, "UP", 200, 0)
+	e.handleStatusChange(site, "UP", 200, 0, "")
 
 	s, _ := getSite(e, 1)
 	if s.Status != "UP" {
@@ -370,7 +373,7 @@ func TestHandleStatusChange_SSLWarning(t *testing.T) {
 	}
 	injectSite(e, site)
 
-	e.handleStatusChange(site, "UP", 200, 0)
+	e.handleStatusChange(site, "UP", 200, 0, "")
 
 	s, _ := getSite(e, 1)
 	if !s.SentSSLWarning {
@@ -393,7 +396,7 @@ func TestHandleStatusChange_SSLWarningNotRepeated(t *testing.T) {
 	}
 	injectSite(e, site)
 
-	e.handleStatusChange(site, "UP", 200, 0)
+	e.handleStatusChange(site, "UP", 200, 0, "")
 
 	waitAsync()
 	if len(ms.getAlertCallsSnapshot()) != 0 {
@@ -412,7 +415,7 @@ func TestHandleStatusChange_SSLWarningReset(t *testing.T) {
 	}
 	injectSite(e, site)
 
-	e.handleStatusChange(site, "UP", 200, 0)
+	e.handleStatusChange(site, "UP", 200, 0, "")
 
 	s, _ := getSite(e, 1)
 	if s.SentSSLWarning {
@@ -433,7 +436,7 @@ func TestHandleStatusChange_SSLWarningSuppressedMaint(t *testing.T) {
 	}
 	injectSite(e, site)
 
-	e.handleStatusChange(site, "UP", 200, 0)
+	e.handleStatusChange(site, "UP", 200, 0, "")
 
 	s, _ := getSite(e, 1)
 	if !s.SentSSLWarning {
@@ -452,7 +455,7 @@ func TestHandleStatusChange_InactiveEngine(t *testing.T) {
 	injectSite(e, site)
 	e.SetActive(false)
 
-	e.handleStatusChange(site, "DOWN", 0, 0)
+	e.handleStatusChange(site, "DOWN", 0, 0, "test error")
 
 	s, _ := getSite(e, 1)
 	if s.Status != "UP" {
@@ -991,7 +994,7 @@ func TestConcurrent_HandleStatusChangeAndGetState(t *testing.T) {
 		wg.Add(2)
 		go func() {
 			defer wg.Done()
-			e.handleStatusChange(site, "DOWN", 500, 0)
+			e.handleStatusChange(site, "DOWN", 500, 0, "test error")
 		}()
 		go func() {
 			defer wg.Done()

internal/server/server.go

@@ -406,6 +406,7 @@ func Start(cfg ServerConfig, s store.Store, eng *monitor.Engine) *http.Server {
 				SiteID      int    `json:"site_id"`
 				LatencyNs   int64  `json:"latency_ns"`
 				IsUp        bool   `json:"is_up"`
+				ErrorReason string `json:"error_reason"`
 			} `json:"results"`
 		}
 		if err := json.NewDecoder(r.Body).Decode(&req); err != nil {
@@ -420,7 +421,7 @@ func Start(cfg ServerConfig, s store.Store, eng *monitor.Engine) *http.Server {
 			if err := s.SaveCheckFromNode(result.SiteID, req.NodeID, result.LatencyNs, result.IsUp); err != nil {
 				log.Printf("Failed to save probe result: %v", err)
 			}
-			eng.IngestProbeResult(req.NodeID, result.SiteID, result.LatencyNs, result.IsUp)
+			eng.IngestProbeResult(req.NodeID, result.SiteID, result.LatencyNs, result.IsUp, result.ErrorReason)
 		}
 		if err := s.UpdateNodeLastSeen(req.NodeID); err != nil {
 			log.Printf("Failed to update node last seen: %v", err)

internal/server/server_test.go

@@ -4,13 +4,14 @@ import (
 	"bytes"
 	"encoding/json"
 	"fmt"
-	"gitea.lerkolabs.com/lerko/uptop/internal/models"
-	"gitea.lerkolabs.com/lerko/uptop/internal/monitor"
 	"net"
 	"net/http"
 	"sync"
 	"testing"
 	"time"
+
+	"gitea.lerkolabs.com/lerko/uptop/internal/models"
+	"gitea.lerkolabs.com/lerko/uptop/internal/monitor"
 )
 
 // --- Mock Store ---
@@ -75,6 +76,8 @@ func (m *mockStore) DeleteMaintenanceWindow(int) error                   { retur
 func (m *mockStore) IsMonitorInMaintenance(int) (bool, error)               { return false, nil }
 func (m *mockStore) GetPreference(string) (string, error)                   { return "", nil }
 func (m *mockStore) SetPreference(string, string) error                     { return nil }
+func (m *mockStore) SaveStateChange(int, string, string, string) error      { return nil }
+func (m *mockStore) GetStateChanges(int, int) ([]models.StateChange, error) { return nil, nil }
 func (m *mockStore) Close() error                                           { return nil }
 
 func (m *mockStore) ExportData() (models.Backup, error) {

internal/store/postgres.go

@@ -72,6 +72,15 @@ func (d *PostgresDialect) CreateTablesSQL() []string {
 			key TEXT PRIMARY KEY,
 			value TEXT NOT NULL
 		)`,
+		`CREATE TABLE IF NOT EXISTS state_changes (
+			id SERIAL PRIMARY KEY,
+			site_id INTEGER NOT NULL,
+			from_status TEXT NOT NULL,
+			to_status TEXT NOT NULL,
+			error_reason TEXT DEFAULT '',
+			changed_at TIMESTAMP DEFAULT NOW()
+		)`,
+		`CREATE INDEX IF NOT EXISTS idx_state_changes_site ON state_changes(site_id, changed_at DESC)`,
 	}
 }
 

internal/store/sqlite.go

@@ -79,6 +79,15 @@ func (d *SQLiteDialect) CreateTablesSQL() []string {
 			key TEXT PRIMARY KEY,
 			value TEXT NOT NULL
 		)`,
+		`CREATE TABLE IF NOT EXISTS state_changes (
+			id INTEGER PRIMARY KEY AUTOINCREMENT,
+			site_id INTEGER NOT NULL,
+			from_status TEXT NOT NULL,
+			to_status TEXT NOT NULL,
+			error_reason TEXT DEFAULT '',
+			changed_at DATETIME DEFAULT CURRENT_TIMESTAMP
+		)`,
+		`CREATE INDEX IF NOT EXISTS idx_state_changes_site ON state_changes(site_id, changed_at DESC)`,
 	}
 }
 

internal/store/sqlstore.go

@@ -347,6 +347,29 @@ func (s *SQLStore) DeleteUser(id int) error {
 	return err
 }
 
+func (s *SQLStore) SaveStateChange(siteID int, fromStatus, toStatus, errorReason string) error {
+	_, err := s.db.Exec(s.q("INSERT INTO state_changes (site_id, from_status, to_status, error_reason) VALUES (?, ?, ?, ?)"),
+		siteID, fromStatus, toStatus, errorReason)
+	return err
+}
+
+func (s *SQLStore) GetStateChanges(siteID int, limit int) ([]models.StateChange, error) {
+	rows, err := s.db.Query(s.q("SELECT id, site_id, from_status, to_status, error_reason, changed_at FROM state_changes WHERE site_id = ? ORDER BY changed_at DESC LIMIT ?"), siteID, limit)
+	if err != nil {
+		return nil, err
+	}
+	defer rows.Close()
+	var changes []models.StateChange
+	for rows.Next() {
+		var sc models.StateChange
+		if err := rows.Scan(&sc.ID, &sc.SiteID, &sc.FromStatus, &sc.ToStatus, &sc.ErrorReason, &sc.ChangedAt); err != nil {
+			return changes, err
+		}
+		changes = append(changes, sc)
+	}
+	return changes, rows.Err()
+}
+
 func (s *SQLStore) SaveCheck(siteID int, latencyNs int64, isUp bool) error {
 	return s.SaveCheckFromNode(siteID, "", latencyNs, isUp)
 }

internal/store/store.go

@@ -38,6 +38,10 @@ type Store interface {
 	SaveCheckFromNode(siteID int, nodeID string, latencyNs int64, isUp bool) error
 	LoadAllHistory(limit int) (map[int][]models.CheckRecord, error)
 
+	// State Changes
+	SaveStateChange(siteID int, fromStatus, toStatus, errorReason string) error
+	GetStateChanges(siteID int, limit int) ([]models.StateChange, error)
+
 	// Nodes
 	RegisterNode(node models.ProbeNode) error
 	GetNode(id string) (models.ProbeNode, error)

internal/tui/tab_sites.go

@@ -309,6 +309,29 @@ func fmtStatus(status string, paused bool, inMaint bool) string {
 	}
 }
 
+func fmtDuration(d time.Duration) string {
+	if d < time.Minute {
+		return fmt.Sprintf("%ds", int(d.Seconds()))
+	}
+	if d < time.Hour {
+		return fmt.Sprintf("%dm", int(d.Minutes()))
+	}
+	if d < 24*time.Hour {
+		h := int(d.Hours())
+		m := int(d.Minutes()) % 60
+		if m > 0 {
+			return fmt.Sprintf("%dh %dm", h, m)
+		}
+		return fmt.Sprintf("%dh", h)
+	}
+	days := int(d.Hours()) / 24
+	hours := int(d.Hours()) % 24
+	if hours > 0 {
+		return fmt.Sprintf("%dd %dh", days, hours)
+	}
+	return fmt.Sprintf("%dd", days)
+}
+
 func (m Model) dynamicWidths() (nameW, sparkW int) {
 	fixed := 6 + 10 + 10 + 8 + 8 + 7 + 9 // #, TYPE, STATUS, LATENCY, UPTIME, SSL, RETRY
 	overhead := 30                       // cell padding + borders
@@ -389,6 +412,14 @@ func (m Model) viewSitesTab() string {
 					name = limitStr(name, nameW)
 				}
 
+				if (site.Status == "DOWN" || site.Status == "SSL EXP") && site.LastError != "" {
+					nameLen := len([]rune(name))
+					errSpace := nameW - nameLen - 1
+					if errSpace > 10 {
+						name = name + " " + subtleStyle.Render(limitStr(site.LastError, errSpace))
+					}
+				}
+
 				hist, _ := m.engine.GetHistory(site.ID)
 				var spark string
 				if site.Type == "push" {
@@ -732,6 +763,25 @@ func (m Model) viewDetailPanel() string {
 	}
 
 	row("Status", fmtStatus(site.Status, site.Paused, m.isMonitorInMaintenance(site.ID)))
+
+	if (site.Status == "DOWN" || site.Status == "SSL EXP") && site.LastError != "" {
+		row("Error", dangerStyle.Render(limitStr(site.LastError, 60)))
+	}
+
+	if site.Type == "http" && site.StatusCode > 0 {
+		row("HTTP Code", strconv.Itoa(site.StatusCode))
+	}
+
+	if !site.StatusChangedAt.IsZero() {
+		dur := time.Since(site.StatusChangedAt)
+		row("State Since", site.StatusChangedAt.Format("2006-01-02 15:04:05")+" ("+fmtDuration(dur)+")")
+	}
+
+	if !site.LastSuccessAt.IsZero() {
+		ago := time.Since(site.LastSuccessAt)
+		row("Last Success", site.LastSuccessAt.Format("15:04:05")+" ("+fmtDuration(ago)+" ago)")
+	}
+
 	if m.isMonitorInMaintenance(site.ID) {
 		for _, mw := range m.maintenanceWindows {
 			if mw.Type == "maintenance" && (mw.MonitorID == 0 || mw.MonitorID == site.ID || mw.MonitorID == site.ParentID) {
@@ -787,7 +837,30 @@ func (m Model) viewDetailPanel() string {
 			}
 			latency := time.Duration(result.LatencyNs).Milliseconds()
 			ago := time.Since(result.CheckedAt).Truncate(time.Second)
-			fmt.Fprintf(&b, "  %-14s %s  %dms  %s ago\n", nodeID, status, latency, ago)
+			line := fmt.Sprintf("  %-14s %s  %dms  %s ago", nodeID, status, latency, ago)
+			if !result.IsUp && result.ErrorReason != "" {
+				line += "  " + dangerStyle.Render(limitStr(result.ErrorReason, 30))
+			}
+			b.WriteString(line + "\n")
+		}
+	}
+
+	stateChanges := m.engine.GetStateChanges(site.ID, 5)
+	if len(stateChanges) > 0 {
+		b.WriteString("\n" + subtleStyle.Render("  STATE CHANGES") + "\n")
+		for _, sc := range stateChanges {
+			ago := fmtDuration(time.Since(sc.ChangedAt))
+			arrow := subtleStyle.Render(sc.FromStatus) + " → "
+			if sc.ToStatus == "UP" {
+				arrow += specialStyle.Render(sc.ToStatus)
+			} else {
+				arrow += dangerStyle.Render(sc.ToStatus)
+			}
+			line := fmt.Sprintf("  %s  %s", arrow, subtleStyle.Render(ago+" ago"))
+			if sc.ErrorReason != "" && sc.ToStatus != "UP" {
+				line += "  " + dangerStyle.Render(limitStr(sc.ErrorReason, 40))
+			}
+			b.WriteString(line + "\n")
 		}
 	}