fix: reduce syslog flooding by downgrading routine logs to debug level

Addresses issue #861 - syslog flooded on docker host

Many routine operational messages were being logged at INFO level,
causing excessive log volume when monitoring multiple VMs/containers.
These messages are now logged at DEBUG level:

- Guest threshold checking (every guest, every poll cycle)
- Storage threshold checking (every storage, every poll cycle)
- Host agent linking messages
- Filesystem inclusion in disk calculation
- Guest agent disk usage replacement
- Polling start/completion messages
- Alert cleanup and save messages

Users can set LOG_LEVEL=debug to see these messages if needed for
troubleshooting. The default INFO level now produces significantly
less log output.

Also updated documentation in CONFIGURATION.md and DOCKER.md to:
- Clarify what each log level includes
- Add tip about using LOG_LEVEL=warn for minimal logging

docs/CONFIGURATION.md

@@ -87,8 +87,19 @@ Environment variables take precedence over `system.json`.
 | Variable | Description | Default |
 |----------|-------------|---------|
 | `FRONTEND_PORT` | Public listening port | `7655` |
-| `LOG_LEVEL` | Log verbosity (`debug`, `info`, `warn`, `error`) | `info` |
+| `LOG_LEVEL` | Log verbosity (see below) | `info` |
 | `LOG_FORMAT` | Log output format (`auto`, `json`, `console`) | `auto` |
+
+#### Log Levels
+
+| Level | Description |
+|-------|-------------|
+| `error` | Only errors and critical issues |
+| `warn` | Errors + warnings (recommended for minimal logging) |
+| `info` | Standard operational messages (startup, connections, alerts) |
+| `debug` | Verbose output including per-guest/storage polling details |
+
+> **Tip**: If your syslog is being flooded with Pulse messages, set `LOG_LEVEL=warn` to significantly reduce log volume while still capturing important events.
 | `PULSE_PUBLIC_URL` | Public URL for notifications/OIDC | `""` |
 | `ALLOWED_ORIGINS` | CORS allowed domains | `""` (Same origin) |
 | `DISCOVERY_ENABLED` | Auto-discover nodes | `false` |

docs/DOCKER.md

@@ -57,7 +57,9 @@ Pulse is configured via environment variables.
 | `API_TOKENS` | Comma-separated API tokens | *(unset)* |
 | `DISCOVERY_SUBNET` | Custom CIDR to scan | *(auto)* |
 | `ALLOWED_ORIGINS` | CORS allowed domains | *(none)* |
-| `LOG_LEVEL` | Log verbosity | `info` |
+| `LOG_LEVEL` | Log verbosity (`debug`, `info`, `warn`, `error`) | `info` |
+
+> **Tip**: Set `LOG_LEVEL=warn` to reduce log volume while still capturing important events.
 
 <details>
 <summary><strong>Advanced: Resource Limits & Healthcheck</strong></summary>

internal/alerts/alerts.go

@@ -2057,7 +2057,7 @@ func (m *Manager) CheckGuest(guest interface{}, instanceName string) {
 
 		// Debug logging for high memory VMs
 		if memUsage > 85 {
-			log.Info().
+			log.Debug().
 				Str("vm", name).
 				Float64("memUsage", memUsage).
 				Str("status", status).
@@ -2100,7 +2100,7 @@ func (m *Manager) CheckGuest(guest interface{}, instanceName string) {
 
 	monitorOnly := settings.MonitorOnly
 	if monitorOnly || m.guestHasMonitorOnlyAlerts(guestID) {
-		log.Info().
+		log.Debug().
 			Str("guest", name).
 			Bool("monitorOnly", monitorOnly).
 			Msg("Pulse monitor-only status applied")
@@ -2185,7 +2185,7 @@ func (m *Manager) CheckGuest(guest interface{}, instanceName string) {
 	}
 
 	// Check each metric
-	log.Info().
+	log.Debug().
 		Str("guest", name).
 		Float64("cpu", cpu).
 		Float64("memory", memUsage).
@@ -4341,7 +4341,7 @@ func (m *Manager) CheckStorage(storage models.Storage) {
 	// Check usage if storage has valid data (even if not currently active on this node)
 	// In clusters, storage may show as inactive on nodes where it's not currently mounted
 	// but we still want to alert on high usage
-	log.Info().
+	log.Debug().
 		Str("storage", storage.Name).
 		Str("id", storage.ID).
 		Float64("usage", storage.Usage).
@@ -8140,7 +8140,7 @@ func (m *Manager) SaveActiveAlerts() error {
 		return fmt.Errorf("failed to rename active alerts file: %w", err)
 	}
 
-	log.Info().Int("count", len(alerts)).Msg("Saved active alerts to disk")
+	log.Debug().Int("count", len(alerts)).Msg("Saved active alerts to disk")
 	return nil
 }
 
@@ -8316,7 +8316,7 @@ func (m *Manager) CleanupAlertsForNodes(existingNodes map[string]bool) {
 	m.mu.Lock()
 	defer m.mu.Unlock()
 
-	log.Info().
+	log.Debug().
 		Int("totalAlerts", len(m.activeAlerts)).
 		Int("existingNodes", len(existingNodes)).
 		Interface("nodes", existingNodes).
@@ -8349,7 +8349,7 @@ func (m *Manager) CleanupAlertsForNodes(existingNodes map[string]bool) {
 	}
 
 	if removedCount > 0 {
-		log.Info().Int("removed", removedCount).Int("remaining", len(m.activeAlerts)).Msg("Cleaned up alerts for non-existent nodes")
+		log.Debug().Int("removed", removedCount).Int("remaining", len(m.activeAlerts)).Msg("Cleaned up alerts for non-existent nodes")
 		// Save the cleaned up state
 		go func() {
 			defer func() {

internal/monitoring/monitor.go

@@ -2219,7 +2219,7 @@ func (m *Monitor) ApplyHostReport(report agentshost.Report, tokenRecord *config.
 	linkedNodeID, linkedVMID, linkedContainerID := m.findLinkedProxmoxEntity(hostname)
 	if linkedNodeID != "" {
 		host.LinkedNodeID = linkedNodeID
-		log.Info().
+		log.Debug().
 			Str("hostId", identifier).
 			Str("hostname", hostname).
 			Str("linkedNodeId", linkedNodeID).
@@ -2227,7 +2227,7 @@ func (m *Monitor) ApplyHostReport(report agentshost.Report, tokenRecord *config.
 	}
 	if linkedVMID != "" {
 		host.LinkedVMID = linkedVMID
-		log.Info().
+		log.Debug().
 			Str("hostId", identifier).
 			Str("hostname", hostname).
 			Str("linkedVmId", linkedVMID).
@@ -2235,7 +2235,7 @@ func (m *Monitor) ApplyHostReport(report agentshost.Report, tokenRecord *config.
 	}
 	if linkedContainerID != "" {
 		host.LinkedContainerID = linkedContainerID
-		log.Info().
+		log.Debug().
 			Str("hostId", identifier).
 			Str("hostname", hostname).
 			Str("linkedContainerId", linkedContainerID).
@@ -5837,7 +5837,7 @@ func (m *Monitor) pollPVEInstance(ctx context.Context, instanceName string, clie
 // When the instance is part of a cluster, the cluster name is used for guest IDs to prevent duplicates
 // when multiple cluster nodes are configured as separate PVE instances.
 func (m *Monitor) pollVMsAndContainersEfficient(ctx context.Context, instanceName string, clusterName string, isCluster bool, client PVEClientInterface, nodeEffectiveStatus map[string]string) bool {
-	log.Info().
+	log.Debug().
 		Str("instance", instanceName).
 		Str("clusterName", clusterName).
 		Bool("isCluster", isCluster).
@@ -6189,7 +6189,7 @@ func (m *Monitor) pollVMsAndContainersEfficient(ctx context.Context, instanceNam
 							}
 
 							if len(includedFS) > 0 {
-								log.Info().
+								log.Debug().
 									Str("instance", instanceName).
 									Str("vm", res.Name).
 									Int("vmid", res.VMID).
@@ -6223,7 +6223,7 @@ func (m *Monitor) pollVMsAndContainersEfficient(ctx context.Context, instanceNam
 								diskFree = totalBytes - usedBytes
 								diskUsage = safePercentage(float64(usedBytes), float64(totalBytes))
 
-								log.Info().
+								log.Debug().
 									Str("instance", instanceName).
 									Str("vm", res.Name).
 									Int("vmid", res.VMID).
@@ -6596,7 +6596,7 @@ func (m *Monitor) pollVMsAndContainersEfficient(ctx context.Context, instanceNam
 		nodesWithResources[res.Node] = true
 	}
 
-	log.Info().
+	log.Debug().
 		Str("instance", instanceName).
 		Int("nodesInResources", len(nodesWithResources)).
 		Int("totalVMsFromResources", len(allVMs)).
@@ -6719,7 +6719,7 @@ func (m *Monitor) pollVMsAndContainersEfficient(ctx context.Context, instanceNam
 
 	m.pollReplicationStatus(ctx, instanceName, client, allVMs)
 
-	log.Info().
+	log.Debug().
 		Str("instance", instanceName).
 		Int("vms", len(allVMs)).
 		Int("containers", len(allContainers)).

internal/monitoring/monitor_polling.go

@@ -208,7 +208,7 @@ func (m *Monitor) pollVMsWithNodes(ctx context.Context, instanceName string, clu
 		}
 	}
 
-	log.Info().
+	log.Debug().
 		Str("instance", instanceName).
 		Int("totalNodes", len(nodes)).
 		Int("onlineNodes", onlineNodes).
@@ -864,7 +864,7 @@ func (m *Monitor) pollVMsWithNodes(ctx context.Context, instanceName string, clu
 	}
 
 	duration := time.Since(startTime)
-	log.Info().
+	log.Debug().
 		Str("instance", instanceName).
 		Int("totalVMs", len(allVMs)).
 		Int("successfulNodes", successfulNodes).
@@ -913,7 +913,7 @@ func (m *Monitor) pollContainersWithNodes(ctx context.Context, instanceName stri
 		}
 	}
 
-	log.Info().
+	log.Debug().
 		Str("instance", instanceName).
 		Int("totalNodes", len(nodes)).
 		Int("onlineNodes", onlineNodes).
@@ -1173,7 +1173,7 @@ func (m *Monitor) pollContainersWithNodes(ctx context.Context, instanceName stri
 	}
 
 	duration := time.Since(startTime)
-	log.Info().
+	log.Debug().
 		Str("instance", instanceName).
 		Int("totalContainers", len(allContainers)).
 		Int("successfulNodes", successfulNodes).
@@ -1236,7 +1236,7 @@ func (m *Monitor) pollStorageWithNodes(ctx context.Context, instanceName string,
 		}
 	}
 
-	log.Info().
+	log.Debug().
 		Str("instance", instanceName).
 		Int("totalNodes", len(nodes)).
 		Int("onlineNodes", onlineNodes).
@@ -1623,7 +1623,7 @@ func (m *Monitor) pollStorageWithNodes(ctx context.Context, instanceName string,
 			Int("failedNodes", failedNodes).
 			Msg("All nodes failed to retrieve storage - check Proxmox API permissions for Datastore.Audit on all storage")
 	} else {
-		log.Info().
+		log.Debug().
 			Str("instance", instanceName).
 			Int("totalStorage", len(allStorage)).
 			Int("successfulNodes", successfulNodes).