Monitoring
The Monitoring page consolidates system observability into a single view with tabs for Audit Logs and System Health.
Audit Logs
The Audit Logs page shows a history of administrative actions across the system.
Audit Table
Each log entry shows:
| Column | Description |
|---|---|
| Status | Success or failure indicator |
| Event Type | Color-coded badge (e.g., Assistant, MCP Server, Page, Configuration) |
| Description | Summary of what happened |
| User | Who performed the action (email or ID) |
| Entity | The affected entity type and ID |
| Timestamp | When the event occurred |
| Details | Expandable section with before/after values |
Filtering
- Event Type — Filter by event category (Assistant, MCP Server, Page, Configuration)
Expandable Details
Click a row to expand its details showing:
- Before Value — JSON representation of the state before the change
- After Value — JSON representation of the state after the change
- Metadata — Additional context about the event
Pagination
Results are shown 20 entries per page with navigation controls.
MCP Errors
MCP error logs are available per instance in the Errors tab of each MCP server's configuration page. This makes it easy to diagnose issues for a specific integration without wading through global logs.
To view errors for an MCP server:
- Navigate to MCP Servers in the sidebar
- Click the server you want to investigate
- Go to the Errors tab
Each instance shows statistics, filterable error logs, and detailed error context. See MCP Configuration > Errors Tab for full documentation.
Errors are automatically classified into categories (Network, Auth, Client Error, Server Error, Circuit Open) and retained for 30 days.
System Health
The System Health page shows real-time status of all system components.
System Status
An overall status badge is displayed at the top:
- Healthy (green) — All systems operational
- Degraded (yellow) — Some components experiencing issues
- Down (red) — Critical components unavailable
- Unknown (gray) — Status cannot be determined
System Info
| Field | Description |
|---|---|
| Environment | Deployment environment (development, staging, production) |
| Version | Application version |
| Last Check | When health was last checked |
| Auto-refresh | Status refreshes automatically every 30 seconds |
Metrics Cards
| Metric | Description |
|---|---|
| Uptime | System uptime formatted as days/hours/minutes |
| Active Sessions | Number of currently active chat sessions |
| Memory Usage | Memory consumption as percentage and bytes |
| Requests/Hour | Current request throughput |
Core Services
A table showing the health of core infrastructure:
| Column | Description |
|---|---|
| Service | Database, Storage, Memory |
| Status | Health badge |
| Response Time | Latency in milliseconds |
| Last Check | When last checked |
MCP Servers
A table showing connected MCP server health:
| Column | Description |
|---|---|
| Name | Server instance name |
| URL | Server endpoint |
| Status | Health badge |
| Latency | Response time |
| Tools | Number of discovered tools |
| Resources | Number of available resources |
| Error | Error message if unhealthy |
Performance Metrics
| Metric | Description |
|---|---|
| Avg Response Time | Average API response latency |
| Error Rate % | Percentage of failed requests |
| Total Requests | Total requests served |
Next Steps
- Usage & Billing — Monitor costs and token usage
- Global Settings — System-wide configuration
- MCP Servers — Manage MCP server instances