Monitoring Overview
Understand the monitoring features available in Devpilot, including server metrics, quota alerts, activity logs, deployment failure patterns, and webhook delivery logs.
Devpilot gives you visibility into the infrastructure and activity inside every workspace you belong to. From real-time server metrics collected by the Devpilot agent to the activity trail for every project, app, deployment, and server, the monitoring area is where you go to see what is happening and what has already happened.
What you can monitor
Monitoring in Devpilot is organised around five areas.
Server metrics
Track CPU, memory, load average, disk usage, network throughput, and uptime for each server.
Quota alerts
Get notified when your workspace approaches or exceeds its plan limits for Projects and Apps.
Logs
Review webhook delivery logs, deployment hook output, and the login history for your account.
Deployment failure patterns
See recurring deployment failures grouped by category, with occurrence counts and affected deployments.
Activity logs
Review every create, update, delete, deploy, login, and access change across a workspace.
Accessing monitoring
Monitoring data is scoped to each workspace. Open a workspace, then use the workspace sidebar to reach:
- Activity logs for the audit trail of that workspace.
- Servers to open any server and see its live metrics and uptime.
- Settings > Usage to review quota usage and quota alerts.
- Apps to open an app and see its deployment failure patterns.
Server metrics
Each server that is provisioned through Devpilot runs the Devpilot agent, which reports metrics back to the platform on a schedule. Open a server to see its monitoring tab.
What the agent collects
| Metric family | Fields |
|---|---|
| CPU | Overall usage percentage, number of cores, per-core usage, 1-minute / 5-minute / 15-minute load averages. |
| Memory | Usage percentage, total, used, free, and available bytes. |
| Disk | For each mounted filesystem: total space, used space, available space, and usage percentage. |
| Network | For each interface: bytes sent/received, packets sent/received, error and drop counters, current upload and download rates. |
| Uptime | Whether the server is online, its uptime in seconds, agent response time, and a status message. |
Each metric carries a recorded-at timestamp so you can line events up across servers.
Time ranges
The analytics view supports four presets: 1h, 6h, 24h, and 7d. Switching the range refreshes CPU, memory, disk, network, and uptime panels together.
Enabling or disabling monitoring
Monitoring is enabled per server via the Monitoring enabled toggle. You can switch it on or off from the server page — when disabled, the agent stops reporting analytics for that server.
If a server has not reported metrics recently, its uptime record shows Offline with a status message describing the last known state.
Activity and access visibility
Beyond host metrics, Devpilot keeps three complementary streams you can review:
- Activity logs capture user and system events against Workspaces, Projects, Apps, Servers, Deployments, Backups, Invitations, and Users (create, update, delete, deploy, backup, login, logout, failed login, 2FA setup, password reset, invitation cancellation, and deployment started/completed).
- Access audit logs record every access granted, revoked, or modified on workspace resources, including old and new role, performer, IP, and user agent.
- Login history records each login attempt with IP, location, device, browser, OS, and whether it was a new device or a failed attempt.
Permissions
Monitoring visibility follows the standard workspace permission model. To see server metrics, activity logs, and quota dashboards you need the workspace.view permission on that workspace. To see access audit entries you need user.view. Roles that do not have these permissions see an access-denied response.
Next steps
- Configure quota alerts so you are warned before hitting plan limits.
- Review activity logs to see exactly who changed what.
- Check logs for webhook deliveries, deployment hooks, and your login history.
- Investigate deployment failure patterns when a build or release keeps breaking.