# Dashboard metrics and logs

**URL:** https://heroiclabs.com/docs/heroic-cloud/operations/dashboard-metrics-logs/
**Summary:** Each Nakama deployment includes built-in graphs and logs to monitor health, spot issues, and investigate incidents before they affect players.
**Keywords:** dashboard metrics and logs, heroic cloud, dashboard metrics logs
**Categories:** heroic-cloud, dashboard-metrics-logs, operations

---


# Dashboard metrics and logs

When something goes wrong in production, or you want to understand how your deployment is performing, the built-in graphs and logs are your first tool. Every Nakama deployment includes them at no additional cost. The **Graphs** tab gives you time-series performance data. The **Logs** tab gives you deployment output in UTC. Both support exporting. Logs also support filtering and search. For integration with your own observability stack, see the [log exporting](../log-exporting/) and [metric exporting](../metric-exporting/) add-ons.

## Graphs

![Nakama Deployment Graphs]({{< fingerprint_image "/images/pages/heroic-cloud/operations/dashboard-metrics-logs/nakama_graphs.png" >}})

### What to watch

* **Load Balancer Request Count (per status code)** shows HTTP request volume broken down by 2xx, 4xx, and 5xx. A sudden spike in 5xx errors means something is wrong on the server side. A spike in 4xx may indicate client-side issues or rate limiting.
* **Nakama CPU Utilization** shows CPU usage per node (for example, nakama-0, nakama-1). If any node consistently exceeds 70-80%, consider scaling up. Uneven utilization across nodes may indicate a load balancing issue.
* **Nakama Memory Utilization** shows memory usage per node. Watch for gradual increases that could indicate a memory leak in your game module.
* **Database CPU Utilization** shows CPU usage for your PostgreSQL database. High sustained usage suggests you need a larger database tier or query optimization.
* **Database Load - all queries (ms/s)** shows total query execution time per second. Useful for spotting workload trends over time.

### Top database queries

The **Top Database Queries** section, beneath the charts, lists the most resource-intensive SQL queries. Each query has an **Impact** indicator (colored bar). If your database CPU is high, start here to identify which queries need optimization.

![Top Database Queries]({{< fingerprint_image "/images/pages/heroic-cloud/operations/dashboard-metrics-logs/database_topbase_indicator.png" >}})

### Date range and export

Choose a date range (for example, **Last 1 hour**) and select the refresh icon to reload. The **Export metrics** button downloads data for the selected range, useful for sharing with your team or importing into external tools.

## Logs

![Nakama Deployment Logs]({{< fingerprint_image "/images/pages/heroic-cloud/operations/dashboard-metrics-logs/nakama_logs.png" >}})

All logs display in **UTC**, regardless of your local timezone or your deployment's region.

### Searching and filtering

Use the search bar to find logs by keyword. Filter by severity using the level dropdown (default: **All**). The date range picker narrows to a specific window. Pagination controls navigate through pages.

### Exporting logs

Select **Export logs** to download logs matching your current filters. This is a one-time download of the visible logs. For continuous shipping to your own infrastructure, see [Log exporting](../log-exporting/).

## Deployment-level audit log

Each deployment also has its own **Audit** tab, separate from the organization-level audit. This scoped log shows all user actions on that specific deployment: who deployed an image, who changed configurations, who triggered a reboot. See [Audit log](../../organizations/audit-log/) for details on categories and filtering.