Log exporting (add-on)
If your studio runs its own observability platform (Elasticsearch, Splunk, Datadog, Grafana, or similar), log exporting lets you ingest Nakama logs into it. Logs are shipped to a dedicated S3 bucket in near real-time, in JSON format. Each entry is enriched with the instance ID and accurate timestamps for when the log was emitted and when the exporter consumed it.
Unlike the built-in dashboard logs, which are truncated and rate-limited to manage volume, the export add-on delivers raw, unsampled log data: every log line your Nakama server produces.
The S3 bucket retains exported logs for a rolling 7-day window. Heroic Labs automatically removes logs older than 7 days.
How it works #
Heroic Labs provisions a dedicated S3 bucket and provides you with credentials (a URL and access keys) to pull from it. You don’t manage the bucket or its configuration: you receive the connection details and read from it using any S3-compatible client.
Each deployment has one log export. You can’t configure multiple export destinations for the same deployment. The export is continuous for the lifetime of the add-on.
Delivery cadence #
The exporter batches log lines rather than writing one object per line. It uploads a file when a batch reaches either a size or an age threshold, whichever comes first, then starts a new batch. A log line typically lands in the bucket 20 to 30 seconds after Nakama emits it, grouped with the lines around it.
The result is a continuous feed of small files rather than a live tail. Use the export for centralized logging, alerting, and analysis in your own platform. To watch a deployment during a release, use the dashboard Logs tab instead.
Getting logs into your platform #
The S3 bucket is the only delivery target. Heroic Cloud doesn’t push logs into third-party APIs, so your log management platform reads from the bucket rather than from Heroic Cloud directly.
Many log management platforms read from S3 through a built-in S3 source. Others need a forwarding agent you run yourself, such as Fluent Bit or Vector. Because entries are already JSON and carry the instance ID and both timestamps, they land as structured fields without custom parsing rules.
When to use it #
Log exporting is useful when you need to ingest logs into your own observability platform (for example, Elasticsearch, Splunk, Datadog, Grafana) for centralized logging, compliance auditing, long-term retention, or custom analysis beyond the 7-day window.
Enabling log exporting #
Contact Heroic Labs (support@heroiclabs.com) to enable this add-on.
Log exporting is a separate, paid add-on to your Heroic Cloud subscription.
