Do I need to manage infrastructure and recovery when using SigNoz?
With SigNoz Cloud, we run the infrastructure that stores and queries your logs. With Bring Your Own Cloud (BYOC), we run SigNoz in your cloud account. If you self-host SigNoz, your team runs the infrastructure and plans for backup and recovery.
What does a self-hosted team need to plan?
Start with enough capacity for your log volume, retention and queries. You'll need to size the OpenTelemetry Collector that receives your logs and ClickHouse, the database that stores them. Our resource planning guide for Self-Hosted Community is a starting point; the right size depends on your workload.
The production readiness checklist covers other work, such as security and scaling. For upgrades, check the upgrade path for required version stops and follow the backup and migration steps in each guide.
What should a self-hosted backup cover?
Your logs are stored in ClickHouse. Saved views, dashboards and alert definitions are stored separately in the Metastore, a SQLite or PostgreSQL database. A saved view keeps the filters and columns you use to investigate logs.
For example, a checkout team might have a saved view for failed payments and an alert for errors. If it loses both databases, restoring the logs alone won't recover the saved view and alert. The team needs the Metastore backup for those too.
Plan backups for both stores, keep your deployment configuration, and test that you can restore what you need. The ClickHouse upgrade guide calls for a data snapshot, and the saved-view upgrade guide calls for a Metastore backup. These guides do not define a complete restore procedure or recovery time for every deployment.
Retention determines how long logs stay available to query in SigNoz. If you need to keep telemetry longer, you can also send it to S3. The exports to SigNoz and S3 are independent, and you cannot query S3 data in the SigNoz UI. The S3 export does not include the Metastore or deployment configuration, so it does not replace a deployment backup.
