Backup & Recovery (MonPG’s own)

How MonPG itself backs up your data + RTO / RPO commitments.

What we back up

The PostgreSQL operational store gets full daily snapshots plus continuous WAL archiving to Azure Blob. Restore window is 35 days; RPO under 5 minutes for that store. ClickHouse gets daily logical backups of the analytical schema plus WAL of system tables; per-tenant data is partition-aligned so a tenant-specific restore can be bounded in scope. Azure Key Vault has soft-delete + purge protection enabled for 90 days, so an accidental key deletion is recoverable.

SLAs

Metric Pro Business Enterprise
Uptime SLA 99.9% 99.95% 99.99% with credit ladder
RTO 4 hours 1 hour 15 min
RPO 1 hour 15 min 5 min
Status page status.monpg.app status.monpg.app + private incident calls

The Enterprise credit ladder is in the contract — short version, every percentage point of downtime past 0.01% gets you a service credit on the next invoice.

DR drills

Quarterly. Full restore from snapshots into a clean isolated environment, then a validation pass over the critical paths: auth still works, ingest accepts a snapshot, alerts fire correctly, billing webhooks process. Results posted to the Trust Center.

Data export

Settings → Data → Request export. We assemble a tar.gz with everything we hold for your org and deliver it within 24 hours. Useful for migration off, archival, or off-platform analysis. The format is documented in the export README inside the archive — JSON for operational data, CSV for time-series.

Account deletion

Delete account triggers a 7-day cool-off (in case you change your mind), then permanent purge from primary, replicas, and backups within 30 days. We retain an audit log of the deletion itself for 7 years for compliance — but it’s just metadata about that the deletion happened, not your data.