Log routing and filtering for managed database fleets

Teams managing multiple managed database instances — across staging, production, and regional deployments — hit a specific problem when they try to centralize logs: getting logs out of Aiven, RDS, or Aurora into whatever log aggregator they already use requires per-service configuration, produces inconsistent field names and formats, and any misconfiguration silently drops logs with no indication anything is wrong. One user described it plainly: 'any export or integrations of logs outside of Aiven are quite hard to use, limited, and introduce difficulties troubleshooting.' The troubleshooting difficulty compounds: when you can't trust that your log export is working correctly, you can't trust the absence of log entries either.

The reason this stays broken is that each managed database vendor builds log export as an afterthought targeting their own ecosystem. Aiven's integrations work best with Aiven Observability. RDS log export targets CloudWatch. Neither is designed to be a clean, reliable source for a team that uses Datadog, Loki, or Papertrail. The buyer here is typically a platform or infrastructure engineer who owns observability tooling, not the developers who write the database queries — and that person rarely shows up in product feedback surveys for a database vendor, so the gap compounds quietly.

What's actually needed is a lightweight agent or connector that sits between one or more managed database instances and any log destination, handles format normalization, gives a live status on whether log export is actually flowing, and exposes per-instance filtering rules so teams aren't paying to ship 'connection closed' noise at full volume to an expensive SaaS log store. Right now, teams either ship everything and pay too much, or they hand-configure filters per service and spend hours when something changes.

This is a business because the need recurs every time a new database instance is provisioned, every time the team changes log aggregators, and every time log volume grows enough that the storage bill forces a conversation about filtering. It's not a one-time setup — it's ongoing management.

What to build

Build a managed log routing agent that connects to RDS, Aurora, and Aiven PostgreSQL instances, normalizes log fields to a consistent schema, applies user-defined per-instance filters, and reliably forwards to any major log destination (Datadog, Loki, S3, Papertrail) with a live delivery status dashboard showing whether each instance is actually flowing.

Where to start

Start with teams already using Datadog for application logs who are frustrated that their database logs arrive in a different format and can't be correlated with app traces — the correlation pain is concrete and immediately visible in their existing tool.

The hard part

You're competing against 'just configure it yourself in CloudWatch Logs subscriptions or Aiven's built-in integrations,' which is painful but free — your onboarding experience has to make the contrast obvious within 20 minutes or the prospect goes back to the manual approach.

How it makes money

Monthly subscription based on number of database instances connected, around $19/instance/month, with volume pricing above 20 instances to match infrastructure team budgets.