We run the hosting for your applications ourselves, on Linux servers in the EU. You choose the availability you need. We test the backups with real restores.
Three levels of availability
Not every application needs a cluster. Together we choose the level that fits the risk of an outage and the budget.
One database server with backups in S3
The database runs on one server, and we back it up regularly to S3 compatible object storage. If the server fails, we restore from the latest backup. Changes since that backup can be lost.
Replica on a second server
A second database server, for example with PostgreSQL, continuously applies the changes of the first. If the first fails, we switch over to the replica. With asynchronous replication, the last transactions before the failure can be missing.
Replica set with automatic failover
Several MongoDB nodes hold the same data. When the primary goes down, the others elect a new one themselves, and the application keeps writing without us having to step in.
Restore tests
We restore from the backup regularly and check that the application starts with that data. We agree the interval with you, and we write the result into the report.
Hosting in the EU
Servers and backups sit in data centres in the EU. We tell you the provider and the location, so you can name both in your data protection records.
Operations included
Hosting includes operations: patching in the maintenance window, monitoring, nginx and TLS, DNS and the host firewall. The details are under Linux server operations, the response times under Service levels.
If your servers sit with another hosting provider or in your own environment, we run them there too.
We do not run Windows Server or databases such as Oracle DB or MS SQL.