Eerdere incidenten

We were working on the b-server terraform config again. As always, only one server down is not an emergency - we remained triple redundant.

Terraform wanted to perform one last reboot and we let it. Server should be stably back online now.

We were experimenting with using Terraform for deployment. It caused b to be offline for an hour. As always: no data was lost, because we need an entire column to be red before that happens.

A deployment took a little too long and looked like a brief outage. Because we never deploy to more than one machine at a time, no service outage was caused by this brief machine outage.