A quick status update to acknowledge that Nord.pub had ~1 hour of downtime yesterday.

This was caused due to an update and configuration change of the underlying infrastructure. Applying the configuration change accidentally unmade another unrelated setting, which caused the ingress controller containers (which direct public HTTPS to the correct backend service) to be scheduled on nodes they were not supposed to run on. This in turn caused them to be unreachable from the internet, causing the outage.

Additionally there was an issue with the redis-ha deployment, causing some exception error messages once the site became accessible again, although that was resolved within a few minutes.