Today was a SHIT day!

Working as ops for my customer, we are maintaining several tools in different environments. Today was the day my fucking Kubernetes Cluster made me rage quit, AGAIN!

We have a MongoDB running on Kubernetes with daily backups, the main node crashed due a full PVC on the cluster.

Full PVC => Pod doesn't start
Pod doesn't start => You can't get the live data
No live data? => Need Backup
Backup is in S3 => No Credentials
Got Backup from coworker
Restore Backup? => No connection to new MongoDB

3 FUCKING HOURS WASTED FOR NOTHING

Got it working at the end... Now we need to make an incident in the incident management software. Tbh that's the worst part.

And the team responsible for the cluster said monitoring wont be supported because it's unnecessary....