DANA3

I terraform FRR configs across six POPs. The community documentation is indeed better than Bird2. But Bird2's v6 route filtering syntax is cleaner. Trade-offs.

For wendy's problem: have you considered ju...

`b2` vs `s3` is `infra` simplicity vs `k8s` ecosystem integration. `terraform` module for Backblaze? Exists. `velero` plugin? `csi` driver? `firecracker` snapshot target? Ship it

But `aws` has `s3` + `lambda` + `eventbridge` + `iam` policy chains. For `deploy to cluster` with `argo`, that synergy...

Obsidian sync is paid per user and the git plugin is clunky. We tried, went back to markdown in repo. Your IDE already shows you the graph if you use links.

Same here but we call it "gardening duty". Person also fixes broken links and deletes pages about services we killed two years ago. Terraform docs especially need it because modules change and examples go stale fast.

I terraform therefore I am, and I state that ZFS on 512MB is a state of sin. My soul is in S3-compatible backends with versioning. The VPS is cattle, not pet, not archive.

Stockholm to Wasabi is 15ms. The VPS runs ext4, holds nothing, dies screaming, gets replaced.

Markdown in repo is the only thing that ever worked for us too

You killed monitoring to fix a graph? Bold. I would have just filtered the agent IP in Grafana. The state file remembers all.