ImpureCreek

This but unironically. Every "cloud native" tool defaults to "we assume you have 16GB and a team of three." You spend first two days just turning things off.

Loki is actually better than most here. Try running Elasticsearch with default...

You should just use a lock file or something

I do this with rclone + sqlite for tracking what already synced

That way you dont hammer apis when dont need to

My thinking: check before request, not after

Same with Hostinger here

Ngl 6 days for bgp analysis is actually fast in carrier land I waited 3 weeks for a vRAM allocation bug on a Vultr gpu instance once they just kept swapping the card til I gave up at least these guys did the work

OVHcloud has a datacenter in Gravelines, not Paris. Their Paris presence is offices only. If you want French anycast you need to look at Online.net (also Scaleway) or build your own in Telehouse Paris. I have done neither because I have a life.

Paris here. I run systemd on Scaleway and I run systemd at work. The only place I see OpenRC is in Docker containers that need an init, and even there people are moving to tini or just using systemd in the container. The init system inside a container debate is its own thread though.

So who's actually moving? Hetzner's €4.51 tier is same ballpark but no object storage. Was looking at Wasabi for that but the egress math hurts.

OVHcloud VPS "instant" is genuinely instant in my experience, but their control panel is from 2003 and makes you want to die. Tradeoffs.

Hostinger's problem is they market to beginners who don't kn...

Endpoint-independent mapping on a budget NAT plan? I'd want to see packet captures