Hacker News Reader

I do something similar with disk usage. Create a dummy file with XX GB and delete it when you are running out of space

I worked at a shop where they didn't bother with a partitioning scheme - everything was just one large volume. The amount of times I've had to do emergency cleanup on production servers...

IIRC ext4 does this already. Saved me a couple of times.

Are load tests not a standard thing people do any more?

When unexpected load causes a production outage, it's often something you wouldn't have predicted using a load test.

Sure, but I just mean sending artificial load to see where you tip over.

Only for the cases you are set up to test. This is about headroom on your golden resources so any kind of overload you can buy yourself time for.

Reminds me a bit of this AWS builders post about doing constant work as variance can lead to instability.

https://aws.amazon.com/builders-library/reliability-and-cons...