Our tests have been timing out like crazy the past couple of days

https://azure.status.microsoft/en-us/status means -- maybe among other things -- tests that download helm binaries are failing

Yeah, one of my clients sites is COMPLETELY down. This brings me back to one of the ideas I had earlier this week, where maybe we should retain a CI cache of the default downloads. Requiring network calls to 200 sites seems kinda silly, grand scheme, when most of the time we're retaining the same files

I can't remember if this is a phantom memory or I actually talked to Tom, but it seems like in principle downloads could be in the REAPI cache?

I also recall looking at this a while back, and im pretty sure there were cached somewhere, but not sure what’s changed

iirc they are only in the local disk cache, not REAPI https://github.com/pantsbuild/pants/blob/1a7da5c5ed29e691b44babf367564881b5f83598/src/rust/engine/src/nodes/downloaded_file.rs#L67

but that would mean they should be in the github cache, I think? Depending on how much we cache anyways

Huh. I never thought of using init-pants in the pants repo. 🤯