Ian Denhardt

Quoting Jason Hennessey (2017-07-28 18:07:16) > From [1]what I can tell, one only gets a single .travis.yml, which has > a single global "sudo" statement. Darn. > Another option would...

This might be worth it at some point, but it strikes me as relatively low priority. I'd say keep it open though. The slowest run we have right now though...

I'm thinking about how a network (switch) driver for this would work, and it seems like it wouldn't need to touch namespaces per-se; it would basically treat the veth devices...

> In theory, other than headnode stuff (which we eventually want to > remove), it should be possible to make HaaS multi-node as the DB is the > point of...

> Running multiple HaaS APIs (as opposed to the network action > daemon) under separate webservers. Ah, I see. Yeah, that should work, though we will have to remove the...

> I thought we supported mysql? Doesn't SQLAlchemy get us support for > several DBs? Almost. It makes it a lot easier to do portability, but there are a few...

I agree we should be consistent. A bit of background: Stuff started off as all markdown, but at somepoint someone suggested we should move to rst, because it's more standard...

I don't think it's worth duplicating the docs, Is there anything besides the README tha has a concrete constraint like this?

I'm okay with that.

My instinct is to keep this out of the server, since as you point out it can be implemented on top of the existing calls.