Jacob Wirth

The toxic needs some work to generalize the usage, since it needs to be controllable via json, but it's definitely something we can put in core. It will also need...

This might be interesting to bring up again. I'm not sure of all the use-cases for this, but a couple things I can think of to put as options are:...

I'm currently looking at it, and I think it's possible to add, it just might be a little messy with matching corresponding upstream/downstream toxics together. The question is mostly if...

Actually, I think most of the work on bidirectional toxics was done, it just needed some review and polishing. I got stalled and stopped working on it, but I could...

There's a couple reasons why things were done the way they are, though that's not to say they couldn't be improved. The first is that having all clients toxic independent...

Updates to `CHANGELOG.md` and `CREATING_TOXICS.md` to come.

I think this should be ready for review now, @Sirupsen Feel free to add some other people to review, I'm not sure who else is interested in Go + Toxiproxy...

Upstream needs to be a hostname + port, not a url. Toxiproxy doesn't handle any protocol-specific proxying. You'll need to set your upstream to `x.y.com:5556` Since you're using HTTPS, you'll...

Your findings are pretty much correct: Toxiproxy 2.0.0's cli uses comma separated attributes, and only the last `-a` flag is used due to how the cli parser works. This was...

To be clear about versioning, currently git/master is "unreleased" and is subject to change. It's effectively development for Toxiproxy 2.1.0 right now. Docker and all release builds are Toxiproxy 2.0.0....