Timo Rothenpieler

I haven't had a chance to give this a test yet, but in theory the code should be complete. The Kernel behaviour in regards to the response address doesn't really...

I have now finally had the chance to test this, and after fixing two small kinks (confused IFLA_BROADCAST and IFA_BROADCAST and Pkt4 refusing the long CHADDR), it's working just fine....

I've been using this on our network/cluster since a while now, and it's been running without issues. So if there is anything further I should do to move forward with...

The basic idea of the way it's implemented in our fork is that instead of a baseline image, we point at a baseline directory, and as long as one image...

I'll need to rebase, or even rewrite, the patch. The introduction of the summary-Mechanism has complicated it quite a bit. Will send it over as PR then (excluding some of...

The issue I see with that is still that it's _A LOT_ of permutations. And it's not always obvious which dependency is causing the change. And sometimes the same set...

Already linked to them above, in my first reply: https://github.com/OGGM/oggm-sample-data/tree/master/baseline_images/freetype_28/test_raster That's probably the most clear demonstration of the random differences that can occur. The other tests show similar differences. Some...

It's primarily telling me that we should investigate using (perceptual?) hashes instead of a growing library of baseline images :D But I'd probably like to have the same option there...

That sounds much more elaborate than anything I had thought of so far, but it'll definitely solve the issue in an elegant way that's hopefully useful for a lot more...

Not entirely sure why the tests are failing. Some of those I'm relatively sure I didn't even touch?