GitHub suddenly rejected my SSH key (the fix was a .pub file?)

GitHub suddenly rejected my SSH key (the fix was a .pub file?)(thorsell.io)

25 points by xGLaDER 8 hours ago | 8 comments

Wish they'd fix what's broken, not fuck with stuff that works.

bluehatbrit 8 hours ago | |

GitHub is big. They'll be doing lots of things all at once. This is probably just a case of some normal component replacement for some maintenance reason. I'm as frustrated as anyone about the lack of stability and the hard pivot to AI, but this at least shows they're doing work on this area of the application.

For all we know this could be a knock on impact from an upgrade or component replacement for improved stability. Maybe even critical security fixes. I don't think it's fair to lambast them for doing work around the core git/ssh systems.

ghusto 6 hours ago | | |

> They'll be doing lots of things all at once

Make all of those things be related to "make the platform stable".

danlitt 8 hours ago |

Interesting fact, but the slop filler is distracting.

gwerbin 8 hours ago | |

I bet the author discovered the problem during an AI coding session. The "mathematically sound" comment sounds like the AI did some deep dive into a possible corrupted private key before checking for a missing public key. Maybe it's a claw-type system that automatically publishes a blog post when something interesting happens. Either way I flagged it.

tzs 41 minutes ago | | |

So let me see if I understand this. Something apparently has recently changed at GitHub that will break a perfectly valid way that clients can configure ssh. Someone hit that problem, tried a bunch of things to diagnose it, figured out what is broken and a workaround (switch to another perfectly valid way to configure ssh on the client).

Rather than going into detail on all the things they tried to figure that out they went straight to concisely explaining the workaround.

They then briefly explain why those two different ways to configure it lead to different flows on the server.

There is absolutely no question that this will be useful to anyone how hits this problem and has this turn up when they search for help. There is nothing in the writing that, if this had been written in 2018, any reasonable person would find to criticize.

But because you see one bit of phrasing that you think might be LLM influenced you flag it?

elashri 7 hours ago | | |

To be honest, the "mathematically sound" is a widely used jargon used in many academic fields (physics, mathematics from first hand experience). It is also more meaningful than people who would use "nice function" to mean a function that follow some criteria in their head (and you would need to guess what these will be).