What’s the most frustrating bug you've ever encountered, and how did you finally fix it (or did you just give up and rewrite everything)?

- Parameter sniffing on mssql
Fix: fuck around, add recompile hints, change a sql query by random to confuse the query store enough. (Using ORM)

- Nuget dependency conflicts
Fix: 10 hours of crying, trial and error and version bindings

Yeah I just rewrote the 100k loc project because I couldn’t fix the bug.

It a was condition race

It would have been something to do with Visual C++ and Windows 3.x in the late 90s, and probably either memory management or GDI related.

Nothing I've worked on in the last 20 years comes close in terms of making you want to gouge out your own eyeballs and pelt them at the screen.

@Lensflare was it autogenerated or do you have like 1k words per minute typing speed

if I can't fix bug I patch it. put in recovery procedures. detect bug automatically, restart the buggy part. can't do anything else

rust has a bunch of weird things i can't fix... before rust worse bug was I think in java something in RAM got corrupted once a year. we were thinking it was like a solar flare bugging out a bit or something so I had ignored it. decided it needed to happen again and it happened again 2 years later but... not after that (granted I ended up leaving)

I don't know if it's exactly frustrating. it's just the nature of things

@Lensflare so you've NEVER rewrote a project? eeee

I once rewrote something that was 20k lines but it took 6 months. wasn't full-time on it either... think I had a 32.5 hour workweek then, 3 days a week... and that was on top of meetings and another "very important" project but there wasn't any work for that just management worried for no reason

had to be rewritten because the first version didn't have unit tests. also whoever did the specs failed to say each part of the code has to run on independent and different machines... so the whole architecture was just wrong

@jestdotty I did, but not because I couldn’t fix a bug.