4 ms·

Ask HN: What was your longest time you've spent debugging a single bug?

11 points by acidfreaks 11y ago

amirouche 11y ago
I spent lot of time (3 weeks) on a bug related to DBUS configuration. The issue was mis-configured DBUS but I was told: "don't look at dbus".
acemarke 11y ago
Might not be my actual longest (and certainly isn't anywhere near the stories that I'm sure others here have to tell), but my favorite debugging problem was one I worked on a couple years ago. I had a Python service that was doing a bunch of number crunching, so I used Cython to speed up the math-intensive portions of the code. Added type definitions, referenced several C math functions from math.h, and it was working great on Windows. Set it up using the side-by-side file approach, which lets you leave your original Python files alone, and add all the type info in a ".pxd" file that has to match the exact name of the original ".py" file. So, if I had "FileA.py", I'd also have "FileA.pxd". Unfortunately, when I went to run this on Linux, it compiled fine, but started throwing some weird exceptions buried deep down in the code. I hadn't actually written the math portion of this myself, but traced it pretty far down. Finally concluded it wasn't finding "cos" somehow, which didn't make any sense, because I was explicitly importing the C function from math.h if using Cython, and importing the normal Python function from the "math" module otherwise. Took me about 6 hours of tracing, investigating, and steadily growing confusion before I finally figured out what was going on: there was a single letter casing mismatch between a .py file and its corresponding .pxd file. One had a capital 'O', the other had a lower-case 'o'. Worked fine on case-insensitive Windows, but under case-sensitive Linux, it never actually found the proper .pxd file, didn't get the right imports, and thus stuff eventually exploded. Definitely a pretty high hours-of-work to number-of-characters-changed ratio :)
- kevinsimper 11y ago
  I also spent a lot of time debugging MySQL table names that on windows is case-insensitive, but not on linux.
jareds 11y ago
Longest so far has been two weeks spend single stepping assembly code and viewing memory output. Turns out in the latest OS release memory was not initialized to 0 which caused junk to get copied allowing the program to execute with out crashing but not give the expected results.
hoodoof 11y ago
Well not a bug really, but working out how to do something never done before, with many technical challenges, maybe six to eight weeks.
- hanniabu 11y ago
  Mind if I ask what it is your working on?
  - hoodoof 11y ago
    Virtualization.
DiabloD3 11y ago
3 years. I still cannot display video on my Chromecast while continuing to play audio on my phone while using any given Chromecast app.
cakes 11y ago
2 - 3 months where I was primarily "focused" on the issue which was ongoing for the place I worked at the time (ongoing being years and multiple people). It was an embedded system problem and I didn't fix it and I don't know if it ever did get fixed - sometimes I wonder if it ever was figured out.
- EricBart 11y ago
  That sounds like hell. Having to leave anything unfixed is terrible, let alone after working on it for as long as you did.

11 points by acidfreaks 11y ago

I spent lot of time (3 weeks) on a bug related to DBUS configuration. The issue was mis-configured DBUS but I was told: "don't look at dbus".

acemarke 11y ago

Might not be my actual longest (and certainly isn't anywhere near the stories that I'm sure others here have to tell), but my favorite debugging problem was one I worked on a couple years ago. I had a Python service that was doing a bunch of number crunching, so I used Cython to speed up the math-intensive portions of the code. Added type definitions, referenced several C math functions from math.h, and it was working great on Windows. Set it up using the side-by-side file approach, which lets you leave your original Python files alone, and add all the type info in a ".pxd" file that has to match the exact name of the original ".py" file. So, if I had "FileA.py", I'd also have "FileA.pxd". Unfortunately, when I went to run this on Linux, it compiled fine, but started throwing some weird exceptions buried deep down in the code. I hadn't actually written the math portion of this myself, but traced it pretty far down. Finally concluded it wasn't finding "cos" somehow, which didn't make any sense, because I was explicitly importing the C function from math.h if using Cython, and importing the normal Python function from the "math" module otherwise. Took me about 6 hours of tracing, investigating, and steadily growing confusion before I finally figured out what was going on: there was a single letter casing mismatch between a .py file and its corresponding .pxd file. One had a capital 'O', the other had a lower-case 'o'. Worked fine on case-insensitive Windows, but under case-sensitive Linux, it never actually found the proper .pxd file, didn't get the right imports, and thus stuff eventually exploded. Definitely a pretty high hours-of-work to number-of-characters-changed ratio :)

kevinsimper 11y ago
I also spent a lot of time debugging MySQL table names that on windows is case-insensitive, but not on linux.

jareds 11y ago

Longest so far has been two weeks spend single stepping assembly code and viewing memory output. Turns out in the latest OS release memory was not initialized to 0 which caused junk to get copied allowing the program to execute with out crashing but not give the expected results.

hoodoof 11y ago

Well not a bug really, but working out how to do something never done before, with many technical challenges, maybe six to eight weeks.

hanniabu 11y ago
Mind if I ask what it is your working on?
- hoodoof 11y ago
  Virtualization.

DiabloD3 11y ago

3 years. I still cannot display video on my Chromecast while continuing to play audio on my phone while using any given Chromecast app.

cakes 11y ago

2 - 3 months where I was primarily "focused" on the issue which was ongoing for the place I worked at the time (ongoing being years and multiple people). It was an embedded system problem and I didn't fix it and I don't know if it ever did get fixed - sometimes I wonder if it ever was figured out.

EricBart 11y ago
That sounds like hell. Having to leave anything unfixed is terrible, let alone after working on it for as long as you did.