A while back I was developing a workflow scheduler. However the run duration kept coming back negative. The runs kept finishing like 15 seconds before they started. Eventually I figured out why; the company had configured a policy to use their custom NTP server, but it was 16.7 seconds off. So my Windows laptop’s time was in the future, but the WSL VM (which ran the program) used another NTP server and was correct. In the end I just added a bodge that made the program request the time from an external server on startup to calculate an offset.
3rd line support for a global IT brand you have heard of, on the kind of security product that people don’t like to mess with too much once it’s installed on their servers.
Bug had been in the tracker over 3 years when I joined, and had defeated anyone who tried it. Segfault that happened to only one customer, unpredictably, and only once or twice a year, and nobody could reproduce it.
After a few weeks of tracing the compiled code line-by-line I found a race condition in the teardown of an object. That’s not the silly part.
I wrote the patch for the version of the code the customer was running and was hailed as a mighty hero and everyone on the bus clapped etc. Then I went to forward-port the patch to the newer versions of the program and… it was already fixed there. Some fucken doofus had either fixed the bug without realizing, or, fixed it and just skipped the standard backporting process that would normally patch all the other versions that were still in support.
Forgot to save. You ever spend 45 minutes looking for a bug you already fixed?
Yep. Last time I used MS Visual Studio it was still all too eager to run an older binary, and I lot plenty of time debugging a binary that doesn’t reflect the current code. And, even more recently, when testing requires you to remember to run separate routines to upload to a testing server and restart a service, it is really easy to forget to do that when you are bouncing between the code and the browser console to figure out what to change.
My system didnt boot for obscure reasons
Found out something had caused my nix store (packages) to explode in size. Chrooting didnt work as it was filled to the last byte. Deleted some files, then chrooted in and did a cleanup
One time I was struggling to understand why a code branch I wrote was being unexpectedly taken, and I found that I had written something like
if pet == 'cat' || 'dog'. Since'dog'was a string, the condition was always true. I laughed out loud at myself.A static type system can be used to make this type of error occur at compile-time rather than runtime. But, not everyone wants to make that tradeoff and some even like the semantics applied to that kind of expression.
||often gets tasked with handlingnull/undefinedin JS, e.g.
Back when I was learning c++ on Windows, I naively jumped right into game dev (as one does) before understanding the purpose of a debugger. I spent 3 months chasing down a period that should have been a comma. Due to function overloading, this was not an error or warning. It’s just silently calling the wrong function and crashing the program.
Thus began my distrust of magic compiler features. To save typing a handful of characters, languages add new ways to blow off your own leg. Every time you blow off your leg, the time you lose is orders of magnitude greater than just being explicit. Just read this thread for more examples. Today, I program almost exclusively in C. I have no desire to use more feature-rich/higher level languages for anything.
ECU (the computer in your car) only came online on the bus at every 3rd restart. No memory was written, yet the bastard somehow knew it anyway.
Turns out the transciever’s enable pin was left unconfigured and supposedly some charge accumulated during two startups to set it.
Love this
Goddamn case sensitive POS JavaScript.
I can’t quite remember what I was debugging exactly, some constant was different in runtime compared to what was in code. Someone added a leading 0 to a const int to make the number align with others in terms of character count just to make it pretty I guess. That is when I learned about octal literals in C. I think that was the only time I’ve seen one in many years and it was used unintentionally. To save you a search leading 0 in C makes number to be base 8 same as 0x makes it hex aka base 16, so that inconspicuous 01000 is not a 1000 it’s a 512.
Haha, I always love that one when I debug some other codebase
I fixed a 5 year old issue that the previous guy claimed was “unfixable”. Turns out it was because someone copy-pasted a value from an Outlook email…and it had a single dash in it…which outlook converted to a double dash.
That’s why I was so confused about the AI emdash connection. MS Office has been replacing dash with emdash for 20+ years.
I was confused too. But now I think it’s specifically em dashes in the middle of words, isn’t it? Like copy—paste instead of copy-paste?
I have had issues of having the same characters but in Unicode failing. But if I type it manually or paste it into Notepad++ forcing it to ASCII first, it works.
Internet Explorer didn’t automatically have “console.log” defined until you opened the console window. So there was an error breaking the web page that would disappear when you tried to open the console to see the error. The error was that console.log wasn’t defined. Fucking IE.
That would genuinely make me clock the fuck out for the day when I figured it out, jesus christ.
/thread, holy shit lmao
This was dumb, and it was on me.
Couldn’t figure out why two obscure software packages were erroring in a very particular way. Thinking there was no way in hell a search would help, I plugged it in Google. There was user on Spiceworks outlining my exact situation and fix. Who is this mad genius?!
It was me. I fixed it two years previously and went back to note so in the forum.
The one and only denvercoder?
Person before flipped a bunch of if checks to put the value before the variable (if 6==x rather then if x==6). They messed up a bunch of the <= and >= checks by forgetting to reverse them. I improved performance and fixed numerous bugs by tracking those down.
Ah yes, yoda conditionals. Because guarding against accidentally typoing
if(x=6)is worth destroying your code’s readability.This is not needed anymore thanks to compiler warnings
Tell that to my colleague who still insists on writing if conditions like that. Maybe it’s just a PHP thing.
I think php had a linter for stuff like this. Defensive programming is good, but having tools to do the defending is better so I can use my brain power where it matters
I was hired to work on a CMS designed for schools. I was tasked with debugging.
I started off with the newest issues and worked my way back.
Well, there was a bug where teachers would upload a video but it wouldn’t play on the page. It was plaguing a team of 8 for YEARS.
It was mentioned in our morning standup, and I decided to fast track it.
The dude who wrote the code was shrinking the display size in half: well how do you think that would work when a dimension of the video isn’t divisible by 2?
So, remember, I am the new guy, quickly fixed it using an if/then and the % modulus operator, and lo and behold, the next morning standup I had to explain the fix in front of the project manager, his boss, and the entire team…
fucking awkward. “You see, there’s this thing in division called a remainder.” This was the first thing in the morning.
But because the bug had been so ever present, that’s why the boss’s boss attended the standup. God, I just wanted to die.
I don’t think I understand what the issue was or what the fix was. Was it a language or system that cannot perform the division with truncation?
Let’s say height and width are 320x214.
(these numbers are arbitrarily chosen)
The code was sizing it down by /2, so now it is 160 by 107.
Because the width being odd, it would display.
But:
if ($width%2) { $width = --$width; }
Hastily written out on the phone, but that was the fix. This was back with php 5.
Sorry, but this comment still doesn’t make sense (or has typos).
You can’t halve a pixel. Truncating would get around the issue but big issue is that you can’t have a float as a video resolution. The browser should just truncate it regardless but the original code is flawed.
OPs fix essentially does the same thing.
For a strong typed language, an integer division should not cast to any other data type. Even assigning a float to an integer should not be allowed (that’s why C is not really strongly typed). I think from the syntax it was PHP. If any of the dimensions were not divisible by 2, then the number changed from integer to float and it could cause an issue somewhere downstream the video processing pipeline.
I don’t think it’s reexporting a video or something, I think it’s the embed bc it is a CMS. PHP 5 isn’t that old so video in webpages isn’t a novel idea.
But you’re right. Can’t halve a pixel.
Do note that PHP5 is old, predating HTML5. Browsers probably still struggle with malformed HTML and in the absence of any sane value, just crash. Also, who knows if it was even HTML. For all I know it could be a flash.
Fuck, I forgot about Flash. I didn’t think it was that old. Odds are it’s just some other internal video player tbh.
OP didn’t say that the problem was with the change in type. I think the interesting part is the downstream issue: what failed when a float was given while an integer was expected?
Dude, this reminds me of my coworker who was setting up a network and was trying to figure out a subnet mask, and he said, “I am trying to find a subnet mask between /22 and /23.”
He spent the entire day on it. I finally had to ask, and once he told me his goal, I of course asked, “how the fuck do you get half of a bit?”
Oh, the blood drained from his face. “Yeah, bro, that’s CIDR.”
How does this not make sense? I am spoon feeding this shit. Are you just vibe coding?
I think there are some versions of Python and/or PHP where the normal division operator silently promotes to IEEE floats, and you have to explicitly cast back to truncate/floor the value.
Back in the 90’s I had to debug a printer routine in a reporting tool. It would occasionally crash. Only on Wednesdays. Only on Wednesdays in September. Only on Wednesdays in September after September 9th.
Code was written in c.
What was the fix. I’m a dumb data scientist, I don’t know C
c requires you to do things like define the maximum length of a string. In fact a string in c is literally defined as an array of characters. There are no protections in c, so you can define a string to be 10 characters long and then copy 20 characters to it. The result is most often a stack overflow that immediately crashes the program.
This was legacy code that was likely written in the early days of DOS before things like Microsoft Windows existed, so memory usage was still pretty important. Whoever wrote this code tried to calculate the maximum size of a string used to print a header in this reporting package as a way to optimize memory usage. But they were off by one character. So when the code created a header with a date containing the longest month name, the longest date name, and a two digit day of the month, then that string overflowed and crashed the program.
Might be the character count. Wednesday has many, September has many, after the 9th has two.
Program filling the date into a form field of width 1 less than was possible with actual dates.this but string length according to OP.
As an even dumber data analyst, something about this comment makes me feel seen. Thank you. <3







