Live data from Hacker News

The weirdest bug I've seen yet

engineering.gusto.com

291–300 of 334 posts

Re: The weirdest bug I've seen yet

#291

Earlier quoted context omitted.

Much better than the Xerox bug that caused numbers in scanned documents to get changed. https://www.zdnet.com/article/xerox-scanners-alter-numbers-i... Ouch

I kind of remember this one. It wasn't really a bug. I think Xerox used software that was known to not be 100% trustworthy to recognize numbers when used at a certain compression level. It was even in the manual if i'm not mistaken.

The manual pointed out this can happen at higher compression levels, but they were able to reproduce it at all levels.

Re: The weirdest bug I've seen yet

#292

As I was reading this I was thinking to myself "I wonder if it is grammarly related" because I experienced a bug some time ago that presented itself in a similar way. It was impossible to reproduce but affecting lots of people internally within certain departments. Eventually we figured out the thing they had in common was that they had the Grammarly extension installed. The other key thing was that the bug only appe…

> It turned out it was because of a bad regex in the grammarly extension that caused the page to hang if the domain name was more than about 100 characters. Just today I debugged a regex that would DoS our backend whenever the user enters the wrong thing in a form. Now I'm reading up on regex engines: https://swtch.com/%7Ersc/regexp/regexp1.html

I had a ReDoS issue at a prior company many years ago; at least they're lintable now:

  https://ota-meshi.github.io/eslint-plugin-regexp/rules/no-super-linear-move.html
  https://www.npmjs.com/package/eslint-plugin-redos/v/1.2.0

Re: The weirdest bug I've seen yet

#293

The weirdest bug I've seen yet was a concurrency race condition which led to 'spooky action at a distance'. Every time I tried to console.log() to observe some value to see the cause of the bug, the bug would go away. But alas, I hadn't discovered some new quantum effect... It turns out to have been because the race condition was close within maybe 1 millisecond and adding the console.log() statement there meant that…

In Qt QML using print can lead to different behavior very very quickly since their engine creates bindings on whatever it comes by so using two variables in a print statement can lead to different behavior than just using one. QML in general is just awful.

Re: The weirdest bug I've seen yet

#294

   alert(banana++)
Should lock things up and not continue until OK is clicked.

If it is just you and your application you can just spray paint from the hip the alerts from top to bottom. After each line of html you can have one more alert(banana++), in the middle of your css alert(banana++) etc

If there are uhh thousands(?) of people actively using the page you put just one alert some place in the middle.

Everyone will have to click on OK, the page crashes. You ask the crashee: Did it crash before or after the "make America great again" alert message. (call it something they would remember)

Now you know the issue is in the top or bottom half. You move the alert to half way the half with the issue in it. If you can get some sort of reasonably consistent crashing you will find it pretty quick even in production.

Hope this helps, or maybe it is a stupid idea and you could explain why.

Re: The weirdest bug I've seen yet

#295
post #73

Earlier quoted context omitted.

I added to all pages in a single-page-web-app after discovering some bug or other with Chrome screwing up the page. Apparently the new incantation to fix an app (can be applied to an element) is (ugly: I presume it isn't CSS to avoid supporting dynamically changing it): https://developer.mozilla.org/en-US/docs/Web/HTML/Global_att... Every now and then I would look at the meta tags for a major single page app and disc…

I first read this as "translate to Norwegian"

In Norway, the word “subject” translates to “fag”. Back in the Usenet days there was a Norwegian group or hierarchy named “no.fag”, which of course regularly got misunderstood.

Re: The weirdest bug I've seen yet

#296

I agree with the others here about the ending being a total letdown. Unfortunately, with access to neither the Chrome source code nor the Grammarly source code, we can only guess. Is this what the "open source" movement has created --- developers who are totally lost without source code and refuse to dig deeper? Of course the corporate interests who don't want us to know the truth, because their profit depends on it,…

Presumably they make a network request to get the options. So it makes sense to have a loading spinner just in case it takes a long time, even if it's usually near-instant.

Re: The weirdest bug I've seen yet

#297

Earlier quoted context omitted.

I've seen the opposite-ish: Someone who was dual booting windows and linux only had working wifi if they booted into windows and rebooted into linux. The linux install didn't have the firmware package for their wifi card installed, so when rebooting from windows into linux it was still all primed and ready to go but not when cold booting into linux directly.

This is my exact problem with a HP envy X360. HP refuses to fix the bug in BIOS. :/ https://bugzilla.redhat.com/show_bug.cgi?id=2107845

Every other day I read some fascinating HP adventure. I've done some questing myself, cant really tell if I did well, are they new multiple issues or the good old same?

Your comment had me wonder if a kickstarter for an rpg game called HP printers would take off. The potential fan base is HUGE, I imagine even people who would never play a game would want to see it.

Let the GUI be like little computer people (only an office)

https://youtu.be/SkTgX1mGmDg?t=17

Then the player has to go though rituals depending on the HP printer type, swap old for new then new for old then old for new ink cartridges after finding the instant ink package in storage, if any spawned of course.

If you try to go to fast the hero starts destroying the office, starting with the printer. Then a new different model is delivered. Ready for setup.... or maybe not?

In some laps of consciousness I purchase one more some years ago. The hours of configuring it made me feel dumber than the wrath of the HP dungeon master.

The ISP has a wifi router fixed on channel 11, half the neighborhood sits on channel 11, range is about 4 meters, the printer doesn't reconnect unless unplugged first. But there is wifi direct, this doesn't have 50 cm of range, it wants to use channel 11.

in the game our hero at least at first doesn't know about channels. The next printer arrives: Something seems off, there is no usb cable in the box? Do you want to A) order a cable B) we will use wifi! Of course if you order the cable the boss will complaint that it takes to long to set up the printer but when you eventually get the cable the boss wants to print from his phone...

https://freesound.org/people/InspectorJ/sounds/402095/

Re: The weirdest bug I've seen yet

#298
post #43

The weirdest one I saw was this: User claims that the wording of the info they enter into a certain form changes when they save. At first I suspected someone else editing the same form at the same time unbeknownst to each other but it wasn't the case according to the logs. And I saw the correct wording on my own computer. Then I noticed in their screenshots that some of the menus had weird wording too. Turns out they…

This is really annoying. I often rely on Google Translate to read Japanese websites (I can read Japanese but only very slowly), and it breaks every website using React [1] because both React and Google Translate try to update DOM nodes without knowing each other. I even seriously looked at Google Translate to see how they are implemented, so that maybe I can recreate the web widget without this issue in the future.

[1] https://bugs.chromium.org/p/chromium/issues/detail?id=872770

Re: The weirdest bug I've seen yet

#299

As I was reading this I was thinking to myself "I wonder if it is grammarly related" because I experienced a bug some time ago that presented itself in a similar way. It was impossible to reproduce but affecting lots of people internally within certain departments. Eventually we figured out the thing they had in common was that they had the Grammarly extension installed. The other key thing was that the bug only appe…

If a website bug is not easily solved, first order of troubleshooting is to disable all extensions. Devs don't often think an extension could be causing the problem, but extensions can do wild things to a webpage. I've caught a few bugs caused by extensions this way.
Post reply on HN