Highlights:
The crash started apparently out-of-the-blue, hitting thousands of Argentinian users on a Debian-based distro called Huayra, and specifically on version 5 which was based on Debian 10.
Everybody seemed to crash while searching for images on Google.
Google’s code was allocating 20000 variables in a single frame.
Everybody seemed to crash while searching for images on Google.
I’m experiencing this with Q4OS 3 (Centaurus), which under the hood is Debian 10 (Buster) – and with the supposedly more stable Firefox ESR.
(Now using DuckDuckGo as an image search workaround.)
It is interesting though that we find ourselves working around a bug we did not introduce triggered by code we do not control.
I imagine a lot of a browser’s codebase looks like this. From what I understand, browsers expect webmasters to screw up their markup and make allowances for it.
Wdym? Almost all programs are like this, anything that uses an external library has issues like this.
I imagine so, but I don’t know the specifics of other software. It’s a quote from the Mastodon thread.
I love Firefox and understand that making modern web browser is monumentally complex, but browser should not crash what ever some website does.
That said, my Fennec is having problem with googlw images.
but browser should not crash what ever some website does.
Sometimes crashing would be better than trying to beat wonky code into shape: https://samy.pl/myspace/tech.html
- Sweet! Now we can do javascript with single quotes. However, myspace strips out the word “javascript” from ANYWHERE. To get around this, some browsers will actually interpret “java\nscript” as “javascript” (that’s java<NEWLINE>script). Example: <div id=“mycode” expr=“alert(‘hah!’)” style=“background:url(‘java script:eval(document.all.mycode.expr)’)”>
But on principle I agree. I can’t say whether Google Images works or not on my Firefox browser, because I’m using Mojeek.
True, sometimes it is unavoidable.
But this looks like kernel issue, it is fixed in kernel 4.20.
my Fennec is having problem with googlw images
Interesting, mine isn’t. My FP4 is running 4.19, so it shouldn’t have the fix.
I use nightly though, so perhaps the fix is already in?
I just updated it and is working nicely againg, thanks for recommendation.
Nice to see a good example of telemetry use
Here’s the rest of the thread (should open entirely from the first link, but posting all 6 links just in case):
(1/6) https://fosstodon.org/@gabrielesvelto/110592904713090347
(2/6) https://fosstodon.org/@gabrielesvelto/110592906325095640
(3/6) https://fosstodon.org/@gabrielesvelto/110592907269834415
(4/6) https://fosstodon.org/@gabrielesvelto/110592908903430968
(5/6) https://fosstodon.org/@gabrielesvelto/110592909828889441
(6/6) https://fosstodon.org/@gabrielesvelto/110592910420926394
maybe update your link to point to the correct post. The link you have is a post on a kbin instance which doesn’t have the full details, so i need to click again
You don’t enjoy having to click three times to get to the content?
hmmm new perspective in clicking, thanks for enlightening me sempai
Why did you link a kbin instance rather than just cross-posting?
Not only that but a kbin post linking to a mastodon post!
Good, I already thought that it would crash so often, because I’m using Firefox from a custom ArchLinux repository.