back
16 comments
>The “code:0xBADDD15C” looks like it could be an arbitrary hexadecimal value, but it’s an example of an error code designed to both uniquely identify and suggest a mnemonic clue to the underlying issue. Apple documents many of these codes, which include 0xc00010ff (cool off), 0xdead10cc (deadlock), and 0xbaadca11 (bad call).

https://developer.apple.com/documentation/xcode/understandin...

Nice!

Editorialized title.

Post title is "Purgeable Mac Apps".

Submitted title is "Apparently macOS sometimes just kills some apps randomly"

macOS has a specific meaning for purgeable, and apps have to opt into it so I think the post is also mislabeled. This sounds not entirely different from the OOM killer in (old?) Linux.
Thank you so much for this! I have been so infuriated the last couple of months with persistent issues like Mail closing and Safari’s cache randomly being purged, losing many website’s login sessions. Two-factor authentication multiple times an hour gets frustrating very fast!

I was really struggling to search for the problem on the web, “Safari forgets website logins” led me nowhere. I was almost at the point of just trying a reinstall of macOS.

Indeed, my MacBook’s disk is pretty low on space. Not dire, but fairly low. I will clean up and hopefully things will return back to normal!

EDIT: Some users mention RAM and the OOM killer on Linux. I don’t think the situations are quite the same. It’s possible for a machine to have low disk space without any RAM pressure. I assume that is what is happening on my machine. When Mail started dying on me I considered and checked the RAM use, closing a lot of backgrounded applications, and even going so far as to reboot the machine without success.

1. This is where power users will always get irritated with Apple's philosophy - instead of informing the user that memory is running out and asking the user to decide which apps to terminate, Apple decides it for you. Good for naive users. Not for superusers.

2. Yet another reason to not go for devices with soldered RAMs from any company. (Remember, we used to run Linux and Windows on just 16/32 MB ram. It wasn't enough. Today, 8/16 GB are common. And yet again, not enough.)

This has nothing to do with RAM though. macOS does this when _disk space_ is low, not RAM.
You are right, my mistake. Still highlights why memory - RAM or HDD - needs to be upgradeable, and why it isn't a good idea to go for devices with soldered RAM and HDD / SSD.
I agree in respect of the SSD. RAM is a little different on M1 machines as there is supposed to be a substantial memory bandwidth improvement resulting from the tight integration with the SoC and processor.
If I’m not mistaken (but I very well might be), that’s true for the SSD too with the M1. The controller of the SSD is a part of the M1 chip; it’s not separate as with traditional SSD. @marcan42 tweeted about this at some point, but I might have misunderstood, it’s too low level for me.
Pour encourager les autres.
Ok? How much space did you have free?
Imagine a car that when running low on gas, randomly ejects passengers to reduce load. Would you choose to use such a car?
The linux oomkiller also does what you are saying. If you are out of memory, it "randomly" picks a process to kill/eject.

People even use the same analogies. https://chase-seibert.github.io/blog/2009/04/23/the-case-of-...

I think a better comparison might be intentional blackouts when there's too much electricity demand. Nobody likes blackouts, but they are necessary to avoid worse problems.
It’s a poor analogy as you’re assuming that the result of the exceptional behaviour isn’t needed to prevent a worse outcome. Rolling blackouts are a better analogy - literally causing the power grid to collapse (which takes a long time to restore) is clearly worse than small, short, localized blackouts.

That’s why things like the OOM killer exist in Linux - there are times where the system as a whole is better off at the cost of a few random users odd processes.

TLDR don’t run out of disk space