Um, you really need to read the entire phrase and not pick out only what you want from it. 😃
Swap can make a system slower to OOM kill, since it provides another, slower source of memory to thrash on in out of memory situations
It means that if you try to use it as a source of memory, when you run out of actual RAM it will make your system almost completely unresponsive due to disk thrash, instead of allowing the kernel to just kill the process that’s eating your RAM. So you’ll just end up hard-booting system.
Yes, and that’s a good thing if you don’t want it to start killing processes. You have that extra time/space to deal with the out-of-memory condition yourself.
Or you can ignore that condition and continue using the system in a degraded state, with swap as “disk RAM”.
Like I said, the system will be almost completely unresponsive due to disk access being several orders of magnitude lower than RAM and allocation thrashing… you won’t be able to do much, the mouse, keyboard and display will react extremely slowly. There may be situations where you’d prefer this to an OOM kill, for example if you’re running a test or experiment where you’d rather have it finish even if it takes a very long time rather than lose the data. But if you’re a regular desktop user or server admin you’ll probably just reboot.
Yep. In fact my comment seemed so clear to me that I assumed it was some kind of joke, but looking at the votes, maybe swapfiles aren’t as well known as I thought.
So, with btrfs on ssd, is there any use case for a swap partition?
Use case is not having enough RAM?
Swap is not “disk RAM”.
https://en.m.wikipedia.org/wiki/Virtual_memory
What would you describe it as? With virtual memory it pretty much functions that way, no?
https://chrisdown.name/2018/01/02/in-defence-of-swap.html
I’m not sure what that post is meant to show, if swap isn’t “disk RAM”. That post even concludes:
Um, you really need to read the entire phrase and not pick out only what you want from it. 😃
It means that if you try to use it as a source of memory, when you run out of actual RAM it will make your system almost completely unresponsive due to disk thrash, instead of allowing the kernel to just kill the process that’s eating your RAM. So you’ll just end up hard-booting system.
Yes, and that’s a good thing if you don’t want it to start killing processes. You have that extra time/space to deal with the out-of-memory condition yourself.
Or you can ignore that condition and continue using the system in a degraded state, with swap as “disk RAM”.
Like I said, the system will be almost completely unresponsive due to disk access being several orders of magnitude lower than RAM and allocation thrashing… you won’t be able to do much, the mouse, keyboard and display will react extremely slowly. There may be situations where you’d prefer this to an OOM kill, for example if you’re running a test or experiment where you’d rather have it finish even if it takes a very long time rather than lose the data. But if you’re a regular desktop user or server admin you’ll probably just reboot.
I think what they mean is that you can just make a swap FILE instead, which you can grow and shrink as needed. No need to mess with partitioning.
Yep. In fact my comment seemed so clear to me that I assumed it was some kind of joke, but looking at the votes, maybe swapfiles aren’t as well known as I thought.
Do you mean that you don’t have to find the LBA of the extents of your swap file, and put that into a kernel argument anymore?
Cuz that is a nasty, skanky hack.
I’ve never heard of that, it’s beyond me. So it’s an increased risk when tweaking the kernel? As an average home user it’s all right?