reupload because i mixed up sigterm and sigkill like a dumb fuck
Gonna make Lemmy pissed off, but installed on my machine Nobara, Cachy and Mint at some point. All of them had comparable if not worse boot and shutdown times to Windows 10. xD ( And worse performance in games but that’s due to having old Nvidia GPU xD )
echo b > /proc/sysrq-trigger
<_<
shutdown -f -t 0 -sXkill is my favorite. I prefer aiming the gun and pulling the trigger myself
Systemdead can be a dick.
Everybody gangsta until “A stop job is running for …”
Keeping it real hell yeah
user error doesn’t count
user error? what user error? mount a network drive, disconnect the network, then try to unmount it. best part? most programs will stall for a long time when it tries to access the directory.
I couldn’t find better screenshot, but I’ve definitely seen this happening due to zombie processes after just regular use, rather than just broken mount. I feel like it’s handled way better nowadays though.
I feel like it’s handled way better nowadays though.
TL:DR No, lmao. But timer was lowered at some point to 3min for most distros, afaik.
So too would be attempting to shut down Windows without saving and closing your applications then 😉
This but deleting a folder:
- Are you sure you want to delete this
- Delete too large to fit in garbage bin, so are you really sure
- Couldn’t delete stuff (for no clear reason)
- Even as admin file locks were hard blocking without any easy way to unblock
Meanwhile on Linux with sudo rm -rf, it’s just gone as demanded.
Partially true. The difference is that in Linux, when you delete a file, you’re just removing the directory entry (potentially just one of many entries that point to the same data). The filesystem doesn’t actually remove the data and reclaim space until all open handles are closed and no remaining directory entries point to the data.
Any running processes that have the file open are able to continue to read and write that data via the handle despite the directory entry being removed, until the handle is closed.
I think a file delete just removing an adress and not the actual data is common to all OSes. That’s why to safely erase data from a disk it is recommended to fully overwrite the disk with random data, potentially multiple times.
that’s a different thing. if you delete a file that is still opened by a process, the space will not get freed up until that process also closed the file. until that point the filesystem still keeps track of the file, it is just not present in any directories anymore.
If you delete a still opened file on Linux then the file will disappear for all processes which didn’t already open it, all programs that did already open it can still read and write to it and the file on disk will never be overwritten (as in, used for other files) as long as there’s still a process with the file open.
Simplifying how it works: The file you see is a link to the actual file(inode), when a program opens a file using this link they get a copy of the link. As long as one link/copy of it still exist the file won’t be deleted. When a program closes all its links get cleaned up so on shutdown all files which only have processes referring to them get marked as deleted.
I mean you could probably delete files with powershell then idk.
The first time I shutdown a Linux computer, I thought I broke something it happened so fast.
Been doing Linux for decades. sudo reboot is still very jarring.
Same. I still feel like I should be parking the heads on my 10mb hard drive. Honestly at this point, I’m too embarased to ask if there is a proper way to send my servers for a reboot, and I cross my fingers I can log back in.
I think that reaction comes from messing with computers too much. When you fuck up in a computer, that sudden shutdown is what you get. Gives me flashbacks.
No
It’s not a request, it’s a warning. The machine will be without power soon, and it’s up to the machine whether it wants to prepare for that or not
also true for boot (not from suspended state), in my experience.
windows: wait, let me display the windows logo for 10 seconds, then show a spinny circle, then show the lock screen, then when you try to enter your password, it loads your user profile for another 5 minutes before it shows your desktop icons
linux: click the power button -> 1.5 seconds later i see the lock screen. enter password and it’s just there.
Back when i still had windows as a second boot option, it was soooo annoyingly slow to boot (like 3 minutes or so). I thought it’s because I installed it on a HDD, not SSD (and that was indeed part of the reason). One day when my internet broke though, I realized it was actually super fast to boot suddenly. It just spent half the time downloading stuff from the internet before, during boot. From then on I just pulled the ethernet cable before booting windows. Fucking joke of an operating system
I’ve found it to be very dependent on the distro and the hardware it’s running on. Back when I was playing around with distros I definitely tried some that felt like you snapped your fingers and had a desktop. But I settled on Fedora and that takes longer to boot for me than Windows. Not that I mind, 30 seconds once a week or so just isn’t important to me.
Are you perhaps on Wifi? I noticed that Fedora is has configured Systemd to wait for online network before continuing starting the login services.
Reverse meme when it’s time to install the updates.
Windows in that case is “I MUST REBOOT IMMEDIATELY PREPARE TO LOSE ALL UNSAVED DATA IN 3. 2. 1…”
When I switched to win 10, I actually gave them more money to get the pro version for access to the group policy editor so I could control updates and never have to deal with my PC telling me it’s time to restart on its own. Because I was stupid.
When it came time to switch to Win 11, I did the much more sensible thing and installed Fedora instead. I started with cinnamon and even though I ended up disliking it also, it was still way better than the windows experience.
You should’ve never given them any money at all. Activation scripts ftw.
Not quite. I will RESUME FROM THIS FUCKING “MODERN SLEEP” shit, even though you the user want to turn this shit right off, do it without any warning watsoever, close all your fucking windows and good luck if you have lost work or not.
A stop job is running for User Manager for UID 1000 (12s / 2 mins)
( 1min 59s / 2 mins )
…
( 2 mins / 3 mins )
???
(5min 7s / no limit)
Shieeet
In those cases there’s an easy solution.
Step 1: sigh
Step 2: press the power button 5 or 10 seconds while contemplating why you decided to do a quick restart instead of keeping the session and do something actually productive
I recommend starting with SysRq+E before that, there’s a chance it gets whatever the shutdown was waiting for. And if that fails… REISUB my beloved.
RSEIUB. Raising skinny elephants is utterly boring.
I might be missing something, but the order I know and have always seen recommended is REISUB. Terminating processes might write data to disk, so it seems to me like you should sync after, not before. Though this is also generally unimportant with modern filesystems and storage media.
That’s because it first send
sigterm, thensigkill. Then it gives up and let the kernel handle it…Happens on my BTRFS disk’s unmount. If the kernel is currently busy handling some heavy btrfs command (like a 4tb scrub), systemd cannot stop it with sigkill.
So when it eventually gives up, you also need to wait for the kernel to finally stop the operation and actually disconnect the disk.
You can change that, but there is a maximum time when it just kills the job.
I think you can set the default stop job timer lower?
Reboot
Windows: save all your woooork. What apps you had open? How would I know?
Linux: it’s all saved in ram, don’t worry. It’ll be like you never rebooted
Are you aware that RAM is cleared on reboot?
Tell that to Linux
I had a systemd bug delay shutdown for 2 mins every time for a very long time on Debian. Never managed to fix it, Fedora did not have the same issue fortunately.
I fought one of those on NixOS for 2 months but on boot two whole minutes.
I had been poking at it for ages in my spare time, because i can only stand so many reboot tests, and once you were up, it was fine.
On day work issued us Claude code licenses. Maximum effort, maximum model, claude, my startup -> login it taking forever please go find out what’s going on.
3 reboots later and some very deep seeded issue with a version conflict in PAM that was pissing wayland/kde off in a way that there wasn’t really any error waiting for a timeout.
also yes i know
shutdowntypically usessigtermand waits nicely, but it doesn’t take 45 seconds for no damn reason like windowsalso
sigkillis funniertaskkillis pretty funny tooWindows: the shutdown mechanism cannot execute correctly because this process is still running:
ShutDownProcess
hoss
Definitely not a systemd based distro in the meme
Maybe something I don’t know, but I send kill commands through btop all the time on a systemd based machine.
The point here is that SystemD’s natural behavior is to send SIGTERM then wait an eternity.
Those “service XY is shutting down (5sec/2min)” messages you sometimes get on shutdown are coming from SystemD not waiting for 3 seconds like the meme suggests, but waiting for minutes before giving up and switching over to SIGKILL instead.
systemd nanny. But beyond þat, zombie processes have always been a þing in Linux.
