Systemd vs Sysvinit

I think the whole Systemd/Init issue has become way too Political. I use which ever works for me and Distro I'm using at the moment. Both systems work. Both have their quirks and strengths and weakness. I've found that most systemd haters have other than technical reasons for hating it. So I take it all with a grain of salt.
 


I went into System Monitor and killed SystemD
You killed systemd? Eh ... you can't! systemd is ID 1, and from
Code:
man 2 kill
we get this:
The only signals that can be sent to process ID 1, the init process, are those for which init has explicitly installed signal handlers. This is done to assure the system is not brought down accidentally.
systemd = init = ID 1, and init can do whatever it likes upon receiving SIGKILL (including exiting), but systemd's init does not install any signal handler for it, so nothing happens. Also, If you somehow managed to do that, I'm pretty sure some very important components, like services and daemons running in your OS for it to work properly, would stop working right there.
systemd is a system and service manager for Linux. It provides aggressive parallelization capabilities, uses socket and D-Bus activation for starting services, offers on-demand starting of daemons, keeps track of processes using Linux control groups, maintains mount and automount points and implements an elaborate transactional dependency-based service control logic. systemd is compatible with SysV and LSB init scripts and can work as a drop-in replacement for sysvinit.
 
You killed systemd? Eh ... you can't! systemd is ID 1, and from
Code:
man 2 kill
we get this:

systemd = init = ID 1, and init can do whatever it likes upon receiving SIGKILL (including exiting), but systemd's init does not install any signal handler for it, so nothing happens. Also, If you somehow managed to do that, I'm pretty sure some very important components, like services and daemons running in your OS for it to work properly, would stop working right there.

I'm not sure what caused my computer to suddenly speed up after I supposedly did that, but I'm not complaining
 
systemd-resolved is a systemd unit/service, and removing that doesn't remove systemd.
From here: https://www.freedesktop.org/software/systemd/man/systemd-resolved.service.html
systemd-resolved is a system service that provides network name resolution to local applications. It implements a caching and validating DNS/DNSSEC stub resolver, as well as an LLMNR and MulticastDNS resolver and responder.

I'm not sure what caused my computer to suddenly speed up after I supposedly did that, but I'm not complaining
Easy; you didn't. :) What you probably did was stop/disable some nasty process/daemon/service that was eating your OS/machine's resources up. But you can't remove/kill systemd, unless you replace it with some other init tool such as runit https://wiki.archlinux.org/title/runit or SysVinit https://wiki.gentoo.org/wiki/Sysvinit, otherwise your OS is completely useless. You shouldn't be doing stuff you don't know what they are, let alone what they do.
 
There's a 'systemd' process in the task manager and you can actually kill it - but it just logs you out and logs you back in again and the 'systemd' process resumes. (Just ran through this to test as I'd never thought to try killing the process before.)

So, yeah...

What you probably did was stop/disable some nasty process/daemon/service that was eating your OS/machine's resources up.

^ that
 
I was thinking about this again today, i don't know nearly enough to comment one way or the other. My Ubuntu distro has an annoying display bug that's probably related to systemd since it happens right after the BIOS splash screen is completed, i will be reading this thread for sure since i like learning about this stuff.
 
There's a 'systemd' process in the task manager and you can actually kill it - but it just logs you out and logs you back in again and the 'systemd' process resumes. (Just ran through this to test as I'd never thought to try killing the process before.)

So, yeah...



^ that

That's what happened
 
That's what happened

I'd sort the task manger by mem or cpu - depending on your needs. I find sorting by mem to be best for me, as it shows me what's consuming all my RAM.
 
I'd sort the task manger by mem or cpu - depending on your needs. I find sorting by mem to be best for me, as it shows me what's consuming all my RAM.

How do I do that?
 
Okay gurus please explain what his tells you and what is state degraded.

Thanks

Code:
● lubuntu
    State: degraded
     Jobs: 0 queued
   Failed: 1 units
    Since: Fri 2022-07-22 05:55:03 CDT; 14h ago
   CGroup: /
           ├─sys-fs-fuse-connections.mount 
           ├─sys-kernel-config.mount 
           ├─sys-kernel-debug.mount 
           ├─dev-mqueue.mount 
           ├─user.slice 
           │ └─user-1000.slice 
           │   ├─[email protected] 
           │   │ ├─session.slice 
           │   │ │ ├─xdg-document-portal.service 
           │   │ │ │ ├─1202 /usr/libexec/xdg-document-portal
           │   │ │ │ └─1211 fusermount3 -o rw,nosuid,nodev,fsname=portal,auto_unmount,subtype=port>
           │   │ │ ├─xdg-desktop-portal.service 
           │   │ │ │ └─1314 /usr/libexec/xdg-desktop-portal
           │   │ │ ├─pipewire-media-session.service 
           │   │ │ │ └─943 /usr/bin/pipewire-media-session
           │   │ │ ├─pulseaudio.service 
           │   │ │ │ └─944 /usr/bin/pulseaudio --daemonize=no --log-target=journal
           │   │ │ ├─plasma-xdg-desktop-portal-kde.service 
           │   │ │ │ └─1329 /usr/lib/x86_64-linux-gnu/libexec/xdg-desktop-portal-kde
           │   │ │ └─pipewire.service                                                              
           │   │ │   └─942 /usr/bin/pipewire                                                       
           │   │ ├─background.slice                                                                
           │   │ │ └─plasma-kscreen.service                                                        
           │   │ │   └─1064 /usr/lib/x86_64-linux-gnu/libexec/kf5/kscreen_backend_launcher         
           │   │ ├─app.slice                                                                       
lines 1-31
 
Bartman wrote:
what is state degraded
It means the system is operational but one or more units failed. It's in the man page. To find out what failed you could sift through the output on the systemd units with a command like the one in post #28. For example, on one of my machines I find:
Code:
 systemd-machine-id-commit.service - Commit a transient machine-id on disk
     Loaded: loaded (/lib/systemd/system/systemd-machine-id-commit.service; static)
     Active: inactive (dead)
  Condition: start condition failed at Sat 2022-07-23 07:35:17 AEST; 4h 41min ago
             └─ ConditionPathIsMountPoint=/etc/machine-id was not met
       Docs: man:systemd-machine-id-commit.service(8)
So it tells me "start condition failed" and refers me the man page. There is no discernible effect of this failure for my use of the machine, so it's not a concern in this case.
 
How do I do that?

Using Mint and the stock task manager as my source:

System Monitor_001.png


Click on 'em to sort by them. You can click on them twice, to sort low to high or high to low. (Down or Up)

I don't think I've seen a GUI task manager that didn't allow you to sort this way? There might be some. I want to say one of the older task manger GUIs from the LXDE would let you sort that way but wouldn't remember the setting, so you had to sort it manually every time you opened the task manager (system monitor in Mint lingo).

Just like sorting columns in other GUIs, it's pretty intuitive seeming - once you know you can. It's how you do things like sort by date/time in Thunderbird. Same principle.
 
If one wishes to know what systemd is doing on a system, run:
Code:
systemctl -a status
The output shows what's active and inactive.

Bartman wrote:

It means the system is operational but one or more units failed. It's in the man page. To find out what failed you could sift through the output on the systemd units with a command like the one in post #28. For example, on one of my machines I find:
Code:
 systemd-machine-id-commit.service - Commit a transient machine-id on disk
     Loaded: loaded (/lib/systemd/system/systemd-machine-id-commit.service; static)
     Active: inactive (dead)
  Condition: start condition failed at Sat 2022-07-23 07:35:17 AEST; 4h 41min ago
             └─ ConditionPathIsMountPoint=/etc/machine-id was not met
       Docs: man:systemd-machine-id-commit.service(8)
So it tells me "start condition failed" and refers me the man page. There is no discernible effect of this failure for my use of the machine, so it's not a concern in this case.

Interest when I enter the command you posted post #28 the output has changed from a red status indicator to a green status indicator.

So it appears all is well now.
Code:
lubuntu@lubuntu:~$ systemctl -a status
● lubuntu
    State: running
     Jobs: 0 queued
   Failed: 0 units
    Since: Fri 2022-07-22 20:33:19 CDT; 1h 16min ago
   CGroup: /
           ├─sys-fs-fuse-connections.mount 
           ├─sys-kernel-config.mount 
           ├─sys-kernel-debug.mount 
           ├─dev-mqueue.mount 
           ├─user.slice 
           │ └─user-1000.slice 
           │   ├─[email protected] 
           │   │ ├─session.slice 
           │   │ │ ├─xdg-document-portal.service 
           │   │ │ │ ├─1480 /usr/libexec/xdg-document-portal
           │   │ │ │ └─1489 fusermount3 -o rw,nosuid,nodev,fsname=portal,auto_unmount,subtype=port>
           │   │ │ ├─xdg-desktop-portal.service 
           │   │ │ │ └─1588 /usr/libexec/xdg-desktop-portal
           │   │ │ ├─pipewire-media-session.service 
           │   │ │ │ └─948 /usr/bin/pipewire-media-session
           │   │ │ ├─pulseaudio.service 
           │   │ │ │ └─949 /usr/bin/pulseaudio --daemonize=no --log-target=journal
           │   │ │ ├─plasma-xdg-desktop-portal-kde.service 
           │   │ │ │ └─1604 /usr/lib/x86_64-linux-gnu/libexec/xdg-desktop-portal-kde
           │   │ │ └─pipewire.service                                                              
           │   │ │   └─947 /usr/bin/pipewire                                                       
           │   │ ├─background.slice                                                                
           │   │ │ └─plasma-kscreen.service                                                        
           │   │ │   └─1065 /usr/lib/x86_64-linux-gnu/libexec/kf5/kscreen_backend_launcher         
           │   │ ├─app.slice                                                                       
lines 1-31
 
Interest when I enter the command you posted post #28 the output has changed from a red status indicator to a green status indicator.
If you take a good at both outputs, in the previous one you can read failed: 1 units, while in this one it reads failed: 0 units. That's why the change from red (degraded) to green (running).
 
If you take a good at both outputs, in the previous one you can read failed: 1 units, while in this one it reads failed: 0 units. That's why the change from red (degraded) to green (running).
That I understand I was going to troubleshoot the reason for what had failed to start as per @osprey post #34.

After running command posted by @osprey in post #28 the status showed green so unable to determine what had failed to start.
 
Ok guys here is your opportunity to put forward your thoughts on these 2 initialization systems.

I used to be very anti-systemd but have found it not to bad when using Debiandog (Based Debian Live with homemade utilities to make it behave like Puppy).

You can also talk about OpenRC or any other alternatives if want.
Systemd is not bad until you fill all the devices slots on your motherboard. Until recently I had 5 storage devices and 1 BluRay burning device. With all that systemd failed to init the system, kept saying "dependency failed for /dev/sda". After I detached sda, it began to init normally. I thought it was a kernel problem but it was in fact a systemd problem. So at this point I don't think it can work with more than 5 devices.
 
Using Mint and the stock task manager as my source:

View attachment 13206

Click on 'em to sort by them. You can click on them twice, to sort low to high or high to low. (Down or Up)

I don't think I've seen a GUI task manager that didn't allow you to sort this way? There might be some. I want to say one of the older task manger GUIs from the LXDE would let you sort that way but wouldn't remember the setting, so you had to sort it manually every time you opened the task manager (system monitor in Mint lingo).

Just like sorting columns in other GUIs, it's pretty intuitive seeming - once you know you can. It's how you do things like sort by date/time in Thunderbird. Same principle.

Got it. Thanks!
 

Members online


Top