You are not logged in.

#26 2012-09-21 18:49:42

alphaniner
Member
From: Ancapistan
Registered: 2010-07-12
Posts: 2,810

Re: SystemD ugly boot

cfr wrote:

It should be configurable for accessibility reasons. Red/green colour blindness is the most common form and people shouldn't have to recompile systemd in order to easily distinguish successes from failures.

You're neglecting to consider the fact that it's not just blobs of color, but colored letters, forming words, presumably in a language understood by the user, such that they can be read, and comprehended, to easily distinguish success from failure.

Which is not to say that, all things being equal, I don't agree that the color should be user-configurable.

Last edited by alphaniner (2012-09-21 18:51:22)


But whether the Constitution really be one thing, or another, this much is certain - that it has either authorized such a government as we have had, or has been powerless to prevent it. In either case, it is unfit to exist.
-Lysander Spooner

Offline

#27 2012-09-21 22:19:01

cfr
Member
From: Cymru
Registered: 2011-11-27
Posts: 7,130

Re: SystemD ugly boot

The messages systemd posts scroll by *fast*. Noticing a patch of a particular colour is nonetheless pretty easy. Scanning the output for a particular word would not be.

These messages are not, moreover, available post-boot as far as I can tell. So routinely checking through them at a slower pace is not an option either.

Of course, evidence of failure will show up elsewhere. But I take it the option to see these messages is intended to provide useful functionality. That particular functionality would be virtually useless to me if I couldn't easily distinguish success from failure *at a glance* i.e. without having to read the text before it disappears.

Also whether it is in a language understood by the user I'm not sure. systemd ignores the locale on my machine even for dates which most stuff manages to pick up whatever else is missing. No idea what it is like at supporting other languages.


CLI Paste | How To Ask Questions

Arch Linux | x86_64 | GPT | EFI boot | refind | stub loader | systemd | LVM2 on LUKS
Lenovo x270 | Intel(R) Core(TM) i5-7200U CPU @ 2.50GHz | Intel Wireless 8265/8275 | US keyboard w/ Euro | 512G NVMe INTEL SSDPEKKF512G7L

Offline

#28 2012-09-21 23:38:44

Stebalien
Member
Registered: 2010-04-27
Posts: 1,237
Website

Re: SystemD ugly boot

cfr wrote:

These messages are not, moreover, available post-boot as far as I can tell. So routinely checking through them at a slower pace is not an option either.

journalctl
systemctl --failed
systemctl status <failed-unit>


Steven [ web : git ]
GPG:  327B 20CE 21EA 68CF A7748675 7C92 3221 5899 410C
Do not email: honeypot@stebalien.com

Offline

#29 2012-09-22 01:23:54

cfr
Member
From: Cymru
Registered: 2011-11-27
Posts: 7,130

Re: SystemD ugly boot

None of those give a record of the output produced during boot, though, do they? As I said, I realise that the information is available but it isn't available in that form. There's no equivalent of just reading /var/log/boot or running dmesg.

I'm not saying it isn't possible to filter the information in some way but I can't find an easy way to get just that output reproduced post-boot.


CLI Paste | How To Ask Questions

Arch Linux | x86_64 | GPT | EFI boot | refind | stub loader | systemd | LVM2 on LUKS
Lenovo x270 | Intel(R) Core(TM) i5-7200U CPU @ 2.50GHz | Intel Wireless 8265/8275 | US keyboard w/ Euro | 512G NVMe INTEL SSDPEKKF512G7L

Offline

#30 2012-09-22 01:35:54

lucke
Member
From: Poland
Registered: 2004-11-30
Posts: 4,018

Re: SystemD ugly boot

"systemctl status" shows whatever the service output, which is also in the journal. I don't know if the boot screen in systemd shows any useful information (or only FAILED), but in the times of initscripts and no /var/log/boot, one had to stop the first vc from clearing to see what was going wrong during boot. That's why I like systemd - quick "systemctl status" shows me what I want to know about the service that failed, I don't even have to see boot messages (with "quiet", I see only fsck output; oh, serenity).

Offline

#31 2012-09-22 01:41:07

WonderWoofy
Member
From: Los Gatos, CA
Registered: 2012-05-19
Posts: 8,414

Re: SystemD ugly boot

Yeah, I think the idea behind "systemctl" "systemctl status" etc. are that the boot info is so rediculously fast that it requires that relevent info be contained somewhere else.  So running a simple systemctl will give you an overview of all started services and systemctl status will tell you about specific services.

Offline

#32 2012-09-22 18:34:56

Haikarainen
Member
Registered: 2012-09-04
Posts: 93

Re: SystemD ugly boot

I actually thought systemd had a better boot than initscripts. The colors and layout looks way less smudged, you can easier get a grasp on whats going on with systemd. Although I dont like to see a lot of uneccessary messages flash before my eyes when I boot so I use the quiet flag smile

Offline

#33 2012-09-24 13:20:29

alphaniner
Member
From: Ancapistan
Registered: 2010-07-12
Posts: 2,810

Re: SystemD ugly boot

cfr wrote:

The messages systemd posts scroll by *fast*.

Fair point.  Anyway, my perspective is probably skewed.  My system is set up to retain the messages, and with KMV very few of them scroll off the screen.


But whether the Constitution really be one thing, or another, this much is certain - that it has either authorized such a government as we have had, or has been powerless to prevent it. In either case, it is unfit to exist.
-Lysander Spooner

Offline

#34 2012-09-25 02:57:33

alexanderthegre
Member
Registered: 2012-07-29
Posts: 66

Re: SystemD ugly boot

Dang... The colors are in a header file... I'm not going to be able to change that without major modifications to the code. Or a better knowledge of C. Probably both. sad

Offline

#35 2012-09-25 03:17:11

2ManyDogs
Forum Moderator
Registered: 2012-01-15
Posts: 4,642

Re: SystemD ugly boot

alexanderthegre wrote:

Dang... The colors are in a header file... I'm not going to be able to change that without major modifications to the code. Or a better knowledge of C. Probably both. sad

I don't understand this. Did you see earlier in this thread where flyingsheep located the colors in the code, and the discussion with Trilby that followed? Constants in a header file should be the easiest thing to change. You don't have to modify the code, just change the header file and rebuild.

Last edited by 2ManyDogs (2012-09-25 03:19:30)


How to post. A sincere effort to use modest and proper language and grammar is a sign of respect toward the community.

Offline

#36 2012-09-25 04:42:43

alexanderthegre
Member
Registered: 2012-07-29
Posts: 66

Re: SystemD ugly boot

2ManyDogs wrote:
alexanderthegre wrote:

Dang... The colors are in a header file... I'm not going to be able to change that without major modifications to the code. Or a better knowledge of C. Probably both. sad

I don't understand this. Did you see earlier in this thread where flyingsheep located the colors in the code, and the discussion with Trilby that followed? Constants in a header file should be the easiest thing to change. You don't have to modify the code, just change the header file and rebuild.

Oops, I probably should have phrased that better. Anyway, what I was trying to say was that I couldn't, (read: don't know how to) make it read from a file within a header such that the user could, edit a file in, say, /etc/systemd-colors.conf and change the colors there.

Offline

#37 2012-09-25 07:56:48

thoho180192
Member
Registered: 2012-05-08
Posts: 57

Re: SystemD ugly boot

alexanderthegre wrote:
2ManyDogs wrote:
alexanderthegre wrote:

Dang... The colors are in a header file... I'm not going to be able to change that without major modifications to the code. Or a better knowledge of C. Probably both. sad

I don't understand this. Did you see earlier in this thread where flyingsheep located the colors in the code, and the discussion with Trilby that followed? Constants in a header file should be the easiest thing to change. You don't have to modify the code, just change the header file and rebuild.

Oops, I probably should have phrased that better. Anyway, what I was trying to say was that I couldn't, (read: don't know how to) make it read from a file within a header such that the user could, edit a file in, say, /etc/systemd-colors.conf and change the colors there.

If it has to be rebuildt, it probably will not work to just have an option in a settings file in etc, right? Then something has to be done with the sourcecode so that systemd actually will read that file.

Last edited by thoho180192 (2012-09-25 07:58:51)

Offline

#38 2012-09-25 10:26:05

WorMzy
Forum Moderator
From: Scotland
Registered: 2010-06-16
Posts: 11,750
Website

Re: SystemD ugly boot

I did create a patch for this some time ago.

https://bugs.archlinux.org/task/31245

I assume it still works. *uses quiet*

You guys might want to "me too" the upstream bug report: https://bugs.freedesktop.org/show_bug.cgi?id=53937


Sakura:-
Mobo: MSI MAG X570S TORPEDO MAX // Processor: AMD Ryzen 9 5950X @4.9GHz // GFX: AMD Radeon RX 5700 XT // RAM: 32GB (4x 8GB) Corsair DDR4 (@ 3000MHz) // Storage: 1x 3TB HDD, 6x 1TB SSD, 2x 120GB SSD, 1x 275GB M2 SSD

Making lemonade from lemons since 2015.

Offline

#39 2012-09-25 13:05:56

thoho180192
Member
Registered: 2012-05-08
Posts: 57

Re: SystemD ugly boot

I think that will work. It is the same code fragment that people in this thread have used earlier. But it will not make systemd read a settings file? Not as I can interpret atleast. It would be nice to actually let the user choose in a settings file which color to use with verbose boot.

Offline

#40 2012-09-25 14:34:00

WorMzy
Forum Moderator
From: Scotland
Registered: 2010-06-16
Posts: 11,750
Website

Re: SystemD ugly boot

Yeah, the patch is just to change the green messages to blue. You'll need to make noise upstream if you want it to be configurable.


Sakura:-
Mobo: MSI MAG X570S TORPEDO MAX // Processor: AMD Ryzen 9 5950X @4.9GHz // GFX: AMD Radeon RX 5700 XT // RAM: 32GB (4x 8GB) Corsair DDR4 (@ 3000MHz) // Storage: 1x 3TB HDD, 6x 1TB SSD, 2x 120GB SSD, 1x 275GB M2 SSD

Making lemonade from lemons since 2015.

Offline

#41 2012-10-14 09:21:14

hellfire[bg]
Member
Registered: 2012-06-08
Posts: 31

Re: SystemD ugly boot

I have no problems with how it looks. However there's something wierd - around the time the desktop manager (lightdm in my case) service starts the log messages are cleared and started again. I mean that all the messages that have appeared until now are cleared and only new messages are shown (like for example how the desktop manager and other services are started). It's not much of a deal but it's kind of annoying.

Last edited by hellfire[bg] (2012-10-14 09:54:12)

Offline

#42 2012-10-17 19:10:26

nixpunk
Member
Registered: 2009-11-23
Posts: 271

Re: SystemD ugly boot

I just hate that the console looks like a mess when suspending (text diagonally strewn across the screen).

Offline

#43 2012-10-17 19:11:40

Unia
Member
From: Stockholm, Sweden
Registered: 2010-03-30
Posts: 2,486
Website

Re: SystemD ugly boot

nixpunk wrote:

I just hate that the console looks like a mess when suspending (text diagonally strewn across the screen).

Agreed, where does that come from?


If you can't sit by a cozy fire with your code in hand enjoying its simplicity and clarity, it needs more work. --Carlos Torres

Offline

#44 2012-10-17 21:50:18

WonderWoofy
Member
From: Los Gatos, CA
Registered: 2012-05-19
Posts: 8,414

Re: SystemD ugly boot

nixpunk wrote:

I just hate that the console looks like a mess when suspending (text diagonally strewn across the screen).

It does?  I definitely don't get this...

Offline

#45 2012-10-20 11:35:51

Šaran
Member
From: Bosnia
Registered: 2011-09-03
Posts: 407

Re: SystemD ugly boot

When suspending to ram, all I see is black screen with cursor line, but when waking up, boot messages about started daemons are shown diagonally.

EDIT: Same thing happens after restarting X.

Last edited by Šaran (2012-10-20 12:01:36)

Offline

#46 2012-10-20 12:08:22

Unia
Member
From: Stockholm, Sweden
Registered: 2010-03-30
Posts: 2,486
Website

Re: SystemD ugly boot

Šaran wrote:

When suspending to ram, all I see is black screen with cursor line, but when waking up, boot messages about started daemons are shown diagonally.

EDIT: Same thing happens after restarting X.

Exactly that, yea. Doesn't bother me too much, but what could be the cause?


If you can't sit by a cozy fire with your code in hand enjoying its simplicity and clarity, it needs more work. --Carlos Torres

Offline

#47 2012-10-20 13:53:22

fsckd
Forum Fellow
Registered: 2009-06-15
Posts: 4,173

Re: SystemD ugly boot

Šaran, that is worth starting a new thread for. I am sure I saw someone else complain of something similar.


aur S & M :: forum rules :: Community Ethos
Resources for Women, POC, LGBT*, and allies

Offline

#48 2012-10-20 15:34:45

Šaran
Member
From: Bosnia
Registered: 2011-09-03
Posts: 407

Re: SystemD ugly boot

Well I have solved it.

I was using this service file (found somewhere around here) to autologin and start X on boot:

[Unit]
Description=Direct login to X
After=systemd-user-sessions.service

[Service]
ExecStart=/bin/su fuchs -l -c "/bin/bash --login -c xinit"

[Install]
WantedBy=graphical.target

Now I'm autologing via

ExecStart=-/sbin/agetty --noclear -a USERNAME %I 38400

in getty@tty1 service and starting X on boot with

[[ $(fgconsole 2>/dev/null) == 1 ]] && startx -- vt1

in .profile as described in wiki.

Artefacts are gone and session is properly registered for authorization with polkit from testing. Only downside is it's not possible to log out on tty1. Even boot time is almost a second faster.

Last edited by Šaran (2012-10-20 15:35:53)

Offline

Board footer

Powered by FluxBB