Scheduled Maintenance: We are aware of an issue with Google, AOL, and Yahoo services as email providers which are blocking new registrations. We are trying to fix the issue and we have several internal and external support tickets in process to resolve the issue. Please see: viewtopic.php?t=158230

 

 

 

What should we do about systemd?

Here you can discuss every aspect of Debian. Note: not for support requests!

What should we do about systemd?

Poll ended at 2014-11-26 08:34

a) give up Debian to use another distribution which respects the *nix tradition
21
24%
b) concentrate on systemd's fork (uselessd) to port it to Debian
10
11%
c) use sysvinit (INIT) irrespective of its limitations with respect to modern software requirements
14
16%
d) use another initialisation system like runit
5
6%
e) accept systemd and continue using Debian
37
43%
 
Total votes: 87

Message
Author
User avatar
keithpeter
Posts: 502
Joined: 2009-06-14 08:06
Location: 5230n 0155w

Re: What should we do about systemd?

#161 Post by keithpeter »

tomazzi wrote:And I think, that with systemd, RedHat is going to "take the leadership" of linux again, by enforcing it's own solution. Since all the systemd developers are paid RedHat employees, then they can take over the service market (in all of the aspects).
All source code available. Oracle can afford to (and do) pay bright people, and historically that is exactly what they have done.

Wider issues: I suspect there will still be Debian in 10 years. Could be 'branding' and development model for a Linux Common Core, could be something a bit different. Time will tell.

Food for thought: What is the median age of Debian developers? How has that statistic changed over the last 5 years? Could the recent turbulence in Debian's internal governance be a generational thing?

mmix
Posts: 205
Joined: 2012-07-14 00:08

Re: What should we do about systemd?

#162 Post by mmix »

google now uses clang as their production compiler for chrome linux builds
http://www.phoronix.com/scan.php?page=n ... px=MTg0MTk

i am afraid that redhat mimic clang as yet another syst**d
and hype something new groundbreaking compiler. :P

because google did it.. heh

User avatar
fleabus
Posts: 98
Joined: 2013-08-28 18:36
Location: Winchester, VA, USA

Re: What should we do about systemd?

#163 Post by fleabus »

keithpeter wrote:Food for thought: What is the median age of Debian developers? How has that statistic changed over the last 5 years? Could the recent turbulence in Debian's internal governance be a generational thing?
Pushing 60 and I have a lot of thoughts that run in that direction these days.

The world belongs to the young; 'Twas ever thus. They know they need only wait a bit, and we'll be gone.
Each generation we older folks, along with our knowledge and views, become irrelevant.
We had our time. It's their time now.
Last edited by fleabus on 2014-11-24 23:07, edited 2 times in total.
Lenovo Legion 5-15IMH05 32GB Ram, 2TB Samsung 970 EVO Plus NVMe SSD

fruitofloom
Posts: 183
Joined: 2014-10-27 21:28

Re: What should we do about systemd?

#164 Post by fruitofloom »

twoflowers wrote:That was to be expected. Well, good bye Debian
Pretty much that.
Give me convenience or give me death.

User avatar
Linadian
Posts: 490
Joined: 2013-12-20 15:25
Location: In a systemd free distro

Re: What should we do about systemd?

#165 Post by Linadian »

fruitofloom wrote:
twoflowers wrote:That was to be expected. Well, good bye Debian
Pretty much that.
This is a very dark day indeed. I see no point in keeping Wheezy on my computer since it has no viable future.
Linux Registered User 533946

User avatar
keithpeter
Posts: 502
Joined: 2009-06-14 08:06
Location: 5230n 0155w

Re: What should we do about systemd?

#166 Post by keithpeter »

Linadian wrote:
fruitofloom wrote:
twoflowers wrote:That was to be expected. Well, good bye Debian
Pretty much that.
This is a very dark day indeed. I see no point in keeping Wheezy on my computer since it has no viable future.
Steady on chaps.

You have a year of Wheezy at least probably 18 months. The Refracta crew are brewing their strange brews. There was a person on Debian-Users a few weeks ago asking for help with an Etch install. One hopes (s)he was sitting the PC behind a decent hardware firewall. If the systemd decliners can muster sufficient resources, there will be a viable 'light' sysvinit desktop under Jessie.

That all buys time. Alternatives will emerge.

User avatar
dasein
Posts: 7680
Joined: 2011-03-04 01:06
Location: Terra Incantationum

Re: What should we do about systemd?

#167 Post by dasein »

keithpeter wrote:Food for thought: What is the median age of Debian developers? How has that statistic changed over the last 5 years? Could the recent turbulence in Debian's internal governance be a generational thing?
Even if the median age hasn't changed, I daresay this is almost certainly a generational thing, at least in part.

It is a luxury of youth to imagine that one is markedly smarter than one's predecessors. Every time I hear some young turk talk about the need to completely redo something from scratch, I cringe. Moreso when it's something crufty and old like *nix init. Because if I learned anything in the 15 years I wrote code for a living, it's that every one of those kludges, every one of those "ugly" patches, every one of those incomprehensible-seeming workarounds represents a solution to a problem that the original designers failed to anticipate.

There is a generation's worth of expertise and wisdom embedded in that crufty old code: throwing it out and starting over is like lobotomizing oneself. Because it ignores the lessons of the past, the newer, shinier gizmo has no meaningful hope of being any better than whatever it's replacing. The "best case" scenario is that shiny new bugs will replace crufty old bugs. (Although in the case of systemd, there's the additional layer of problems inherent in any monolithic middleware design.)

(Side note: Oracle Linux is RedHat.)

I have to say that I'm disappointed but not at all surprised by the voting results. As soon as I saw the "no GR needed" amendment offered by the current DPL, I recognized it as a subtle way of providing "political cover" for everyone who wanted to avoid the hard choice.

I'm saddened. Haven't quite figured out yet exactly what I'll do, but to my mind, "Lennax" is really no different from Windows, much less any better.

One last thought. My posting history makes it clear that I've been an active systemd hater for several years now. But keithpeter is quite right. The impotent rage, the endless name-calling, the unnecessary melodrama and empty invective that permeated every systemd thread on this board over the last month ultimately served only to weaken the folks who had (and no doubt still have) legitimate technical reasons for mistrusting systemd. (Just sayin')

User avatar
fleabus
Posts: 98
Joined: 2013-08-28 18:36
Location: Winchester, VA, USA

Re: What should we do about systemd?

#168 Post by fleabus »

dasein wrote:Haven't quite figured out yet exactly what I'll do, but to my mind, "Lennax" is really no different from Windows, much less any better.
I really wish they'd created their own thing, their "CoreOS" if that's what they wanted. I wish they'd done it from the ground up and left Linux alone.
I don't know what I'll ultimately do either.
Last edited by fleabus on 2014-11-25 02:53, edited 7 times in total.
Lenovo Legion 5-15IMH05 32GB Ram, 2TB Samsung 970 EVO Plus NVMe SSD

mmix
Posts: 205
Joined: 2012-07-14 00:08

Re: What should we do about systemd?

#169 Post by mmix »

Ian Jackson Resigns From The Debian Technical Committee
http://www.phoronix.com/scan.php?page=n ... px=MTg0NDA

if you want to avoid systemd distro, look here
http://forums.debian.net/viewtopic.php?f=3&t=118319

KBD47
Posts: 87
Joined: 2011-09-04 09:07

Re: What should we do about systemd?

#170 Post by KBD47 »

mmix wrote:Ian Jackson Resigns From The Debian Technical Committee
http://www.phoronix.com/scan.php?page=n ... px=MTg0NDA

if you want to avoid systemd distro, look here
http://forums.debian.net/viewtopic.php?f=3&t=118319
These guys are dropping like flies. Is anyone left in charge of Debian?

User avatar
buntunub
Posts: 591
Joined: 2011-02-11 05:23

Re: What should we do about systemd?

#171 Post by buntunub »

KBD47 wrote:
mmix wrote:Ian Jackson Resigns From The Debian Technical Committee
http://www.phoronix.com/scan.php?page=n ... px=MTg0NDA

if you want to avoid systemd distro, look here
http://forums.debian.net/viewtopic.php?f=3&t=118319
These guys are dropping like flies. Is anyone left in charge of Debian?
I think we are seeing the exodus of all those who value freedom of choice, since there is now none in Debian. Perhaps we will see if that fork takes shape, and those Devs who are leaving should have a refuge there.

User avatar
golinux
Posts: 1579
Joined: 2010-12-09 00:56
Location: not a 'buntard!
Been thanked: 1 time

Re: What should we do about systemd?

#172 Post by golinux »

buntunub wrote:I think we are seeing the exodus of all those who value freedom of choice, since there is now none in Debian. Perhaps we will see if that fork takes shape, and those Devs who are leaving should have a refuge there.
Several of the resignations I've read stated that they were leaving the TC but continuing to work on Debian related projects.
May the FORK be with you!

tomazzi
Posts: 730
Joined: 2013-08-02 21:33

Re: What should we do about systemd?

#173 Post by tomazzi »

keithpeter wrote:
tomazzi wrote:And I think, that with systemd, RedHat is going to "take the leadership" of linux again, by enforcing it's own solution. Since all the systemd developers are paid RedHat employees, then they can take over the service market (in all of the aspects).
All source code available. Oracle can afford to (and do) pay bright people, and historically that is exactly what they have done.
You don't understand what's going on...
Yes, in case (and until) the OS is build with simple and stable "bricks", then it's sufficient to just hire a programmer who knows C and has sufficient knowledge of GNU/Linux.

But now we're dealing with never-yet-frozen complex software, which has thousants of non-trivial inter-process dependancies, which additonally can change "just-like-that" in the next version (breaking another packages/rules "just-like-that") , tests are simple and limitted, not to say: dumb and insufficient.
f.e. systemd DNS daemon is totally insecure piece of crap - everyone knows that, and there are thousants of links to pages, where detailed explanation of this fact can be found...

However RedHat have managed to force this crap on Linux - and nobody knows what will change in systemd tommorow - nobody can provide paid support for it, unless someone would like to risk loosing customers or even getting sued.

Currently, noone besides RedHat can provide paid systemd support. Besides, even if systemd will eventually get frozen, then sill corporations will prefer to have support directly from the authors/developers...

got it?
Odi profanum vulgus

User avatar
edbarx
Posts: 5401
Joined: 2007-07-18 06:19
Location: 35° 50 N, 14 º 35 E
Been thanked: 2 times

Re: What should we do about systemd?

#174 Post by edbarx »

On debianfork chat, someone said that he was using his own OS initialisation daemon, which only runs a script at system startup. This looks like an irresistible temptation, what do you expect from a daemon, eh? :twisted:

The procedure is as easy as to write a simple daemon in C/C++ that calls a script at startup. Let me begin. This will be yet another adventure for me... :)

I found this is a daemon's skeleton. How come, daemons are spirits, if I recall correctly?

Code: Select all

#include <sys/types.h>
#include <sys/stat.h>
#include <stdio.h>
#include <stdlib.h>
#include <fcntl.h>
#include <errno.h>
#include <unistd.h>
#include <syslog.h>
#include <string.h>

int main(void) {
        
        /* Our process ID and Session ID */
        pid_t pid, sid;
        
        /* Fork off the parent process */
        pid = fork();
        if (pid < 0) {
                exit(EXIT_FAILURE);
        }
        /* If we got a good PID, then
           we can exit the parent process. */
        if (pid > 0) {
                exit(EXIT_SUCCESS);
        }

        /* Change the file mode mask */
        umask(0);
                
        /* Open any logs here */        
                
        /* Create a new SID for the child process */
        sid = setsid();
        if (sid < 0) {
                /* Log the failure */
                exit(EXIT_FAILURE);
        }
        

        
        /* Change the current working directory */
        if ((chdir("/")) < 0) {
                /* Log the failure */
                exit(EXIT_FAILURE);
        }
        
        /* Close out the standard file descriptors */
        close(STDIN_FILENO);
        close(STDOUT_FILENO);
        close(STDERR_FILENO);
        
        /* Daemon-specific initialization goes here */
        
        /* The Big Loop */
        while (1) {
           /* Do some task here ... */
           
           sleep(30); /* wait 30 seconds */
        }
   exit(EXIT_SUCCESS);
Added Later 1:
I am still stuck at trying to figure out how zombie processes are cleared. After that, it will be testing time. This is the transformed daemon code, although it needs more scrutiny before it will(will it?) spring into life.

Code: Select all

#include <sys/types.h>
#include <sys/wait.h>
#include <sys/stat.h>
#include <stdio.h>
#include <stdlib.h>
#include <fcntl.h>
#include <errno.h>
#include <unistd.h>
#include <syslog.h>
#include <string.h>

int main(void) {
       
        // Our process ID and Session ID 
        pid_t pid, sid;
       
        // Fork off the parent process 
        pid = fork();
        if (pid < 0) return -1; //failure
	
	
        // If we got a good PID, then we can exit the parent process.
        if (pid > 0) return 0;

        // Change the file mode mask
        umask(0);
               
        // Open any logs here       
               
        // Create a new SID for the child process
        sid = setsid();
        if (sid < 0) {
                // Log the failure
                return -2;
        }
       

       
        // Change the current working directory
        if ((chdir("/")) < 0) {
                // Log the failure
                return -3;
        }
       
        // Close out the standard file descriptors 
        close(STDIN_FILENO);
        close(STDOUT_FILENO);
        close(STDERR_FILENO);
       
        // Daemon-specific initialization goes here
        system("/os-loader");
        // The Infinite Loop
        while (1) {
           // Do some task here
           
           sleep(1);
        }
   return 0;
}  
Debian == { > 30, 000 packages }; Debian != systemd
The worst infection of all, is a false sense of security!
It is hard to get away from CLI tools.

Naron
Posts: 28
Joined: 2011-08-29 11:48
Has thanked: 2 times
Been thanked: 1 time

Re: What should we do about systemd?

#175 Post by Naron »

I do not understand why people hate so much systemd. There are objective (technical) arguments against systemd?
I'm a simple user, without any technical knowledge, but I feel, after I read the discussions from here, that the arguments against systemd are more emotional than technical.
I read what other people say about systemd and it does not seem so bad.
That's why I ask here.

User avatar
aicardi
Posts: 388
Joined: 2009-11-18 01:30
Location: Chicago

Re: What should we do about systemd?

#176 Post by aicardi »

Naron wrote:I do not understand why people hate so much systemd. There are objective (technical) arguments against systemd?
I'm a simple user, without any technical knowledge, but I feel, after I read the discussions from here, that the arguments against systemd are more emotional than technical.
I read what other people say about systemd and it does not seem so bad.
That's why I ask here.
tomazzi explained this very well just (2) posts prior to yours.
Jessie/Xfce

User avatar
golinux
Posts: 1579
Joined: 2010-12-09 00:56
Location: not a 'buntard!
Been thanked: 1 time

Re: What should we do about systemd?

#177 Post by golinux »

aicardi wrote:
Naron wrote:I do not understand why people hate so much systemd. There are objective (technical) arguments against systemd?
I'm a simple user, without any technical knowledge, but I feel, after I read the discussions from here, that the arguments against systemd are more emotional than technical.
I read what other people say about systemd and it does not seem so bad.
That's why I ask here.
tomazzi explained this very well just (2) posts prior to yours.
He's just trolling. Any id10t could get those answers with one simple search.
May the FORK be with you!

User avatar
buntunub
Posts: 591
Joined: 2011-02-11 05:23

Re: What should we do about systemd?

#178 Post by buntunub »

I think the real question is, WHY?!

Why this rush to get Systemd into Jessie?
Why is it nobody who voted to derail Ian's GR can explain to all of us why they did so?
Why is it the DDs feel the need to table any discussion of decoupling Systemd dependancies so that people can run other things without being forced to run Systemd, even without their knowing about it happening, as in the case with libpam?

When I started down this road right here, I asked, begged even, for some Systemd fanbois to explain to me why they are doing what they are doing to Debian. I am still waiting.

User avatar
dasein
Posts: 7680
Joined: 2011-03-04 01:06
Location: Terra Incantationum

Re: What should we do about systemd?

#179 Post by dasein »

buntunub wrote:I am still waiting.
I'm still waiting one step further back, for someone to identify a specific problem that systemd solves. No one can rewrite decades worth of code from scratch without screwing a lot of things up. So there needs to be some specific, identifiable (and, I'm sorry, utterly compelling) benefit to doing so.

(And no, design "aesthetics" do not constitute a genuine problem.)

mmix
Posts: 205
Joined: 2012-07-14 00:08

Re: What should we do about systemd?

#180 Post by mmix »

may be unification, lindows. :P
in lindows world, there will be no more debian, because it called "Lindows[TM]".

Post Reply