Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > alt.os.linux > #52637 > unrolled thread

This is what good old systemd has done to me...

Back to article view | Back to alt.os.linux

Contents

  This is what good old systemd has done to me... Ike <Ike@google.com> - 2019-01-13 15:23 -0600
    Re: This is what good old systemd has done to me... William Unruh <unruh@invalid.ca> - 2019-01-13 22:39 +0000
      Re: This is what good old systemd has done to me... "Ike" <Ike@google.com> - 2019-01-13 18:28 -0600
        Re: This is what good old systemd has done to me... "Carlos E.R." <robin_listas@es.invalid> - 2019-01-14 02:08 +0100
          Re: This is what good old systemd has done to me... "Tech Systems" <querry@tci.com> - 2019-01-14 06:52 -0600
            Re: This is what good old systemd has done to me... "Carlos E.R." <robin_listas@es.invalid> - 2019-01-14 15:50 +0100
            Re: This is what good old systemd has done to me... Richard Kettlewell <invalid@invalid.invalid> - 2019-01-14 16:09 +0000
              Re: This is what good old systemd has done to me... Yrrah <Yrrah-aol@aol.invalid> - 2019-01-14 19:35 +0100
            Re: This is what good old systemd has done to me... Peter Köhlmann <peter-koehlmann@t-online.de> - 2019-01-15 00:05 +0100
          Re: This is what good old systemd has done to me... Bud Frede <frede@mouse-potato.com> - 2019-01-19 08:33 -0500
      Re: This is what good old systemd has done to me... Yrrah <Yrrah-aol@aol.invalid> - 2019-01-14 14:00 +0100
        Re: This is what good old systemd has done to me... "Tech Systems" <querry@tci.com> - 2019-01-14 07:07 -0600
          Re: This is what good old systemd has done to me... "Carlos E.R." <robin_listas@es.invalid> - 2019-01-14 15:53 +0100
            Re: This is what good old systemd has done to me... "Tech Systems" <querry@tci.com> - 2019-01-14 09:53 -0600
            Re: This is what good old systemd has done to me... Yrrah <Yrrah-aol@aol.invalid> - 2019-01-14 18:37 +0100
              Re: This is what good old systemd has done to me... "Carlos E.R." <robin_listas@es.invalid> - 2019-01-14 19:53 +0100
        Re: This is what good old systemd has done to me... Aragorn <thorongil@telenet.be> - 2019-01-14 13:08 +0000
          Re: This is what good old systemd has done to me... Johnny <johnny@invalid.net> - 2019-01-14 08:35 -0600
        Re: This is what good old systemd has done to me... "Carlos E.R." <robin_listas@es.invalid> - 2019-01-14 15:53 +0100
    Re: This is what good old systemd has done to me... Dan Espen <dan1espen@gmail.com> - 2019-01-13 17:46 -0500
      Re: This is what good old systemd has done to me... "Ike" <Ike@google.com> - 2019-01-13 18:30 -0600
        Re: This is what good old systemd has done to me... Peter Köhlmann <peter-koehlmann@t-online.de> - 2019-01-15 00:03 +0100
    Re: This is what good old systemd has done to me... "Carlos E.R." <robin_listas@es.invalid> - 2019-01-13 23:50 +0100
    Re: This is what good old systemd has done to me... "F. Russell" <fr@random.info> - 2019-01-14 01:02 +0000
      Re: This is what good old systemd has done to me... bad sector <forgetski@postit_INVALID_.gov> - 2019-01-13 20:14 -0500
      Re: This is what good old systemd has done to me... Roger Blake <rogblake@iname.invalid> - 2019-01-14 01:21 +0000
      Re: This is what good old systemd has done to me... "J.O. Aho" <user@example.net> - 2019-01-14 07:05 +0100
        Re: This is what good old systemd has done to me... "Carlos E.R." <robin_listas@es.invalid> - 2019-01-14 13:25 +0100
          Re: This is what good old systemd has done to me... "J.O. Aho" <user@example.net> - 2019-01-14 20:00 +0100
            Re: This is what good old systemd has done to me... "Carlos E.R." <robin_listas@es.invalid> - 2019-01-14 20:39 +0100
              Re: This is what good old systemd has done to me... "J.O. Aho" <user@example.net> - 2019-01-14 21:15 +0100
      Re: This is what good old systemd has done to me... Tech Systems <querry@tci.com> - 2019-01-14 07:00 -0600
      Re: This is what good old systemd has done to me... Eli the Bearded <*@eli.users.panix.com> - 2019-01-14 19:51 +0000
      Re: This is what good old systemd has done to me... queequeg@trust.no1 (Queequeg) - 2019-01-18 11:42 +0000
        Re: This is what good old systemd has done to me... Richard Kettlewell <invalid@invalid.invalid> - 2019-01-18 11:59 +0000
          Re: This is what good old systemd has done to me... Eric Pozharski <whynot@pozharski.name> - 2019-01-20 18:04 +0200
        Re: This is what good old systemd has done to me... Aragorn <thorongil@telenet.be> - 2019-01-18 14:24 +0000
          Re: This is what good old systemd has done to me... "Cybe R. Wizard" <cybe_r_wizard@WizardsTower.invalid> - 2019-01-18 09:23 -0600
            Re: This is what good old systemd has done to me... Aragorn <thorongil@telenet.be> - 2019-01-18 15:26 +0000
          Re: This is what good old systemd has done to me... Melzzzzz <Melzzzzz@zzzzz.com> - 2019-01-18 19:14 +0000
            Re: This is what good old systemd has done to me... "J.O. Aho" <user@example.net> - 2019-01-19 10:07 +0100
        Re: This is what good old systemd has done to me... Melzzzzz <Melzzzzz@zzzzz.com> - 2019-01-18 19:16 +0000
          Re: This is what good old systemd has done to me... William Unruh <unruh@invalid.ca> - 2019-01-18 23:02 +0000
            Re: This is what good old systemd has done to me... "Carlos E.R." <robin_listas@es.invalid> - 2019-01-19 05:43 +0100
              Re: This is what good old systemd has done to me... William Unruh <unruh@invalid.ca> - 2019-01-19 04:56 +0000
                Re: This is what good old systemd has done to me... "Carlos E.R." <robin_listas@es.invalid> - 2019-01-19 06:07 +0100
              Re: This is what good old systemd has done to me... "J.O. Aho" <user@example.net> - 2019-01-19 10:24 +0100
                Re: This is what good old systemd has done to me... "Carlos E.R." <robin_listas@es.invalid> - 2019-01-19 14:39 +0100
                  Re: This is what good old systemd has done to me... "J.O. Aho" <user@example.net> - 2019-01-19 20:57 +0100
          Re: This is what good old systemd has done to me... "J.O. Aho" <user@example.net> - 2019-01-19 10:08 +0100

Page 3 of 3 — ← Prev page 1 2 [3]

#52779

On 1/18/19 8:14 PM, Melzzzzz wrote:
> On 2019-01-18, Aragorn <thorongil@telenet.be> wrote:
>> On Fri, 18 Jan 2019 11:42:42 +0000, Queequeg scribbled:
>>
>>> In comp.os.linux.misc F. Russell <fr@random.info> wrote:
>>>
>>>> For a personal workstation systemd is a TOTALLY UNNECESSARY and massive
>>>> layer of pure junk.  Anyone stupid enough to accept it will reject all
>>>> counterarguments for its usefulness.
>>>
>>> I'm still wondering what was the reason for replacing sysvinit with
>>> systemd in default Debian installations despite users' protests.
>>
>> For most part, the fact that Debian wishes to continue supporting GNOME,
>> which has more or less always been the default Debian desktop
>> environment.
>>
>> systemd, GNOME/GDM and GTK+ are all developed at RedHat, and Lennart
>> Poettering, the developer of systemd, asked the GNOME/GDM developers to
>> make systemd into a hard dependency of GNOME/GDM.  So without systemd --
>> and most specifically, logind, which is part of the systemd suite -- you
>> cannot have GNOME 3 on your system anymore.  Given that systemd is tied
>> to the Linux kernel, this also implies that GNOME 3 cannot be used on any
>> other UNIX variants than GNU/Linux anymore.
> 
> Not true.
> https://www.freebsd.org/gnome/

There been projects like systembsd and elogind to come around the hard 
dependency to systemd when you want to install gnome3 on systems with a 
proper multi os supporting init system, so even GNU/Linux systems can 
run gnome3, not just systemd/linux.

-- 

  //Aho

[toc] | [prev] | [next] | [standalone]

#52771

On 2019-01-18, Queequeg <queequeg@trust.no1> wrote:
> In comp.os.linux.misc F. Russell <fr@random.info> wrote:
>
>> For a personal workstation systemd is a TOTALLY UNNECESSARY and
>> massive layer of pure junk.  Anyone stupid enough to accept it
>> will reject all counterarguments for its usefulness.
>
> I'm still wondering what was the reason for replacing sysvinit with 
> systemd in default Debian installations despite users' protests.

Because systemd is better...
>


-- 
press any key to continue or any other to quit...

[toc] | [prev] | [next] | [standalone]

#52773

On 2019-01-18, Melzzzzz <Melzzzzz@zzzzz.com> wrote:
> On 2019-01-18, Queequeg <queequeg@trust.no1> wrote:
>> In comp.os.linux.misc F. Russell <fr@random.info> wrote:
>>
>>> For a personal workstation systemd is a TOTALLY UNNECESSARY and
>>> massive layer of pure junk.  Anyone stupid enough to accept it
>>> will reject all counterarguments for its usefulness.
>>
>> I'm still wondering what was the reason for replacing sysvinit with 
>> systemd in default Debian installations despite users' protests.
>
> Because systemd is better...

Of course the discussion is over what the definition of "better" is. 
Fron what I have read there are various advantages and disadvantages of
systemd. Faster boot because of parallel booting, easier to make the booting of subsystem A depend
on the state of B,C,D. 
logging is incomprehensible binary database file which requires a
separate program to read. systemd boot programs binary programs which
are hard to either understand or change. Dependencies hard to
understand. 

So, better or worse may well depend on one's point of view.

>>
>
>

[toc] | [prev] | [next] | [standalone]

#52775

On 19/01/2019 00.02, William Unruh wrote:
> On 2019-01-18, Melzzzzz <Melzzzzz@zzzzz.com> wrote:
>> On 2019-01-18, Queequeg <queequeg@trust.no1> wrote:
>>> In comp.os.linux.misc F. Russell <fr@random.info> wrote:
>>>
>>>> For a personal workstation systemd is a TOTALLY UNNECESSARY and
>>>> massive layer of pure junk.  Anyone stupid enough to accept it
>>>> will reject all counterarguments for its usefulness.
>>>
>>> I'm still wondering what was the reason for replacing sysvinit with 
>>> systemd in default Debian installations despite users' protests.
>>
>> Because systemd is better...
> 
> Of course the discussion is over what the definition of "better" is. 
> Fron what I have read there are various advantages and disadvantages of
> systemd. Faster boot because of parallel booting,

openSUSE had parallel booting with initd. Clear dependencies.

> easier to make the booting of subsystem A depend
> on the state of B,C,D. 
> logging is incomprehensible binary database file which requires a
> separate program to read.

You can use traditional log files if you wish, or dump the journal to
text later.

> systemd boot programs binary programs which
> are hard to either understand or change. Dependencies hard to
> understand. 

Services are defined by simple configuration files easy to change. No
need to debug boot scripts.

> So, better or worse may well depend on one's point of view.

Yes.

-- 
Cheers, Carlos.

[toc] | [prev] | [next] | [standalone]

#52776

On 2019-01-19, Carlos E.R. <robin_listas@es.invalid> wrote:
> On 19/01/2019 00.02, William Unruh wrote:
>> On 2019-01-18, Melzzzzz <Melzzzzz@zzzzz.com> wrote:
>>> On 2019-01-18, Queequeg <queequeg@trust.no1> wrote:
>>>> In comp.os.linux.misc F. Russell <fr@random.info> wrote:
>>>>
>>>>> For a personal workstation systemd is a TOTALLY UNNECESSARY and
>>>>> massive layer of pure junk.  Anyone stupid enough to accept it
>>>>> will reject all counterarguments for its usefulness.
>>>>
>>>> I'm still wondering what was the reason for replacing sysvinit with 
>>>> systemd in default Debian installations despite users' protests.
>>>
>>> Because systemd is better...
>> 
>> Of course the discussion is over what the definition of "better" is. 
>> Fron what I have read there are various advantages and disadvantages of
>> systemd. Faster boot because of parallel booting,
>
> openSUSE had parallel booting with initd. Clear dependencies.
>
>> easier to make the booting of subsystem A depend
>> on the state of B,C,D. 
>> logging is incomprehensible binary database file which requires a
>> separate program to read.
>
> You can use traditional log files if you wish, or dump the journal to
> text later.

Sure as long as you have a  TB disk just for booting. (Yes, I exagerate
a little bit but to have two parallel log systems is not exactly
efficient.)

>
>> systemd boot programs binary programs which
>> are hard to either understand or change. Dependencies hard to
>> understand. 
>
> Services are defined by simple configuration files easy to change. No
> need to debug boot scripts.

Well, simple may also be in the eyes of the beholder, and when the boot
file does not do what you want it to, configuration files are almost
never enough. Or do you really believe Pottinger thought of all
possibilites?

>
>> So, better or worse may well depend on one's point of view.
>
> Yes.
>

[toc] | [prev] | [next] | [standalone]

#52777

On 19/01/2019 05.56, William Unruh wrote:
> On 2019-01-19, Carlos E.R. <robin_listas@es.invalid> wrote:
>> On 19/01/2019 00.02, William Unruh wrote:
>>> On 2019-01-18, Melzzzzz <Melzzzzz@zzzzz.com> wrote:
>>>> On 2019-01-18, Queequeg <queequeg@trust.no1> wrote:
>>>>> In comp.os.linux.misc F. Russell <fr@random.info> wrote:
>>>>>
>>>>>> For a personal workstation systemd is a TOTALLY UNNECESSARY and
>>>>>> massive layer of pure junk.  Anyone stupid enough to accept it
>>>>>> will reject all counterarguments for its usefulness.
>>>>>
>>>>> I'm still wondering what was the reason for replacing sysvinit with 
>>>>> systemd in default Debian installations despite users' protests.
>>>>
>>>> Because systemd is better...
>>>
>>> Of course the discussion is over what the definition of "better" is. 
>>> Fron what I have read there are various advantages and disadvantages of
>>> systemd. Faster boot because of parallel booting,
>>
>> openSUSE had parallel booting with initd. Clear dependencies.
>>
>>> easier to make the booting of subsystem A depend
>>> on the state of B,C,D. 
>>> logging is incomprehensible binary database file which requires a
>>> separate program to read.
>>
>> You can use traditional log files if you wish, or dump the journal to
>> text later.
> 
> Sure as long as you have a  TB disk just for booting. (Yes, I exagerate
> a little bit but to have two parallel log systems is not exactly
> efficient.)

Then disable the journal. I do sometimes.
If not, you can limit the size of the journal, or limit it to the
current boot. Plenty of alternatives.

> 
>>
>>> systemd boot programs binary programs which
>>> are hard to either understand or change. Dependencies hard to
>>> understand. 
>>
>> Services are defined by simple configuration files easy to change. No
>> need to debug boot scripts.
> 
> Well, simple may also be in the eyes of the beholder, and when the boot
> file does not do what you want it to, configuration files are almost
> never enough. Or do you really believe Pottinger thought of all
> possibilites?

I did not say that I understand all those files ;-)

> 
>>
>>> So, better or worse may well depend on one's point of view.
>>
>> Yes.
>>


-- 
Cheers, Carlos.

[toc] | [prev] | [next] | [standalone]

#52781

On 1/19/19 5:43 AM, Carlos E.R. wrote:
> On 19/01/2019 00.02, William Unruh wrote:
>> On 2019-01-18, Melzzzzz <Melzzzzz@zzzzz.com> wrote:
>>> On 2019-01-18, Queequeg <queequeg@trust.no1> wrote:
>>>> In comp.os.linux.misc F. Russell <fr@random.info> wrote:
>>>>
>>>>> For a personal workstation systemd is a TOTALLY UNNECESSARY and
>>>>> massive layer of pure junk.  Anyone stupid enough to accept it
>>>>> will reject all counterarguments for its usefulness.
>>>>
>>>> I'm still wondering what was the reason for replacing sysvinit with
>>>> systemd in default Debian installations despite users' protests.
>>>
>>> Because systemd is better...
>>
>> Of course the discussion is over what the definition of "better" is.
>> Fron what I have read there are various advantages and disadvantages of
>> systemd. Faster boot because of parallel booting,
> 
> openSUSE had parallel booting with initd. Clear dependencies.

Init systems which supports parallel start up of services (some predates 
systemd): sysvinit, openrc, upstart, smf*, launched*, finit, runit, and 
BSD rc.d*

* no Linux version.

The list is in no way complete.

>> easier to make the booting of subsystem A depend
>> on the state of B,C,D.
>> logging is incomprehensible binary database file which requires a
>> separate program to read.
> 
> You can use traditional log files if you wish, or dump the journal to
> text later.
> 
>> systemd boot programs binary programs which
>> are hard to either understand or change. Dependencies hard to
>> understand.
> 
> Services are defined by simple configuration files easy to change. No
> need to debug boot scripts.

One issue with the "configuration file" is that in some parts it poorly 
documented and what you find may not be up to date.
Many of the "configuration files" do call on a script which does all the 
things the old init script did due of limitations what you can do in 
systemd.


The main reasons I do not like systemd is that it keeps on eating up 
other project and that it's full of security issues (will just be more 
and more for each thing incorporated into it) and assumptions that 
things went well when something goes badly (initialization of some 
hardware is checked, if things fails, it's assumed the hardware in 
question is working).

-- 

  //Aho

[toc] | [prev] | [next] | [standalone]

#52783

On 19/01/2019 10.24, J.O. Aho wrote:
> On 1/19/19 5:43 AM, Carlos E.R. wrote:
>> On 19/01/2019 00.02, William Unruh wrote:
>>> On 2019-01-18, Melzzzzz <Melzzzzz@zzzzz.com> wrote:
>>>> On 2019-01-18, Queequeg <queequeg@trust.no1> wrote:
>>>>> In comp.os.linux.misc F. Russell <fr@random.info> wrote:

...

>>> easier to make the booting of subsystem A depend
>>> on the state of B,C,D.
>>> logging is incomprehensible binary database file which requires a
>>> separate program to read.
>>
>> You can use traditional log files if you wish, or dump the journal to
>> text later.
>>
>>> systemd boot programs binary programs which
>>> are hard to either understand or change. Dependencies hard to
>>> understand.
>>
>> Services are defined by simple configuration files easy to change. No
>> need to debug boot scripts.
> 
> One issue with the "configuration file" is that in some parts it poorly
> documented and what you find may not be up to date.
> Many of the "configuration files" do call on a script which does all the
> things the old init script did due of limitations what you can do in
> systemd.
> 
> 
> The main reasons I do not like systemd is that it keeps on eating up
> other project and that it's full of security issues (will just be more
> and more for each thing incorporated into it) and assumptions that
> things went well when something goes badly (initialization of some
> hardware is checked, if things fails, it's assumed the hardware in
> question is working).
> 

I don't like it, but not enough to blame it of "my email is not working"
with no proof, or to jump distro, when all my computers are working fine.

And those services that I needed to change to my liking I managed to.

-- 
Cheers, Carlos.

[toc] | [prev] | [next] | [standalone]

#52787

On 1/19/19 2:39 PM, Carlos E.R. wrote:
> On 19/01/2019 10.24, J.O. Aho wrote:

>> The main reasons I do not like systemd is that it keeps on eating up
>> other project and that it's full of security issues (will just be more
>> and more for each thing incorporated into it) and assumptions that
>> things went well when something goes badly (initialization of some
>> hardware is checked, if things fails, it's assumed the hardware in
>> question is working).
>>
> 
> I don't like it, but not enough to blame it of "my email is not working"
> with no proof,

As noted before, the claim is from a microsoft windows user who don't 
use Linux and that subject been left behind a long time ago in this part 
of the thread,

> or to jump distro, when all my computers are working fine.

Kind of the same excuse all current microsoft windows xp users uses.


-- 

  //Aho

[toc] | [prev] | [next] | [standalone]

#52780

On 1/18/19 8:16 PM, Melzzzzz wrote:
> On 2019-01-18, Queequeg <queequeg@trust.no1> wrote:
>> In comp.os.linux.misc F. Russell <fr@random.info> wrote:
>>
>>> For a personal workstation systemd is a TOTALLY UNNECESSARY and
>>> massive layer of pure junk.  Anyone stupid enough to accept it
>>> will reject all counterarguments for its usefulness.
>>
>> I'm still wondering what was the reason for replacing sysvinit with
>> systemd in default Debian installations despite users' protests.
> 
> Because systemd is better...

Yes, in having severe security issues, just the last couple of months 
there been at least 4 discovered that can make your system pawned.

-- 

  //Aho

[toc] | [prev] | [standalone]

Page 3 of 3 — ← Prev page 1 2 [3]

Back to top | Article view | alt.os.linux

csiph-web