Author Topic: VSIDO systemd and upstart discussion  (Read 91757 times)

PackRat

  • Posts: 2285
    • View Profile
    • PackRat configs on github
Re: VSIDO systemd and upstart discussion
« Reply #165 on: September 03, 2014, 03:28:46 PM »
^ No kidding.

Then there is this

link might be a double post; lost track of this thread.

I for one, am really waiting to see what path Slackware takes.

Edit - click on the "No Seriously" link in item 2 of the above link and enjoy the read - priceless.
I am tired of talk that comes to nothing.
-- Chief Joseph

...the sun, the darkness, the winds are all listening to what we have to say.
-- Geronimo

zbreaker

  • Posts: 240
    • View Profile
Re: VSIDO systemd and upstart discussion
« Reply #166 on: September 03, 2014, 03:38:25 PM »
As a slacker I can tell you alternative answers are being pursued.

http://www.linuxquestions.org/questions/slackware-14/systemd-216-a-4175515577/

PackRat

  • Posts: 2285
    • View Profile
    • PackRat configs on github
Re: VSIDO systemd and upstart discussion
« Reply #167 on: September 03, 2014, 05:39:27 PM »
Nice thread, thanks.

There is an interesting post in there that Void linux has switched from systemd to runit as the init system.
I am tired of talk that comes to nothing.
-- Chief Joseph

...the sun, the darkness, the winds are all listening to what we have to say.
-- Geronimo

VastOne

  •      v-ger
  • Posts: 4225
    • View Profile
    • VSIDO Community
Re: VSIDO systemd and upstart discussion
« Reply #168 on: September 03, 2014, 06:22:28 PM »
I am going to test converting a build from systemd back to sysVinit and then implement runit and see how it performs

Thanks for that Rat Man
VSIDO      VSIDO FB       

    I dev VSIDO

zbreaker

  • Posts: 240
    • View Profile
Re: VSIDO systemd and upstart discussion
« Reply #169 on: September 03, 2014, 08:17:06 PM »
I've got a few spare partitions and will be more than glad to test drive  8)

VastOne

  •      v-ger
  • Posts: 4225
    • View Profile
    • VSIDO Community
Re: VSIDO systemd and upstart discussion
« Reply #170 on: September 04, 2014, 01:40:51 AM »
^ I will most definitely hold you to that offer ...  8)

Once done I will post a link for downloads for anyone wanting to test

I appreciate the offer  ;D
VSIDO      VSIDO FB       

    I dev VSIDO

PackRat

  • Posts: 2285
    • View Profile
    • PackRat configs on github
Re: VSIDO systemd and upstart discussion
« Reply #171 on: September 04, 2014, 12:27:13 PM »
Interesting read here - http://www.infoworld.com/d/data-center/choose-your-side-the-linux-divide-248950

and the comments don't have all the drama.
I am tired of talk that comes to nothing.
-- Chief Joseph

...the sun, the darkness, the winds are all listening to what we have to say.
-- Geronimo

VastOne

  •      v-ger
  • Posts: 4225
    • View Profile
    • VSIDO Community
Re: VSIDO systemd and upstart discussion
« Reply #172 on: September 04, 2014, 12:57:23 PM »
^ Thanks for that PackRat...

I have several issues with that whole article...

1. Where the hell has he been?  Same old tired arguments again and again... I mean, he said NOTHING new

2. if enough established and learned folks disagree with the change, then perhaps it bears further inspection before going to production. Clearly, that hasn't happened with systemd. ... WHAT?  Was that whole conversation that the debian selection committee had a figment of our imaginations?

3.  Is debian to be taken so lightly?  Whether you like them or not, they have been doing this as long as anyone, over 20 years now

4, DAFUQ is he saying that sysVinit is not broke and needs no fix?  This is an old and stupid argument, it is clearly lacking and no longer developed as such

5. Comparing any data or design to 1984, 1994 or 2004 is ridiculous...  Things change, it is a fundamental reason why Linux was created to begin with.  It is why the kernel changes every month or so, it is who we are

I give no shits on any of it any more.  It is like a bunch of children who have never been taught how to resolve differences and only know how to argue or fight.  I see the same things on most forums and comments I read and it scares me how uninformed we are now and still feel obligated to demand OUR insights are THE ONLY WAY...

I blame it all on the cowardice of anonymity... 

Lets get in a room and discuss it and resolve it, not hope it breaks and sit back and say.. 'na na na na na, I told you it was shit and was going to break' as we stroke our fucking neckbeards and yell at our mothers to turn down her fucking soap operas so we can design something the world will cherish

I'll throw away my pedestal now
VSIDO      VSIDO FB       

    I dev VSIDO

statmonkey

  • Posts: 330
    • View Profile
Re: VSIDO systemd and upstart discussion
« Reply #173 on: September 04, 2014, 10:34:40 PM »
FWIW. Clearly have some catching up to do.  Imagine my surprise after all this time to see that this discussion is still alive, amazing Thailand Debian as they say.  I have been away on other projects and will get up to speed before putting my foot into it.

jeffreyC

  • Posts: 59
    • View Profile
Re: VSIDO systemd and upstart discussion
« Reply #174 on: September 04, 2014, 11:30:14 PM »
My impression is that systemd would get a lot further if they told Kay S. to take a hike. He seems to have made a vocation of rubbing people the wrong way.

jedi

  • DEFCON 1
  • Posts: 1639
  • Not all those who wander are lost.
    • View Profile
    • NixNut.com
Re: VSIDO systemd and upstart discussion
« Reply #175 on: September 05, 2014, 01:35:15 AM »
I too can not believe we are still having this discussion.  Most major distro's have adopted systemd and there hasn't been an end-of-the-world crisis because of it.  I'm just going to post here how I, as a daily, longtime Linux user regard the issue.  Over 2 years with VSIDO.  A year of VSIDO (appx) using systemd.  VastOne and I, and maybe one other fella gave it a serious run before settling on it.  It wasn't a split second decision on VastOne's part, and I don't believe it has been a split second decision coming from any of the other 'major' distro's.

Feel free to correct me if I'm wrong here, but, you have Arch, OpenSUSE, Fedora/Redhat, Debian, Ubuntu, etc all deciding to use systemd.

Old guard Unix versus new guard Linux?  Huh?  If something makes my system work better, tbh, I don't care about the politics or the "philosophy" (more rant on that below!) of "why it is better".  It is a better solution than sysVinit, and we all know it whether we want to admit it is so, or not.  That said, sparking heated discussion isn't the "only" thing systemd does well.

Quote
Fundamental changes in the structure of most Linux distributions should not be met with such fervent opposition. It indicates that no matter how reasonable a change may seem, if enough established and learned folks disagree with the change, then perhaps it bears further inspection before going to production. Clearly, that hasn't happened with systemd.

HUH?  For the last year and a half, AT LEAST, systemd has borne "further inspection".  Further inspection by some of the best minds in the Linux community.

Holding Linux up in the air, and claiming "hey this isn't following the Unix philosophy" is to me, stupid.  It ain't Unix, it's Linux.  I don't run a single program from 1998 that depends on anything from Unix.  Do any of you?

No, sysVinit was not broken, but as technology evolves, so does the software that is written that drives it.  sysVinit while not broken, definitely had used up it's lifespan and then some.  I'm not a developer.  I'm not a programmer.  I can barely work my way through a lua script.  With the growth of Linux in the last 10 years, (if I'm not mistaken Linux is the fastest growing OS in the world today) it only makes sense to keep making it better.  sysVinit was at the end of it's practicality and usability while trying to keep up with newer, faster and ever more evolving technology.  In one fell swoop, systemd came charging in.  And you know what?  It works.  It works really well.  It is going to work for a long time to come.  Does anyone really believe that sysVinit was going to be around in 2024?  Gimme a break.

The very word 'philosophy' when ascribed to Linux, Unix, or anything else 'technology' related is asinine.  "Philosophy is the study of general and fundamental problems, such as those connected with reality, existence, knowledge, values, reason, mind, and language."  (definition grabbed from Google) Understand that?  In other words, an ever evolving view of how something is, was, or comes to be.  Or, better yet, and even simpler way to put it;  Oh f*^k!  I changed my mind!

Linus Torvalds created the Linux Kernel.  He didn't create a "UNIX KERNEL".  This silly argument of trying to pretend that if you never used Unix, you have no business doing any kind of dev work for Linux is utter stupidity IMO.  This is not an argument about systemd anymore.  It is an argument about who has the biggest EGO.
 
systemd has been implemented on an "official" basis for less than 6 mos now.  (I'm speaking of Debian making that decision.  Other big name distro's already had)  Not every distro is using it.  Not every distro will use it.  The fact that there is still an argument going on over systemd is an overwhelming case AGAINST Linux altogether.  It is doing nothing other than stir unease in someone new to Linux or considering Linux.  This is not a "distro war", or Linux trying to replace Unix.  (or Windows and Mac for that matter) It is a case of a LOT of dev's creating a new init system, and because of ego's, attitude, and prejudice, stamping their foot and saying something like "fine, I'm taking my toys and going home".  Get the F*^K home already! (Kay)

In the end, it comes down to numbers.  Plain and simple.  Do the math.  What system/program works/lasts the longest?  The one with the most dev's/maintainers behind it.  Guess which init system has the most dev's/maintainers working on it daily.  It ain't Upstart, OpenRC, runit, Epoch, initng, eINIT, etc etc etc...  It is, hands down, systemd.

It is time for the old neckbeards to buy a razor and come out of the cave.  There is a better world out here!

Quote from: VastOne
I blame it all on the cowardice of anonymity...

Priceless ^^

And now, I'll hand VastOne's pedestal back to him...
Forum Netiquette

"No matter how smart you are you can never convince someone stupid that they are stupid."  Anonymous

jedi

  • DEFCON 1
  • Posts: 1639
  • Not all those who wander are lost.
    • View Profile
    • NixNut.com
Re: VSIDO systemd and upstart discussion
« Reply #176 on: September 05, 2014, 01:36:40 AM »
Holy crap!  That should count for like 7 posts!  Sorry, didn't mean to bloviate and end up writing a novel...
Forum Netiquette

"No matter how smart you are you can never convince someone stupid that they are stupid."  Anonymous

zbreaker

  • Posts: 240
    • View Profile
Re: VSIDO systemd and upstart discussion
« Reply #177 on: September 05, 2014, 01:48:11 AM »
Well put, Jedi, quite well put!



jedi

  • DEFCON 1
  • Posts: 1639
  • Not all those who wander are lost.
    • View Profile
    • NixNut.com
Re: VSIDO systemd and upstart discussion
« Reply #178 on: September 05, 2014, 02:54:01 AM »
As stated at the beginning of my post, I wrote that from the perspective of an end-user running Linux on a modern laptop.  Perhaps as a sys-admin, you know, the guy running the 'server-farms' or the company network, systemd might be being harangued and argued about from a security standpoint.  Or because it runs on PID1.  For every single argument about systemd and the "server environment", there is an equal and just as valid explanation.  This is not the 'dumbing down' of Linux, or trying to turn Linux into Windows.  This is moving forward, and allowing Linux to mature into a viable and real alternative to Unix.  (or any other OS, server or otherwise)  I don't think anyone is seeing it from that perspective.  Unix has been around almost 50 years, and Linux almost 30.  This is all about choice, and being able to rely on an already reliable framework.  In the end, one of the other init systems I mention in the post above may become the "Big Dog on The Porch".  Until then, the arguing that is going on in the community is alienating a lot of folks who just might have chosen Linux, but then saw all of the distasteful discourse happening as a result of systemd.  You all do know, you CAN choose whichever init system you want.  systemd is not a vast chasm that cannot ever be traversed.  If you don't like it, then don't use it.  Isn't that why we all use Linux in the first place?  Unbelievable...
Forum Netiquette

"No matter how smart you are you can never convince someone stupid that they are stupid."  Anonymous

statmonkey

  • Posts: 330
    • View Profile
Re: VSIDO systemd and upstart discussion
« Reply #179 on: September 05, 2014, 03:02:23 AM »
And now a word from our contrarian sponsor.  No offense to anyone intended... this is a discussion.

My impression is that systemd would get a lot further if they told Kay S. to take a hike. He seems to have made a vocation of rubbing people the wrong way.

Yes, definitely rubs people the wrong way but that is shadows and fog.  The issues most central to the discussions core are the (for lack of a better way to put it) "black box" design approach used in putting together systemd.  The developers feel that there are some things that should be out of users control and even their view.

There are few people on this site that I respect and enjoy more than Jedi and it is with some trepidation that I mention the following. In most things he says here my esteemed friend is correct, to the average user and even the power user systemd is great, it just works and indeed it works well. The problem lies deeper, it rests on that fact that systemd closes doors for the developers and coders. To use an analogy it is like having a house where every night the food just comes to your table. It's good and it's just there. Great for most people but if you want to go in the kitchen the door is locked and there is someone saying "you don't need to go in there, tell me what you want and it will just be there". Some of us want to go in that room, some of us want sorbet made with a spoon and not a blender, etc. or something we may want some day that today we don't even realize. We want to be able to have the option of going into that room and painting it green or whatever. I personally don't like someone telling me I can't look in the kitchen in my house, it's why I came to Linux in the first place.  Now, I would say that is a philosophical difference, my friend tells me that is stupid and he is probably right.  I am sure he is right, I am stupid and antiquarian and pedantic, etc. but I yam what I yam.

I have been down this road before, someone telling me "no, you just don't get it, it's safer for the majority of users to not see this stuff.  We can't really explain what we have done but you don't need to make alterations to how certain calls are made and the information and tools we give you are all you need".  This is a very slippery slope, first it's system calls and then it is I/O interfaces and then ... I have seen this movie before and so have many others who are objecting to what is included in systemd and what is available to developers for interfaces and calls.  As The Who once said "the new boss is the old boss, don't get fooled again".

I have no choice but to accept systemd at least in the interim but I am pretty sure that as I get back to doing things on a box again there will be a time where I run up against the locked door to the kitchen I am going to get pretty po'd. I also remain pretty sure that greater minds than mine are looking at this as well. It might be that I personally go back to sysv or some alternative.  It would not be the first time, after all  I still use a cli mail system, dmenu instead of a gui launcher, fluxbox instead of gnome etc. etc.  Because I have more control and the ability to do goofy things with it should I want to.  There are plenty of people who have told me this or that is "better" and I don't really need to see all that stuff that is going on behind the curtain.

I too can not believe we are still having this discussion.  Most major distro's have adopted systemd and there hasn't been an end-of-the-world crisis because of it.  I'm just going to post here how I, as a daily, longtime Linux user regard the issue.  Over 2 years with VSIDO.  A year of VSIDO (appx) using systemd.  VastOne and I, and maybe one other fella gave it a serious run before settling on it.  It wasn't a split second decision on VastOne's part, and I don't believe it has been a split second decision coming from any of the other 'major' distro's.

Feel free to correct me if I'm wrong here, but, you have Arch, OpenSUSE, Fedora/Redhat, Debian, Ubuntu, etc all deciding to use systemd.

Not a very compelling argument for me.  Everyone of those distro's have lots of crap in them.  I don't want unity for example or rpm or ...


HUH?  For the last year and a half, AT LEAST, systemd has borne "further inspection".  Further inspection by some of the best minds in the Linux community.

Holding Linux up in the air, and claiming "hey this isn't following the Unix philosophy" is to me, stupid.  It ain't Unix, it's Linux.  I don't run a single program from 1998 that depends on anything from Unix.  Do any of you?

Firstly, if it had borne further inspection this discussion would be over. It's not and that should be a signal that there is something deeper here.

Secondly, I have no idea but I would bet bash is older than that and probably pine and ... probably many things I use but I don't use them because they are old.  I use them because they work, fit my needs and allow me the control that I want to do creative things with them.  I can use them to create new things and often do.

No, sysVinit was not broken, but as technology evolves, so does the software that is written that drives it.  sysVinit while not broken, definitely had used up it's lifespan and then some.  I'm not a developer.  I'm not a programmer.  I can barely work my way through a lua script.  With the growth of Linux in the last 10 years, (if I'm not mistaken Linux is the fastest growing OS in the world today) it only makes sense to keep making it better.  sysVinit was at the end of it's practicality and usability while trying to keep up with newer, faster and ever more evolving technology.  In one fell swoop, systemd came charging in.  And you know what?  It works.  It works really well.  It is going to work for a long time to come.  Does anyone really believe that sysVinit was going to be around in 2024?  Gimme a break.

I am not aware of anything in sysVinit being broken either while I am aware of it's limitations and the work-arounds becoming harder and harder I am not aware of any urgent compelling reason that it could not continue to be used efficiently and without the user knowing much different. I would guess it might well be around in 2024 in some form or other. 

The very word 'philosophy' when ascribed to Linux, Unix, or anything else 'technology' related is asinine.  "Philosophy is the study of general and fundamental problems, such as those connected with reality, existence, knowledge, values, reason, mind, and language."  (definition grabbed from Google) Understand that?  In other words, an ever evolving view of how something is, was, or comes to be.  Or, better yet, and even simpler way to put it;  Oh f*^k!  I changed my mind!

asinine that was the word I was looking for above :) I would point out "connected with reality, existence, knowledge, values, reason, mind and language" and say that your definition of philosophy might be the same but our definitions of reality, existence, reason might be different and that we for sure have different values, knowledge and minds.

Linus Torvalds created the Linux Kernel.  He didn't create a "UNIX KERNEL".  This silly argument of trying to pretend that if you never used Unix, you have no business doing any kind of dev work for Linux is utter stupidity IMO.  This is not an argument about systemd anymore.  It is an argument about who has the biggest EGO.

Oh come now this is semantics at best and a conceptual deceit that is beneath you sir. What he called the kernel was not the point, where he got the kernel is and he got it from Unix.  Linus has always been very upfront about stating the importance of maintaining that "asinine" Unix philosophy and he is probably not someone you want to cite in defense of either systemd or Kay. LOL.
 
systemd has been implemented on an "official" basis for less than 6 mos now.  (I'm speaking of Debian making that decision.  Other big name distro's already had)  Not every distro is using it.  Not every distro will use it.  The fact that there is still an argument going on over systemd is an overwhelming case AGAINST Linux altogether.  It is doing nothing other than stir unease in someone new to Linux or considering Linux.  This is not a "distro war", or Linux trying to replace Unix.  (or Windows and Mac for that matter) It is a case of a LOT of dev's creating a new init system, and because of ego's, attitude, and prejudice, stamping their foot and saying something like "fine, I'm taking my toys and going home".  Get the F*^K home already! (Kay)

Could not disagree with this blurb any more.  I think this discussion is a fine testament to the strength of the Linux community.  I am one of those looney's who think discussion and discourse are healthy and that debate furthers knowledge and understanding.  I have no dog in this fight other than some pet projects that won't be moving forward. Yes I understand I have a choice etc. but if I am ripping Debian apart again to get what I want then that is a bit of a painful choice.  That again is the problem here there are parts to systemd that are outside of the core.

Remember when Vsido started part of it was to take the core of Debian and build up from it what you wanted but Debian was at the core.  Now you are making it so that the core of Debian is more than just that core, it means you have to take this and that and the other thing.  This is an important difference to some of us.

I think it is fine for Kay to say "Get the F*^K home already!"  but it's the equivalent of someone burning your house down and saying here take this fine house that I built for you and if you don't like it go the F" home.  But then you have no home anymore because the b3i#ard burnt your house down.  I am not saying I am taking my toys and going home but there are certain projects that I participated in that I am not going to redevelop to work with what I perceive is a very limited set of tools, it violates the intent and integrity of what I was doing and frankly my reason for doing it.  What I did is open and free for all to do as they wish there are no limitations can Kay say the same?

In the end, it comes down to numbers.  Plain and simple.  Do the math.  What system/program works/lasts the longest?  The one with the most dev's/maintainers behind it.  Guess which init system has the most dev's/maintainers working on it daily.  It ain't Upstart, or OpenRC etc etc etc...  It is, hands down, systemd.

Oh my, the get on the bandwagon, everybody is doing it argument.  Really Jed?  Really?  You were apparently at your Germans bombing Pearl Harbor section of your rant.  :D Um, I can think of very few worthwhile projects that were adopted by the masses immediately.  Easier does not mean better, I won't bore any of you who might take the time to read this with the examples.  Just take a few minutes and consider things that were the end all be all, everybody was developing for ...

...

...

...

and we are back.  We would all be surfing with Netscape right now for example.


It is time for the old neckbeards to buy a razor and come out of the cave.  There is a better world out here!

Quote from: VastOne
I blame it all on the cowardice of anonymity...

Priceless ^^

And now, I'll hand VastOne's pedestal back to him...

Hmm.  I am a double edge guy myself.  Can't grow a beard to save my life.  Jedi I hope you take this with the respect that is intended (with a little playfulness) and you consider that there might be some other ways to see  this.  Again, I will use systemd for the interim (mainly because I have little time or inclination to do differently) but I am extremely wary of this path and will be watching closely.  If things turn as I fear or if something I need gets screwed up over it I reserve the right to change my mind (philosophically of course)