[arch-general] Systemd : Analysis of reactions of Users
Hi, DISCLAIMER: I support systemd but haven't switched to it yet, because I haven't had time till now, and also because I have some concerns. I like the ease-of-use that systems like PA/Systemd brings but I sincerely appreciate issues like the ones Ralf Mardorf and others have and I sincerely hope there are ways to address their problems well. I have been reading a lot about systemd discussions everywhere, on Fedora, on Arch and everywhere and I guess while the developers have been clear about why they would like to switch to systemd, the users have not been clear 1. Many times in discussions, some valid points by the user have been covered with a "resistance to change" flavor and hence, led to a rebuttal that has not addressed the valid point hidden within. 2. Also, developers work as a team while the users are scattered and hence, every user tries to make a point which is essentially similar to others and hence, gets a same rebuttal but the actual reason he posted the question was because he wanted to know something which was not clarified due to point 1. 3. However more failproof the new system might be, whenever you are shipping it on such a large scale, the guarantee that the users want is for the system to fail nicely. It doesn't matter to me if the systemd has less bugs than the init code, if I encounter that 1 bug, I want have a reasonable guarantee to get around it, and this is what is scaring most of the users I guess. So, I guess here are the points --------------------------- Apparent Simplicity --------------------------- Init scripts had text files while systemd will have binary files to load from. So, there is a point of error in converting the .conf to binary files and I guess, this worries users because, even if they do not know it, they are fearing that the text-to-binary conversion will not be proper (due to various reasons such as some user accidentally editing a file with gedit and setting different encoding and thousands of other reasons) and hence, will end up borking the system. Now I am not sure, since I have not actually used systemd, about how much this point is correct, since it depends a lot on the design of the systemd parser and how the output is, but I would like the following features for the same: 1. Systemd should overwrite the current binary files if and only if it can verify that the new files are correct. So, it can do something like regenerate the text from the binary and then check with the original text or some as the user if the information is correct. 2. Systemd should have a mechanism by which it can fallback to reading the text files and booting from it. I do not know if this feature is really useful since init scripts already do this and hence, you can make it fall back on initscripts instead, but the point is then I would then have to maintain two systems, one which I use regularly (systemd) and one which I do not, and when systemd fail, I would have to use the initscripts which I do not use regularly and hence is a big error-prone exercise. Hence, if the users can have that guarantee (which I think there is) that the systemd configuration files can be used by initscripts to boot an exact system (though slower then I guess) they will be satisfied. and hence, when people say that if systemd fails or you don't like it, just switch back to initscripts, they are not addressing the real concern of the users. ----------------------------------- Single File vs Multiple File ------------------------------------ Most of the users will access their rc.conf files once in a month or so once their system has been setup considerably. I on my personal laptop nowadays only look at rc.conf when there is a pacnew notification. For these users, till now rc.conf was the one-stop service so to say. This concept for them is similar to the one-desk customer service that banks provide where you can get almost all types of manipulations to your accounts at a single desk. This fact is very comforting. Point is at one glance, you can get all you want to know about the system. This is especially true in case of daemons, now with separate service files, you have to look in every service file to check if the files are correct. If there was a single command to print out all the boot info, then probably it would be great, but I guess this tool would be the same one that I described in point 1 in Apparent Simplicity. Then, after every edit and regenerate, I can test my files and satisfy myself that what I have typed is correct. ---------------- DAEMONS ---------------- With respect to daemons, the BEFORE and AFTER in the service files is redundant and though not likely to cause errors, likely to be inconsistent, because for every service file where a daemon "xyz" appears in AFTER, the corresponding daemon must appear in BEFORE in the service file for xyz. I am not quiet sure why this redundancy is there, you can simply have just "AFTER" variables and they should take care of all the dependencies I guess. Thanks for reading such a long mail if you have reached this far! -- Cheers Jayesh Badwaik stop html mail | always bottom-post www.asciiribbon.org | www.netmeister.org/news/learn2quote.html
On Thu, Jul 26, 2012 at 4:30 PM, Jayesh Badwaik <jayesh.badwaik90@gmail.com> wrote:
Thanks for reading such a long mail if you have reached this far!
I read it, and all I have to say is that you obviously haven't done much (or any?) reading on systemd. That should be a pre-requisite to posting a request for information, and it IS if you're an Arch (DIY) user. Just a short summary of the misconceptions there, systemd does not 'replace text files with binary equivalents', it is not a piece-by-piece replacement (which invalidates quite a few of the hopeful suggestions you have there). Actually, re-reading that, I'm not sure you understand too much about how initscripts work (and what they do) either. Not that I'm an expert myself, but when you say 'booting from text files' that does give a bad impression.... those are bash scripts, to start with.
On Thu, 2012-07-26 at 16:48 +0800, Oon-Ee Ng wrote:
those are bash scripts
Exactly, but what is better when we need to use irrational cryptic text files to set up or Linux, instead of easy to understand bash scrips?
On Thu, Jul 26, 2012 at 5:17 PM, Ralf Mardorf <ralf.mardorf@alice-dsl.net> wrote:
On Thu, 2012-07-26 at 16:48 +0800, Oon-Ee Ng wrote:
those are bash scripts
Exactly, but what is better when we need to use irrational cryptic text files to set up or Linux, instead of easy to understand bash scrips?
Yeah, because key=value pairs are more complicated then, you know, a programming language? Its not like systemd even prevents you from USING bash if you feel it should be an integral part of the init system.
On Thu, 2012-07-26 at 17:22 +0800, Oon-Ee Ng wrote:
On Thu, Jul 26, 2012 at 5:17 PM, Ralf Mardorf <ralf.mardorf@alice-dsl.net> wrote:
On Thu, 2012-07-26 at 16:48 +0800, Oon-Ee Ng wrote:
those are bash scripts
Exactly, but what is better when we need to use irrational cryptic text files to set up or Linux, instead of easy to understand bash scrips?
Yeah, because key=value pairs are more complicated then, you know, a programming language? Its not like systemd even prevents you from USING bash if you feel it should be an integral part of the init system.
I'm just a user today, I'm able to program 65xx assembler and similar. However, I'm a dummy. So in the future Linux is only for experts? Regards, Ralf
On Thursday, July 26, 2012 11:31:49 AM Ralf Mardorf wrote:
On Thu, 2012-07-26 at 17:22 +0800, Oon-Ee Ng wrote:
On Thu, Jul 26, 2012 at 5:17 PM, Ralf Mardorf
<ralf.mardorf@alice-dsl.net> wrote:
On Thu, 2012-07-26 at 16:48 +0800, Oon-Ee Ng wrote:
those are bash scripts
Exactly, but what is better when we need to use irrational cryptic text files to set up or Linux, instead of easy to understand bash scrips?
Yeah, because key=value pairs are more complicated then, you know, a programming language? Its not like systemd even prevents you from USING bash if you feel it should be an integral part of the init system.
I'm just a user today, I'm able to program 65xx assembler and similar. However, I'm a dummy. So in the future Linux is only for experts?
Regards, Ralf
We may have to then move to *BSD
On Thursday, July 26, 2012 05:22:09 PM Oon-Ee Ng wrote:
On Thu, Jul 26, 2012 at 5:17 PM, Ralf Mardorf
<ralf.mardorf@alice-dsl.net> wrote:
On Thu, 2012-07-26 at 16:48 +0800, Oon-Ee Ng wrote:
those are bash scripts
Exactly, but what is better when we need to use irrational cryptic text files to set up or Linux, instead of easy to understand bash scrips?
Yeah, because key=value pairs are more complicated then, you know, a programming language? Its not like systemd even prevents you from USING bash if you feel it should be an integral part of the init system.
If one would choose to use bash with systemd then what would be the point of changing to systemd, should not one just leave well enough alone?
On Thu, Jul 26, 2012 at 2:58 PM, Baho Utot <baho-utot@columbus.rr.com> wrote:
On Thursday, July 26, 2012 05:22:09 PM Oon-Ee Ng wrote:
On Thu, Jul 26, 2012 at 5:17 PM, Ralf Mardorf
<ralf.mardorf@alice-dsl.net> wrote:
On Thu, 2012-07-26 at 16:48 +0800, Oon-Ee Ng wrote:
those are bash scripts
Exactly, but what is better when we need to use irrational cryptic text files to set up or Linux, instead of easy to understand bash scrips?
Yeah, because key=value pairs are more complicated then, you know, a programming language? Its not like systemd even prevents you from USING bash if you feel it should be an integral part of the init system.
Because in 99.9% of cases you do not need to use bash... However, just in case some people have some requirements that are not covered by systemd yet, then it is possible to fall back to bash as before. Meaning we get huge benefits in almost all cases, and are no worse than before in the remaining few.... -t
On Thu, Jul 26, 2012 at 8:00 AM, Tom Gundersen <teg@jklm.no> wrote:
On Thu, Jul 26, 2012 at 2:58 PM, Baho Utot <baho-utot@columbus.rr.com> wrote:
On Thursday, July 26, 2012 05:22:09 PM Oon-Ee Ng wrote:
On Thu, Jul 26, 2012 at 5:17 PM, Ralf Mardorf
<ralf.mardorf@alice-dsl.net> wrote:
On Thu, 2012-07-26 at 16:48 +0800, Oon-Ee Ng wrote:
those are bash scripts
Exactly, but what is better when we need to use irrational cryptic text files to set up or Linux, instead of easy to understand bash scrips?
Yeah, because key=value pairs are more complicated then, you know, a programming language? Its not like systemd even prevents you from USING bash if you feel it should be an integral part of the init system.
Because in 99.9% of cases you do not need to use bash... However, just in case some people have some requirements that are not covered by systemd yet, then it is possible to fall back to bash as before. Meaning we get huge benefits in almost all cases, and are no worse than before in the remaining few....
-t
Tom: I've decided you may as well save your breath. Everyone here is still kicking "a dead horse". I will admit to being far below the level of most Arch users and I made the transition will minimal problems, one being the network setup. The only consistency in this whole conversation is the lack of perception of what is happening, the difference between the two systems, and the benefits to all. Initially I was a naysayer, 30 minutes reading and editing changed that. Most Arch users are very linux and system savvy (I hate nerds, geeks, etc. Reminds me off the late 70's and 80's in Silicon Valley.) who quickly adapt to these types of changes. Many don't seem to grasp the concept that both systems are still supported. In my case because of my bull head and set ways, I still use the /etc/rc.d/network script for starting my network and let systemd do the rest. With the current setup one can mix and match as necessary then try to get the system improved over time. I said something about saving your breath earlier. That's so you can keep up the good work you ( and all those involved ) are doing on Arch and not go crazy trying to defend the choices that have been made to support certain software and ways of doing things. To all the rest flame if you must, but please keep the noise down. IMHO there are are enough threads about this topic already. Myra -- Life's fun when your sick and psychotic!
On Thu, 2012-07-26 at 17:21 -0500, Myra Nelson wrote:
To all the rest flame if you must
OT: I like this list, because I didn't notice "real" flame. Different levels of knowledge and being a little bit rough etc. is far away from flame.
On Thu, Jul 26, 2012 at 5:29 PM, Ralf Mardorf <ralf.mardorf@alice-dsl.net> wrote:
On Thu, 2012-07-26 at 17:21 -0500, Myra Nelson wrote:
To all the rest flame if you must
OT: I like this list, because I didn't notice "real" flame. Different levels of knowledge and being a little bit rough etc. is far away from flame.
Ralf: I agree with you whole heartedly, so far. This discussion had fairly decent. My comment was meant to let people know those are my opinions and were not directed at anyone specific. They were more for the crowd that believes a user with my level of knowledge should be using Ubuntu, Mint, etc. Myra -- Life's fun when your sick and psychotic!
On Thu, 2012-07-26 at 17:38 -0500, Myra Nelson wrote:
On Thu, Jul 26, 2012 at 5:29 PM, Ralf Mardorf <ralf.mardorf@alice-dsl.net> wrote:
On Thu, 2012-07-26 at 17:21 -0500, Myra Nelson wrote:
To all the rest flame if you must
OT: I like this list, because I didn't notice "real" flame. Different levels of knowledge and being a little bit rough etc. is far away from flame.
Ralf:
I agree with you whole heartedly, so far. This discussion had fairly decent. My comment was meant to let people know those are my opinions and were not directed at anyone specific. They were more for the crowd that believes a user with my level of knowledge should be using Ubuntu, Mint, etc.
Myra
:) Nice! Sorry, my English isn't good, I just want to clarify that we are not objective but far away from name-calling or being much to rough. I can't contribute to facts, but I like to read the threads to learn. :) Ralf
May be this is a silly question but: will there be a general announcement when systemd became officially adopted? As T. G. said in the Dev list: "If a move should happen, I suggest waiting a bit longer until more unit files have been added to our various packages. And to allow some more time to see if problems crop up." I'm as a final user would like to make the leap ony after systemd is already adopted as the new AL official init manager. Kind regards, Martin -- -msx
On Jul 27, 2012 6:21 AM, "Martin Cigorraga" <msx@archlinux.us> wrote:
May be this is a silly question but: will there be a general announcement when systemd became officially adopted?
That would be announced, yes.
As T. G. said in the Dev list: "If a move should happen, I suggest waiting a bit longer until more unit files have been added to our various packages. And to allow some more time to see if problems crop up."
I'm as a final user would like to make the leap ony after systemd is already adopted as the new AL official init manager.
Kind regards, Martin
-- -msx
May be this is a silly question but: will there be a general announcement when systemd became officially adopted?
That would be announced, yes.
As you have said the timescale for initscripts being replaced by systemd comes down to the challenges of maintenance and I guess this is a question for that day as systemd is obviously a moving target but as I would expect like ipv6 some people think systemd, event starts and parallelism and it's binary size are more mature and sane than I do (Of course without those users these things will have no chance of becoming mature, compilation bug free, improved, unix-like, secure and completely hackable in my eyes). I expect those maintenance challenges coem down to working for every package so can even an unmaintained initscripts package be kept for those of us who want to maintain just the scripts we use or do you expect openrc to be a better base to work from or do you think a switch to another tiny init binary distro to then be a better option? Thanks, Kc -- _______________________________________________________________________ 'Write programs that do one thing and do it well. Write programs to work together. Write programs to handle text streams, because that is a universal interface' (Doug McIlroy) _______________________________________________________________________
On Jul 27, 2012 3:16 PM, "Kevin Chadwick" <ma1l1ists@yahoo.co.uk> wrote:
May be this is a silly question but: will there be a general
announcement
when systemd became officially adopted?
That would be announced, yes.
As you have said the timescale for initscripts being replaced by systemd comes down to the challenges of maintenance and I guess this is a question for that day as systemd is obviously a moving target but as I would expect like ipv6 some people think systemd, event starts and parallelism and it's binary size are more mature and sane than I do (Of course without those users these things will have no chance of becoming mature, compilation bug free, improved, unix-like, secure and completely hackable in my eyes).
I expect those maintenance challenges coem down to working for every package so can even an unmaintained initscripts package be kept for those of us who want to maintain just the scripts we use or do you expect openrc to be a better base to work from or do you think a switch to another tiny init binary distro to then be a better option?
I don't see any reason for initscripts to stop working (might be that stuff like gnome will require systemd at some point, but that's our of our hands). Especially if the people who are so adamantly against systemd contribute to initscripts (by way of testing and bug reports). Oddly, that does not seem to happen much. Cheers, Tom
On Thu, Jul 26, 2012 at 6:21 PM, Myra Nelson <myra.nelson@hughes.net> wrote:
quickly adapt to these types of changes. Many don't seem to grasp the concept that both systems are still supported. In my case because of my bull head and set ways, I still use the /etc/rc.d/network script for starting my network and let systemd do the rest. With the current
I'd recommend either netcfg, or networkmanager for networking under systemd depending on your needs. They each suit me well, and are well documented on the wiki. =-Jameson
On Jul 27, 2012 3:25 PM, "Jameson" <imntreal@gmail.com> wrote:
On Thu, Jul 26, 2012 at 6:21 PM, Myra Nelson <myra.nelson@hughes.net>
wrote:
quickly adapt to these types of changes. Many don't seem to grasp the concept that both systems are still supported. In my case because of my bull head and set ways, I still use the /etc/rc.d/network script for starting my network and let systemd do the rest. With the current
I'd recommend either netcfg, or networkmanager for networking under systemd depending on your needs. They each suit me well, and are well documented on the wiki.
=-Jameson
Yet hardly no one seems to be working on those systems outside of their respective distros. Makes one wonder...
On Jul 27, 2012 3:57 PM, "Tom Gundersen" <teg@jklm.no> wrote:
On Jul 27, 2012 3:25 PM, "Jameson" <imntreal@gmail.com> wrote:
On Thu, Jul 26, 2012 at 6:21 PM, Myra Nelson <myra.nelson@hughes.net>
wrote:
quickly adapt to these types of changes. Many don't seem to grasp the concept that both systems are still supported. In my case because of my bull head and set ways, I still use the /etc/rc.d/network script for starting my network and let systemd do the rest. With the current
I'd recommend either netcfg, or networkmanager for networking under systemd depending on your needs. They each suit me well, and are well documented on the wiki.
=-Jameson
Yet hardly no one seems to be working on those systems outside of their respective distros. Makes one wonder...
Sorry, answered the wrong message. This was aimed at the alternative init systems.
On Thu, Jul 26, 2012 at 05:22:09PM +0800, Oon-Ee Ng wrote:
Yeah, because key=value pairs are more complicated then, you know, a programming language?
Apples and oranges. What you read in a bash script is what actually gets executed. And this is being done by a tool that is not specific for the task at hand, but one that is used for thousands of other things and that can therefore be considered both reliable and well-known. The key=value pairs in a service file are just the input to some ad-hoc code that remains hidden unless you care to read the systemd sources. A service file may be easier to read than a script, but it it is purely declarative and doesn't provide any hint at all as to how things really work. Which means that if they don't work, you're damned. Ciao, -- FA A world of exhaustive, reliable metadata would be an utopia. It's also a pipe-dream, founded on self-delusion, nerd hubris and hysterically inflated market opportunities. (Cory Doctorow)
On Thu, Jul 26, 2012 at 11:17 AM, Ralf Mardorf <ralf.mardorf@alice-dsl.net> wrote:
On Thu, 2012-07-26 at 16:48 +0800, Oon-Ee Ng wrote:
those are bash scripts
Exactly, but what is better when we need to use irrational cryptic text files to set up or Linux, instead of easy to understand bash scrips?
I don't understand how you can claim the systemd syntax to be cryptic. Sure, if you never saw it before you need to look up the meaning in the manpage to be certain you got it right. This is not so different from someone never seeing a bash script before, even if the syntax looks intuitive, there are lots of subtleties that you better look up. Even worse, lots of the idioms used in bash scripts are not documented in bash's monster manpage, or are at least difficult to find. The benefit of the systemd syntax is that it is much more restricted, so every possible usage is well documented. Of course, you can then complain that it might not be powerful enough (as it is not a programing language), however for these cases, we can simply fall back to using a bash script like in the old days. Note: this is almost never necessary. -t
I read it, and all I have to say is that you obviously haven't done much (or any?) reading on systemd. That should be a pre-requisite to posting a request for information, and it IS if you're an Arch (DIY) user.
I read it and there are valid points for and against that some choose to ignore and others may not care for all those features. You haven't offered any analysis. -- ________________________________________________________ Why not do something good every day and install BOINC. ________________________________________________________
On Thursday 26 Jul 2012 16:48:30 Oon-Ee Ng wrote:
I read it, and all I have to say is that you obviously haven't done much (or any?) reading on systemd. That should be a pre-requisite to posting a request for information, and it IS if you're an Arch (DIY) user.
At one point, I had read that initscripts is slow compared to systemd because it reads all the damn text files and has to parse it first, while with systemd, it is all binary. Now that I look back, I realize, the author must have been talking about initscripts and not the configuration files. My bad there, I agree. I was talking about text configuration file. I had this mistaken idea all the time that since now, we have systemd which is binary implementation, the text files are also converted to binary to reduce the time in fetching and parsing them and hence, reduce the boot times even more. I guess, the readahead implementation was what I was confused with, which is just prefetching, not binary.
Just a short summary of the misconceptions there, systemd does not 'replace text files with binary equivalents', it is not a piece-by-piece replacement (which invalidates quite a few of the hopeful suggestions you have there).
Actually, re-reading that, I'm not sure you understand too much about how initscripts work (and what they do) either. Not that I'm an expert myself, but when you say 'booting from text files' that does give a bad impression.... those are bash scripts, to start with.
I know those are bash scripts, but my above point explains it I guess. I was talkign about reading configuration "text files" vs binary files. I still believe that there should be a script/program which can output all the configurations from different file onto the terminal describing the currently configured boot process. -- Jayesh Badwaik stop html mail | always bottom-post www.asciiribbon.org | www.netmeister.org/news/learn2quote.html
I still believe that there should be a script/program which can output all the configurations from different file onto the terminal describing the currently configured boot process.
I nice perhaps ncurses gui or any config displaying binary always comes along but shouldn't be required for fast interpretation. -- ________________________________________________________ Why not do something good every day and install BOINC. ________________________________________________________
On Thu, Jul 26, 2012 at 11:53 AM, Kevin Chadwick <ma1l1ists@yahoo.co.uk> wrote:
I still believe that there should be a script/program which can output all the configurations from different file onto the terminal describing the currently configured boot process.
I nice perhaps ncurses gui or any config displaying binary always comes along but shouldn't be required for fast interpretation.
"systemctl list-unit-files" gives a very nice and quick overview. Otherwise you can use "tree /etc/systemd/system" if that is what you prefer. I don't see the point of doing that, but if that's what floats your boat, it should give you the same information without the use of a tool. -t
"systemctl list-unit-files" gives a very nice and quick overview. Otherwise you can use "tree /etc/systemd/system" if that is what you prefer. I don't see the point of doing that, but if that's what floats your boat, it should give you the same information without the use of a tool.
The point as I have said is a universal textual interface that even strings /dev/sda | grep DAEMONS could pickup without difficulty. It may not be such an important point but if I am editing configs on one system I shouldn't need to ssh to a system with systemctl or boot said system. tree: command not found but I could use find etc.. So fair enough in this context but I still believe it leaves the issue of seperating daemon executions from anything else etc.. On Arch we still have DAEMONS perhaps for a good while or forever, so that's cool. -- ________________________________________________________ Why not do something good every day and install BOINC. ________________________________________________________
On Thursday, July 26, 2012 04:48:30 PM Oon-Ee Ng wrote:
On Thu, Jul 26, 2012 at 4:30 PM, Jayesh Badwaik
<jayesh.badwaik90@gmail.com> wrote:
[putolin]
Actually, re-reading that, I'm not sure you understand too much about how initscripts work (and what they do) either. Not that I'm an expert myself, but when you say 'booting from text files' that does give a bad impression.... those are bash scripts, to start with.
To be more technically correct bash scripts are ascii text files.
On Thu, Jul 26, 2012 at 11:30 AM, Jayesh Badwaik <jayesh.badwaik90@gmail.com> wrote:
With respect to daemons, the BEFORE and AFTER in the service files is redundant and though not likely to cause errors, likely to be inconsistent, because for every service file where a daemon "xyz" appears in AFTER, the corresponding daemon must appear in BEFORE in the service file for xyz. I am not quiet sure why this redundancy is there, you can simply have just "AFTER" variables and they should take care of all the dependencies I guess.
This is certainly not true – it is enough for /one/ unit to have Before or After. -- Mantas Mikulėnas
On Thu, 2012-07-26 at 12:43 +0300, Mantas Mikulėnas wrote:
On Thu, Jul 26, 2012 at 11:30 AM, Jayesh Badwaik <jayesh.badwaik90@gmail.com> wrote:
With respect to daemons, the BEFORE and AFTER in the service files is redundant and though not likely to cause errors, likely to be inconsistent, because for every service file where a daemon "xyz" appears in AFTER, the corresponding daemon must appear in BEFORE in the service file for xyz. I am not quiet sure why this redundancy is there, you can simply have just "AFTER" variables and they should take care of all the dependencies I guess.
This is certainly not true – it is enough for /one/ unit to have Before or After.
Sorry, what kind of new logic philosophy/math do users need to learn? http://en.wikipedia.org/wiki/Logic Pardon, I only know the German Wiki, since my English is broken.
On Thu, Jul 26, 2012 at 11:52:49AM +0200, Ralf Mardorf wrote:
On Thu, 2012-07-26 at 12:43 +0300, Mantas Mikulėnas wrote:
On Thu, Jul 26, 2012 at 11:30 AM, Jayesh Badwaik <jayesh.badwaik90@gmail.com> wrote:
With respect to daemons, the BEFORE and AFTER in the service files is redundant and though not likely to cause errors, likely to be inconsistent, because for every service file where a daemon "xyz" appears in AFTER, the corresponding daemon must appear in BEFORE in the service file for xyz. I am not quiet sure why this redundancy is there, you can simply have just "AFTER" variables and they should take care of all the dependencies I guess.
This is certainly not true – it is enough for /one/ unit to have Before or After.
Sorry, what kind of new logic philosophy/math do users need to learn? http://en.wikipedia.org/wiki/Logic Pardon, I only know the German Wiki, since my English is broken.
https://en.wikipedia.org/wiki/Redundancy By the way, I, for one, am increasingly annoyed by (not only) your style of discussion. Not that it'd matter in any way, but I miss the times of productive and helpful threads on this list. HTH, Dennis -- "Den Rechtsstaat macht aus, dass Unschuldige wieder frei kommen." Dr. Wolfgang Schäuble, Bundesinnenminister (14.10.08, TAZ-Interview) 0D21BE6C - F3DC D064 BB88 5162 56BE 730F 5471 3881 0D21 BE6C
On Thu, 2012-07-26 at 11:57 +0200, Dennis Herbrich wrote:
On Thu, Jul 26, 2012 at 11:52:49AM +0200, Ralf Mardorf wrote:
On Thu, 2012-07-26 at 12:43 +0300, Mantas Mikulėnas wrote:
On Thu, Jul 26, 2012 at 11:30 AM, Jayesh Badwaik <jayesh.badwaik90@gmail.com> wrote:
With respect to daemons, the BEFORE and AFTER in the service files is redundant and though not likely to cause errors, likely to be inconsistent, because for every service file where a daemon "xyz" appears in AFTER, the corresponding daemon must appear in BEFORE in the service file for xyz. I am not quiet sure why this redundancy is there, you can simply have just "AFTER" variables and they should take care of all the dependencies I guess.
This is certainly not true – it is enough for /one/ unit to have Before or After.
Sorry, what kind of new logic philosophy/math do users need to learn? http://en.wikipedia.org/wiki/Logic Pardon, I only know the German Wiki, since my English is broken.
https://en.wikipedia.org/wiki/Redundancy
By the way, I, for one, am increasingly annoyed by (not only) your style of discussion. Not that it'd matter in any way, but I miss the times of productive and helpful threads on this list.
HTH, Dennis
I don't claim to be an expert, I already mentioned that I'm a dummy. So again: Is Linux in the future for experts only?
On 26 July 2012 12:07, Ralf Mardorf <ralf.mardorf@alice-dsl.net> wrote:
I don't claim to be an expert, I already mentioned that I'm a dummy. So again: Is Linux in the future for experts only?
Arch has always been targeted towards a competent userbase, if you're not that kind of person, there's still distros that don't require you to tinker with your system (ubuntu, fedora, suse, mint, mandriva / mageia etc...)
On Thu, 2012-07-26 at 12:31 +0200, Alexandre Ferrando wrote:
On 26 July 2012 12:07, Ralf Mardorf <ralf.mardorf@alice-dsl.net> wrote:
I don't claim to be an expert, I already mentioned that I'm a dummy. So again: Is Linux in the future for experts only?
Arch has always been targeted towards a competent userbase, if you're not that kind of person, there's still distros that don't require you to tinker with your system (ubuntu, fedora, suse, mint, mandriva / mageia etc...)
I run Ubuntu, Debian, Suse and Arch. Until now I'm able to handle this, but I don't know any German city or company using Linux anymore ... of course, there are a few exceptions, I'm aware of them. Arch is the less painful distro I know, so the recommendation to switch to another distro, especially to distros I already use is stupid. Did you ever take a look how many men and how many women are subscribed to Linux mailing lists? ... Sorry, I tend to become OT, but back to the topic. Are you and some others really interested in getting an OS for everybody, or do you enjoy to be one of the few experts? I'm not against systemd, I just wish to be able to get some explanation, in German! Thank you for the links you'll send me, that are understandable and short enough so that everybody is able to read and understand them within three hours. Regards, Ralf
I'd like to know how many "basic unskilled" users have noticed when fedora moved to systemd ? That should be a good way to inform the community about this technical switch. Best, Antoine 2012/7/26 Ralf Mardorf <ralf.mardorf@alice-dsl.net>:
On Thu, 2012-07-26 at 12:31 +0200, Alexandre Ferrando wrote:
On 26 July 2012 12:07, Ralf Mardorf <ralf.mardorf@alice-dsl.net> wrote:
I don't claim to be an expert, I already mentioned that I'm a dummy. So again: Is Linux in the future for experts only?
Arch has always been targeted towards a competent userbase, if you're not that kind of person, there's still distros that don't require you to tinker with your system (ubuntu, fedora, suse, mint, mandriva / mageia etc...)
I run Ubuntu, Debian, Suse and Arch. Until now I'm able to handle this, but I don't know any German city or company using Linux anymore ... of course, there are a few exceptions, I'm aware of them. Arch is the less painful distro I know, so the recommendation to switch to another distro, especially to distros I already use is stupid.
Did you ever take a look how many men and how many women are subscribed to Linux mailing lists? ... Sorry, I tend to become OT, but back to the topic. Are you and some others really interested in getting an OS for everybody, or do you enjoy to be one of the few experts?
I'm not against systemd, I just wish to be able to get some explanation, in German!
Thank you for the links you'll send me, that are understandable and short enough so that everybody is able to read and understand them within three hours.
Regards, Ralf
On Thu, Jul 26, 2012 at 12:59:29PM +0200, Ralf Mardorf wrote:
On Thu, 2012-07-26 at 12:31 +0200, Alexandre Ferrando wrote:
On 26 July 2012 12:07, Ralf Mardorf <ralf.mardorf@alice-dsl.net> wrote:
I don't claim to be an expert, I already mentioned that I'm a dummy. So again: Is Linux in the future for experts only?
Arch has always been targeted towards a competent userbase, if you're not that kind of person, there's still distros that don't require you to tinker with your system (ubuntu, fedora, suse, mint, mandriva / mageia etc...)
I run Ubuntu, Debian, Suse and Arch. Until now I'm able to handle this, but I don't know any German city or company using Linux anymore ... of course, there are a few exceptions, I'm aware of them. Arch is the less painful distro I know, so the recommendation to switch to another distro, especially to distros I already use is stupid.
In fact, the company I work at uses Arch Linux for it's development virtual machines. Yes, *before* I started to work here. ;) However, as I wrote earlier, I fail to see the merit in using "potential user base" as a metric for the validity of strategic discussions. OSS projects DO need to reach a critical user base to stay alive and healthy, but Arch Linux is well beyond that point, so why bother? "Arch Linux to rule them all"? Waste of time and brain cycles. Arch Linux is one of the "less painful" distributions, as you rightly put it. Let's keep it that way, along with the "early adopter" spirit.
Did you ever take a look how many men and how many women are subscribed to Linux mailing lists? ... Sorry, I tend to become OT, but back to the topic. Are you and some others really interested in getting an OS for everybody, or do you enjoy to be one of the few experts?
I find it rather rude to imply that people here are striving to maintain an elitarian status of expertism. Quite the contrary is the case, with Arch being one of the few distributions openly and explicitly sticking to what one might call a consensus of "standard" linux concepts. I hope you see the difference to blatant obfuscation. *cough*YaST anno '99*cough*. ;)
I'm not against systemd, I just wish to be able to get some explanation, in German!
Well, that shouldn't be too hard to find in the long run. It might not be available now, but if systemd has merits and becomes more widely adopted, volunteers will crop up and translate/create appropriate guides. If that's too slow for someone, they may choose to speed things up by paying for such services. "Time, Cost, Quality - Choose two." Best wishes, Dennis -- "Den Rechtsstaat macht aus, dass Unschuldige wieder frei kommen." Dr. Wolfgang Schäuble, Bundesinnenminister (14.10.08, TAZ-Interview) 0D21BE6C - F3DC D064 BB88 5162 56BE 730F 5471 3881 0D21 BE6C
On Thu, 2012-07-26 at 13:29 +0200, Dennis Herbrich wrote:
Arch Linux is one of the "less painful" distributions, as you rightly put it. Let's keep it that way, along with the "early adopter" spirit.
I promised to be quiet, forgive me for replying again, in this case I've to apologize, since Arch really is user friendly regarding to it's good Wikis. I became aware of Arch because of the Wikis. :)
On Thu, Jul 26, 2012 at 12:07:02PM +0200, Ralf Mardorf wrote:
I don't claim to be an expert, I already mentioned that I'm a dummy. So again: Is Linux in the future for experts only?
I'm at a loss what kind of answer you are expecting, nobody has any "right" or even the ability to answer such a purely political statement. If you want my opinion, for whatever reason: Yes, please. Optimize Linux for experts at the very core, and leave the subjective and political interpretation to those experts who feel compelled to create Linux variants that serve specific needs. Historically, such variants were called "distributions", and I like the concept. Now, if you were really wondering if *Arch Linux* shall be for experts only, I'd tentatively give a yes as well, as the vision for Arch has been, and still is, to provide a system with "few surprises" to the archetypical, generic linux administrator. However, the specification of this often cited archetype has been subject of discussion uncountable times, and I personally am quite tired of this, to be perfectly honest. If you feel any distribution's goals diverge too much from yours, you can and should(!) look for something better suited to your needs, or adapt the distribution. There's nothing wrong with that, and nobody in their right might will take offense. However, if the benevolent dictators of Arch decide that systemd is technically "good enough" and will be (or already is) the "way of least surprise" for seasoned linux admins, then it's a sensible decision to follow that route. Please excuse me starting to ramble. I would have made my point more concisely, but I lack the time to do so. If there's a point I'd like to make, it should be "Linux is whatever you make of it.", and this basic statement implicitly precludes "dummies", as you put it, from taking part as well as "experts". Best regards, Dennis -- "Den Rechtsstaat macht aus, dass Unschuldige wieder frei kommen." Dr. Wolfgang Schäuble, Bundesinnenminister (14.10.08, TAZ-Interview) 0D21BE6C - F3DC D064 BB88 5162 56BE 730F 5471 3881 0D21 BE6C
On Thu, 2012-07-26 at 12:52 +0200, Dennis Herbrich wrote: [snip] Fair play. So we've got Arch for experts and Ubuntu using Unity for idiots, but no Linux for averaged people?! I'm kidding! For good reasons I still recommend Ubuntu/Debian and Arch, depending to the users needs. Until now everything excepted of a few issues is ok, I just fear that this could change. My postings for this thread regarding to technical facts belong to /dev/null, but my intension is to take care of the user base. Are users wanted? Or is it just for the group of regulars? Rhetorical questions. There's much enthusiasm by all the developers, so I guess in the end they'll find the best way to go. With or without systemd. At least I can be silent now :p. Ralf
On Thursday, July 26, 2012 12:07:02 PM Ralf Mardorf wrote:
On Thu, 2012-07-26 at 11:57 +0200, Dennis Herbrich wrote:
On Thu, Jul 26, 2012 at 11:52:49AM +0200, Ralf Mardorf wrote:
On Thu, 2012-07-26 at 12:43 +0300, Mantas Mikulėnas wrote:
On Thu, Jul 26, 2012 at 11:30 AM, Jayesh Badwaik
<jayesh.badwaik90@gmail.com> wrote:
With respect to daemons, the BEFORE and AFTER in the service files is redundant and though not likely to cause errors, likely to be inconsistent, because for every service file where a daemon "xyz" appears in AFTER, the corresponding daemon must appear in BEFORE in the service file for xyz. I am not quiet sure why this redundancy is there, you can simply have just "AFTER" variables and they should take care of all the dependencies I guess.
This is certainly not true – it is enough for /one/ unit to have Before or After.
Sorry, what kind of new logic philosophy/math do users need to learn? http://en.wikipedia.org/wiki/Logic Pardon, I only know the German Wiki, since my English is broken.
https://en.wikipedia.org/wiki/Redundancy
By the way, I, for one, am increasingly annoyed by (not only) your style of discussion. Not that it'd matter in any way, but I miss the times of productive and helpful threads on this list.
HTH,
Dennis
I don't claim to be an expert, I already mentioned that I'm a dummy. So again: Is Linux in the future for experts only?
So much for world domination!
On Thu, 2012-07-26 at 11:57 +0200, Dennis Herbrich wrote:
By the way ...
... is there the need to improve something that already works, while there are other things that still don't work? Do you give a guarantee that systemd won't make things more complicated and that everything, every user used before still will work? Do you expect that every Linux user is a computer nerd? And are you aware of all usages that computer users have got? Again, do you give a guarantee that everything still will work? Pulseaudio already borked Linux for the averaged, non-expert-computer user. Please think out of the box. Are you aware how many people wish to get rid of Windows, but they keep it because less Linux-nonsense has more weight than much more available Linux-advantages? Did you ever try to establish a Linux at a school? It's nearly impossible in Germany, regarding to unsettled state. You need not to win me over, I'm using Linux. Do you take care about "normal humans", who wish to use a computer and who are completely uninterested in any problems? All my Linux do run, without systemd. I can send you a list off-list what all my Linux are not able to do. "Never change a winning team" as long as your alternative for sure does improve the computer usage for everybody.
On Thu, Jul 26, 2012 at 12:45:28PM +0200, Ralf Mardorf wrote:
"Never change a winning team" as long as your alternative for sure does improve the computer usage for everybody.
I assume you wanted to say "unless" instead of "as long as". First off, "everybody" is not Arch Linux' target demography. No distribution targets 'everybody', and that'd be close to impossible to achieve, anyway. Please just accept that Arch Linux, as one distribution among literally hundreds, targets "experienced linux users". Don't take this as an official definition, but it's a good rule of thumb to work with that renders most points you're making moot. Teachers or students, are not "experienced linux users", usually. There are distributions much better suited for these people's needs, which would be proper subjects to your argumentation, but not Arch Linux. They don't care how their system boots, but we Arch users do. That's why we've got these threads clogging the MLs right now. Arch Linux is not specifically designed to "win over" the desktop users, or the grandmas of the nation. There's not even a goal to win over *anybody* in the first place! We're not running a business here, but offer a service for a specific niche. Take it or leave it, voice your opinion, help develop a better solution for that niche. Pretty please, with sugar and cherry on top, lose the arguments "it's too complicated for X" with X!="experience linux users" and "status quo works for me". Those are simply not largely relevant when debating the inclusion of another, optional(!) init system. Laters, Denni -- "Den Rechtsstaat macht aus, dass Unschuldige wieder frei kommen." Dr. Wolfgang Schäuble, Bundesinnenminister (14.10.08, TAZ-Interview) 0D21BE6C - F3DC D064 BB88 5162 56BE 730F 5471 3881 0D21 BE6C
The 26/07/12, Ralf Mardorf wrote:
On Thu, 2012-07-26 at 11:57 +0200, Dennis Herbrich wrote:
By the way ...
... is there the need to improve something that already works
As I've already said, it does NOT work. Systems based on init scripts are BROKEN because some of them scripts won't give you any chance to catch all the failures. -- Nicolas Sebrecht
On 26/07/12 16:35, Nicolas Sebrecht wrote:
The 26/07/12, Ralf Mardorf wrote:
On Thu, 2012-07-26 at 11:57 +0200, Dennis Herbrich wrote:
By the way ... ... is there the need to improve something that already works As I've already said, it does NOT work. Systems based on init scripts are BROKEN because some of them scripts won't give you any chance to catch all the failures.
Instead of fixing such problems we need something new that's broken too?
On Fri, Jul 27, 2012 at 11:07 AM, Mike <mkgmafbt@gmail.com> wrote:
On 26/07/12 16:35, Nicolas Sebrecht wrote:
The 26/07/12, Ralf Mardorf wrote:
On Thu, 2012-07-26 at 11:57 +0200, Dennis Herbrich wrote:
By the way ...
... is there the need to improve something that already works
As I've already said, it does NOT work. Systems based on init scripts are BROKEN because some of them scripts won't give you any chance to catch all the failures.
Instead of fixing such problems we need something new that's broken too?
NEW IS ALWAYS BETTER
On Thursday, July 26, 2012 11:13:42 AM Nicholas MIller wrote:
On Fri, Jul 27, 2012 at 11:07 AM, Mike <mkgmafbt@gmail.com> wrote:
On 26/07/12 16:35, Nicolas Sebrecht wrote:
The 26/07/12, Ralf Mardorf wrote:
On Thu, 2012-07-26 at 11:57 +0200, Dennis Herbrich wrote:
By the way ...
... is there the need to improve something that already works
As I've already said, it does NOT work. Systems based on init scripts are BROKEN because some of them scripts won't give you any chance to catch all the failures.
Instead of fixing such problems we need something new that's broken too?
NEW IS ALWAYS BETTER
Then you had better throw away all the gnu tools
The 27/07/12, Mike wrote:
Instead of fixing such problems we need something new that's broken too?
You have to know that fixing was tried more than once with various approaches along time. Parallelism is one of the best-known of these failed attempts. Smart logs, correct error reporting are other examples. It's very hard to fix all the issues of init scripts. Some issues even appeared to be nearly impossible to solve. -- Nicolas Sebrecht
On 27/07/12 09:18, Nicolas Sebrecht wrote:
The 27/07/12, Mike wrote:
Instead of fixing such problems we need something new that's broken too? You have to know that fixing was tried more than once with various approaches along time. Parallelism is one of the best-known of these failed attempts. Smart logs, correct error reporting are other examples.
It's very hard to fix all the issues of init scripts. Some issues even appeared to be nearly impossible to solve.
I'm aware of that, but that doesn't mean one can't fix them. Nobody said, that the code base of sysvinit shouldn't be modified. The problem with parallelism is to make sure that all dependencies are met. That works in systemd if and only if the unit files are done right (tm), this isn't any different to, e.g. upstart
The 27/07/12, Mike wrote:
I'm aware of that, but that doesn't mean one can't fix them. Nobody said, that the code base of sysvinit shouldn't be modified.
It would have been fixed for a long time if it were easy enough. :-) -- Nicolas Sebrecht
On 27/07/12 13:57, Nicolas Sebrecht wrote:
The 27/07/12, Mike wrote:
I'm aware of that, but that doesn't mean one can't fix them. Nobody said, that the code base of sysvinit shouldn't be modified. It would have been fixed for a long time if it were easy enough. :-)
Uhm there are init systems available that are baesd on or using sysvinit or at least are trying to stay compatible (e.g. upstart), without reinventing the wheel or declare the unix philosophy obsolote.
On 27/07/2012 9:29 AM, Mike wrote:
On 27/07/12 13:57, Nicolas Sebrecht wrote:
The 27/07/12, Mike wrote:
I'm aware of that, but that doesn't mean one can't fix them. Nobody said, that the code base of sysvinit shouldn't be modified. It would have been fixed for a long time if it were easy enough. :-)
Uhm there are init systems available that are baesd on or using sysvinit or at least are trying to stay compatible (e.g. upstart), without reinventing the wheel or declare the unix philosophy obsolote.
AFAIK systemd is trying to stay backwards compatible at least in the sense that upstart is. It can parse old initscripts, and there is even a target in archlinux that will read your DAEMONS array so you can pretend you never switched. All this other stuff is just a more powerful option in systemd that people can move to as they're ready.
On 27/07/12 15:45, Stephen E. Baker wrote:
On 27/07/2012 9:29 AM, Mike wrote:
On 27/07/12 13:57, Nicolas Sebrecht wrote:
The 27/07/12, Mike wrote:
I'm aware of that, but that doesn't mean one can't fix them. Nobody said, that the code base of sysvinit shouldn't be modified. It would have been fixed for a long time if it were easy enough. :-)
Uhm there are init systems available that are baesd on or using sysvinit or at least are trying to stay compatible (e.g. upstart), without reinventing the wheel or declare the unix philosophy obsolote.
AFAIK systemd is trying to stay backwards compatible at least in the sense that upstart is. It can parse old initscripts, and there is even a target in archlinux that will read your DAEMONS array so you can pretend you never switched. All this other stuff is just a more powerful option in systemd that people can move to as they're ready. Not exactly, upstart is an relativly easy to handle sysvinit replacement, systemd is a lot more work. I don't want to pretend anything, systemd doesn't suit my needs, but I have no problems maintaining an init system on my own. I don't care if Arch would declare systemd as default today ;)
On Thursday, July 26, 2012 11:57:26 AM Dennis Herbrich wrote:
On Thu, Jul 26, 2012 at 11:52:49AM +0200, Ralf Mardorf wrote:
On Thu, 2012-07-26 at 12:43 +0300, Mantas Mikulėnas wrote:
On Thu, Jul 26, 2012 at 11:30 AM, Jayesh Badwaik
<jayesh.badwaik90@gmail.com> wrote:
[putolin]
Sorry, what kind of new logic philosophy/math do users need to learn? http://en.wikipedia.org/wiki/Logic Pardon, I only know the German Wiki, since my English is broken.
https://en.wikipedia.org/wiki/Redundancy
By the way, I, for one, am increasingly annoyed by (not only) your style of discussion. Not that it'd matter in any way, but I miss the times of productive and helpful threads on this list.
HTH, Dennis
What is wrong with Ralf? That is just his style as you have your own style. I read his posts and sometimes even learn from it.
On Thu, Jul 26, 2012 at 10:30 AM, Jayesh Badwaik <jayesh.badwaik90@gmail.com> wrote:
DISCLAIMER: I support systemd but haven't switched to it yet, because I haven't had time till now, and also because I have some concerns. I like the ease-of-use that systems like PA/Systemd brings but I sincerely appreciate issues like the ones Ralf Mardorf and others have and I sincerely hope there are ways to address their problems well.
I have been reading a lot about systemd discussions everywhere, on Fedora, on Arch and everywhere and I guess while the developers have been clear about why they would like to switch to systemd, the users have not been clear
1. Many times in discussions, some valid points by the user have been covered with a "resistance to change" flavor and hence, led to a rebuttal that has not addressed the valid point hidden within.
2. Also, developers work as a team while the users are scattered and hence, every user tries to make a point which is essentially similar to others and hence, gets a same rebuttal but the actual reason he posted the question was because he wanted to know something which was not clarified due to point 1.
3. However more failproof the new system might be, whenever you are shipping it on such a large scale, the guarantee that the users want is for the system to fail nicely. It doesn't matter to me if the systemd has less bugs than the init code, if I encounter that 1 bug, I want have a reasonable guarantee to get around it, and this is what is scaring most of the users I guess.
The above probably has a lot of truth to it :-)
--------------------------- Apparent Simplicity --------------------------- Init scripts had text files while systemd will have binary files to load from.
This is incorrect. Initscripts use bash scripts, systemd uses .ini-like text files.
and hence, when people say that if systemd fails or you don't like it, just switch back to initscripts, they are not addressing the real concern of the users.
Even though your premise was wrong, there is a point that it should be possible to switch back and forth between systemd and initscripts easily. This is the case. If you just install "systemd" and keep the "initscripts" package on your system, then in order to start from systemd just append "init=/bin/systemd" to your kerenl commandline to try it out. systemd will not write anything to your disk by default, so this is completely safe. If you don't like what you get, or there is a problem, simply reboot and remove the init= parameter and you get back initscripts. If ever you decide you don't want initscripts any more, then install systemd-sysvcompat which will make /sbin/init point to systemd, and remove initscripts and sysvinit from your system.
----------------------------------- Single File vs Multiple File ------------------------------------ Most of the users will access their rc.conf files once in a month or so once their system has been setup considerably. I on my personal laptop nowadays only look at rc.conf when there is a pacnew notification. For these users, till now rc.conf was the one-stop service so to say. This concept for them is similar to the one-desk customer service that banks provide where you can get almost all types of manipulations to your accounts at a single desk. This fact is very comforting.
Point is at one glance, you can get all you want to know about the system. This is especially true in case of daemons, now with separate service files, you have to look in every service file to check if the files are correct.
If there was a single command to print out all the boot info, then probably it would be great, but I guess this tool would be the same one that I described in point 1 in Apparent Simplicity. Then, after every edit and regenerate, I can test my files and satisfy myself that what I have typed is correct.
Systemd configures what daemons to start by creating symlinks under /etc/systemd/system. However, "systemctl" gives a nice interface for the user to see what the status is without having to read the symlinks. Have a look at its manpage for details. It is similar (but much more powerful) than our /usr/bin/rc.d.
---------------- DAEMONS ---------------- With respect to daemons, the BEFORE and AFTER in the service files is redundant and though not likely to cause errors, likely to be inconsistent, because for every service file where a daemon "xyz" appears in AFTER, the corresponding daemon must appear in BEFORE in the service file for xyz. I am not quiet sure why this redundancy is there, you can simply have just "AFTER" variables and they should take care of all the dependencies I guess.
This is not true. You only need to specify either Before= or After=, not both. The reason that both exist, is that you should have the choice of which .service file to add the dependency. Cheers, Tom
On Thu, 2012-07-26 at 13:19 +0200, Tom Gundersen wrote:
This is not true. You only need to specify either Before= or After=, not both. The reason that both exist, is that you should have the choice of which .service file to add the dependency.
And step by step it becomes easier to understand. But I (perhaps others too) were confused. I suspect a clean introduction would smooth ruffled feelings. All I experienced are divide the minds and powerpoint presentation like neuro linguistic programming show crap. For example, how should (others and) I know that just "Before" or "After" is needed? Sorry, now I'll shut up.
On Thu, Jul 26, 2012 at 1:47 PM, Ralf Mardorf <ralf.mardorf@alice-dsl.net> wrote:
For example, how should (others and) I know that just "Before" or "After" is needed?
By reading the manpage: "If a unit foo.service contains a setting Before=bar.service and both units are being started, bar.service's start-up is delayed until foo.service is started up." It says nothing about having to put After=foo.service in bar.service, so why would you think it is needed? If you want easy-to-understand information, please read Lennart's blog posts, then read the relevant manpages to get the nitty gritty details. Then ask questions if something is unclear (and hopefully the answers will lead to patches to the manpages). -t
On Thu, 2012-07-26 at 13:53 +0200, Tom Gundersen wrote:
On Thu, Jul 26, 2012 at 1:47 PM, Ralf Mardorf <ralf.mardorf@alice-dsl.net> wrote:
For example, how should (others and) I know that just "Before" or "After" is needed?
By reading the manpage: "If a unit foo.service contains a setting Before=bar.service and both units are being started, bar.service's start-up is delayed until foo.service is started up." It says nothing about having to put After=foo.service in bar.service, so why would you think it is needed?
If you want easy-to-understand information, please read Lennart's blog posts, then read the relevant manpages to get the nitty gritty details. Then ask questions if something is unclear (and hopefully the answers will lead to patches to the manpages).
-t
I was a little bit forward :S. Sorry, anyway, such important changes are not that easy, reading manpages as a dyslexic in a foreign language isn't fun for me, while the old style without systemd perfectly works. I'm not against progress, it's just that I'm (was) scary when I read systemd is from the same guy as pulseaudio is. Pardon, Ralf PS: I anyway will not switch to systemd now ;).
On Thu, 2012-07-26 at 14:03 +0200, Ralf Mardorf wrote:
On Thu, 2012-07-26 at 13:53 +0200, Tom Gundersen wrote:
On Thu, Jul 26, 2012 at 1:47 PM, Ralf Mardorf <ralf.mardorf@alice-dsl.net> wrote:
For example, how should (others and) I know that just "Before" or "After" is needed?
By reading the manpage: "If a unit foo.service contains a setting Before=bar.service and both units are being started, bar.service's start-up is delayed until foo.service is started up." It says nothing about having to put After=foo.service in bar.service, so why would you think it is needed?
If you want easy-to-understand information, please read Lennart's blog posts, then read the relevant manpages to get the nitty gritty details. Then ask questions if something is unclear (and hopefully the answers will lead to patches to the manpages).
-t
I was a little bit forward :S. Sorry, anyway, such important changes are not that easy, reading manpages as a dyslexic in a foreign language isn't fun for me, while the old style without systemd perfectly works. I'm not against progress, it's just that I'm (was) scary when I read systemd is from the same guy as pulseaudio is.
Pardon, Ralf
PS: I anyway will not switch to systemd now ;).
Oops, wrong button :D
participants (18)
-
Alexandre Ferrando
-
Antoine Jardin
-
Baho Utot
-
Dennis Herbrich
-
Fons Adriaensen
-
Jameson
-
Jayesh Badwaik
-
Kevin Chadwick
-
Mantas Mikulėnas
-
Martin Cigorraga
-
Mike
-
Myra Nelson
-
Nicholas MIller
-
Nicolas Sebrecht
-
Oon-Ee Ng
-
Ralf Mardorf
-
Stephen E. Baker
-
Tom Gundersen