[ale] for all you systemd haters...

[Jim Kinney via Ale]

heh, heh,.

Edit your grub kernel line and append init=/bin/bash

no systemd.

[William Wylde via Ale]

Do you still get all your drivers & crap loaded that way?

Post by Jim Kinney via Ale
heh, heh,.
Edit your grub kernel line and append init=/bin/bash
no systemd.
--
James P. Kinney III Every time you stop a school, you will have to build a
jail. What you gain at one end you lose at the other. It's like feeding a
dog on his own tail. It won't fatten the dog. - Speech 11/23/1900 Mark
Twain http://heretothereideas.blogspot.com/
_______________________________________________
Ale mailing list
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Jim Kinney via Ale]

It's pretty much the same as run level 1. If the initrd loads the driver, it in. If it's loaded by a startup process, nope.

So it's a rather crap-free environment.

Note that this won't work without an unlocked encrypted filesystem and/or a password locked grub. But it's a great way to get a kernel only environment. Sort of like Linux circa 1992 but modern stuff installed.

Post by William Wylde via Ale
Do you still get all your drivers & crap loaded that way?

Post by Jim Kinney via Ale
heh, heh,.
Edit your grub kernel line and append init=/bin/bash
no systemd.
--
James P. Kinney III Every time you stop a school, you will have to

build a

Post by Jim Kinney via Ale
jail. What you gain at one end you lose at the other. It's like

feeding a

Post by Jim Kinney via Ale
dog on his own tail. It won't fatten the dog. - Speech 11/23/1900

Mark

Post by Jim Kinney via Ale
Twain http://heretothereideas.blogspot.com/
_______________________________________________
Ale mailing list
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

--
--
“Keep away from people who try to belittle your ambitions. Small people
always do that, but the really great make you feel that you, too, can
become great.”
― Mark Twain

[Steve Litt via Ale]

Post by William Wylde via Ale

Post by Jim Kinney via Ale
heh, heh,.
Edit your grub kernel line and append init=/bin/bash
no systemd.

On February 15, 2018 9:06:05 PM EST, William Wylde

Post by William Wylde via Ale
Do you still get all your drivers & crap loaded that way?

On Thu, 15 Feb 2018 22:01:53 -0500

It's pretty much the same as run level 1. If the initrd loads the
driver, it in.

That's a big "if". Systemd is usually interwoven with the initramfs and
usually implies big, complicated initramfs'.

If it's loaded by a startup process, nope.
So it's a rather crap-free environment.

init=/bin/bash is an old time diagnostic to debug early boot and init.
It's not a substitute for any init system.


One more thing: This is a diagnostic, not a solution for "systemd
haters". And speaking of the phrase "systemd haters", a more accurate
representation would be "lovers of simplicity, modularity, and
functionality."

SteveT
_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[leam hall via Ale]

Post by Steve Litt via Ale
One more thing: This is a diagnostic, not a solution for "systemd
haters". And speaking of the phrase "systemd haters", a more accurate
representation would be "lovers of simplicity, modularity, and
functionality."

I was thinking "People who actually like Linux."
_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Solomon Peachy via Ale]

Post by leam hall via Ale
I was thinking "People who actually like Linux."

What does that even mean?

BTW, "Linux" is not a hand-wavey concept or a state of being. It's just
an operating system kernel, and is rather useless on its own without the
rest of an operating system around it. The TIMTOWTDI principle
definitely applies.

At the same time, one shouldn't discount TIMTOWTDIBSCINABTE either.

- Solomon

[Jim Kinney via Ale]

I prefer 'my way or the highway'. Maybe I took too many of the BOFH
practices to daily use.
NAH!

Post by Solomon Peachy via Ale

Post by leam hall via Ale
I was thinking "People who actually like Linux."

What does that even mean?
BTW, "Linux" is not a hand-wavey concept or a state of being. It's just
an operating system kernel, and is rather useless on its own without the
rest of an operating system around it. The TIMTOWTDI principle
definitely applies.
At the same time, one shouldn't discount TIMTOWTDIBSCINABTE either.
- Solomon
_______________________________________________
Ale mailing list
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[leam hall via Ale]

Post by Solomon Peachy via Ale

Post by leam hall via Ale
I was thinking "People who actually like Linux."

What does that even mean?
BTW, "Linux" is not a hand-wavey concept or a state of being. It's just
an operating system kernel, and is rather useless on its own without the
rest of an operating system around it. The TIMTOWTDI principle
definitely applies.
At the same time, one shouldn't discount TIMTOWTDIBSCINABTE either.
- Solomon

There are many ways not to do it; systemd is one of those ways.

If you need containers and micro-virtualization below the level of
"OS" then something like systemd could be useful. From a standpoint of
the many things that made Linux great, as a kernel and a concept,
systemd is not a good thing.

Leam
_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Solomon Peachy via Ale]

Post by leam hall via Ale
If you need containers and micro-virtualization below the level of
"OS" then something like systemd could be useful. From a standpoint of
the many things that made Linux great, as a kernel and a concept,
systemd is not a good thing.

Pray tell, what exactly is the "concept" of Linux?

I can all but guarantee that no matter how you answer that, the
overwhelming majority of the >2 billion Linux systems deployed don't
conform to or otherwise embody any of it. (Nor, I might add, do they
use systemd)

And a follow-up question -- With your answer in mind, why did Linux
succeed (or achieve "greatness") as opposed to, say, one or more of the
BSDs? After all, the latter had a considerable head start, in both
mindshare and technical capability.

(I'm asking these questions as someone who is responsible for ~20K lines
of code in the Linux kernel, has put together embedded distributions
from scratch, and done Cloudy DevOps things a decade before those
terms entered the lexicon.)

- Solomon

[Steve Litt via Ale]

On Fri, 16 Feb 2018 11:52:03 -0500

Post by Solomon Peachy via Ale

Post by leam hall via Ale
I was thinking "People who actually like Linux."

What does that even mean?

It's pretty clear to me that his use of "Linux" was shorthand for
GNU/Linux, which is a modern UNIX lookalike. The GNU part includes ls,
tr, grep, cut, sort, and a whole bunch of other tools that "do one
thing and do it well."

The "do one thing and do it well" philosophy, when well implemented, has
many positive outcomes:

* Easy interchangeability of parts.
* Easy troubleshooting without relying on special tools custom made for
specific software.
* Easy DIY.
* More modularity, especially with respect to encapsulation.

The preceding list describes pre-systemd Unix/Linux, with contained
exceptions such as KDE, Kmail, Gnome, and Unity, which can simply be
excised from Linux (I removed all KDE programs and libraries in 2013).

Now of course, some folks claim that systemd has all those features
because it's "made of separate modules". Yeah, but in practice, those
modules are very hard to remove, and almost impossible to use, alone,
on non-systemd systems.

So I think Liam was saying people who reject systemd are people who
like concepts such as "do one thing and do it well", with all its
positive outcomes.

SteveT
_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[leam hall via Ale]

Post by Steve Litt via Ale
On Fri, 16 Feb 2018 11:52:03 -0500

Post by Solomon Peachy via Ale

Post by leam hall via Ale
I was thinking "People who actually like Linux."

What does that even mean?

It's pretty clear to me that his use of "Linux" was shorthand for
GNU/Linux, which is a modern UNIX lookalike. The GNU part includes ls,
tr, grep, cut, sort, and a whole bunch of other tools that "do one
thing and do it well."
The "do one thing and do it well" philosophy, when well implemented, has
* Easy interchangeability of parts.
* Easy troubleshooting without relying on special tools custom made for
specific software.
* Easy DIY.
* More modularity, especially with respect to encapsulation.
The preceding list describes pre-systemd Unix/Linux, with contained
exceptions such as KDE, Kmail, Gnome, and Unity, which can simply be
excised from Linux (I removed all KDE programs and libraries in 2013).
Now of course, some folks claim that systemd has all those features
because it's "made of separate modules". Yeah, but in practice, those
modules are very hard to remove, and almost impossible to use, alone,
on non-systemd systems.
So I think Liam was saying people who reject systemd are people who
like concepts such as "do one thing and do it well", with all its
positive outcomes.
SteveT

Pretty sure I'm covered by "prior art" like "Linux World" conferences,
"Linux Journal" magazine, and at least a few "Linux" books". While the
GNU folks have my undying appreciation I find Richard Stallman's
insistence on prepending "GNU/" in front of everything to be sad. It
seems like someone doing a great job and then yelling at you about how
great a job they did instead of letting you look and say "Wow, what a
great job!"

Leam
_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Jim Kinney via Ale]

Post by leam hall via Ale
Pretty sure I'm covered by "prior art" like "Linux World"
conferences,
"Linux Journal" magazine, and at least a few "Linux" books". While the
GNU folks have my undying appreciation I find Richard Stallman's
insistence on prepending "GNU/" in front of everything to be sad. It
seems like someone doing a great job and then yelling at you about how
great a job they did instead of letting you look and say "Wow, what a
great job!"

Were it not for the work of GNU, there would be no Linux. Stallman has
been rather clear in my recollection that a GNU/Linux name refers to
the system as a whole. Linux refers to the kernel.
Stallman may present himself as a crazy old coot, but he's "our crazy
old coot". Compare him to Bill Gates, Larry Ellison, Steve Jobs, Woz,
Joy, etc. All the others had a huge impact on the business side of
technology, i.e., they made a shit ton of money. Stallman changed the
world in a tiny, subtle way that led to a paradigm shift of development
that led to Linux that led to now.
I think it should be proudly and loudly called GNU/Linux as long as he
is alive and probably for years afterwards. I've fed my family thanks
to what he started.

Post by leam hall via Ale
Leam
_______________________________________________
Ale mailing list
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Solomon Peachy via Ale]

Post by Jim Kinney via Ale
Were it not for the work of GNU, there would be no Linux. Stallman has
been rather clear in my recollection that a GNU/Linux name refers to
the system as a whole. Linux refers to the kernel.

Yep. Without GNU (the project and the GPL) Linux as it exists today
would have never come to pass.

Post by Jim Kinney via Ale
I think it should be proudly and loudly called GNU/Linux as long as he
is alive and probably for years afterwards. I've fed my family thanks
to what he started.

It's worth keeping in mind that these days only a minority of Linux
deployments are GNU/Linux. Even excluding Android, consider how many
appliances/boxes run Linux with uClibc and busybox, ie no GNU userspace
at all.

But yeah, I owe my professional career to what Stallman started.

- Solomon

[Solomon Peachy via Ale]

Post by Steve Litt via Ale
It's pretty clear to me that his use of "Linux" was shorthand for
GNU/Linux, which is a modern UNIX lookalike. The GNU part includes ls,
tr, grep, cut, sort, and a whole bunch of other tools that "do one
thing and do it well."

Since you brought up GNU.. GNU's Not UNIX, and their respective
underlying philosophies are generally orthogonal.

GNU isn't necessarily Linux, nor is Linux necessarily GNU either.

Post by Steve Litt via Ale
The "do one thing and do it well" philosophy, when well implemented, has
* Easy interchangeability of parts.

....For functionally equivalent pieces, or the whole system falls apart.

Post by Steve Litt via Ale
* Easy troubleshooting without relying on special tools custom made for
specific software.

I already have an extensive toolbox of special tools optimized for
different tasks. What's one more? (None of them are "custom made"
either)

Post by Steve Litt via Ale
* Easy DIY.

I don't recall if I asked this before (or how you answered), but have
you ever built a Linux distribution from source? It's quite an
undertaking, even today.

Post by Steve Litt via Ale
* More modularity, especially with respect to encapsulation.

Generally a worthy goal, but all abstractions are ultimately leaky.

Post by Steve Litt via Ale
The preceding list describes pre-systemd Unix/Linux, with contained
exceptions such as KDE, Kmail, Gnome, and Unity, which can simply be
excised from Linux (I removed all KDE programs and libraries in 2013).

I don't think it's fair to compare interactive GUI things with
non-interative non-GUI things as they are designed for fundamentally
different purposes.

But putting that aside, you can replace parts of systemd, you can
ascertain state/status and troubleshoot more easily than with the old
mess (since there are fewer moving parts requiring fewer tools to muck
with it all), it's easier to build a supportable distro using systemd
than without it (due to having to reimplement far fewer wheels), and
finally, systemd is modular to a fault, with all interactions between
components explitly defined.

Post by Steve Litt via Ale
Now of course, some folks claim that systemd has all those features
because it's "made of separate modules". Yeah, but in practice, those
modules are very hard to remove, and almost impossible to use, alone,
on non-systemd systems.

In practice, you can't go replacing bits of _any_ system willy-nilly
without problems, because a system is more than the sum of its parts.

(Case in point -- Just try and replace bash with csh, and see how much
joy ensues as you end up needing to rewrite all but the most trivial of
shell scripts. Including those invoked via init..)

Post by Steve Litt via Ale
So I think Liam was saying people who reject systemd are people who
like concepts such as "do one thing and do it well", with all its
positive outcomes.

You may be correct, but I'd still prefer to hear his reasoning.

IMO it's hard to argue that folks who have spent literally decades
elbows-deep in Linux (both the kernel and various operating systems
built on it) somehow don't "like" it.

- Solomon

[Steve Litt via Ale]

On Fri, 16 Feb 2018 11:52:03 -0500

Post by Solomon Peachy via Ale

Post by leam hall via Ale
I was thinking "People who actually like Linux."

What does that even mean?

I already posited one theory on what Leam meant, and now a second
possible explanation occurs to me:

Systemd originator and evangelizer in chief Lennart Poettering has
never been shy in stating his dislike of Linux, often going so far as
to negatively compare Linux to Mac and even Windows. Yeah Windows. And
it's clear that systemd makes Linux systems it has been installed on
more like Windows, in terms of reduced diy, and increased program to
program interdependencies. Poettering has several times suggested
ignoring and breaking POSIX, an underpinning of all things UNIXy
including Linux, for the last 30 years.

Poettering doesn't like Linux and created systemd to make Linux less
Linux. A person who likes Linux would therefore not like systemd.

SteveT
_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[leam hall via Ale]

Post by Steve Litt via Ale
On Fri, 16 Feb 2018 11:52:03 -0500

Post by Solomon Peachy via Ale

Post by leam hall via Ale
I was thinking "People who actually like Linux."

What does that even mean?

I already posited one theory on what Leam meant, and now a second
Systemd originator and evangelizer in chief Lennart Poettering has
never been shy in stating his dislike of Linux, often going so far as
to negatively compare Linux to Mac and even Windows. Yeah Windows. And
it's clear that systemd makes Linux systems it has been installed on
more like Windows, in terms of reduced diy, and increased program to
program interdependencies. Poettering has several times suggested
ignoring and breaking POSIX, an underpinning of all things UNIXy
including Linux, for the last 30 years.
Poettering doesn't like Linux and created systemd to make Linux less
Linux. A person who likes Linux would therefore not like systemd.
SteveT

That makes me wonder; what does systemd do that non-systemd mechanisms
cannot? Granted, it may make some things easier or more to someone's
liking, but what does it do well that cannot be replicated otherwise?

Leam
_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Lightner, Jeffrey via Ale]

I ntractables
N ever
I ntend
T ransistioning

:p

-----Original Message-----
From: Ale [mailto:ale-***@ale.org] On Behalf Of leam hall via Ale
Sent: Friday, February 16, 2018 2:21 PM
To: Atlanta Linux Enthusiasts
Subject: Re: [ale] for all you systemd haters...

Post by Steve Litt via Ale
On Fri, 16 Feb 2018 11:52:03 -0500

Post by Solomon Peachy via Ale

Post by leam hall via Ale
I was thinking "People who actually like Linux."

What does that even mean?

I already posited one theory on what Leam meant, and now a second
Systemd originator and evangelizer in chief Lennart Poettering has
never been shy in stating his dislike of Linux, often going so far as
to negatively compare Linux to Mac and even Windows. Yeah Windows. And
it's clear that systemd makes Linux systems it has been installed on
more like Windows, in terms of reduced diy, and increased program to
program interdependencies. Poettering has several times suggested
ignoring and breaking POSIX, an underpinning of all things UNIXy
including Linux, for the last 30 years.
Poettering doesn't like Linux and created systemd to make Linux less
Linux. A person who likes Linux would therefore not like systemd.
SteveT

That makes me wonder; what does systemd do that non-systemd mechanisms cannot? Granted, it may make some things easier or more to someone's liking, but what does it do well that cannot be replicated otherwise?

Leam
_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo
_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Solomon Peachy via Ale]

Post by leam hall via Ale
That makes me wonder; what does systemd do that non-systemd mechanisms
cannot? Granted, it may make some things easier or more to someone's
liking, but what does it do well that cannot be replicated otherwise?

Given that systemd uses public Linux (and/or glibc) APIs, there's
nothing it does that another tool (or set of tools) couldn't also
accomplish.

An example of something that systemd does well is logind. Sure, you can
recreate the external service API, but replicating the functionality
"well" might be impossible without also replicating much of the systemd
functionality that logind relies upon. Consequently, anything that
seeks to replicate logind's underlying functionality (ie session
management using Linux's cgroup features) is likely to end up looking
subtantially similar to systemd's implementation.

Are there other approaches to take? Sure, but someone who cares has to
actually step up, make it work, and most importantly, keep supporting
it. (That's why everyone was so happy to drop ConsoleKit like a hot
potato)

- Solomon

[Alex Carver via Ale]

Post by Solomon Peachy via Ale

Post by leam hall via Ale
That makes me wonder; what does systemd do that non-systemd mechanisms
cannot? Granted, it may make some things easier or more to someone's
liking, but what does it do well that cannot be replicated otherwise?

Given that systemd uses public Linux (and/or glibc) APIs, there's
nothing it does that another tool (or set of tools) couldn't also
accomplish.
An example of something that systemd does well is logind. Sure, you can
recreate the external service API, but replicating the functionality
"well" might be impossible without also replicating much of the systemd
functionality that logind relies upon. Consequently, anything that
seeks to replicate logind's underlying functionality (ie session
management using Linux's cgroup features) is likely to end up looking
subtantially similar to systemd's implementation.

What exactly is logind supposed to handle? I've already searched
multiple times and most sites regurgitate the manual without really
discussing what it's supposed to be doing and how it's different or
better than other implementations. I've not come across anything that
explains it well.

I ask in all seriousness because the few things I've been able to find
are worded in ways suggesting that logind has abilities that have never
existed before. For example, one site says that logind provides the
ability to track user logins but that was already possible for both
individual machines and for central authentication (Kerberos and more
recently Active Directory). Again, in context (with paragraphs before
and after) it extends the claim of an ability to "this has not been
possible before" (paraphrased).
_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Solomon Peachy via Ale]

Post by Alex Carver via Ale
What exactly is logind supposed to handle? I've already searched
multiple times and most sites regurgitate the manual without really
discussing what it's supposed to be doing and how it's different or
better than other implementations. I've not come across anything that
explains it well.

Logind manages user sessions. It ensues that when a user logs out all of
their detritus is cleaned up properly, or that if you switch to a
different user then appropriate permissions are set up and revoked --
this can even extends to stuff like network authentication, which may
require per-user authentiction.

It also (for all practical purposes) made multi-seat Linux systems
feasible; that is a single box with different users simultaneously
logged in using different sets of displays/keyboards/etc. Granted,
other things like rootless X (via KMS) were also necessary, but logind
tied it all together and finally made it work.

Post by Alex Carver via Ale
I ask in all seriousness because the few things I've been able to find
are worded in ways suggesting that logind has abilities that have never
existed before. For example, one site says that logind provides the
ability to track user logins but that was already possible for both
individual machines and for central authentication (Kerberos and more
recently Active Directory). Again, in context (with paragraphs before
and after) it extends the claim of an ability to "this has not been
possible before" (paraphrased).

logind is not "tracking" logins so much as ensuring that everything a
user gets up to is properly lumped in together so that when they log out
you don't end up with stray processes or permissions out of whack.

AD and Kerberos are both much lower-level mechanisms that are orthogonal
to logind. Kerberos can easily be the authentication mechanism, while
AD might only care that a given user is logged into on machine X for
accounting purposes (eg there may be a security policy to prevent a user
from being logged into more than one machine at a time. logind only
knows or otherwise cares about the local system)

- Solomon

[Leam Hall via Ale]

Post by Solomon Peachy via Ale
It also (for all practical purposes) made multi-seat Linux systems
feasible; that is a single box with different users simultaneously
logged in using different sets of displays/keyboards/etc. Granted,
other things like rootless X (via KMS) were also necessary, but logind
tied it all together and finally made it work.

I'm confused on this, but it's likely just the pain from a recently
pulled tooth. What is the use case for multiple physically logged in
users? Didn't that already work?


_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Solomon Peachy via Ale]

I'm confused on this, but it's likely just the pain from a recently pulled
tooth. What is the use case for multiple physically logged in users? Didn't
that already work?

Multiple simultaneous physical (graphical) sessions didn't really work,
at least not without a lot of careful hand-configuration and even then
the secrity between them was quite brittle due to requiring a lot of
things to run as root.

Two features in particluar made this work more recently -- proper cgroup
support up and down the stack (eg systemd/logind in userspace, and
more kernel subsystems implementing cgroups properly) and KMS allowing
for the X server to not have to run as root.

(Personally I think the whole idea is a bit daft, but evidently there
are a lot who disagree with me...)

- Solomon

[Alex Carver via Ale]

Post by Solomon Peachy via Ale

Post by Alex Carver via Ale
What exactly is logind supposed to handle? I've already searched
multiple times and most sites regurgitate the manual without really
discussing what it's supposed to be doing and how it's different or
better than other implementations. I've not come across anything that
explains it well.

Logind manages user sessions. It ensues that when a user logs out all of
their detritus is cleaned up properly, or that if you switch to a
different user then appropriate permissions are set up and revoked --
this can even extends to stuff like network authentication, which may
require per-user authentiction.

Now currently a non-systemd machine tracks what belongs to whom, right?
I can pull up a list of PIDs and their owners (top already can do it) so
in theory I should have always been able to clean up behind a user even
if unceremoniously with a giant killall. So that part isn't new as
described by some of these sites.

Permissions for the network is interesting. Network authentication I
can see since that currently would require some glue logic to pull off
(like feeding current environment to wpasupplicant or an 802.1X EAP
interface).

Post by Solomon Peachy via Ale
It also (for all practical purposes) made multi-seat Linux systems
feasible; that is a single box with different users simultaneously
logged in using different sets of displays/keyboards/etc. Granted,
other things like rootless X (via KMS) were also necessary, but logind
tied it all together and finally made it work.

But isn't this what thin-clients did ages ago? You had a keyboard,
mouse, local GPU, local display manager and everything else ran on the
central machine. Some older versions of Windows had that and I remember
thin clients for using X as well.

_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Jim Kinney via Ale]

Post by Alex Carver via Ale

Post by Solomon Peachy via Ale
It also (for all practical purposes) made multi-seat Linux systems
feasible; that is a single box with different users simultaneously
logged in using different sets of
displays/keyboards/etc. Granted,
other things like rootless X (via KMS) were also necessary, but logind
tied it all together and finally made it work.

But isn't this what thin-clients did ages ago? You had a keyboard,
mouse, local GPU, local display manager and everything else ran on the
central machine. Some older versions of Windows had that and I remember
thin clients for using X as well.

Different stuff. It's literally a multi-port video card with a screen
attached to display only one users' stuff. Picture a desktop with 6
users all with their own keyboard, display and mouse. Multi-console
makes more sense.

https://linuxgazette.net/124/smith.html

Post by Alex Carver via Ale
_______________________________________________
Ale mailing list
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Alex Carver via Ale]

Post by Jim Kinney via Ale

Post by Alex Carver via Ale

Post by Solomon Peachy via Ale
It also (for all practical purposes) made multi-seat Linux systems
feasible; that is a single box with different users simultaneously
logged in using different sets of
displays/keyboards/etc. Granted,
other things like rootless X (via KMS) were also necessary, but logind
tied it all together and finally made it work.

But isn't this what thin-clients did ages ago? You had a keyboard,
mouse, local GPU, local display manager and everything else ran on the
central machine. Some older versions of Windows had that and I remember
thin clients for using X as well.

Different stuff. It's literally a multi-port video card with a screen
attached to display only one users' stuff. Picture a desktop with 6
users all with their own keyboard, display and mouse. Multi-console
makes more sense.

How is that different from a thin client? I'm picturing what I saw
before, a screen, a keyboard, a mouse and no real computer in sight. It
didn't require that the computer be next to the keyboard/monitor nor
that the four people were at the same desk but it still all ran from a
single machine. The only real difference is that you've pushed the
hardware deeper into the central machine's box.
_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Leam Hall via Ale]

Post by Jim Kinney via Ale
Different stuff. It's literally a multi-port video card with a screen
attached to display only one users' stuff. Picture a desktop with 6
users all with their own keyboard, display and mouse. Multi-console
makes more sense.
https://linuxgazette.net/124/smith.html

That version of Mandriva predates the distro adoption of systemd,
doesn't it?
_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Jim Kinney via Ale]

Oh yeah. But that was just a fast example of the multi-seat linux that logind from systemd helps manage.

It's still used in schools (not mandriva I hope!)

Post by Leam Hall via Ale

Post by Jim Kinney via Ale
Different stuff. It's literally a multi-port video card with a screen
attached to display only one users' stuff. Picture a desktop with 6
users all with their own keyboard, display and mouse. Multi-console
makes more sense.
https://linuxgazette.net/124/smith.html

That version of Mandriva predates the distro adoption of systemd,
doesn't it?
_______________________________________________
Ale mailing list
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Solomon Peachy via Ale]

Post by Alex Carver via Ale
Now currently a non-systemd machine tracks what belongs to whom, right?
I can pull up a list of PIDs and their owners (top already can do it) so
in theory I should have always been able to clean up behind a user even
if unceremoniously with a giant killall. So that part isn't new as
described by some of these sites.

Yes and no -- you don't want to blindly kill all of a user's processes.

They may have more than one login session (eg via ssh), or may have
requested something to persist.

At the same time, you may want to reap stuff that has done the
double-fork detachment game.

It's not necessarily "new" in a pure capability sense, but what is new
is that it JustWorks(tm).

Post by Alex Carver via Ale
Permissions for the network is interesting. Network authentication I
can see since that currently would require some glue logic to pull off
(like feeding current environment to wpasupplicant or an 802.1X EAP
interface).

Yep, there's intereaction with NetworkManager (via dbus) to handle this
sort of thing.

Post by Alex Carver via Ale
But isn't this what thin-clients did ages ago? You had a keyboard,
mouse, local GPU, local display manager and everything else ran on the
central machine. Some older versions of Windows had that and I remember
thin clients for using X as well.

You've been around long enough to see what's old become new become old
become new again. Meanwhile the old IBM greybeards are still chuckling
about today's kids not knowing anything about anything. :)

- Solomon

[Steve Litt via Ale]

On Fri, 16 Feb 2018 14:20:38 -0500

Post by leam hall via Ale

Post by Steve Litt via Ale
On Fri, 16 Feb 2018 11:52:03 -0500

On Fri, Feb 16, 2018 at 11:22:20AM -0500, leam hall via Ale

Post by leam hall via Ale
I was thinking "People who actually like Linux."

What does that even mean?

I already posited one theory on what Leam meant, and now a second
Systemd originator and evangelizer in chief Lennart Poettering has
never been shy in stating his dislike of Linux, often going so far
as to negatively compare Linux to Mac and even Windows. Yeah
Windows. And it's clear that systemd makes Linux systems it has
been installed on more like Windows, in terms of reduced diy, and
increased program to program interdependencies. Poettering has
several times suggested ignoring and breaking POSIX, an
underpinning of all things UNIXy including Linux, for the last 30
years.
Poettering doesn't like Linux and created systemd to make Linux less
Linux. A person who likes Linux would therefore not like systemd.
SteveT

That makes me wonder; what does systemd do that non-systemd mechanisms
cannot?
Granted, it may make some things easier or more to someone's
liking, but what does it do well that cannot be replicated otherwise?

Hi Liam,

It looks like you're asking 2 questions:

1) What does systemd do that other mechanisms can't do?

2) Can these benefits be replicated by other systems?


As far as your question 1, it depends who you ask. Here's the systemd
party line:

http://0pointer.de/blog/projects/the-biggest-myths.html

The preceding is so chock full of logical fallacies, but not a day goes
by when I don't hear someone with absolutely no idea of the
functionality of an init system parroting its fallacies.

Here's the rebuttal to same, by Jude Nelson, author of the vdev
replacement for systemd-entangled udev:

http://judecnelson.blogspot.com/2014/09/systemd-biggest-fallacies.html

And here's why musl C standard library author Rich Felker says systemd
is "broken by design." Be sure to go to the bottom and see the 14 line
PID1 written by Felker. Later, Dimitris Papastamos made an 83 line
version, Suckless Tools Init (sinit), that responded to all the
necessary interrupts for a PID1. Then I bolted on daemontools-encore,
and some custom shellscripts I called LittKit to bring up services in a
given order. The sinit/daemontools-encore/LittKit combo was better,
faster and more understandable than sysvinit, and it was better and
more understandable than systemd (systemd *does* boot pretty fast, if
that's your priority and you're a one issue voter). I'd still be using
the combo today, but I found out that the runit init system is pretty
much the same thing, from a single free software vendor, with a lot of
improvements bestowed not by dependencies or kilolines of code, but by
a few symlinks and a few lightweight utilities.

So, both sinit+Daemontools-encore+LittKit and runit do the init work of
systemd, in a different way, much more simply. Another init system does
it in a slightly more complex and featureful way than runit and my
combo, but it's as good or superior to the init section of systemd,
and much simpler. These have to do with your question #2.


LOL, here's the systemd sales process, as shown time and time again by
Poettering:

1 Find or invent a need.

2 Write a messy, entangled solution with a design you might expect from
a group of DP101 students. This solution is inseparably interwoven
with systemd.

3 When people say it's a mess, he says "well, how else are you going to
do ________?"

4 When people voice realistic ideas, he proves they won't work by
showing how they conflict with the existing Rube Goldberg systemd
he's already built. He does this by getting into nitty gritty details
of his code that nobody outside the systemd/freedesktop developer
community could possibly understand.

5 Finally, after somebody shows a working example of how to do ______
that involves no help from systemd, Poettering writes it off as
"simple", the worst insult he knows.

Steps 4 and 5 are Poettering's answer to your question: They say you
can't. And if you *do* make a superior mechanism, they'll say it
doesn't count because it's "simple."

Throughout all those steps he Poettering (and systemd fans who use his
techniques) sprinkle in all sorts of personal insults.

Here's a video showing Poettering, using the steps above, ambushing and
bullying a presenter who had little skill in English.

At 23:02
Poettering asks the presenter "do you hate handicapped people"
because the presenter didn't want the display manager to be a
preload of Gnome. He's fully in step 4 here, explaining why with his
software, you need all that trash to be accessible.

At 23:53 Poettering appears to score some major points when he
defends NetworkManager as part of the Display manager and the
presenter unwisely suggests static configuration. But of course, a very
simple program with a limited X interface can manipulate wpa_cli and
wpa_password within a shellscript to provide that function: You don't
need NetworkManager, which is extremely promiscuous with its
communications with other programs. Poettering's kicking ass on step 4.

And then at 24:51 the presenter makes Poettering look silly with the
sentence "do you know what shellscripts are?", thereby kicking it up to
step 5. Poettering mutters an insult and changes the subject.

Many projects are led by dicks. If I refused to use software led by
dicks, I'd still be using CPM. Thing is though, the systemd posse
parrots the words of Poettering as if it's gospel instead of a
combination of insults and fallacies.

True story: A pro-systemd guy assured me I needed "socket activation"
to read and automount thumb drives. After two hours experimentation I
presented a shellscript automounter based on inotifywait. Pro-systemd
guy didn't like it, too simple, didn't work with systemd (that was the
whole point), and too buggy (did I mention I wrote it in 2 hours?).

I have yet to hear of anything systemd does that could not be done in a
more modular, simpler and easy to troubleshoot way.

SteveT
_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Solomon Peachy via Ale]

Post by Steve Litt via Ale
Systemd originator and evangelizer in chief Lennart Poettering has
never been shy in stating his dislike of Linux, often going so far as
to negatively compare Linux to Mac and even Windows. Yeah Windows. And
it's clear that systemd makes Linux systems it has been installed on
more like Windows, in terms of reduced diy, and increased program to
program interdependencies.

(If you're referring to what I think you are..)

Windows Vista introduced seamless shared audio output, capable of
switching to different output devices, per-application volume control,
even for applications written against the very first WaveOut audio APIs
introduced in the Win16 days. It Just Worked(tm).

So is it somehow bad to acknowledge that Windows or Macs genuinely did
something better and provided a superior user experience, and attempt to
address that deficiency in Linux systems? (With two major audio APIs,
three or four incompatible sound servers, and a whole load of
buggy-with-not-quite-the-same-quirks audio drivers to wrangle )

Another example. Is it somehow bad to acknowlede that Windows had
objectively superior multi-desktop-users-logged-in-simultaneously
support, and attempt to improve that?

Meanwhile, I'd caution you against blaming Pottering for "increased
program to program interdependencies" as that's a property of all
systems and it well predates his birth. You might as well try to blame
him for water being wet, for that would make as about much sense.

Post by Steve Litt via Ale
Poettering has several times suggested ignoring and breaking POSIX, an
underpinning of all things UNIXy including Linux, for the last 30
years.

POSIX and UNIX are not some holy documents beyond reproach, nor do they
represent the absolute pinnacle of what can ever be achieved.

They both have many warts, in design and implementation, and their
abstrations break down all the damn time. Even the creators of both
acknoweldge that.

As an aside, Linus Torvalds often advocates for ignoring or otherwise
breaking POSIX. Or are you seriously going to claim that Torvalds also
doesn't know what underpins Linux?

Post by Steve Litt via Ale
Poettering doesn't like Linux and created systemd to make Linux less
Linux. A person who likes Linux would therefore not like systemd.

By that same logic dropping a high output crate motor into my truck to
make it more capable means that I don't like my truck, instead of
showing that I like it more than is probably healthy.

Yeah, Pottering hates Linux so much he's spent the past fifteen years
improving it. We should all be so lucky to have folks like him that
hate the things we use every day.

- Solomon

[Alex Carver via Ale]

Post by Solomon Peachy via Ale

Post by Steve Litt via Ale
Systemd originator and evangelizer in chief Lennart Poettering has
never been shy in stating his dislike of Linux, often going so far as
to negatively compare Linux to Mac and even Windows. Yeah Windows. And
it's clear that systemd makes Linux systems it has been installed on
more like Windows, in terms of reduced diy, and increased program to
program interdependencies.

(If you're referring to what I think you are..)
Windows Vista introduced seamless shared audio output, capable of
switching to different output devices, per-application volume control,
even for applications written against the very first WaveOut audio APIs
introduced in the Win16 days. It Just Worked(tm).
So is it somehow bad to acknowledge that Windows or Macs genuinely did
something better and provided a superior user experience, and attempt to
address that deficiency in Linux systems? (With two major audio APIs,
three or four incompatible sound servers, and a whole load of
buggy-with-not-quite-the-same-quirks audio drivers to wrangle )
Another example. Is it somehow bad to acknowlede that Windows had
objectively superior multi-desktop-users-logged-in-simultaneously
support, and attempt to improve that?
Meanwhile, I'd caution you against blaming Pottering for "increased
program to program interdependencies" as that's a property of all
systems and it well predates his birth. You might as well try to blame
him for water being wet, for that would make as about much sense.

Post by Steve Litt via Ale
Poettering has several times suggested ignoring and breaking POSIX, an
underpinning of all things UNIXy including Linux, for the last 30
years.

POSIX and UNIX are not some holy documents beyond reproach, nor do they
represent the absolute pinnacle of what can ever be achieved.
They both have many warts, in design and implementation, and their
abstrations break down all the damn time. Even the creators of both
acknoweldge that.
As an aside, Linus Torvalds often advocates for ignoring or otherwise
breaking POSIX. Or are you seriously going to claim that Torvalds also
doesn't know what underpins Linux?

Post by Steve Litt via Ale
Poettering doesn't like Linux and created systemd to make Linux less
Linux. A person who likes Linux would therefore not like systemd.

By that same logic dropping a high output crate motor into my truck to
make it more capable means that I don't like my truck, instead of
showing that I like it more than is probably healthy.
Yeah, Pottering hates Linux so much he's spent the past fifteen years
improving it. We should all be so lucky to have folks like him that
hate the things we use every day.

Just remember that Pottering was primarily responsible for Pulse Audio
and, although perhaps it's working better now, he pushed it hard
claiming it was ready and it was not leading to a severe hatred for it.
He pushed hard enough that, in his very typical style, was calling out
any distribution that did not install it by default. Later he
backtracked saying he never claimed it was ready.

I think if systemd had been spearheaded by anyone other than Pottering
it probably would have gotten an improved reputation or at minimum would
have had an opportunity to demonstrate itself and sell the idea on the
merits. There are plenty of people that improve the things we use but
are not even remotely as sour as Pottering. For systemd, Pottering's
previous actions with Pulse Audio basically guaranteed that he burned
the bridge before he crossed it.
_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Jim Kinney via Ale]

Post by Alex Carver via Ale
I think if systemd had been spearheaded by anyone other than
Pottering
it probably would have gotten an improved reputation or at minimum would
have had an opportunity to demonstrate itself and sell the idea on the
merits. There are plenty of people that improve the things we use but
are not even remotely as sour as Pottering. For systemd, Pottering's
previous actions with Pulse Audio basically guaranteed that he burned
the bridge before he crossed it.
__________________________

And that sentiment is a real shame. I've managed to mellow (slightly)
as I've become older. I don't instantly start waving the pitchfork and
torch when things don't go my way. But I do keep them close :-)
At the time, pulseaudio was a difficult project, a needed project as an
attempt to create a Linux-wide workable audio system. Several people
worked on it and made it happen. Far fewer people pitch pull requests
that complain about things not being the way _they_ want it. I found
the combo of pulsaudio and esd quite workable especially during the
thin client project in 2006-2007. It just worked. Did it work 2 years
earlier? Um, not well.
Things improve. Sometimes people do to.

Post by Alex Carver via Ale
_____________________
Ale mailing list
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Alex Carver via Ale]

Post by Jim Kinney via Ale

Post by Alex Carver via Ale
I think if systemd had been spearheaded by anyone other than
Pottering
it probably would have gotten an improved reputation or at minimum would
have had an opportunity to demonstrate itself and sell the idea on the
merits. There are plenty of people that improve the things we use but
are not even remotely as sour as Pottering. For systemd, Pottering's
previous actions with Pulse Audio basically guaranteed that he burned
the bridge before he crossed it.
__________________________

And that sentiment is a real shame. I've managed to mellow (slightly)
as I've become older. I don't instantly start waving the pitchfork and
torch when things don't go my way. But I do keep them close :-)
At the time, pulseaudio was a difficult project, a needed project as an
attempt to create a Linux-wide workable audio system. Several people
worked on it and made it happen. Far fewer people pitch pull requests
that complain about things not being the way _they_ want it. I found
the combo of pulsaudio and esd quite workable especially during the
thin client project in 2006-2007. It just worked. Did it work 2 years
earlier? Um, not well.
Things improve. Sometimes people do to.

Sometimes people improve but I think that Pottering has shown he won't
(there's plenty of very recent messages from him where his overall tone
suggest this).

If you believe you have a great product, go out and sell the product.
However, if you try to sell the product by yelling at potential clients
that they're idiots if they don't buy it, your sales will not be as
large as they would otherwise. Pottering may have a great product but
he's an awful salesman (and not such a wonderful person).
_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Solomon Peachy via Ale]

Post by Alex Carver via Ale
Just remember that Pottering was primarily responsible for Pulse Audio
and, although perhaps it's working better now, he pushed it hard
claiming it was ready and it was not leading to a severe hatred for it.
He pushed hard enough that, in his very typical style, was calling out
any distribution that did not install it by default. Later he
backtracked saying he never claimed it was ready.

That's fair to say.

My personal experience with PA is that most of the end-user pain was due
to Ubuntu utterly botching its integration and not really doing anything
to improve things. (Fedora, despite using what essentially amounted to
the same PA releases, suffered far fewer problems.)

Well, that and buggy drivers, buggy applications [often relying on
quirks of individual drivers], and bugs bugs bugs and even more bugs
everywhere. Moral of the story: Linux Audio was a complete festering
mess before PA came along, and becoming an even greater mess until
everyone finally got serious about deploying bugfixes. I'm personally
glad he pushed so hard, we all benefited in the end. (And amazed he
keeps coming back for more. He must have skin made of asbestos..)

(This situation paralleled NetworkManager and the multitude of wifi
drivers that all behaved subtly or not so subtly differently...)

Post by Alex Carver via Ale
I think if systemd had been spearheaded by anyone other than Pottering
it probably would have gotten an improved reputation or at minimum would
have had an opportunity to demonstrate itself and sell the idea on the
merits. There are plenty of people that improve the things we use but
are not even remotely as sour as Pottering. For systemd, Pottering's
previous actions with Pulse Audio basically guaranteed that he burned
the bridge before he crossed it.

A big difference is that systemd wasn't disruptive to the typical
end-user, could be incrementally utilized, and didn't require folks to
rewrite anything except muscle memory.

(It's worth noting that systemd paying attention to return codes rapidly
demonstrated a whole ton of bugs in supposedly-mature-and-robust
distro init scripts. But guess who got blamed...)

The average joe wouldn't have noticed any outward difference beyond
faster boot times, distro writers and a decent chunk of their userbase
went "hell yeah this is better", a smaller subset of their userbase
still hasn't stopped screaming about ethics in init systems, and most
users were like "meh, whatever.."

Anyway. Simulations finally about to finish. I guess I'll have to go
off and be productive now.

- Solomon

[Alex Carver via Ale]

Post by Solomon Peachy via Ale

Post by Alex Carver via Ale
Just remember that Pottering was primarily responsible for Pulse Audio
and, although perhaps it's working better now, he pushed it hard
claiming it was ready and it was not leading to a severe hatred for it.
He pushed hard enough that, in his very typical style, was calling out
any distribution that did not install it by default. Later he
backtracked saying he never claimed it was ready.

That's fair to say.
My personal experience with PA is that most of the end-user pain was due
to Ubuntu utterly botching its integration and not really doing anything
to improve things. (Fedora, despite using what essentially amounted to
the same PA releases, suffered far fewer problems.)
Well, that and buggy drivers, buggy applications [often relying on
quirks of individual drivers], and bugs bugs bugs and even more bugs
everywhere. Moral of the story: Linux Audio was a complete festering
mess before PA came along, and becoming an even greater mess until
everyone finally got serious about deploying bugfixes. I'm personally
glad he pushed so hard, we all benefited in the end. (And amazed he
keeps coming back for more. He must have skin made of asbestos..)
(This situation paralleled NetworkManager and the multitude of wifi
drivers that all behaved subtly or not so subtly differently...)

I remember having to deal with Pulse Audio early on. ALSA happened to
work better and I stuck with it.

NetworkManager on the other hand tends to stick its nose into places
that make things impossible. I think it's poorly implemented in some
respects and am very glad it's not a mandatory component because I rip
it out anytime I do a new installation. I know I mentioned the
integration of EAP with the current session but I think the systemic
annoyances that NM causes for some other cases outweighed it for me.

Post by Solomon Peachy via Ale

Post by Alex Carver via Ale
I think if systemd had been spearheaded by anyone other than Pottering
it probably would have gotten an improved reputation or at minimum would
have had an opportunity to demonstrate itself and sell the idea on the
merits. There are plenty of people that improve the things we use but
are not even remotely as sour as Pottering. For systemd, Pottering's
previous actions with Pulse Audio basically guaranteed that he burned
the bridge before he crossed it.

A big difference is that systemd wasn't disruptive to the typical
end-user, could be incrementally utilized, and didn't require folks to
rewrite anything except muscle memory.

I think it depended on the location of that end-user. If it was the
end-user is the administrator, it did have somewhat of a disruptive
effect. This became especially obvious when certain things got jammed
up due to systemd taking control of sockets for example (some
applications are not expecting that an entirely different and unrelated
piece of software is going to open and hold the socket). Some of the
pain here is caused by package maintainers that blindly install every
last bit of systemd configuration (sockets, etc.) without checking what
the daemon actually expects or uses. But that ends up reflecting poorly
on systemd because that was the "new thing" that appeared on the
end-user's machine and suddenly broke whatever was working. And with
something new like that, it's hard to get a grasp of all the details to
unwind it. For one machine that I had to use systemd due to a
dependency, it took me the better part of a day to figure out there was
a socket defined that shouldn't have been there, find it and remove it.

Post by Solomon Peachy via Ale
(It's worth noting that systemd paying attention to return codes rapidly
demonstrated a whole ton of bugs in supposedly-mature-and-robust
distro init scripts. But guess who got blamed...)
The average joe wouldn't have noticed any outward difference beyond
faster boot times, distro writers and a decent chunk of their userbase
went "hell yeah this is better", a smaller subset of their userbase
still hasn't stopped screaming about ethics in init systems, and most
users were like "meh, whatever.."
Anyway. Simulations finally about to finish. I guess I'll have to go
off and be productive now.
- Solomon

_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[DJ-Pfulio via Ale]

Post by Solomon Peachy via Ale
The average joe wouldn't have noticed any outward difference beyond
faster boot times, <snip>

Exactly when will that happen?

My systems used to boot in 15-25 seconds, usually just under 20s.

$ systemd-analyze
Startup finished in 26.383s (kernel) + 13.483s (userspace) = 39.866s

On systems with static IPs, networking.service takes over 20 seconds!!!
Huh?

I keep hearing systemd makes booting faster and I suppose it does for
people with 150 LUNs, but not here.
_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Alex Carver via Ale]

Post by DJ-Pfulio via Ale

Post by Solomon Peachy via Ale
The average joe wouldn't have noticed any outward difference beyond
faster boot times, <snip>

Exactly when will that happen?
My systems used to boot in 15-25 seconds, usually just under 20s.
$ systemd-analyze
Startup finished in 26.383s (kernel) + 13.483s (userspace) = 39.866s
On systems with static IPs, networking.service takes over 20 seconds!!!
Huh?
I keep hearing systemd makes booting faster and I suppose it does for
people with 150 LUNs, but not here.

I tried it out on one system. It gave me a login prompt fairly quickly
(maybe five to ten seconds) but what I didn't know is that it was
holding things off in the background. I expected to have all background
services/daemons up and running by the time I was able to log in (i.e.
presented with the login prompt) and that was not the case. A lot of
the boot process was hidden away. I had no idea what was going on.
Took me a while to figure out how to have all the boot messages show up
on the screen again while the boot was in progress.
_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Steve Litt via Ale]

On Fri, 16 Feb 2018 19:00:16 -0500

Post by DJ-Pfulio via Ale

Post by Solomon Peachy via Ale
The average joe wouldn't have noticed any outward difference beyond
faster boot times, <snip>

Exactly when will that happen?
My systems used to boot in 15-25 seconds, usually just under 20s.
$ systemd-analyze
Startup finished in 26.383s (kernel) + 13.483s (userspace) = 39.866s
On systems with static IPs, networking.service takes over 20
seconds!!! Huh?
I keep hearing systemd makes booting faster and I suppose it does for
people with 150 LUNs, but not here.

I personally made a 2 second boot systemd system while writing my
Manjaro Experiments
http://www.troubleshooters.com/linux/init/manjaro_experiments.htm

The best I ever did with runit was about 6 to 8 seconds. However, from
what I read on mailing lists, real systemd systems that do real things
take between 30 seconds and a minute: Similarly practical runit systems
take roughly the same amount of time. I haven't used sysvinit for years,
but as I remember it might have taken even longer than systemd and runit
when used on a practical system.

On any of these systems, if you're very knowledgeable about your init,
you can substantially speed up the boot by parallelizing and/or timing
stuff, and by making sure something like broken reverse dns doesn't
give you timeouts.

SteveT
_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Solomon Peachy via Ale]

Post by DJ-Pfulio via Ale
Exactly when will that happen?
My systems used to boot in 15-25 seconds, usually just under 20s.
$ systemd-analyze
Startup finished in 26.383s (kernel) + 13.483s (userspace) = 39.866s

If the kernel itself is taking more time to boot than the entire system
used to in the past, it seems silly to blame systemd for that.

(And wtf is going on to make the kernel take 26s to hand off to userspace?)

Anectdotally, Fedora 24 on a high-end P700 workstation I have at the
office booted noticably faster from spinning rust than CentOS6 did from
an SSD in the same system, despite having more to start up.

Post by DJ-Pfulio via Ale
On systems with static IPs, networking.service takes over 20 seconds!!!
Huh?

You can drill down deeper into that to see where the time is actually
spent.

network.target on my main server took 22 seconds to start on its last
boot. It's also a fully static setup, but that time includes *every*
network-facing service. Mailman accounted for a ludicrous 10s of that.

- Solomon

[Jonathan Meek via Ale]

I am just going wander over here and drink my coffee because I match the
case that someone pointed out in the email chain about a user. As a user, I
have rarely messed around with checking into what's running in the init
scripts. One failed systemd unit script kept me from messing with it again.
I am pretty that was me not understanding what in the wide world of sports
was going on there.

The biggest issue I have seen with systemd has been its logging into binary
instead of text. It makes me a little nervous because most Linux distros
has some program to view text in the command line (vi, emacs, nano, cat)
while viewing the binaries could be a tricky affair.

Just my thoughts on the matter at hand. Back to my Puerto Rican coffee
sipping.

Jonathan

Post by Solomon Peachy via Ale

Post by DJ-Pfulio via Ale
Exactly when will that happen?
My systems used to boot in 15-25 seconds, usually just under 20s.
$ systemd-analyze
Startup finished in 26.383s (kernel) + 13.483s (userspace) = 39.866s

If the kernel itself is taking more time to boot than the entire system
used to in the past, it seems silly to blame systemd for that.
(And wtf is going on to make the kernel take 26s to hand off to userspace?)
Anectdotally, Fedora 24 on a high-end P700 workstation I have at the
office booted noticably faster from spinning rust than CentOS6 did from
an SSD in the same system, despite having more to start up.

Post by DJ-Pfulio via Ale
On systems with static IPs, networking.service takes over 20 seconds!!!
Huh?

You can drill down deeper into that to see where the time is actually
spent.
network.target on my main server took 22 seconds to start on its last
boot. It's also a fully static setup, but that time includes *every*
network-facing service. Mailman accounted for a ludicrous 10s of that.
- Solomon
--
Solomon Peachy pizza at shaftnet dot org
Coconut Creek, FL ^^ (email/xmpp) ^^
Quidquid latine dictum sit, altum videtur.
_______________________________________________
Ale mailing list
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[James Sumners via Ale]

I'm pretty sure this thread was a joke. I certainly hope the majority were
aware of how to specify the program to run after kernel booting. I'm sure
we've all had to reset a root password that last guy didn't leave us.

Post by Steve Litt via Ale

Post by William Wylde via Ale

Post by Jim Kinney via Ale
heh, heh,.
Edit your grub kernel line and append init=/bin/bash
no systemd.

On February 15, 2018 9:06:05 PM EST, William Wylde

Post by William Wylde via Ale
Do you still get all your drivers & crap loaded that way?

On Thu, 15 Feb 2018 22:01:53 -0500

It's pretty much the same as run level 1. If the initrd loads the
driver, it in.

That's a big "if". Systemd is usually interwoven with the initramfs and
usually implies big, complicated initramfs'.

If it's loaded by a startup process, nope.
So it's a rather crap-free environment.

init=/bin/bash is an old time diagnostic to debug early boot and init.
It's not a substitute for any init system.
One more thing: This is a diagnostic, not a solution for "systemd
haters". And speaking of the phrase "systemd haters", a more accurate
representation would be "lovers of simplicity, modularity, and
functionality."
SteveT
_______________________________________________
Ale mailing list
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Damon L. Chesser via Ale]

What's a root password?  Is that like a certificate thingy?

Post by James Sumners via Ale
I'm pretty sure this thread was a joke. I certainly hope the majority
were aware of how to specify the program to run after kernel booting.
I'm sure we've all had to reset a root password that last guy didn't
leave us.

On Thu, Jan 11, 2018 at 5:11 PM, Jim Kinney via Ale

Post by Jim Kinney via Ale
heh, heh,.
Edit your grub kernel line and append init=/bin/bash
no systemd.

On February 15, 2018 9:06:05 PM EST, William Wylde

Do you still get all your drivers & crap loaded that way?

On Thu, 15 Feb 2018 22:01:53 -0500

It's pretty much the same as run level 1. If the initrd loads the
driver, it in.

That's a big "if". Systemd is usually interwoven with the
initramfs and
usually implies big, complicated initramfs'.

If it's loaded by a startup process, nope.
So it's a rather crap-free environment.

init=/bin/bash is an old time diagnostic to debug early boot and init.
It's not a substitute for any init system.
One more thing: This is a diagnostic, not a solution for "systemd
haters". And speaking of the phrase "systemd haters", a more accurate
representation would be "lovers of simplicity, modularity, and
functionality."
SteveT
_______________________________________________
Ale mailing list
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo
--
James Sumners
http://james.sumners.info/ (technical profile)
http://jrfom.com/ (personal site)
http://haplo.bandcamp.com/ (music)
_______________________________________________
Ale mailing list
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Phil Turmel via Ale]

Yes, it is the first level of certification for landscapers and gardeners.

What's a root password?  Is that like a certificate thingy?

Post by James Sumners via Ale
I'm pretty sure this thread was a joke. I certainly hope the majority
were aware of how to specify the program to run after kernel booting.
I'm sure we've all had to reset a root password that last guy didn't
leave us. 

[Jim Kinney via Ale]

I thought that was network trunking...

Post by Phil Turmel via Ale
Yes, it is the first level of certification for landscapers and
gardeners.

Post by Damon L. Chesser via Ale
What's a root password?  Is that like a certificate thingy?

Post by James Sumners via Ale
I'm pretty sure this thread was a joke. I certainly hope the

majority

Post by Damon L. Chesser via Ale

Post by James Sumners via Ale
were aware of how to specify the program to run after kernel

booting.

Post by Damon L. Chesser via Ale

Post by James Sumners via Ale
I'm sure we've all had to reset a root password that last guy didn't
leave us. 

_______________________________________________
Ale mailing list
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Phil Turmel via Ale]

No, trunking is the second level, and network trunking is the third
level. The root is a prereq for the trunk.

Post by Jim Kinney via Ale
I thought that was network trunking...
Yes, it is the first level of certification for landscapers and gardeners.
What's a root password?  Is that like a certificate thingy?
I'm pretty sure this thread was a joke. I certainly hope the
majority
were aware of how to specify the program to run after kernel
booting.
I'm sure we've all had to reset a root password that last
guy didn't
leave us. 
------------------------------------------------------------------------
Ale mailing list
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo
--
Sent from my Android device with K-9 Mail. All tyopes are thumb related
and reflect authenticity.

[leam hall via Ale]

Very low level. Mostly an underground topic.

Post by Phil Turmel via Ale
Yes, it is the first level of certification for landscapers and gardeners.

What's a root password? Is that like a certificate thingy?

Post by James Sumners via Ale
I'm pretty sure this thread was a joke. I certainly hope the majority
were aware of how to specify the program to run after kernel booting.
I'm sure we've all had to reset a root password that last guy didn't
leave us.

_______________________________________________
Ale mailing list
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[Jim Kinney via Ale]

I'm trying to dig up something on it but the root of the issue is when the trunk failed, they didn't leave anything behind.

Post by leam hall via Ale
Very low level. Mostly an underground topic.

Post by Phil Turmel via Ale
Yes, it is the first level of certification for landscapers and

gardeners.

Post by Phil Turmel via Ale

What's a root password? Is that like a certificate thingy?

Post by James Sumners via Ale
I'm pretty sure this thread was a joke. I certainly hope the

majority

Post by Phil Turmel via Ale

Post by James Sumners via Ale
were aware of how to specify the program to run after kernel

booting.

Post by Phil Turmel via Ale

Post by James Sumners via Ale
I'm sure we've all had to reset a root password that last guy

didn't

Post by Phil Turmel via Ale

Post by James Sumners via Ale
leave us.

_______________________________________________
Ale mailing list
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

_______________________________________________
Ale mailing list
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo

[DJ-Pfulio via Ale]

If you leave it alone long enough, it will rot out and the chipmunks
will move into the tunnels where the trunks and roots used to be. I was
surprised how quickly that rot happened.

Post by Jim Kinney via Ale
I'm trying to dig up something on it but the root of the issue is when
the trunk failed, they didn't leave anything behind.
Very low level. Mostly an underground topic.
Yes, it is the first level of certification for landscapers and gardeners.
What's a root password? Is that like a certificate thingy?
I'm pretty sure this thread was a joke. I certainly hope
the majority
were aware of how to specify the program to run after
kernel booting.
I'm sure we've all had to reset a root password that
last guy didn't
leave us.

_______________________________________________
Ale mailing list
***@ale.org
http://mail.ale.org/mailman/listinfo/ale
See JOBS, ANNOUNCE and SCHOOLS lists at
http://mail.ale.org/mailman/listinfo