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

Groups > alt.folklore.computers > #213902 > unrolled thread

SYSTEMD? EGAD!

Back to article view | Back to alt.folklore.computers

Contents

  SYSTEMD? EGAD! gareth evans <headstone255@yahoo.com> - 2020-09-08 12:08 +0100
    Re: SYSTEMD? EGAD! Niklas Karlsson <anksil@yahoo.se> - 2020-09-08 11:13 +0000
      Re: SYSTEMD? EGAD! J. Clarke <jclarke.873638@gmail.com> - 2020-09-08 07:57 -0400
        Re: SYSTEMD? EGAD! Niklas Karlsson <anksil@yahoo.se> - 2020-09-08 13:27 +0000
          Re: SYSTEMD? EGAD! Douglas Miller <durgadas311@gmail.com> - 2020-09-08 06:40 -0700
            Re: SYSTEMD? EGAD! Thomas Koenig <tkoenig@netcologne.de> - 2020-09-08 16:21 +0000
              Re: SYSTEMD? EGAD! Douglas Miller <durgadas311@gmail.com> - 2020-09-08 09:27 -0700
          Re: SYSTEMD? EGAD! Bob Eager <news0073@eager.cx> - 2020-09-08 14:31 +0000
            Re: SYSTEMD? EGAD! Niklas Karlsson <anksil@yahoo.se> - 2020-09-08 14:34 +0000
        Re: SYSTEMD? EGAD! gareth evans <headstone255@yahoo.com> - 2020-09-08 16:14 +0100
        Re: SYSTEMD? EGAD! antispam@math.uni.wroc.pl - 2020-09-10 11:20 +0000
          Re: SYSTEMD? EGAD! Bob Eager <news0073@eager.cx> - 2020-09-10 11:57 +0000
            Re: SYSTEMD? EGAD! J. Clarke <jclarke.873638@gmail.com> - 2020-09-10 08:23 -0400
            Re: SYSTEMD? EGAD! gareth evans <headstone255@yahoo.com> - 2020-09-10 17:37 +0100
              Re: SYSTEMD? EGAD! Bob Eager <news0073@eager.cx> - 2020-09-10 17:37 +0000
          Re: SYSTEMD? EGAD! J. Clarke <jclarke.873638@gmail.com> - 2020-09-10 08:19 -0400
            Re: SYSTEMD? EGAD! Stefan Möding <Sep2020.5.kill-9@spamgourmet.com> - 2020-09-10 14:32 +0200
              Re: SYSTEMD? EGAD! Dan Espen <dan1espen@gmail.com> - 2020-09-10 09:15 -0400
                Re: SYSTEMD? EGAD! Niklas Karlsson <anksil@yahoo.se> - 2020-09-10 13:30 +0000
                  Re: SYSTEMD? EGAD! Dan Espen <dan1espen@gmail.com> - 2020-09-10 09:34 -0400
                    Re: SYSTEMD? EGAD! Niklas Karlsson <anksil@yahoo.se> - 2020-09-10 13:35 +0000
                      Re: SYSTEMD? EGAD! Dan Espen <dan1espen@gmail.com> - 2020-09-10 10:00 -0400
                        Re: SYSTEMD? EGAD! Douglas Miller <durgadas311@gmail.com> - 2020-09-10 07:15 -0700
                          Re: SYSTEMD? EGAD! Niklas Karlsson <anksil@yahoo.se> - 2020-09-10 14:22 +0000
                          Re: SYSTEMD? EGAD! scott@slp53.sl.home (Scott Lurndal) - 2020-09-10 16:01 +0000
                        Re: SYSTEMD? EGAD! Chris <xxx.syseng.yyy@gfsys.co.uk> - 2020-09-10 17:18 +0100
                          Re: SYSTEMD? EGAD! Andy Leighton <andyl@azaal.plus.com> - 2020-09-10 11:59 -0500
                            Re: SYSTEMD? EGAD! Chris <xxx.syseng.yyy@gfsys.co.uk> - 2020-09-10 18:20 +0100
                              Re: SYSTEMD? EGAD! scott@slp53.sl.home (Scott Lurndal) - 2020-09-10 23:12 +0000
                                Re: SYSTEMD? EGAD! Niklas Karlsson <anksil@yahoo.se> - 2020-09-11 06:35 +0000
                              Re: SYSTEMD? EGAD! Bob Eager <news0073@eager.cx> - 2020-09-10 23:23 +0000
                                Re: SYSTEMD? EGAD! Chris <xxx.syseng.yyy@gfsys.co.uk> - 2020-09-11 01:11 +0100
                                Re: SYSTEMD? EGAD! JimP <chucktheouch@gmail.com> - 2020-09-11 09:11 -0500
                      Re: SYSTEMD? EGAD! Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2020-09-10 16:24 +0000
                        Re: SYSTEMD? EGAD! Niklas Karlsson <anksil@yahoo.se> - 2020-09-10 17:03 +0000
                        Re: SYSTEMD? EGAD! Andreas Kohlbach <ank@spamfence.net> - 2020-09-10 17:55 -0400
                        Re: SYSTEMD? EGAD! Mike Spencer <mds@bogus.nodomain.nowhere> - 2020-09-11 02:12 -0300
                          Re: SYSTEMD? EGAD! Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2020-09-11 20:41 +0000
                            Re: SYSTEMD? EGAD! Bob Eager <news0073@eager.cx> - 2020-09-11 21:11 +0000
                              Re: SYSTEMD? EGAD! Chris <xxx.syseng.yyy@gfsys.co.uk> - 2020-09-12 15:06 +0100
                                Re: SYSTEMD? EGAD! Bob Eager <news0073@eager.cx> - 2020-09-12 16:03 +0000
                                  Re: SYSTEMD? EGAD! Chris <xxx.syseng.yyy@gfsys.co.uk> - 2020-09-26 16:05 +0100
                        Re: SYSTEMD? EGAD! Jorgen Grahn <grahn+nntp@snipabacken.se> - 2020-09-25 23:15 +0000
                      Re: SYSTEMD? EGAD! Melzzzzz <Melzzzzz@zzzzz.com> - 2020-09-10 23:55 +0000
                        Re: SYSTEMD? EGAD! Stefan Möding <Sep2020.5.kill-9@spamgourmet.com> - 2020-09-11 08:09 +0200
                    Re: SYSTEMD? EGAD! Melzzzzz <Melzzzzz@zzzzz.com> - 2020-09-10 23:54 +0000
                Re: SYSTEMD? EGAD! Stefan Möding <Sep2020.5.kill-9@spamgourmet.com> - 2020-09-10 15:54 +0200
                  Re: SYSTEMD? EGAD! Dan Espen <dan1espen@gmail.com> - 2020-09-10 10:14 -0400
                  Re: SYSTEMD? EGAD! Melzzzzz <Melzzzzz@zzzzz.com> - 2020-09-10 23:57 +0000
                Re: SYSTEMD? EGAD! scott@slp53.sl.home (Scott Lurndal) - 2020-09-10 15:56 +0000
              Re: SYSTEMD? EGAD! Peter Flass <peter_flass@yahoo.com> - 2020-09-10 09:53 -0700
              Re: SYSTEMD? EGAD! Jorgen Grahn <grahn+nntp@snipabacken.se> - 2020-09-25 23:03 +0000
                Re: SYSTEMD? EGAD! Niklas Karlsson <anksil@yahoo.se> - 2020-09-26 08:35 +0000
                  Re: SYSTEMD? EGAD! scott@slp53.sl.home (Scott Lurndal) - 2020-09-26 18:37 +0000
                Re: SYSTEMD? EGAD! Stefan Möding <Sep2020.5.kill-9@spamgourmet.com> - 2020-09-26 17:41 +0200
            Re: SYSTEMD? EGAD! antispam@math.uni.wroc.pl - 2020-09-10 16:09 +0000
              Re: SYSTEMD? EGAD! J. Clarke <jclarke.873638@gmail.com> - 2020-09-10 12:17 -0400
                Re: SYSTEMD? EGAD! Peter Flass <peter_flass@yahoo.com> - 2020-09-10 09:53 -0700
                  Re: SYSTEMD? EGAD! Thomas Koenig <tkoenig@netcologne.de> - 2020-09-11 13:22 +0000
                    Re: SYSTEMD? EGAD! ted@loft.tnolan.com (Ted Nolan <tednolan>) - 2020-09-11 13:39 +0000
                      Re: SYSTEMD? EGAD! Bob Eager <news0073@eager.cx> - 2020-09-11 14:11 +0000
                    Re: SYSTEMD? EGAD! Mike Spencer <mds@bogus.nodomain.nowhere> - 2020-09-11 20:24 -0300
                Re: SYSTEMD? EGAD! scott@slp53.sl.home (Scott Lurndal) - 2020-09-10 17:21 +0000
                Re: SYSTEMD? EGAD! Bob Eager <news0073@eager.cx> - 2020-09-10 17:36 +0000
              Re: SYSTEMD? EGAD! Bob Eager <news0073@eager.cx> - 2020-09-10 17:34 +0000
          Re: SYSTEMD? EGAD! Ahem A Rivet's Shot <steveo@eircom.net> - 2020-09-10 17:14 +0100
          Re: SYSTEMD? EGAD! Peter Flass <peter_flass@yahoo.com> - 2020-09-10 09:53 -0700
            Re: SYSTEMD? EGAD! Niklas Karlsson <anksil@yahoo.se> - 2020-09-10 17:05 +0000
              Re: SYSTEMD? EGAD! Stefan Möding <Sep2020.5.kill-9@spamgourmet.com> - 2020-09-10 19:13 +0200
                Re: SYSTEMD? EGAD! ted@loft.tnolan.com (Ted Nolan <tednolan>) - 2020-09-10 19:19 +0000
              Re: SYSTEMD? EGAD! Bob Eager <news0073@eager.cx> - 2020-09-10 17:39 +0000
                Re: SYSTEMD? EGAD! Ahem A Rivet's Shot <steveo@eircom.net> - 2020-09-10 20:25 +0100
                  Re: SYSTEMD? EGAD! Bob Eager <news0073@eager.cx> - 2020-09-10 20:23 +0000
                Re: SYSTEMD? EGAD! Jorgen Grahn <grahn+nntp@snipabacken.se> - 2020-09-26 05:48 +0000
                  Re: SYSTEMD? EGAD! Bob Eager <news0073@eager.cx> - 2020-09-26 11:24 +0000
                    Re: SYSTEMD? EGAD! Jorgen Grahn <grahn+nntp@snipabacken.se> - 2020-09-26 16:14 +0000
                      Re: SYSTEMD? EGAD! Bob Eager <news0073@eager.cx> - 2020-09-26 20:53 +0000
                  Re: SYSTEMD? EGAD! maus <maus@dmaus.org> - 2020-09-26 12:44 +0000
                    Re: SYSTEMD? EGAD! Bob Eager <news0073@eager.cx> - 2020-09-26 13:03 +0000
    Re: SYSTEMD? EGAD! Douglas Miller <durgadas311@gmail.com> - 2020-09-08 06:04 -0700
      Re: SYSTEMD? EGAD! gareth evans <headstone255@yahoo.com> - 2020-09-08 16:16 +0100
        Re: SYSTEMD? EGAD! gareth evans <headstone255@yahoo.com> - 2020-09-08 16:22 +0100
        Re: SYSTEMD? EGAD! Bob Eager <news0073@eager.cx> - 2020-09-08 15:40 +0000
          Re: SYSTEMD? EGAD! gareth evans <headstone255@yahoo.com> - 2020-09-08 19:42 +0100
            Re: SYSTEMD? EGAD! gareth evans <headstone255@yahoo.com> - 2020-09-08 19:52 +0100
              Re: SYSTEMD? EGAD! Bob Eager <news0073@eager.cx> - 2020-09-08 19:38 +0000
                Re: SYSTEMD? EGAD! gareth evans <headstone255@yahoo.com> - 2020-09-08 21:02 +0100

Page 3 of 5 — ← Prev page 1 2 [3] 4 5  Next page →

#214192

On Sat, 12 Sep 2020 15:06:10 +0100, Chris wrote:

> On 09/11/20 22:11, Bob Eager wrote:

>> I have a hardback A4 log book for each computer. And I keep the old
>> ones.

> Similar thing here, 4 ring binders, indexed, tab for each system.
> Different ring binder for each class of system worked with.
> Some hand scribbled and some properly typed records of systems and
> configurations, experiments etc, but have been caught out by being a
> little too terse at times...

Yes, I end up sticking in small printed bits sometimes.

I've also started putting non-trivial config files into a subversion 
repository.

-- 
Using UNIX since v6 (1975)...

Use the BIG mirror service in the UK:
 http://www.mirrorservice.org

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

#214579

On 09/12/20 17:03, Bob Eager wrote:
> On Sat, 12 Sep 2020 15:06:10 +0100, Chris wrote:
>
>> On 09/11/20 22:11, Bob Eager wrote:
>
>>> I have a hardback A4 log book for each computer. And I keep the old
>>> ones.
>
>> Similar thing here, 4 ring binders, indexed, tab for each system.
>> Different ring binder for each class of system worked with.
>> Some hand scribbled and some properly typed records of systems and
>> configurations, experiments etc, but have been caught out by being a
>> little too terse at times...
>
> Yes, I end up sticking in small printed bits sometimes.
>
> I've also started putting non-trivial config files into a subversion
> repository.
>

That's a good idea. I have a backup raid array that has a
/machines directory, then subdirs for just about every system
used here in decades. Each machine has it's subdirs, with
config, notes, drivers and backups saved there. It could be a
single point of failure, but have a second array to rsync
everything to every so often. Both arrays only get powered up
when needed.

Only thing missing from the archive section is the data from the
Apple II days, but need to find a way to read a stack of
A2 floppies to retrieve that...

Chris

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

#214569

On Thu, 2020-09-10, Charlie Gibbs wrote:
> On 2020-09-10, Niklas Karlsson <anksil@yahoo.se> wrote:
>
>> On 2020-09-10, Dan Espen <dan1espen@gmail.com> wrote:
>>
>>> Are you sure there isn't some GUI to go into and just click on static
>>> IP?  That would seem rather elementary for any distro.
>>
>> No doubt there is, but most servers don't have X.
>
> Who needs X?  For me, the one must-have when setting up a new system
> is a second computer with a working network interface and web browser.
> Then I can look up the instructions for setting up /etc/network/interfaces.

"man 5 interfaces" on the system you're setting up should also be
enough.  You need something like two lines for DHCP or IPv6, or four
lines for static IPv4.

/Jorgen

-- 
  // Jorgen Grahn <grahn@  Oo  o.   .     .
\X/     snipabacken.se>   O  o   .

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

#214128

On 2020-09-10, Niklas Karlsson <anksil@yahoo.se> wrote:
> On 2020-09-10, Dan Espen <dan1espen@gmail.com> wrote:
>>
>> Are you sure there isn't some GUI to go into and just click on static
>> IP?  That would seem rather elementary for any distro.
>
> No doubt there is, but most servers don't have X.

Exactly. And configuring it ain't rocket science as well :)

>
> Niklas


-- 
current job title: senior software engineer
skills: c++,c,rust,go,nim,haskell...

press any key to continue or any other to quit...
U ničemu ja ne uživam kao u svom statusu INVALIDA -- Zli Zec
Svi smo svedoci - oko 3 godine intenzivne propagande je dovoljno da jedan narod poludi --  Zli Zec
Na divljem zapadu i nije bilo tako puno nasilja, upravo zato jer su svi 
bili naoruzani. -- Mladen Gogala

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

#214138

Melzzzzz <Melzzzzz@zzzzz.com> writes:

> On 2020-09-10, Niklas Karlsson <anksil@yahoo.se> wrote:
>> On 2020-09-10, Dan Espen <dan1espen@gmail.com> wrote:
>>>
>>> Are you sure there isn't some GUI to go into and just click on static
>>> IP?  That would seem rather elementary for any distro.
>>
>> No doubt there is, but most servers don't have X.

> Exactly. And configuring it ain't rocket science as well :)

I’ve worked at a place where many Linux servers were running X only because
some sysadmin was unable to do the job without all the GUI tools.

-- 
Stefan

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

#214127

On 2020-09-10, Dan Espen <dan1espen@gmail.com> wrote:
> Niklas Karlsson <anksil@yahoo.se> writes:
>
>> On 2020-09-10, Dan Espen <dan1espen@gmail.com> wrote:
>>> Stefan Möding <Sep2020.5.kill-9@spamgourmet.com> writes:
>>>> I consider Linux today almost as fragmented as Unix was 20-30 years ago.
>>>> Running mostly on one or two CPU architectures seems to make it better now.
>>>> But if you look at the differences in setting up network interfaces for
>>>> example:
>> [various solutions]
>>>
>>> Huh?
>>>
>>> You just install it and your network works.
>>> What is this "setting up network interfaces"
>>> you speak of?
>>
>> That's fine for a workstation with DHCP available on the network. For
>> servers, yes, you can use DHCP (likely with a particular address/MAC
>> pair configured on the DHCP server), but so far as I've seen, the most
>> common solution by far is to simply configure a static IP locally on
>> your server.
>>
>> That requires digging rather deeper into whatever network configuration
>> solution your distro is using at the moment.
>
> Are you sure there isn't some GUI to go into and just click on static
> IP?  That would seem rather elementary for any distro.

If you are doing it on server, you probably don't want GUI...
>
>


-- 
current job title: senior software engineer
skills: c++,c,rust,go,nim,haskell...

press any key to continue or any other to quit...
U ničemu ja ne uživam kao u svom statusu INVALIDA -- Zli Zec
Svi smo svedoci - oko 3 godine intenzivne propagande je dovoljno da jedan narod poludi --  Zli Zec
Na divljem zapadu i nije bilo tako puno nasilja, upravo zato jer su svi 
bili naoruzani. -- Mladen Gogala

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

#214063

Dan Espen <dan1espen@gmail.com> writes:

> You just install it and your network works.
> What is this "setting up network interfaces"
> you speak of?

I need to do do automatic installations & configurations for multiple distros
and I wish there would be a uniform way.


> Apparently 'systemd network config" is an example of Linux
> doing better.

> Excuse me I need to put on my asbestos suit now.

I like the systemd way of network configuration.  After getting rid of some
legacy machine I hoped for doing it the systemd way.

Unfortunately RedHat (I believe they pay for much of the systemd work) decided
to not support systemd network configuration for RedHat-8.  The binary is not
available for RedHat-8/CentOS-8 (unless you compile it yourself).

-- 
Stefan

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

#214065

Stefan Möding <Sep2020.5.kill-9@spamgourmet.com> writes:

> Dan Espen <dan1espen@gmail.com> writes:
>
>> You just install it and your network works.
>> What is this "setting up network interfaces"
>> you speak of?
>
> I need to do do automatic installations & configurations for multiple distros
> and I wish there would be a uniform way.
>
>> Apparently 'systemd network config" is an example of Linux
>> doing better.
>
>> Excuse me I need to put on my asbestos suit now.
>
> I like the systemd way of network configuration.  After getting rid of some
> legacy machine I hoped for doing it the systemd way.
>
> Unfortunately RedHat (I believe they pay for much of the systemd work) decided
> to not support systemd network configuration for RedHat-8.  The binary is not
> available for RedHat-8/CentOS-8 (unless you compile it yourself).

Well seems like a necessary consequence of changing things to make them better.

Looks like you can get a package, but not from Redhat:

https://bugzilla.redhat.com/show_bug.cgi?id=1650342

-- 
Dan Espen

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

#214129

On 2020-09-10, Stefan Möding <Sep2020.5.kill-9@spamgourmet.com> wrote:
> Dan Espen <dan1espen@gmail.com> writes:
>
>> You just install it and your network works.
>> What is this "setting up network interfaces"
>> you speak of?
>
> I need to do do automatic installations & configurations for multiple distros
> and I wish there would be a uniform way.

You have ifconfig and ip...
>
>
>> Apparently 'systemd network config" is an example of Linux
>> doing better.
>
>> Excuse me I need to put on my asbestos suit now.
>
> I like the systemd way of network configuration.  After getting rid of some
> legacy machine I hoped for doing it the systemd way.
>
> Unfortunately RedHat (I believe they pay for much of the systemd work) decided
> to not support systemd network configuration for RedHat-8.  The binary is not
> available for RedHat-8/CentOS-8 (unless you compile it yourself).
>


-- 
current job title: senior software engineer
skills: c++,c,rust,go,nim,haskell...

press any key to continue or any other to quit...
U ničemu ja ne uživam kao u svom statusu INVALIDA -- Zli Zec
Svi smo svedoci - oko 3 godine intenzivne propagande je dovoljno da jedan narod poludi --  Zli Zec
Na divljem zapadu i nije bilo tako puno nasilja, upravo zato jer su svi 
bili naoruzani. -- Mladen Gogala

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

#214072

Dan Espen <dan1espen@gmail.com> writes:
>Stefan Möding <Sep2020.5.kill-9@spamgourmet.com> writes:
>
>> J. Clarke <jclarke.873638@gmail.com> writes:
>>
>>> Or the Unix world would have continued to be fragmented.
>>
>> I consider Linux today almost as fragmented as Unix was 20-30 years ago.
>> Running mostly on one or two CPU architectures seems to make it better now.
>> But if you look at the differences in setting up network interfaces for
>> example:
>>
>> - legacy config files
>> - systemd network config 
>> - netplan
>> - NetworkManager
>> - ...?
>>
>
>Huh?
>
>You just install it and your network works.
>What is this "setting up network interfaces"
>you speak of?

Well, one of my linux boxes has a trunking connection
to a managed switch which carries two VLANs.  There
are four virtual machines on that linux system for which
one of the VLANs is a DMZ containing internet-facing
web servers and mail relays.  The other VLAN is internal.

You can't set up that configuration using the NetworkManager
or systemd crap.

With pre-systemd, it was adding four or five
environment variables to /etc/sysconfig/network-scripts/ifup-eth0.

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

#214086

Stefan Möding <Sep2020.5.kill-9@spamgourmet.com> wrote:
> J. Clarke <jclarke.873638@gmail.com> writes:
> 
>> Or the Unix world would have continued to be fragmented.
> 
> I consider Linux today almost as fragmented as Unix was 20-30 years ago.
> Running mostly on one or two CPU architectures seems to make it better now.
> But if you look at the differences in setting up network interfaces for
> example:
> 
> - legacy config files
> - systemd network config 
> - netplan
> - NetworkManager
> - ...?
> 
> Linux could have done better.
> 

Agreed. As I’ve said elsewhere, everyone thinks they have a better of how
to do things. Maybe they do, but sometimes consistent beats better.

-- 
Pete

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

#214568

On Thu, 2020-09-10, Stefan Möding wrote:
> J. Clarke <jclarke.873638@gmail.com> writes:
>
>> Or the Unix world would have continued to be fragmented.
>
> I consider Linux today almost as fragmented as Unix was 20-30 years ago.
> Running mostly on one or two CPU architectures seems to make it better now.

Why would the CPU be an issue?  I used to run Debian on x86, AMD64 and
PowerPC, and of course they worked exactly the same way.  Free
software tends not to have endianness bugs, or 32/64-bit bugs.

(In the past few years though, there's some kind of force away from
big-endian.  I haven't fully understood what's behind that, apart the
fact that big-endian architectures are less successful right now.)

/Jorgen

-- 
  // Jorgen Grahn <grahn@  Oo  o.   .     .
\X/     snipabacken.se>   O  o   .

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

#214573

On 2020-09-25, Jorgen Grahn <grahn+nntp@snipabacken.se> wrote:
> (In the past few years though, there's some kind of force away from
> big-endian.  I haven't fully understood what's behind that, apart the
> fact that big-endian architectures are less successful right now.)

Are ARM Macs going to be big- or little-endian, I wonder?

Niklas
-- 
Must confess, i'd feel just a little bit
conspicuous ordering a large Espresso and an enema-bag at my local
coffee-house... 
        -- Tanuki in asr

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

#214589

Niklas Karlsson <anksil@yahoo.se> writes:
>On 2020-09-25, Jorgen Grahn <grahn+nntp@snipabacken.se> wrote:
>> (In the past few years though, there's some kind of force away from
>> big-endian.  I haven't fully understood what's behind that, apart the
>> fact that big-endian architectures are less successful right now.)
>
>Are ARM Macs going to be big- or little-endian, I wonder?

Little, however, ARM Aarch64 supports running both LE and
BE applications while the OS and Firmware are LE.

Whether Apple supports bi-endian in OSX is up to them.

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

#214582

Jorgen Grahn <grahn+nntp@snipabacken.se> writes:

> On Thu, 2020-09-10, Stefan Möding wrote:

>> I consider Linux today almost as fragmented as Unix was 20-30 years ago.
>> Running mostly on one or two CPU architectures seems to make it better now.

> Why would the CPU be an issue?  I used to run Debian on x86, AMD64 and
> PowerPC, and of course they worked exactly the same way.  Free
> software tends not to have endianness bugs, or 32/64-bit bugs.

If I remember correctly the Jave JRE contains some non-trivial assembler
code for performance reasons.  Doing this once for AMD64 was a big win for
the majority of Linux users no matter what ditro as long as they used
AMD64.

I tried to upgrade an Puppet 3 environment to Puppet 4 some years ago.
The software is free but the build system needed some binary-only library
maintained by Google.  No luck for Debian on S/390...

But yes, endian or 64-bit bugs are probably gone by now.

-- 
Stefan

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

#214075

J. Clarke <jclarke.873638@gmail.com> wrote:
> On Thu, 10 Sep 2020 11:20:04 +0000 (UTC), antispam@math.uni.wroc.pl
> wrote:
> 
> >J. Clarke <jclarke.873638@gmail.com> wrote:
> >> On 8 Sep 2020 11:13:43 GMT, Niklas Karlsson <anksil@yahoo.se> wrote:
> >> 
> >> >On 2020-09-08, gareth evans <headstone255@yahoo.com> wrote:
> >> >>
> >> >> My impression, clearly quite wrong, was that LINUX
> >> >> was little more than a vehicle for launching programs,
> >> >> ie, an uppity MSDOS, but with provision for multi-tasking
> >> >> (multi-programming?)
> >> >
> >> >Not to mention vastly better memory management.
> >> 
> >> Calling LINUX "an uppity MSDOS", is kind of like calling MSDOS a
> >> "crippled Unix".
> >> 
> >> LINUX provided pretty much the full functionality of the UNIX kernel,
> >> which had up until that time been the missing piece for an open source
> >> UNIX clone.  Prior to that time, any UNIX implementation that was full
> >> enough to be useful for anything beyond recreational and student use
> >> used AT&T licensed code and the license could be quite costly.
> >
> >Before Linux were Minix,
> 
> Linux started out as a rewrite of Minix to address its shortcomings.

This is rather inprecise.  Minix had its shortcomings.  I am not
aware of serious use of Minix, but my lack of info does not
prove that there were none.  Certainly was not replacement for
UNIX(TM) but was solid enough to usable in _some_ serious setups.
 
> > Coherent 
> 
> Was a proprietary commercial product.  It did not become open source
> until 2015, long, long after Linux became well established.

Well, you wrote about AT&T, Coherent was independent developement.
And quite a bit cheaper.
 
> >and 386BSD.
> 
> I was not aware that 386BSD was unencumbered.

You are aware of AT&T versus Berkeley process, arent you?  386BSD
contained code that was released under BSD style licence, most
came from official Berkeley release.
 
> >Your claim
> >probably is about Minux, but even in this case you exagerate.
> 
> So would you have tried to run Amazon on Minix?

There are applications intermediate between running Amazon and
"recreational and student use".  One resonable use would be
for lab equipement (AFAIK intersting intrument used to
run UNIX(TM) at that time).  Or a machine for few users.

> >I used 386BSD and it was quite buggy, but had all what is
> >needed to run wide variety of useful programs.  I never used
> >Coherent, but given descriptions it is pretty clear that
> >that Coherent had much more functionality than Linux-0.01.
> 
> >The difference was that Linux and 386BSD had free licence and
> >consequently potential to grow. 
> 
> What part of "open source" are you having trouble with?

I have no trouble with open source.  You seem to have trouble
with reading or writing.  Is it that hard to write
"I agree" if it is what you mean?

Or maybe you just look for opportunity to start a flame war?

> >One can speculate why Linux
> >developed faster than 386BSD (and successors like Free BSD):
> >- creator of 386BSD after free realease wanted to switch
> >  to proprietary licence
> >- famus ATT versus Berkely process
> >- GPL
> >but without Linux we would probably live in FreeBSD world.
> 
> Or the Unix world would have continued to be fragmented.

-- 
                              Waldek Hebisch

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

#214076

On Thu, 10 Sep 2020 16:09:13 +0000 (UTC), antispam@math.uni.wroc.pl
wrote:

>J. Clarke <jclarke.873638@gmail.com> wrote:
>> On Thu, 10 Sep 2020 11:20:04 +0000 (UTC), antispam@math.uni.wroc.pl
>> wrote:
>> 
>> >J. Clarke <jclarke.873638@gmail.com> wrote:
>> >> On 8 Sep 2020 11:13:43 GMT, Niklas Karlsson <anksil@yahoo.se> wrote:
>> >> 
>> >> >On 2020-09-08, gareth evans <headstone255@yahoo.com> wrote:
>> >> >>
>> >> >> My impression, clearly quite wrong, was that LINUX
>> >> >> was little more than a vehicle for launching programs,
>> >> >> ie, an uppity MSDOS, but with provision for multi-tasking
>> >> >> (multi-programming?)
>> >> >
>> >> >Not to mention vastly better memory management.
>> >> 
>> >> Calling LINUX "an uppity MSDOS", is kind of like calling MSDOS a
>> >> "crippled Unix".
>> >> 
>> >> LINUX provided pretty much the full functionality of the UNIX kernel,
>> >> which had up until that time been the missing piece for an open source
>> >> UNIX clone.  Prior to that time, any UNIX implementation that was full
>> >> enough to be useful for anything beyond recreational and student use
>> >> used AT&T licensed code and the license could be quite costly.
>> >
>> >Before Linux were Minix,
>> 
>> Linux started out as a rewrite of Minix to address its shortcomings.
>
>This is rather inprecise.  Minix had its shortcomings.  I am not
>aware of serious use of Minix, but my lack of info does not
>prove that there were none.  Certainly was not replacement for
>UNIX(TM) but was solid enough to usable in _some_ serious setups.

In what way is it imprecise?  Linus Torvalds had Minix.  He wasn't
happy with it, he set out to redo the kernel, he did, what he came up
with was a significant improvent, and so we have Linux.
 
>> > Coherent 
>> 
>> Was a proprietary commercial product.  It did not become open source
>> until 2015, long, long after Linux became well established.
>
>Well, you wrote about AT&T, Coherent was independent developement.
>And quite a bit cheaper.

For certain values.  It sold for something like 500 bucks.  That's
about what I paid for System V for a 286.
 
>> >and 386BSD.
>> 
>> I was not aware that 386BSD was unencumbered.
>
>You are aware of AT&T versus Berkeley process, arent you?  386BSD
>contained code that was released under BSD style licence, most
>came from official Berkeley release.

I am aware of multiple lawsuits between AT&T and Berkeley.  That BSD
did something does not mean that what they did was legal, as they
found out the hard way.
 
>> >Your claim
>> >probably is about Minux, but even in this case you exagerate.
>> 
>> So would you have tried to run Amazon on Minix?
>
>There are applications intermediate between running Amazon and
>"recreational and student use".  One resonable use would be
>for lab equipement (AFAIK intersting intrument used to
>run UNIX(TM) at that time).  Or a machine for few users.

Well that's the thing, Linux is robust enough to run Amazon or Google.
Minix never was.

>> >I used 386BSD and it was quite buggy, but had all what is
>> >needed to run wide variety of useful programs.  I never used
>> >Coherent, but given descriptions it is pretty clear that
>> >that Coherent had much more functionality than Linux-0.01.
>> 
>> >The difference was that Linux and 386BSD had free licence and
>> >consequently potential to grow. 
>> 
>> What part of "open source" are you having trouble with?
>
>I have no trouble with open source.  You seem to have trouble
>with reading or writing.  Is it that hard to write
>"I agree" if it is what you mean?

Pot, kettle.  You started out disagreeing with me, now I'm supposed to
agree with your disagreement?

>Or maybe you just look for opportunity to start a flame war?

Et tu?
>
>> >One can speculate why Linux
>> >developed faster than 386BSD (and successors like Free BSD):
>> >- creator of 386BSD after free realease wanted to switch
>> >  to proprietary licence
>> >- famus ATT versus Berkely process
>> >- GPL
>> >but without Linux we would probably live in FreeBSD world.
>> 
>> Or the Unix world would have continued to be fragmented.

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

#214087

J. Clarke <jclarke.873638@gmail.com> wrote:
> On Thu, 10 Sep 2020 16:09:13 +0000 (UTC), antispam@math.uni.wroc.pl
> wrote:
> 
>> J. Clarke <jclarke.873638@gmail.com> wrote:
>>> On Thu, 10 Sep 2020 11:20:04 +0000 (UTC), antispam@math.uni.wroc.pl
>>> wrote:
>>> 
>>>> J. Clarke <jclarke.873638@gmail.com> wrote:
>>>>> On 8 Sep 2020 11:13:43 GMT, Niklas Karlsson <anksil@yahoo.se> wrote:
>>>>> 
>>>>>> On 2020-09-08, gareth evans <headstone255@yahoo.com> wrote:
>>>>>>> 
>>>>>>> My impression, clearly quite wrong, was that LINUX
>>>>>>> was little more than a vehicle for launching programs,
>>>>>>> ie, an uppity MSDOS, but with provision for multi-tasking
>>>>>>> (multi-programming?)
>>>>>> 
>>>>>> Not to mention vastly better memory management.
>>>>> 
>>>>> Calling LINUX "an uppity MSDOS", is kind of like calling MSDOS a
>>>>> "crippled Unix".
>>>>> 
>>>>> LINUX provided pretty much the full functionality of the UNIX kernel,
>>>>> which had up until that time been the missing piece for an open source
>>>>> UNIX clone.  Prior to that time, any UNIX implementation that was full
>>>>> enough to be useful for anything beyond recreational and student use
>>>>> used AT&T licensed code and the license could be quite costly.
>>>> 
>>>> Before Linux were Minix,
>>> 
>>> Linux started out as a rewrite of Minix to address its shortcomings.
>> 
>> This is rather inprecise.  Minix had its shortcomings.  I am not
>> aware of serious use of Minix, but my lack of info does not
>> prove that there were none.  Certainly was not replacement for
>> UNIX(TM) but was solid enough to usable in _some_ serious setups.
> 
> In what way is it imprecise?  Linus Torvalds had Minix.  He wasn't
> happy with it, he set out to redo the kernel, he did, what he came up
> with was a significant improvent, and so we have Linux.

The vast improvements in hardware made the difference. I think I first
looked at Minix on an XT where it barely fit. I think by the time Linux
became popular most machines were Pentium or better, with a lot more RAM
and disk.

-- 
Pete

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

#214142

Peter Flass <peter_flass@yahoo.com> schrieb:
> I think by the time Linux
> became popular most machines were Pentium or better, with a lot more RAM
> and disk.

I fist ran Linux on a 486 (I think) from the first version of
Slackware ever, downloaded via the the only PC in the computer
center that could connect to the mainframe, which was the only
way into the Internet.

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

#214143

In article <rjftmu$2ng$1@newsreader4.netcologne.de>,
Thomas Koenig  <tkoenig@netcologne.de> wrote:
>Peter Flass <peter_flass@yahoo.com> schrieb:
>> I think by the time Linux
>> became popular most machines were Pentium or better, with a lot more RAM
>> and disk.
>
>I fist ran Linux on a 486 (I think) from the first version of
>Slackware ever, downloaded via the the only PC in the computer
>center that could connect to the mainframe, which was the only
>way into the Internet.

Of course the epochal V7 ran in 64K..

Interestingly IBM had a commercial System-III based Unix called, I think,
PC-IX that ran on the 8088 IBM-PC.  You did have to have a hard drive.
It fudged memory protection with the segment registers, and if you compiled
your programs with the provided C compiler, you were "unlikely" to be able
to crash the system with a user program.

I wonder how many copies they actually sold?
-- 
columbiaclosings.com
What's not in Columbia anymore..

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

Page 3 of 5 — ← Prev page 1 2 [3] 4 5  Next page →

Back to top | Article view | alt.folklore.computers

csiph-web