> google is compiled into your operating system. > Th

File: 1481916546254.jpg (47KB, 645x968px) Image search: [Google]

47KB, 645x968px

>>62376270
>google is compiled into your operating system.

>>62376270
You don't know how DNS servers work

>>62376289
> doesnt configure dns
> dns queries get sent to google

>>62376323
Yes, Google DNS servers are used by many users, companies and ISPs themselves. You are a misguided fucktard.

>>62376352

> I send everything I can to google

t. google shill

>>62376352
Just because loads of people use something doesn't mean everyone should you fucking retard.

>>62376370
Avoiding google is not entirely possible. You are free to try.

I wonder how you are posting in 4chan, you really didn't buy the pass, did you?

>>62376394
Which is why you are free to tweak it. What's your point?

>>62376397
Bought a pass with Bitcoin, blocking Google with uMatrix. Nice try kid.

>>62376411
Prove it then bitch, show your pass icon

>>62376411
>Bought a pass with Bitcoin,
LMAO nice job, fuckwit. Now you are on "the list"

File: 1461192251151.jpg (446KB, 1000x664px) Image search: [Google]

446KB, 1000x664px

>>62376397
> not hosting your own dns server
> not querying root dns directly.

systemd was a mistake.

>>62376411
nice mang. Congratulations on fighting the botnet.

>>62376270
Is this real

File: lennart-poettering.jpg (239KB, 1280x960px) Image search: [Google]

239KB, 1280x960px

>>62376456
bro I linked the github issue. click it, it's halfway down and comes from harry potter himself.

>>62376456
man resolved.conf
check the fallback dns

>>62376471
better hope your dns servers dont go "down" and systemd "remembers" for you. Otherwise its straight to big brother google for your poor dns queries.

>>62376270
FUCK YOU POETTERING

File: 2130732.jpg (22KB, 460x460px) Image search: [Google]

22KB, 460x460px

>>62376456
> Jun 7 10:37:38 t470 systemd-resolved[2161]: Switching to fallback DNS server 8.8.8.8.

Is it possible to switch from system D on an existing install, or does one need to reinstall the entire system?

>>62376531
what distribution are you using fellow concerned citizen?

>>62376557
Ubuntu 16 with unity, currently

File: linus-torvalds.jpg (39KB, 770x578px) Image search: [Google]

39KB, 770x578px

>>62376567
this might help you, but i'd give it a good look before diving in.

https://medium.com/@gdm85/runit-as-your-init-on-ubuntu-16-xenial-55d18513aac0


otherwise:

http://without-systemd.org/wiki/index.php/Main_Page

>>62376585
Much appreciated

File: 1412774805965.png (213KB, 700x700px) Image search: [Google]

213KB, 700x700px

>>62376270
How do you guys not trust this face?

>just configure it XD
way to miss the point. Lennart has coded it to silently fail and fallback to Google's botnet DNS. Unless you pay attention and scour the log files, you may NEVER know that you've got a problem with your DNS. Before systemd you would have had no working DNS -> easy to notice and fix. But no no no.. Lenny knows better, let uncle Lenny set it up for you

>>62376456
You looking at the screen and deciphering OPs message from a series of pixels on the screen and then being confused at the message is real.
What's also real is how retarded and lazy you are. Look it up yourself, never believe anything you hear on 4chan.

File: 1505156301223.png (260KB, 483x368px) Image search: [Google]

260KB, 483x368px

>>62376653
>>62376456

Its in the link in the OP

The github issue might be worth archiving.

>>62376270
Pootering really is just a normie with C knowledge who has no concept of privacy or security, and just wants to turn Linux into OSX.
He doesn't give a shit about who gets sent what data as long as it "Just Works™" in the end.

What was wrong with resolv.conf?

>>62376669
>archiving
This is old news
https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/1449001
>reported on 2015-04-27

>>62376742
my b

>>62376719
>>62376719
resolve.conf is comfy imo. Using it right now to point at a DNS i have running inside a docker container.

captcha: bari safety

File: 1396645988903.png (218KB, 743x625px) Image search: [Google]

218KB, 743x625px

hes banned from the kernel, he should be banned from your machine.

File: index.png (5KB, 225x225px) Image search: [Google]

5KB, 225x225px

one last bump, gnite folks

File: 1504342639548.png (429KB, 571x420px) Image search: [Google]

429KB, 571x420px

>There are people still using systemd distros in 2017

>>62377078
99% of /g/ uses sysemd

>>62376270
poettering is a CIA shill

File: merkel.png (86KB, 300x300px) Image search: [Google]

86KB, 300x300px

>>62376270

>>62376352
I blocked 8.8.8.8 using iptables now

>>62376411
>Bitcoin
>Dubs
Why is this allow, criminal?

>>62376644

> Unless you pay attention and scour the log files, you may NEVER know that you've got a problem with your DNS
Yes, clearly someone advanced enough to use systemd-resolvd won't look at its configuration first to edit it before enabling it.
How are you still alive?

#  This file is part of systemd.
#
#  systemd is free software; you can redistribute it and/or modify it
#  under the terms of the GNU Lesser General Public License as published by
#  the Free Software Foundation; either version 2.1 of the License, or
#  (at your option) any later version.
#
# Entries in this file show the compile time defaults.
# You can change settings by editing this file.
# Defaults can be restored by simply deleting this file.
#
# See resolved.conf(5) for details

[Resolve]
#DNS=
#FallbackDNS=8.8.8.8 8.8.4.4 2001:4860:4860::8888 2001:4860:4860::8844
#Domains=
#LLMNR=yes
#MulticastDNS=yes
#DNSSEC=no
#Cache=yes
#DNSStubListener=udp

Joke's on you I blocked port 53

y not opendns tho

>>62376471
I just did, mine is =(), but I already use google DNS like a good goy.

>>62378227
opendns is owned by cisco nowadays, i wouldnt use it
opennic is comfy

Yeah it wouldn't surprise me if it's straight up malware at this point.
Thank God I use BSD.

File: Screen Shot 2017-09-12 at 1.13.18 AM.png (52KB, 764x363px) Image search: [Google]

52KB, 764x363px

/thread

File: cia.jpg (105KB, 1280x702px) Image search: [Google]

105KB, 1280x702px

>>62376270
This will blow your mind:

Firefox has a compiled-in fallback to use Google search if you don't configure it to something else.

>>62376270
>Get a real distro.
Install Windows.

>>62376270
>his ISP don't have DNS server

>>62376270
> real distro
> void linux
> devuan
Even more of a joke than Gentoo.

>>62378694
>macfags

>>62378744
That's botnet, not distro.

File: rob.jpg (57KB, 508x400px) Image search: [Google]

57KB, 508x400px

>>62378694
So it's a meme that only true for retarded distros (maybe). Sadly the """public""" NTP of google is real, but can be disabled with ease, but still comfy.

File: poettering.jpg (92KB, 1200x800px) Image search: [Google]

92KB, 1200x800px

You are compiling the Linux kernel and this guy slaps your girlfriend's ass - what do?

>>62379144
Slap his ass.

File: 1500510438597.png (10KB, 133x30px) Image search: [Google]

10KB, 133x30px

>>62376270
>Ping 8.8.8.8
>Google has now received that entire 32 bytes of juicy random data from your computer
>LiterallySpyware.jpg

File: 1475824387621.png (1MB, 927x960px) Image search: [Google]

1MB, 927x960px

>>62379144
He will regret it anally.

File: 1495140299573.png (780KB, 1001x1001px) Image search: [Google]

780KB, 1001x1001px

>>62379144
Pull down my pants to see if he's serious or not. If he backs away like a fag, I win.

>>62379184
>pinging is the same as broadcasting every website you visit from your IP

>>62379270
You don't need to worry about that. Your DNS records every connection you don't make through a VPN anyway, it doesn't matter if Googlel has that information. Unless you're a literal Jihadi John, google isn't going to care enough to put the FBI on the phone.

>>62379318
>You have nothing to worry about if you've got nothing to hide :^)

>>62379404
I don't like mass surveillance as much as anyone else, but I'm being serious. Your feeble attempts to hide yourself are actually fly paper designed to concentrate and look even closer at "stragglers".

maybe I'm stupid, but is there even a reason systemd is touching networking at all?
why is this even potentially an issue?

File: I just use this one for every image post I make.jpg (78KB, 640x640px) Image search: [Google]

78KB, 640x640px

>>62376270
>There are people who are using operating systems that have systemd

File: djb-systemd.png (57KB, 586x336px) Image search: [Google]

57KB, 586x336px

>>62379801
>muh boot times

>>62377078
And which distros don't have systemd? Oh that's right, shit that doesn't fucking matter

File: systemd_dns.png (25KB, 778x258px) Image search: [Google]

25KB, 778x258px

>poettering-haters can't read
I'm not even surprised.

File: 1504902467186.jpg (56KB, 645x773px) Image search: [Google]

56KB, 645x773px

>>62379853
>Judging how much a distro "matters" by its popularity rather than whether it delivers a consistent, easy and excellent computing experience

Linux isn't the Oscars, anon. Sure, there are "celebrity" distros like Fedora and Ubuntu, but it doesn't help the fact that they're loaded with shit.

File: medium.jpg (8KB, 64x64px) Image search: [Google]

8KB, 64x64px

>>62379838
When I realized I could have great audio on GNU+Linux w/out Pulseaudio, I was forever distrustful of the maintainers of all distribitions.

>>62379887

/thread

>>62379887
>>62379956
>missing the point entirely.

>>62378160
Notice how the line is commented by default.
Anti-systemd cucks still try to find all excuses. It's comfy and it's free software, deal with it.

>>62379887
It shouldn't be there in the first place. It should start up the system then let the system get on with what you want it to do, not call, ping or schedule every fucking thing or flush your toilet for you.

>>62380021
It's free software. Modify it if you don't like it.

Daily reminder that systemd-haters are useless poseurs: https://blog.finnix.org/2017/04/18/the-future-of-finnix/

>systemd init: Finnix currently uses runit as a system init, which was small and let me sidestep sysvinit and Debian’s distro rc assumptions. (Prior to runit, Finnix build procedures involved cutting out most of the sysvinit rc functionality and putting Finnix replacements in.) But it’s very much a manual process, and as all distros are or are in the process of transitioning to systemd, now would be a good time to do the same.
>systemd is large and has a reputation of wanting to consume everything, but as I’ve found out with a few days of research and testing, it’s actually rather conducive to distribution management. For example, one thing Finnix needs to be able to do is attempt DHCP on all network interfaces, but not block on it (getting to the command prompt quickly is the most important consideration). I was able to quickly write a service which depended on systemd-udev-settle.service, ensuring all boot interfaces were available, write interfaces.d definitions and trigger starts of ifup@$INT.service without blocking on the getty process. It even has the ability to pivot-root back to an initramfs at the end of system shutdown, something I needed to patch manually into runit.

>>62376352
Lmao dude you guys are not even hiding it anymore.

>>62380031
I've never used it and I am not going to start using it.

>>62380021
>It shouldn't be there in the first place.
You seem to think systemd-resolved is a part of PID 1/init system, which is wrong. It's a stand-alone daemon what may or may not be included with systemd, and it makes distro maintainers' job easier by providing basic DNS functionality out-of-box. It's like critiquing busybox for including all the tools it has.

>>62380076
Don't complain then.

Fucking great. Just another piece of systemd shitware I have to contend with when deploying to RHEL servers.

Fuck poettering and fuck red hat for promoting his cancer. This is incredibly distressing.

>>62377207
Probably 80% of /g/ uses windows

>>62380108
WHY IS MY INIT PROCESS RESOLVING DOMAIN NAME QUERIES HOLY SHIT

>>62380084
>It's a stand-alone daemon what may or may not be included with systemd, and it makes distro maintainers' job easier by providing basic DNS functionality out-of-box

until something that requires systemd also relies on systemd-resolved quirky behaviours.

So that in 20 years time all our DNS resolvers can have systemd-resolved quirks flags in them.

Because resolver(5) is so fucking complicated.

Because dnsmasq is so fucking complicated.

Because this clearly solves a problem that existed.

>>62380088
It is part of Red Hat's strategy to Embrace - Extend - Extinguish.

It is not just an engineering problem it is a business problem for FOSS.

>>62380084

Having a massive blob lump of shit totally breaks any UNIX idea of system engineering.

>>62376397
>you really didn't buy the pass, did you?
Suck a bag of dicks, cheapo.

>>62380010
And if it's commented out it uses Google DNS as fallback. Unless you build with "-D dns-servers=something else" (which no distro does).

btw Systemd also defaults to Google's NTP server which is only for Google's internal use and shouldn't be used by others:
https://github.com/systemd/systemd/issues/437

>>62380310
You know you can also edit lines other than commenting them out, right? Or are you this retarded?

>>62376704
God damn right

>>62380319
Now you're just moving the goal post.
You or whoever implied systemd doesn't use it by default because it's commented out.
And that is simply wrong - accept it.

>>62380226
Gee, you must really hate GNU coreutils, findutils and binutils then, because it's exactly the same scheme - a single project providing a lot of different tools. Or take any BSD - the entire world is built from a single repository, I guess BSD guys know nothing about Unix Philosophy.
> UNIX idea of system engineering
Wake up, grandpa, UNIX is dead, "Unix philosophy" is a stale meme for teenagers. In the real world, no one cares about abstract ideology, only about results, and systemd provides them.
>>62380138
>Because dnsmasq is so fucking complicated.
Wait, are you implying we should use full-blown caching DNS/DHCP-server for DNS resolving?

>>62380363
>Gee, you must really hate GNU coreutils, findutils and binutils then, because it's exactly the same scheme - a single project providing a lot of different tools.
These tools aren't completely interdependent.
I don't need GNU grep installed to run GNU ls, etc.

>>62380379
You don't need to have systemd-resolved installed to run systemd-timesyncd either.

>>62376704
Sounds fine with me. Why do you think Linux desktop share is up? Y'all could use some simplification.

>>62380363
Have BSD taken to systemd?

I suspect I was using computers long before you were born, senpai. Unix philosophy is not a meme it is sound engineering principle. Without it the worlds servers would not be stable systems.

>>62380439
If the world servers are so stable why do *nix servers often have more failovers?

>>62380424
I do need systemd installed and running though.

Meanwhile I can run GNU grep on a BSD/Windows/OS X/...

>>62376270
This is simply false. You can edit /etc/systemd/resolved.conf to remove/replace the fallback DNS.

File: lolololololololololol.png (23KB, 830x255px) Image search: [Google]

23KB, 830x255px

I have to scream but I have no mouth

>>62376835
the article is referring to Kay Sievers, not Poettering

File: bcc_tracing_tools_2017.png (288KB, 1500x1050px) Image search: [Google]

288KB, 1500x1050px

>>62380451
So what? These are systemd-specific tools that require systemd. It's just like you need Linux kernel to run Linux-specific stuff like `iproute2`, `perf`, `cpupower` and such, or stuff like https://github.com/iovisor/bcc which provides dozens of tools all requiring Linux kernel and bcc itself. How exactly is this approach wrong?

Poettering got mad and deleted lifeboy's latest comment along the lines of:

>Why does this man keep breaking things that make the world go round and why do we keep letting him get away with it? Void/Devuan ftw

>>62380531
>>62380569
>Why does he delete inflammatory/trolling posts that have nothing to do with the project and the issue?
Systemd-haters are pathetic.

>>62380565
Because that actually needs the kernel?
No one is complaining that a tool for systemd-specific debugging and configuring needs systemd, of course it does.

Lol linux fags got botnet too xD

>>62378160
>resolved.conf
>not /etc/resolv.conf

>>62379318
Yeah remember that whole PRISM thing and how they're literally storing all your data? kys

>>62380605
And these tools actually need systemd to reduce code duplication and be more useful.

>>62379887
>dumbing shit down for retards is a good thing
>adding a single line less than 20 characters long in resolv.conf is too hard

>>62376270
Which distro then tho

>>62376278
it is though, if it's compiled into the system daemon to redirect all traffic, by default, to a Google server who's specific job it is to analyse and make decisions based on exactly where the traffic came from and where it is supposed to go, then carry out those actions

>>62376270
enjoy being stuck with xorg and having no wayland. enjoy living in the early 2000's forever.

>>62377748
What about 8.8.4.4?

>>62380108
>He's using RHEL/CentOS 7
You asked for this, stay on 6.

I bet none of the people bitching about systemd in this thread have flashed their BIOS with libreboot and are getting cucked hard each and every day by the Intel ME/AMD PSP backdoor sponsored by the NSA.

So? If you're so bothered by this, why don't you patch it out? Or I don't know, maybe configure your fucking dns

>>62381010
libreboot removes IME?

>>62381145
Linux users have too much free time on their hands; it's why they use Linux and also the reason why they can complain so often and go on such long autistic tirades.

>>62380848
>IMPLYING PEOPLE HAVE WAYLAND NOW
HAHAHAHAHAHA

>>62381157
Yes.

>>62380848
>enjoy living in the early 2000's forever.
how early are you talking? like 9/11 groundhog day? either way this sounds glorious

>>62376270
This feels like a shitpost, but you're right about everything you said.

t. fellow void user

>>62376397
do you really think the cost of a pass is not worth the freedom it gives you? sad.

>>62378160
Why the fuck wouldn't it simply use the DNS root servers? I get it, query times will be longer, but that'll prompt someone to take a look at config and notice something is wrong.

>>62376270
Void / OBSD here. Yeah fuck Poettering and his botnet.

>>62379186
> bummed to within an inch of their lives

>>62381167
>IMPLYING PEOPLE HAVE WAYLAND NOW
what did he mean by this?

>>62378725
actually if you delete all your search plugins it defaults to yahoo.

>>62381164
So they do everything but fix their own problem?

>>62382047
Sounds like it. I thought the whole thing about open source software with a permissible license was that you can fork it and edit it yourself. Seems like Linux users only know how to complain; which is funny because they tout being "able to change anything, just edit the source" as a major advantage of the Linux system. Weird, senpai.

>>62379838
>(((Bernstein)))

>>62376523
How do I check this hasn't happened to me?

File: 1328563496631.jpg (117KB, 311x311px) Image search: [Google]

117KB, 311x311px

>>62382735

>>62382747
prob something like `journalctl -u systemd-resolved` and look it up
systemd-resolved isn't used by default so you probably aren't affected

>>62382798
>(((Stallman))

>>62382814
Okay thank, I am okay.

>>62380715
Void linux

>>62380363
Resource optimization is dead, no one needs good performance as long as it works.

Reminder he born in Ecuador, he is a fake german

>>62376835
Linus did the right thing, Poettering tried to shove systemd dependencies to the kernel by introducing kdbus.

>>62379827
Garfield is real.

File: 1500306102024.jpg (59KB, 680x535px) Image search: [Google]

59KB, 680x535px

No place to hide from the big G, embrace it.

>>62376270
in what fucking case would you not have a dns server configured
this is a pointless fallback

>>62378855
>gentoo
>joke

>>62379827
>There are people who are using operating systems other than systemd

>>62379318
>implying the reptilians aren't going to kill us all

>>62376423
You lost little contrarian asshat, get over it.

>>62384251
Do you have imaginary friends too?

>>62380759
>analyse and make decisions based on exactly where the traffic came from and where it is supposed to go

No it isn't, that isn't what DNS is, nor what it does, DNS receives a domain name, and returns and IP address associated with the domain name. Is it possible that google compromised it's own server's name database in order to point traffic to sites where data could be collected? Yeah, it'd be trivially simple for them to do so. It'd also be incredibly ineffective as this is why we have SSL, to verify that the site we THINK we're connecting to is the site we're actually connecting to.

You've confused DNS with a fucking routing protocol you damn brainlet

>>62382735
>>62382833
Go back to /pol/

File: 1e843692a5ecbc55526ca4659d9cab3e.jpg (65KB, 800x450px) Image search: [Google]

65KB, 800x450px

>>62385195
DNS is a routing protocol you fucking dumbass

File: hqdefault.jpg (10KB, 480x360px) Image search: [Google]

10KB, 480x360px

>>62385769
You fucking what mate?

Are you seriously this retarded?

>>62379948
How does one achieve this? Explain. Using Xubuntu and it runs Pulseaudio, no problems.

Why isn't there an Ubuntu fork without systemd? I want Xubuntu without systemd, damn it.

>>62386287
DIY

>>62386664
>>62386287
Fuck Xubuntu with systemd would be perfect

>>62376270
>https://github.com/systemd/systemd/issues/5755
>No. It won't. If any DNS configuration is configured at all, it is used, regardless if it actually works or doesn't. If no DNS configuration exists at all, then the default DNS configuration specified at systemd buildtime is used, using the DNS_SERVERS meson build parameters. We encourage distributions to set these servers towhatever they like, but many just leave it at 8.8.8.8. If you don't like that please politely try to convince your distribution to change them to better suited servers. Note that these fallback servers are exclusively used if no DNS configuration exists at all, and resolved immediately switches to whatever is configured as soon as something is configured again.

Thank god school is opening soon so we can go back to having intelligent posters having intelligent conversations.

>>62385660
>(((Anonymous)))

>>62387347
Thank you!

File: 1497696170501.jpg (95KB, 768x1024px) Image search: [Google]

95KB, 768x1024px

>>62376411
OS? Phone? Method of carrying out recent transaction?

All those things are linked to the botnet. Much larger scale. No way is it limited to computer use. That is simply an insignificant leg.

File: high_on_chromosomes.jpg (116KB, 1053x731px) Image search: [Google]

116KB, 1053x731px

>google aborts stormfags from the internets
>/pol/ reees and installs the commie OS
>/pol/ now leads the systemd-uprising and loves Stallman, albeit ((())) tattoo
i'm loving this timeline

>>62387514
>not jewish, but actually made me lol

Take an upvote

systemd is a botnet

File: NSA.jpg (55KB, 400x400px) Image search: [Google]

55KB, 400x400px

>>62386287

>>62378694
This is resolved which is part of systemd you retard.

Read the github issue everyone, he clarified how it works. Yes it is still stupid and it breaks a lot of peoples shit.

>>62376270
Systemd is 100% confirmed botnet with backdoors.

>>62387347
So if my system gets its address through DHCP, and the DHCP server on my router says "I'm your default gateway, send DNS queries to me", does that count as "configuration" that'd cause systemd to say "well, I have a DNS server, don't need to use any of this fallback stuff"?

>>62394101
yes

>>62378694
https://github.com/systemd/systemd/blob/72d17ce680a22b958b4c074f62d9fa44bc1f54e0/configure.ac#L1193

>[DNS_SERVERS="8.8.8.8

>>62394101
depends

>>62376585
as far as i remember, network-manager has a dependency on systemd

i guess he uses wicd

or no gui

>>62376270
THANKS for linking that issue, I read all trough it! It's the first time I actually see systemd devs doing something so terribly wrong that it could affect the functionallity of my production environment

>The main problem here is that the first local DNS server is regually switched trough behaviour of UDP and traffic so the remote DNS like 8.8.8.8 is used randomly which can be a huge problem with VPN connections or Domains only avaible on local DNS servers

fuck it this is huge

>>62376468
>https://github.com/systemd/systemd/issues/5755
from the horse's mouth

>>62395595
We all fucking told you. Go read that Arguments Against Systemd wiki page.

>>62396479
wiki is down
if anyone want's to read an archived version of the wiki, here is it
https://web.archive.org/web/20170809021655/https://wiki.installgentoo.com/index.php/Main_Page

>>62396656
God bless you

>>62396681
got it from another anon, spread the link for others too

>>62376352
Just because many people use google's dns doesn't make it a good choice, especially for an open source project. They should switch to either not providing a dns resolver or at least defaulting to something open like opennic.

>>62394334
Lots of red hat related projects now pull in systemd for no reason. Polkit and NetworkManager don't actually need systemd as far as I can tell (they work just fine on systemd free distros like void and gentoo) but systemd based distros like debian have them depend on systemd. Also wicd works well too, I use it on my laptop.

>>62380635
/etc/resolv.conf is what sane dns implementations use. resolved.conf is for systemd's special snowflake dns resolve systemd-resolved.

>>62378725
No one said mozilla was any better than systemd team. Also mozilla really needs to be restructed and actually fullfill their privacy promises.

>>62379144
Shoot him because It's my god-given right as an American to shoot trespassers.

>>62379801
There is no reason for systemd to touch half the shit it touches, and that's why so many people don't like it. systemd should only handle init and service management, but the systemd team wants it to handle everything they consider backend system stuff.

>>62379853
>shit that doesn't matter
Gentoo doesn't use systemd by default, and is heading the development of things like eudev and elogind to remove systemd dependencies. Are you trying to tell me (on /g/ of all places) that gentoo is not the most important distro?

>>62379948
Like >>62386236 asked, please explain how you eliminated pulse from your setup. I want to get rid of pulse, but as far as I can tell you can't do per application volume without it.

>>62380038
Just because some distros are switching to it doesn't make it good.

>>62376352
Why would Google's DNS be queried if there were no DNS servers set up?

Seems like the proper, expected thing to do in that case is to not make any DNS queries.

>>62380084
We already have several DNS clients, and distro maintainers have been including them in distros for years now, and most of them actually respect the c library standard of using /etc/resolv.conf for configuration. Making a new special snowflake dns resolver tied to the init system is just ridiculous.

>>62380123
Which interestingly enough lines up about right with the around 80% of /g/ that is /v/ cross over.

>>62377207
popularity != good

>>62380211
I don't think red hat's strategy is quite embrace extend extinguis, I think they prefer to introduce something news as a "linux standard" then push it into other distros using their army of developers which in turn makes their support more valuable as their shit is being used everywhere.

>>62376270
Jesus christ, this issue is frustrating to read. Lennart and co. decide to just break conventional, default resolv behavior for no good reason and fuck up a lot of people's setups. When politely asked if they could just make the rotate DNS thing an option they could disable or something, Poettering completely ignores that and keeps arguing that his way is better completely ignoring that other people might have different needs. I used to not mind systemd so much a few years ago, but shit like this makes me glad I got rid of that shit.

>>62376634
Thank you Lennart!

>>62380363
Having things in the same repo doesn't automatically mean that it somehow breaks unix philosophy. the GNU coreutils are still mostly portable, for example you can run gnu grep just fine in a BSD userland. Also plenty of bsd software can run on gnu/linux like openssh for example.

>>62380449
Because perfect stability isn't really possible as long as imperfect humans are the ones making software. Also lots of big servers run Linux so failovers are always good to have.

>>62397415
A lot of those set ups were utterly incoherent and a mess. It's good they were broken.

>>62380679
An init and service management system shouldn't have code relating to dns resolution or ntp. Those should be seperate, mostly portable things.

>>62396656
I meant the without-systemd wiki you nigger

This meme link that gets spammed in every systemd thread: http://without-systemd.org/wiki/index.php/Arguments_against_systemd
It's actually pretty informative and you should read it. At least pick some links that look interesting and look over them.

>>62397659
poettering pls

>>62397770
will read it, this thread was a bit eye opening to me

>>62397801
i wish

>>62380715
Void, Gentoo and Devuan are all good for avoiding systemd.

>>62397856
are these the only ones left mentioning besides funtoo and maybe sourcemage? looks like a dark future

>>62397898
Well there's a few more obscure forks around. I'm on Artix which is a WIP semi-fork of Arch.

>>62380848
Wayland is nowhere near ready anyway, it lacks a tool like xrandr for scripted control of displays and it lacks drivers. Also as far as I know wayland doesn't depend on systemd.

>>62381010
If my systems were compatible with libreboot I would use it.

>>62397921
>Artix
neat, didn't know about an arch clone with openrc

File: 82dqG.jpg (60KB, 600x450px) Image search: [Google]

60KB, 600x450px

>>62397985
>wayland doesn't depend on systemd.

wtf man don't give him any ideas

>>62381145
The problem isn't just that it's default. The problem is that an init system shouldn't be implementing a dns resolver and open source projects shouldn't default to google dns.

>>62397985
Wayland is pretty usable with Sway although you probably won't be able to anything terrible fancy with your displays. You don't need systemd either.

>>62381164
We have alternatives, like OpenRC and runit, the problem is that systemd keeps being pushed harder and harder.

>>62398005
There were some guys that maintained openrc repos for Arch and Manjaro for a couple of years and then they decided to merge their efforts and make a fork. Basically, they're using elogind to track sessions and compiling some of the core packages without systemd.

one of you spergs is going to assassinate him over loli aren't you...

>>62386287
It's not too hard to make xfce look like xubuntu on other distros, just install the greybird theme.

I think its a great feature.

You guys complaining are either frightened or neurotic.

The only thing I would change is to make it use Google's encrypted dns instead.

>>62398074
so there aren't (nearly) all arch packages avaible?

>>62398112
then I guess you are not dealing with local DNS in a large network

>>62398147
You are a moron and your implication that this causes problems with local DNS on a large network is laughable.

>>62398118
You still pull from the main repos and AUR. It's just that some of the essential packages like base-devel are pulled from their repos instead.

>>62398195
did you read the whole issue?

>>62398200
thanks for the info, I'll look into that, once I have the time. is there anyhing else I need to know before I make the switch?

>>62398254
The migration on the webpage is pretty simple.
https://artixlinux.org/migrate.php

The only thing is that they default to the lts kernel and not mainline (this may or may not annoy you I dunno). And naturally, sometimes they'll be a tad behind Arch since that's just the nature of the thing. It all works fine for me, but it's still pretty much a beta. Plus you'll probably have to mess around with your configuration a bit to accommodate the switch.

>>62376270
For the fucks sake, read the Lennart response instead of shitposting.

>>62398254
>is there anyhing else I need to know before I make the switch?
Its a waste of time and doesn't accomplish very much.

>>62398353
shut up tripfag you're not welcome here

>>62385195
>server who's specific job it is to analyse and make decisions based on exactly where the traffic came from and where it is supposed to go, then carry out those actions

i said:
>server who's specific job it is to
>analyse and
>make decisions
>based on exactly where the traffic came from
>and where it is supposed to go,
>then carry out those actions

are you retarded or something?

>>62380310
No distro I used enables systemd-resolvd by default, nigger.
You have to comply with 4 conditions before your DNS fallback to google ones.

>>62396479
>muh unix philosophy
And plain lies.
That page is literally FUD.

>>62398443
>plain lies
Did you actually read the links on the page?

>>62376270
Unless you make your OS yourself you should assume that it's not secure.

>>62398481
Countless times. They insist on push bullshit like systemd runs all shit on pid1.

What's wrong with OpenRC?

>>62398670
Literally its becoming systemd and still uses sysvinit.

>>62398699
>its becoming systemd
What are you talking about? OpenRC is a great init system.

>>62398699
>Literally its becoming systemd
Now that is FUD

>>62398699
>still uses sysvinits
systemd zealots complain about systemd haters for spreading FUD but I see even more FUD from the zealots. OpenRC has had openrc-init etc. for many releases now. sysvinit is entirely optional even though it's still part of the default Gentoo profile

>>62376411
I'm calling bullshit on that one. You didn't "block google"

It's all or nothing. Blocking google means blocking ALL google subnets. The web would break on your network if you did this.

>>62398859
Using remote mounts with openrc and sysv sucks to the point that it doesn't work.

systemd's .mount and .automount files are exponentially better than using fstab or bash like openrc script hacks.

>>62398720
>>62398725
>openrc is implementing features from systemd
>it's not systemd, lads, I swear.

>>62399046
Last time I checked, openrc didn't fuck with my resolv.conf.

>>62399046
>openrc is an init system
>systemd contains an init system
>openrc is turning into systemd!

>>62399046
Do you have any actual examples of that or are you just making shit up?

>>62398995
>using remote mounts with openrc and sysv sucks to the point that it doesn't work
OpenRC comes with a service specifically for netmounts, and all you have to do is add the option _netdev in your fstab to register a mount as one.
>systemd's .mount and .automount files are exponentially better than using fstab or bash
You have got to be kidding me. fstab is the standard way to layout filesystems mounted at boot and has been since the dawn of unix. fstab is a well formatted table of the filesystems you want to mount, why the fuck would you want an individual file to configure every mount? Also if you want automounts you can just use autofs.

>>62399119
>You have got to be kidding me. fstab is the standard way to layout filesystems mounted at boot and has been since the dawn of unix.
And?
>>62399063
OpenRC implemented a method to parallelize their daemon start, a la systemd (but shittier)

>>62399119
Don't bother. I've seen maybe one semi-coherent post from him.

>>62399143
>implementing a feature to configure how you start up services is becoming systemd
wow, I didn't know Lennart invented parallel daemons.

>>62399143
fstab is a well established and perfectly functional way to mount filesystems, there is no reason to replace it.
>OpenRC implemented a mathod to parallelize their daemon start, a la systemd (but shittier)
Parallel daemon startup is not a systemd exclusive feature, runit does parallel daemon starting by default and was created before systemd, also there have been hacks that even made sysvinit do parallel startup. OpenRC adding parallel startup is a good thing and hardly means it's becoming systemd.

>>62399191
OpenRC is becoming what systemd should have been

>>62399191
>fstab is a well established and perfectly functional way to mount filesystems, there is no reason to replace it.
Except that it asumes a static system, like the systems in the dawn of Unix, where hardware changes weren't existent in first place.
>hardly means it's becoming systemd.
Okey, I can accept that.

>>62399211
Exactly, it's becoming a good init and service management system that does those things well and does them efficiently.

>>62399211
systemd should have never existed. Everyone should have transitioned to openrc and maybe down the road we all switch to runit or s6 or some shit.

>>62399218
I'll admin that fstab isn't the best for mounting things like samba or nfs shares, but fstab is still just fine for system disks. I use autofs to mount autofs shares.

>>62399219
>that does those things well and does them efficiently
and (importantly) nothing else.

>>62399240
No.
Systemd will unify GNU/Linux finally and will purge the shitfest in that ecosystem.
OpenRC can't do that in first place.

>>62399271
GNU/Linux is not and should never be a monoculture.

>>62399271
>Systemd will unify GNU/Linux
Terrible idea. This isn't Windows. There is no need for GNU/Linux to be "unified."

I will admit I ping 8.8.8.8 to see if I have a network connection

but use it as my dns?

>>62399271
>Systemd will unify GNU/Linux finally and will purge the shitfest in that ecosystem
Why are people so obsessed with unifying GNU/Linux? It is not a good thing to strive for, as it leads to a monoculture which can lead to security problems (openssl anyone?). Besides a "unified" GNU/Linux eliminates one of the beautiful things about free software, the freedom to run what you want. One of the reasons why GNU/Linux is so widely used today is because of the choice available due to its modularity, which meant that companies, sysadmins and just people running stuff in their basement could make a setup just right for them by combining multiple free software projects. If you "unify" GNU/Linux then it will probably end up a bloated mess like Windows or Mac OS X as it will have to try to cover every use case.

>>62399355
I don't think we should trust people who tell us to Strive for Unity or to Trust thy Brother (i.e. Lennart).

>>62399384
Yeah. The main benefactor from unity would seem to be companies, especially red hat. If we all use red hats blessed software then they can support any distro and get plenty more customers.

>>62399355
It's too hard for normalfags to learn how to setup a working environment, so we need to "unify" it to make it easier.

>>62399479
Or we could write better documentation and not compromise our principles.

>>62399507
Oy vey, we can't have people needing to read documentation. They just want things to work!

>>62399537
Perhaps we could take a leaf out of the book of old computer game copy protection techniques and ask the user questions during install that can only be answered if the user has read the instruction manual.

>>62399289
It will not be a monoculture, even with systemd.
>>62399479
Yes. Most people aren't ricer archfags.

>>62399384
I'm ashamed that nobody got this

>>62399384
Why not?

>>62399479
It's not hard for normalfags to setup a working environment. A plain old ubuntu install works fine for normalfags (even when they used upstart). If you are talking about configuring system/server stuff, then you need to actually learn things about that because it is a skill.

>>62399723
>It will not be a monoculture, even with systemd
systemd is trying to replace everything even remotely related to "system" functionality, I'm pretty sure that is a monoculture. Sure I can choose between chrome and firefox, but if every GNU/Linux system is running systemd, systemd-resolved, systemd-timesyncd, etc then how is that not a monoculture.

>>62399191
>fstab is a well established and perfectly functional way to mount filesystems
But it isn't.

.mount and .automount are far better.

>>62399822
It can't be a monoculture if they are alternatives and activism towards these alternatives. The fact that Gentoo, Void, Devuan, Alpine, etc, exists proves it.

>>62400006
I'm not saying it's a monoculture right now, I use gentoo myself. I'm just saying that the systemd team and red hat seem to want a monoculture and that's a bad thing.

>>62400025
Your a fucking moron.

>>62400042
How much does red hat pay you?

>>62400042
tripfags being tripfags as per usual

>>62399947
In what way is fstab not a well established and functional way to mount file systems? How are .mount and .automount better?

>>62396986
why he did that useless shit then?

>>62380848
>enjoy living in the early 2000's forever.
i wish

>>62400144
Because Lennart and friends are convinced that they can implement everything better.

>>62400124
When you use dns resolution to find your ssh server and want your mount to keep working when you resume from suspending your laptop.

Just off the top of my head.
Something that has basically never worked and nobody gave a shit until systemd improved it.

>>62400183
Have you never heard of autofs? It does exactly what you are describing. I'm pretty sure systemd based .automount units on autofs because of their whole obsession with everything integrated into systemd.

>>62400237
systemd uses autofs

>>62400277
I'm pretty sure it doesn't. I'm pretty sure .automount is something they implemented themselves.

>>62400300
Probably isn't the same autofs then.

Just install devuan, works great.

>>62376270
>Mozilla Firefox defaults to Google Search unless the user changes it
WOOOOOOOOOW WHAT A BOTNET

>>62400824
Not the same thing Lennart.

>>62376456
Is this jist fantasy?

>>62400885
Yes, you have to configure that in first place. With firefox that configuration can't be made.

>>62401001
Why do you start wildly pointing fingers every which way when valid criticisms of systemd crop up?

Perhaps you don't want people looking too closely at its many already-revealed flaws, hmmmm?

File: best plan ever.jpg (64KB, 704x480px) Image search: [Google]

64KB, 704x480px

>>62401064
>bawww stop interrupting my hateboner circlejerk with annoying facts

>>62401098
>that fact that you use shill tactics and try to divert the issue is a fact
thank you for admitting this

>>62401110
Its a sane improvement.

Just like how power breaks and steering have a backup, except the backup in this case has no drawbacks and the improvement takes no maintenance.

So no, not really like that, but still good.

>>62401167
stop tripfagging, it's not relevant to the topic at hand

File: 1486835330543.jpg (23KB, 275x284px) Image search: [Google]

23KB, 275x284px

>>62376270
I just read that whole issue page and the replies and you are retarded. The only reason that 8.8.8.8 is brought up is because the OP of the issue used it and 192.168.0.1 as example DNS servers. AKA 8.8.8.8 was manually entered to the settings by the user.

Quit being stupid, whether on purpose or by accident.

>>62401199
Neither is your butthurt whining.

>>62401255
Do you have to use that to get paid?

>>62401448
Yeah Redhat has been paying me to shill systemd for 10 years.

>>62401493
so do you use multiple proxies or just the one, because this board doesn't have IDs?

>>62401512
I'm having trouble understanding you.
Is English your native language or are you just off your meds?

>>62401573
Why would I be on meds?

Anyway the question was, how many proxies do you use in addition to your tripfag when you argue with yourself?

>>62401110
Oh, the irony.

>>62401587
Dude you suck at insults. I'm surprised you have the capacity to use punctuation.

>FOSS world has an issue where the bugs that aren't easily reproducible by maintainers, aren't that interesting to fix, or simply are differences of opinion about the design are ignored.
>but somehow, when systemd does it is a FUCKING GIANT MORTAL ISSUE

>muh sysvinit
>muh minimalism

>>62401721
Yes it's so weird that Unix people like it when their Unix-like OS follows the Unix philosophy when possible don't like it when extremely complicated, huge, bloated botnets are shoehorned into the place of something small and simple.

>>62401834
>Unix people
I'm a Linux person.

UNIX is proprietary.

>>62401834
>yeah is so weird that religious nutjobs about their shitty os are trying to boycott the only thing that can bring some sanity to the shitfest thats the backbone of the internet

>>62401665
I don't think he was trying to insult you. I think he was asking if you argue with yourself in threads using proxies to try and spread the red hat narrative.

>>62401669
A bug in a file manager that happens every once in a while isn't that big of a deal as long it doesn't delete files. A bug in systemd that happens every once in a while is still a big deal as systemd runs in PID 1 as root, which means it has permissions to do literally anything on the system and if it crashes it brings down the whole system.

>>62401876
He was basically trying to call me a samefag in a roundabout way from what I could tell.

Pretty unimaginative, hes probably melanin enriched.

>>62401721
Only a subset of systemd haters actually think sysvinit is the answer. Plenty of us support other init systems like openrc and runit.

>>62401910
>Plenty of us support other init systems like openrc and runit.

OpenRC has pretty good support.
You lost all credibility when you added runit though sorry.

Into the dipshit contrarian systemd hater bin.

>>62401894
>systemd is a giant binary running on pid 1
For the fucks sake.
Systemd-init also has mitigation measurements. For example, it can recover itself of crashes before the kernel reachs panic state.
Sysvinit and runit don't have those mitigations, and nobody think that its bad.

>>62401894
What about the kernel?
It suffers the same issues.

>>62401961
sysvinit and runit don't need those mitigations as they are small enough to not crash unless something is actually wrong.

>>62401932
I was just suggesting runit as something people might prefer. I use openRC myself.

>>62401976
The kernel is a much more mature project and has many more eyes on it. Also the kernel has plenty of things relegated to kernel modules to mitigate problems.

>>62401998
Why?
No distribution is using runit and fucking around with init is pretty pointless when systemd works better than all alternatives.

I don't even know why I'm asking, its pretty obvious you subscribe to the cat-v suckless shit and instantly resort to what amounts to propaganda to explain why people should use shitty software because high LOC counts stress you out.

>>62401983
Are you implying that pid 1 only crashes because its bad/good written?
They are other problems that can cause init to crash.
Trying to pass a feature as a bug is why you lose the init wars.
>>62402013
>The kernel is a much more mature project and has many more eyes on it.
And still has annoying or dangerous bugs, like the degradation of interactivy on high IO load.
Also, you can't argue in 2017 for the "many eyes" bash and openssl can tell you that isn't a guarantee.
Even considering that anything has to be approved by Torvalds et al. And they have some questionable practices. (For example: nobody rembembers why Torvalds said the "masturbating monkey" line)

>>62402053
Void uses runit which is why I mentioned it, as openrc, runit and sysv are the only init systems that I know of that are used in distros (besides systemd of course). I don't subscribe to cat-v suckless shit, I use stuff that they wouldn't approve of. I'm not worried about LOC counts, I'm worried about the amount of things systemd is trying to cover. systemd should just be init and service management, if that takes 200k loc of whatever fine but they shouldn't being doing things like dns resolution, ntp, qr code generation, web servers, etc.

>>62402085
You say you don't subscribe to the sys-v suckless shit and then quite literally start bitching about too many features and high LOC.

>>62402085
Why they have to limit their scope?
Proyects evolve all the time. The kernel was initially a kernel, but now its also a virtual machine manager, a program itself, the virtual terminal, etc, etc, etc.
Why its so bad that someone is offering a suite to offer the basic plumbing?

>>62402063
>Are you implying that pid 1 only crashes because its bad/good written?
>They are other problems that can cause init to crash
I'm not saying there aren't other problems that can cause init to crash, but the fact is that systemd exposes many more interfaces that can lead to it crashing (remember that tweet with a command that could be run as any user that would crash systemd?).
>And still has annoying or dangerous bugs, like the degradation of interactivity on high IO load
I heard they are working on a new scheduler to fix that. Also yes the kernel isn't perfect, and a microkernel like hurd might be the better solution in the end, but it's still the best kernel we have now (the same can't be said for systemd).

Reminder:
All bad shit that can be said of systemd also can be said for the kernel.

>>62402110
I literally just said that I don't care about the lines of code. Caring about a program doing too many things is more of a unix philosophy thing that suckless (they do say that but they also focus on other shit).

>>62402128
>the same can't be said for systemd
I literally said that a few posts ago and most major distributions are using systemd for that exact reason.

Fucking textbook cognitive dissonance.

>>62402158
OpenRC is proof that there are better alternatives. I don't see any viable alternatives to the Linux kernel right now.

>>62402128
>I'm not saying there aren't other problems that can cause init to crash
You implied that in first place
> but the fact is that systemd exposes many more interfaces that can lead to it crashing
Yes, it has a cost of increased complexity and things can gone wrong easily. However, systemd improves so many shit that this issues are even worth.
>remember that tweet with a command that could be run as any user that would crash systemd?
I don't know that tweet. Care to show it to me?
>>62402175
seL4/L4 was/is the best alternative. Literally the wet dream of the "unix philosophy" fags, but they insist to stick to Linux.

>>62402175
OpenRC is only better in your shallow egotistical opinionated hugbox built from other peoples ignorant opinions.

File: Faggot Alert.png (44KB, 534x400px) Image search: [Google]

44KB, 534x400px

>>62379853

>>62402189
There was a bug where an unprivileged user could send an empty string to systemd via systemctl that would cause it to stop responding.

It was immediately patched and probably discovered by accident. Equivalent of a fork bomb.

>>62402215
Does systemd team practice correct security things like disclosure and publish mitigation measures?
The kernel fuckers, as I know, they like to avoid those.

>>62402117
Projects should limit there scope because having a clear goal helps to design and implement a piece of software.
>The kernel was initially a kernel but now its also a virtual machine manage, a program itself, the virtual terminal, etc, etc, etc.
Linux isn't a virtual machine manager, all KVM does is expose the hardware virtualization functionality to userspace programs like qemu. Kernels are always programs, what do you mean it's now a program itself? The virtual terminal is just an interface to the kernel anyway so I'd say that's still in scope. Linux is still a kernel and still aims to be one.
>Why its so bad that someone is offering a suite to offer the basic plumbing?
Because we already have seperate applications that offer that plumbing, and with seperate applications you can switch out the parts for programs that fit your needs better. If every "system plumbing" function is handled by systemd then you can't replace parts freely.

>>62402189
>>62402197
Please enlighten me as to how systemd improves so much over OpenRC

>>62402234
Doubt it.
Gentoo has you covered though.
https://security.gentoo.org/glsa/201612-34

>>62402189
>I'm not saying there aren't other problems that can cause init to crash
>You implied that in the first place
I didn't mean to imply that, sorry if I did.
>systemd improves so many shit that this issues are even worth
What does systemd actually improve?
>I don't know that tweet. Care to show it to me?
Here's the tweet that can crash systemd: https://www.agwa.name/blog/post/how_to_crash_systemd_in_one_tweet

>>62402266
>Projects should limit there scope because having a clear goal helps to design and implement a piece of software.
you are implying that goals can't change or the goals can be driven by timeframes.
Systemd evolved from a "init system" to a "suite of programs that can form the basic plumbing of a Linux system"
Why? Because server and embedded needs.
>Linux isn't a virtual machine manager, all KVM does is expose the hardware virtualization functionality to userspace programs like qemu.
Still adds complexity that, by the words of Unix autists, it doesn't have to exist.
Why not a type 1 hypervisor instead of a kernel that exposes shit to a user mode program?
> Kernels are always programs, what do you mean it's now a program itself?
Linux can be ported to itself. Google "usermode linux"
>Because we already have seperate applications that offer that plumbing, and with seperate applications you can switch out the parts for programs that fit your needs better.
Except that time proved that these "already existent programs" didn't do their work pretty well, like Unix weirdos like to claim. Only the 80% if those worked well. The strongest and most sucessfull flavors of Unix (BSD, Solaris) in fact they did have lots of programs with they own scope, but they also weren't repleaceable easily, because where designed to play well by themselves.
>every "system plumbing" function is handled by systemd then you can't replace parts freely.
If I have to repleace the organds of my system instead of working with it, something is bad in the design of my system.

>>62402277
I don't really give a shit if you are enlightened all the distributions use systemd already, you can argue why openrc is better if you want or you could just open your mouth as wide as it can go and wonder if you should jump out of your nest before your flight feathers come in.

I can almost fucking guarantee the only thing you can think of is that changing inconveniences you and systemd is bad, without any actual reasons for believing OpenRC is better.

And I'm not faulting you for believing that, I thought the same thing, just admit to yourself that you are being fucking lazy and stop assigning value to the hours you spent trying to get some fucking runscript to work. That shit will never happen with systemd because its feature complete by comparison.

>>62402321
>What does systemd actually improve?
For the first time, the OS can be aware of the reality of modern computers: Constantly changing systems where resources appear and dissapear, instead of giant boxes where hardware and software changes werent inexistent.
Copy pasting an answer in le reddit:

With hardware becoming more dynamic and asynchronous initialization of drivers in the kernel, it was impossible to say when a certain piece of hardware would be available. For a long time, this was solved by first triggering uevents, then waiting for udev to "settle". This often took a very long time and still gave no guarantee that all required hardware was available. Working around this in shell code would be very complex, slow and error-prone: You'd have to retry all kinds of operations in a loop until they succeed. Solution: An system that can perform actions based on events - this is one of the major features of systemd.

Initscripts had no dependency handling for daemons. In times where only a few services depended on dbus and nothing else, that was easy to handle. Nowadays, we have daemons with far more complex dependencies, which would make configuration in the old initscripts-style way hard for every user. Handling dependencies is a complex topic and you don't want to deal with it in shell code. Systemd has it built-in (and with socket-activation, a much better mechanism to deal with dependencies).

Complex tasks in shell scripts require launching external helper program A LOT. This makes things very slow. Systemd handles most of those tasks with builtin fast C code, or via the right libraries. It won't call many external programs to perform its tasks.
The whole startup process was serialized. Also very slow. Systemd can parallelize it and does so quite well.

>>62402432
The whole startup process was serialized. Also very slow. Systemd can parallelize it and does so quite well.
No indication of whether a certain daemon was already started. Each init script had to implement some sort of PID file handling or similar. Most init scripts didn't. Systemd has a 100% reliable solution for this based on Linux cgroups.

Race conditions between daemons started via udev rules, dbus activation and manual configuration. It could happen that a daemon was started multiple times (maybe even simultaneously), which lead to unexpected results (this was a real problem with bluez). Systemd provides a single instance where all daemons are handled. Udev or dbus don't start daemons anymore, they tell systemd that they need a specific daemon and systemd takes care of it.

Lack of confiurability. It was impossible to change the behaviour of initscripts in a way that would survive system updates. Systemd provides good mechanisms with machine-specific overrides, drop-ins and unit masking.

Burden of maintenance: In addition to the aforementioned design problems, initscripts also had a large number of bugs. Fixing those bugs was always complicated and took time, which we often did not have. Delegating this task to a larger community (in this case, the systemd community) made things much easier for us.

https://www.reddit.com/r/archlinux/comments/4lzxs3/why_did_archlinux_embrace_systemd/

And most important thing IMO: Development for linux becomes a lot easier thanks to a common plumbing.

>>62402391
>you are implying that goals can't change or the goals can be driven by timeframes
Sure goals can be changed, but you should try and maintain that goal, and unless the new functionality is related to that goal or necessary to achieve it you should probably make a different project for different goals.
>Systemd evolved from a "init system" to a "suite of programs that can form the basic plumbing of a Linux system"
>Why? Because server and embedded needs.
Clearly server and embedded didn't need it because they were doing just fine with Linux long before systemd. Sure some improvements over sysvinit were needed like parallel startup and better service management but that's it.
>Still adds complexity that, by the words of Unix autists, it doesn't have to exist.
>Why not a type 1 hypervisor instead of a kernel that exposes shit to a user mode program?
It's not adding anything outside of the scope of a kernel. The kernels job is to handle the communication between software and hardware. Exposing hardware virtualization features to software falls into that purpose.
>Linux can be ported to itself. Google "usermode linux"
Oh, you mean containers. That just allows for multiple OS instances to be run on one OS, kind of like a hypervisor but less overhead. It's still functioning as a kernel in for those containers.
>Except that time proved that these "already existent programs" didn't do their work pretty well, like Unix weirdos like to claim. Only the 80% if those worked well.
There are always going to be low enough level programs that aren't easily replaceable, but that doesn't mean that all system programs should be that way.
>If I have to repleace the organds of my system instead of working with it, something is bad in the design of my system.
Just because you can doesn't mean you need to, having the option is still good though.

>>62402432
>>62402460
Event based startup is also in OpenRC, as is parallel startup. All of your points are only comparing to sysvinit.

>>62402421
>I can almost fucking guarantee the only thing you can think of is that changing inconveniences you and systemd is bad, without any actual reasons for believing OpenRC is better
The thing is I made a change coming to OpenRC. I was running systemd before that and had even written a couple .service files. I'm not stuck in the past, I just don't think that systemd is a good thing. I haven't spent hours getting runscripts to work because OpenRC has a simple syntax, it's not fucking sysvinit you just define a start function, a stop function and a depend function.

>>62402594
Que they daid that OpenRC isnt becoming systemd.

>>62401234
Confirmed can't read

>>62399337
If you want to ping something that isn't Google, you can use 4.2.2.2, which is a Level 3 DNS server IIRC.

>>62379887
That wad posted after OP.

>>62376523
wow, nice taking something out of fucking context,
it is true that he said that by default 8.8.8.8 is the default address for a DNS server in systemd source, but the place you took it from it was CONFIGURED that way, kill yourself, lying faggot

>>62376397
useragent "android"