Arch & "Rolling Releases"

I had never heard of the concept of "rolling releases" that Arch uses until I started using it. (In case anyone doesn't know, you never have to do an OS upgrade with Arch because new versions of the OS are merely snapshots of the current packages.) However, this completely baffles my mind. I've searched all over for information about this but have come across very little, so hopefully some people can entertain my questions a bit.
Basically, how does Arch get away with this? Why does every other OS in the world seem to not use this concept such that you have to update their OS every so often? It seems like others (provided they're not selling their OS, of course ) would want to do this as well. Do any other linux distros follow this model? Why can't Red Hat Fedora, for example, do this as well? Or maybe, more accurately, why would they choose not to?
I'm not a linux guru but I just don't understand why this isn't more common with linux distros. Now that I've been fortunte enough to have been introduced to this concept with Arch, it pains me to think about using a distro that forces me to download their newest CDs and update every 6 months or so. (Don't worry, I have no intention of leaving )

It is a nice system that is composed by aggregating components. The components themselves need to be as modular as possible, and able to be moved in and out arbitrarily. This works well on a *nix based os.
Problems inherent in the system can occur when large changes are required to some founding element of the system. For instance, if the arch devs decided to change the hashing algorithm on packages for instance..*cough*..it would take quite a bit of overhead to make such a modification. With a classical release based system, the overhead would be there, but it would be mitigated through the ability to "freeze" creation of new packages in order to make a single "one shot" change-over. My example might not best illustrate what I am trying to say, but hopefully it is clear enough that you get some of my murky meaning.
For projects that are not so componentized, or that have rapidly changing interfaces (like the linux kernel, or openoffice, for instance), a rolling release would be harder to do. Things are not broken up into discreet elements, so updating one thing without effecting another is hard. Similarly, the inability to effectively freeze one feature in relation to another until code stability can again be achieved might be serious detriment to project advancement..
those are just ideas off the top of my head. When it works, it works. When it doesn't, it doesn't.
I don't know why other distros don't use more of a rolling release. It seems well suited to linux. Now that I think of it, debian does in its own way. They just tag a release, and effectively have serveral simultaneous rolling releases and just migrate data between them (waterfalling downwards towards stability).
*end babbling*

Similar Messages

Arch as a semi-rolling release ? [Solved]

Hi,
I'm considering setting an "old" server (bi-xeon) for my students in computer science -- mainly to learn programming languages and tools. They only needs to log in via an ssh -Y session from their Windows accounts.
Arch rolling release is perfect for me as it allows to use last versions of programming languages and tools but, for the system itself (kernel and core software) i wonder if i could safely ignore updates of the core set (i prefer to update this component once a year, during student holidays, just in case something goes wrong).
I suppose i could use IgnoreGroup = core in pacman.conf, but i don't know what could be the possible side-effects.
Last edited by jaco (2012-01-07 22:59:33)

You could do as you suggested and let your students fix the server. Perfect training.

How is rolling release possible?

how do you maintain a rolling release that rolls this fast? Allow me to clarify with a hypothetical example:
i am running my favorite text based rpg called 'foo-1.0', it depends on lib-foo-1. i also play bar-1.0, which depends on lib-bar-1, and lib-foo-1.
bar 1.0 is pretty cool and its under heavy development... eventually bar 2.0 is released. yay! bar-2.0 comes with a new lib called lib-bar-2, but still depends the old lib-foo-1.
all of a sudden foo gets upgraded to foo-2.0, lib-foo also bumps up to lib-foo-2.0. is bar broken now? how do the arch devs keep up with all this?
is it up to the package maintainer to regularly check for new versions and hold them back if there is some kind of software conflict? is it up to him to make it work?
maybe all it takes to fix the broken "bar-2.0", is point the old lib-foo-1 at lib-foo-2, who would do that?
i am a little bewildered by the development of arch linux. i have used every distro under the sun, they all suck. except arch. arch has also turned out to be very stable for me, i run a full desktop, xfce, multimedia, stuff from the aur, how does it not break ?
could someone maybe explain this to me, or point me the appropriate wikki page
Last edited by machinecrusade (2011-03-12 06:42:58)

hardywu wrote:
As a new user of linux system, I noticed that "pacman" check the package dependences each time I want to remove a package.
So what would happen if there are packages
A, B and C such that A depends on B , B depends on C and C depends on A? They form a loop like
A <-B <-C <-A <-B .....
Does that means I will never be able to delete any of A, B and C unelse I format the disk ?
That would not happen. In any case there's an option to remove ignoring dependencies.

Noob question concerning rolling releases

Hi,
I'm little confused on how the rolling release /snapshot iso downloads that are available, works.
For instance, if I look into 'Don't panic 2007.08', it comes with the latest kernel found in 'Don't panic 2007.08-2'.
What are then the differences between 2007.08 and 2007.08-2?
Also, the core ISO available for download point to the latest kernel - ftp://ftp.free.fr/mirrors/ftp.archlinux … 07.08/i686
but the date indicates it was uploaded: 2007-10-04 06:26:00. Does it still mean that if I download an ISO *now* from 'core', I will get the latest packages = same as the ones I would get if I chose to do a FTP-installation?
Last edited by new2arch (2008-02-27 10:18:47)

new2arch wrote:What are then the differences between 2007.08 and 2007.08-2?
I think this is just a revision to remove bugs in the actual installer program.
new2arch wrote:Does it still mean that if I download an ISO *now* from 'core', I will get the latest packages = same as the ones I would get if I chose to do a FTP-installation?
No, but as long as you have a good internet connection without a download limit, then it shouldn't matter. With rolling release, you simply upgrade the system with the 'pacman -Syu' command and everything suddenly becomes up to date (well, by suddenly i mean a few hundred megabytes later. this is only for the first update after an install though. updating regularly will mean each update will be usually no more than 15MB)
From the arch wiki
Arch Linux releases coincide with each major kernel version, but are merely a snapshot of the /core repository, combined with various features or modifications to the installer script itself. The rolling release model keeps every Arch Linux system current and on the bleeding edge by issuing one command.
For this reason, releases are not terribly important in Arch, because the rolling-release system makes new releases out of date as soon as a package has been updated. If you are looking to obtain the latest Arch Linux release, you do not need to reinstall. You simply run the pacman -Syu command and your system will be identical to what you would get with a brand-new install.
For this same reason, new Arch Linux releases are not typically full of new and exciting features. New and exciting features are released as needed with the packages that are updated, and can be obtained immediately via pacman -Syu.

Arch Kernel Release Cycle

Hi guys,
I recently started compiling my own kernels, but I still use the arch packages as a guideline. For example I use the patches provided by Arch Linux, and at the moment I stay with the version which is in the arch repositories.
So for example if I want to compile 2.6.35.7, the PKGBUILD I use just downloads the source for 2.6.35, and the patch for .7 from Arch. Would it be 'possible' to use the stock kernel patches? What is different in the Arch packages (I know I can just diff, but why are there Arch specific patches)?
Arch gernerally updates the kernel rather fast. But some versions are skipped. Is there any policy on that? For example at the moment 2.6.36 is out, yet Arch packages 2.6.35.8 as the latest.
Thanks and kind regards
Julian

2.6.36 is in [testing]. Kernels usually do not make it to [core] until the around .2 version.
The Arch specific patches are rather small... http://projects.archlinux.org/linux-2.6 … ee/patches AUFS2 and some logo replacements.

Gnome-cups-manager and rolling release

can anyone make an update of gnome-cups-manager for voodoo? it as been 6 months since the last one...some help here please

I [03/Jan/2007:07:36:20 -0100] Configured for up to 100 clients.
I [03/Jan/2007:07:36:20 -0100] Allowing up to 100 client connections per host.
I [03/Jan/2007:07:36:20 -0100] Using policy "default" as the default!
I [03/Jan/2007:07:36:20 -0100] Full reload is required.
I [03/Jan/2007:07:36:20 -0100] Loaded MIME database from '/etc/cups': 34 types, 39 filters...
I [03/Jan/2007:07:36:20 -0100] Loading NextJobId from job cache file "/var/cache/cups/job.cache"...
I [03/Jan/2007:07:36:20 -0100] Full reload complete.
I [03/Jan/2007:07:36:20 -0100] Listening to ::1:631 on fd 2...
I [03/Jan/2007:07:36:20 -0100] Listening to 127.0.0.1:631 on fd 3...
E [03/Jan/2007:07:36:20 -0100] Unable to set ACLs on root certificate "/var/run/cups/certs/0" - Operation not supported
here is the log of cups...please some help here...

48 Hours of Arch (again)

Here's the text of a blog posting I did yesterday. I posted the link earlier, and was imformed by a mod that this was not the appropriate way to do it and I should just copy/paste the text here instead. So here it is, and I apologize for the indiscretion:
It's a huge leap from farting around on a VM to actually installing and using something new as a primary desktop OS, especially the way I use computers. My computer usage is complex and multi-faceted; I do multi-channel recording, web development, bash shell scripting, document creation using desktop publishing tools, and all of the usual day-to-day web browsing and emailing that everyone else does... all from a single desktop machine. I make full use of KDE's desktop activities in an attempt to keep my computing life somewhat organized, and do a lot of personalizing and tweaking. Here's a report of my first couple of days using Arch on the desktop.
The installation was as straightforward as it gets. As always I preserved my /home directory so as to avoid having to transfer all of my data or redo all of my desktop and app configurations. Arch does an extremely minimal install by default... absolutely nothing non-essential is installed! I like this model; with a fresh install of Arch I have a basic platform from which I can build any computer I can imagine. At the end of the installation process the Arch installer offers the user the opportunity to edit some system configuration files to his or her liking. This feature would probably be a bit off-putting for noobs, but for an experienced user it's extraordinarily cool!
The first thing I did was to run updates. Once that's done your basic platform is now a very up-to-date basic platform, in keeping with the Arch rolling release model. My next move was to install the stuff I needed in order to launch a KDE desktop. I had to install xorg, hal, phonon-xine and the KDE metapackage. I decided to install the KDE 4.6 RC2 package that was available in the KDE-unstable repository (I'd been running KDE 4.6 RC2 on this box with Alien Bob's Slackware packages already, so it made sense). I also had to add the hal and udev daemons to the /etc/rc.conf file... Arch never does stuff like that automatically; the only non-essential daemons that start at boot time are the ones I specify. I like it. I went ahead and installed the proprietary Nvidia graphics drivers while I was at it; why screw around? Once I had all that stuff taken care of I created my regular user account, making sure to add my user to all the appropriate groups. Finally I rebooted my box and switched over to runlevel five (multi-user graphical mode).
It wasn't exactly what I was expecting. Because I'd saved my /home partition I figured I'd be logging right into my familiar old desktop, but it wasn't quite that simple. Turns out Arch uses a .kde4 directory for KDE config files instead of .kde. Since I had no such directory it created one and launched me into a default KDE desktop. Fortunately it didn't take a rocket scientist to figure out what the problem was, and all I had to do was drop back into a terminal, get rid of the directory that KDE had just created for me, and rename my old .kde folder .kde4. I then logged back in and voila! Problem solved.
Well, almost. It still took some time to get my workspace back into order. I use a weather widget called YAWP and a window decoration program called Dekorator, neither of which are part of the standard KDE bundle. They weren't in the official repos either, but fortunately Arch packages for both were available from the unofficial AUR repo. Downloading and installing them was kind of similar to getting Slackware packages from Slackbuilds.org, except the process is a bit more automated and the Pacman package manager handles the dependencies for you. It turned out I had to install a few other workspace essentials as well; the plugin package for Konqueror is not a part of the default KDE bundle and had to be installed, as did Ktorrent and a few other things. That stuff was all available from the official repos. Some configuration details also had to be handled. Once all those things were taken care of the look and feel of my desktop was indistiguishable from what it had been with Slackware... a testament to both distros' commitment to providing unaltered, vanilla packages.
Then I set about installing software. Lots of software! Despite the fairly well-rounded software bundle the Slackware full install offers I'd had a ton of extra stuff installed, and my needs were compounded by the Arch minimal install approach. I installed VLC, Kaffeine, and Amarok for multi-media fun (and I shouldn't have been surprised to learn that Arch comes with no codecs installed at all!), Koffice and Scribus for document creation, qemu-kvm for virtualization, Digikam for photo management, the Gcompris educational game suite for my 3-year-old daughter, Jack, Ardour, Jamin, Audacity, Soundkonverter, and a slew of associated plugins for recording... and some other stuff, too. Some of that software had configuration associated with it; quemu-kvm requires kernel modules to be loaded at boot time, and realtime privileges had to be configured for Jack.
All of that sounds like a lot, but in truth it didn't take all that long. For one thing I'm an experienced Linux user who knows what he's doing and what he wants, and for another I had the excellent Arch Wiki to help me along when my unfamiliarity with the Arch way of doing things momentarily bogged me down. Really, that's one superb piece of documentation... I didn't have to ask a single question at the Arch forums. In any event, I had the whole thing up and running in about half a day... and I wasn't exactly devoting my full attention to it.
The rest of the time I've spent using my computer. In the last 24 hours or so I have:
1)recorded a demo using Jack/Ardour, and converted it to MP3 for web upload (no, it's not up yet)
2)Run a pre-existing Window 7 VM using quemu-kvm. I didn't do anything with it; I just wanted to make sure that it worked.
3)Used Scribus to create a brochure to send out to venues that will hopefully pay me money to play my guitar.
4)Printed said brochures.
5)Watched a movie using Kaffeine (I'm not feeling well).
6)Composed this blog posting.
7)Listened to lots of music using Amarok.
In short, Arch Linux is 100% functional on my desktop, and I'm doing all of the same things with it that I was doing with Slackware two days ago. So far it's been fast and stable, and it seems every time I turn around there's a newer version of some package or another available for download; I think this rolling release thing is going to be a lot of fun. I also kind of like the leaness of the system; there's nothing on this machine that doesn't contribute directly to it's use on my desktop. So far so good, and I'll keep you posted.

ngoonee wrote:As Inxsible said, welcome to Arch. Always interesting to have people doing linux-related audio work here, perhaps you'd have heard of archaudio? No idea what their current status is, but they do get mentioned on the LAU once in a while.
No I hadn't heard of it, thanks! Just dropped in over at their webpage; looks intriguing.

How can I send a suggestion to GNOME developers?

Hi. I have to use GNOME 3.10 because of Arch rolling release nature (which is a good thing), but there is some problems with the new design of GNOME which need to be solved for ones who use it on daily basis.
With the new merge of titlebar and toolbar there is no way to not accidently maximize the window while you clicking fast on a Previous/Next buttons of nautilus, for example. Can someone point me on how can I hint the developers about the problem? Is it nautilus or gnome guy's responsibility to fix this kind of issue?
I thought as a workaround I could disable double clicking feature in gnome tweak tool, but unfortunatelly even if I disabling them nothing happens and double clicking is still working.
Last edited by Procedural (2013-10-20 18:25:06)

if you look down the bottom of the gnome.org page, there are links for mailing lists, bug tracker and irc channels
any of them would work, but can't really tell which would be the best option in this case. maybe you could ask in some of the irc channels
Last edited by ooo (2013-10-20 18:52:07)

[SOLVED] Grub2 and LVM -- "Couldn't find physical volume `pv1'"

Hello Folks
I'm trying to upgrade from grub-legacy to grub2, following the instructions at https://wiki.archlinux.org/index.php/GRUB2
I've installed grub-bios, and run this without problem:
# modprobe dm-mod
# grub-install --recheck /dev/sda
But this command
# grub-mkconfig -o /boot/grub/grub.cfg
gives this:
Generating grub.cfg ...
/usr/sbin/grub-probe: warning: Couldn't find physical volume `pv1'. Some modules may be missing from core image..
/usr/sbin/grub-probe: warning: Couldn't find physical volume `pv1'. Some modules may be missing from core image..
/usr/sbin/grub-probe: warning: Couldn't find physical volume `pv1'. Some modules may be missing from core image..
/usr/sbin/grub-probe: warning: Couldn't find physical volume `pv1'. Some modules may be missing from core image..
/usr/sbin/grub-probe: warning: Couldn't find physical volume `pv1'. Some modules may be missing from core image..
Found linux image: /boot/vmlinuz-linux
Found initrd image: /boot/initramfs-linux.img
/usr/sbin/grub-probe: warning: Couldn't find physical volume `pv1'. Some modules may be missing from core image..
done
So now I'm reluctant to try to reboot the system because it seems likely to be broken. Should I ignore the warnings, or fix something?
I'm using LVM2 as you can see. /boot is on a separate non-LVM partition (/dev/sdc1). root is on LVM. This is all on a recently-updated 64-bit Arch installation using systemd.
Here's a load of information -- I hope it's relevant.
# fdisk -lu
Disk /dev/sdb: 250.1 GB, 250059350016 bytes, 488397168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0xea22bb30
Device Boot Start End Blocks Id System
/dev/sdb1 63 488392064 244196001 83 Linux
Disk /dev/sda: 250.0 GB, 250000000000 bytes, 488281250 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x00000080
Device Boot Start End Blocks Id System
/dev/sda1 2048 488281249 244139601 8e Linux LVM
Disk /dev/sdc: 500.1 GB, 500107862016 bytes, 976773168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x00000000
Device Boot Start End Blocks Id System
/dev/sdc1 * 63 481949 240943+ 83 Linux
/dev/sdc2 481950 12482504 6000277+ 82 Linux swap / Solaris
/dev/sdc3 12482505 976773167 482145331+ 8e Linux LVM
Disk /dev/mapper/vg1-root: 64.4 GB, 64424509440 bytes, 125829120 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk /dev/mapper/vg1-home: 583.0 GB, 583008256000 bytes, 1138688000 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
# pvdisplay
--- Physical volume ---
PV Name /dev/sdc3
VG Name vg1
PV Size 459.81 GiB / not usable 1.05 MiB
Allocatable yes (but full)
PE Size 4.00 MiB
Total PE 117711
Free PE 0
Allocated PE 117711
PV UUID zaLJiO-1LCH-TGi6-hwBr-OyNs-Sjlm-HggrMo
--- Physical volume ---
PV Name /dev/sda1
VG Name vg1
PV Size 232.83 GiB / not usable 1.58 MiB
Allocatable yes
PE Size 4.00 MiB
Total PE 59604
Free PE 22955
Allocated PE 36649
PV UUID P05c2d-1d2i-bf0M-u6BX-EEq0-fvZW-VkTLhY
# lvdisplay
--- Logical volume ---
LV Path /dev/vg1/root
LV Name root
VG Name vg1
LV UUID Z68H3p-VvbC-ZNau-7Ds7-GptS-Hpl0-VZNjo4
LV Write Access read/write
LV Creation host, time ,
LV Status available
# open 1
LV Size 60.00 GiB
Current LE 15360
Segments 1
Allocation inherit
Read ahead sectors auto
- currently set to 256
Block device 254:0
--- Logical volume ---
LV Path /dev/vg1/home
LV Name home
VG Name vg1
LV UUID uUfmS9-C4CK-Vw3V-cmwD-hEC1-VcwD-90yAyO
LV Write Access read/write
LV Creation host, time ,
LV Status available
# open 1
LV Size 542.97 GiB
Current LE 139000
Segments 2
Allocation inherit
Read ahead sectors auto
- currently set to 256
Block device 254:1
Last edited by Chris Dennis (2013-04-03 19:04:58)

Chris Dennis wrote:
Oh well, I took a punt on the word 'Warning' in the message, and rebooted anyway.
It worked!
I've just completed a series of experiments involving LVM and GRUB2. The short story is that such warnings are innocuous and arise from extending a volume group.
Now in some detail, here's what happens (all of which was performed in VirtualBox with the current Arch rolling release just to make it easy to add and remove disk devices):
a). pvcreate /dev/sde1 /dev/sdf1
* Use partitions of type 8e, spanning the whole drive, for BOTH devices comprising the physical
volume to prove that partitioning is irrelevant to the matter.
b). vgcreate vg_x /dev/sde1
* Start with just one device in the volume group.
c). lvcreate --extents 100%VG --name boot vg_x
d). mkfs.ext4 /dev/vg_x/boot && mount /dev/vg_x/boot /mnt/other
e). grub2-install --boot-directory=/mnt/other /dev/sde
Installation finished. No error reported.
All is well...but now let's extend the vg_x volume group with the pre-allocated device, /dev/sdf1:
f). vgextend vg_x /dev/sdf1
g). grub2-install --boot-directory=/mnt/other /dev/sde
/usr/sbin/grub2-probe: warning: Couldn't find physical volume `pv1'. Some modules may be missing from core
/usr/sbin/grub2-probe: warning: Couldn't find physical volume `pv1'. Some modules may be missing from core
Installation finished. No error reported.
...and boom goes the dynamite. As Chris Dennis stated, GRUB2 installs fine and the system is bootable in spite of the warning. The grub-2.00 source where the warning arises is in ./grub-core/disk/diskfilter.c and has this comment:
/* TRANSLATORS: This message kicks in during the detection of
which modules needs to be included in core image. This happens
in the case of degraded RAID and means that autodetection may
fail to include some of modules. It's an installation time
message, not runtime message. */
I haven't tried to hack the GRUB code but, based upon my experimentation and the ease of replicating the problem, my guess is that somehow a volume group that extended in manner shown above is mishandled by GRUB. It's arguably a bug, IMHO, since a volume group, even when extended, is still a valid entity.

What happened to Nvidia 3D effects in kdemod4.3?

Recently I re-installed Arch using the 08/2009 iso onto my P4(no hyperthreading)/AsusP4S800DXmob/ GeForce6200 desktop with HannspreeHF237 monitor. The video card(8XAGP) and monitor were recent additions.
The prior 2mo old install updated until midAug. gave me the 3D effects. The new install following the new wiki and my old notes did not give me the 3D stuff. Ubu and MDK(mandriva) still did. But Ubu log files mentioned ACPI not found by nvidia after the last kernel update. So I checked the var/log for xorg and I noticed that nvidia could not find ACPI... so 3D desktop was changed to 2D.
Reading the recent posts about the nvidia driver(from extra) and my GeForce6200 having problems with 3D I said EEK! I've been hit by the Arch rolling release engine!
Could there be a problem with the new kernel's built-in ACPI that nvidia wasn't seeing the ACPI module it expects? So I installed ACPID and now the kernel was showing ACPI messages and hal loaded the acpid module. /var/log/Xorg.0.log shows nvidia now seeing ACPI and loading 3D. Just to be sure I ran nvidia-xconfig( since kernel was recently upgraded) and xorg logs showed same ACPI loading. Following the log messages that AGPFastWrite and PageFlip were not used and that Freetype module doesn't exist I commented those lines from /etc/X11/xorg.conf. And now nvidia loads dri and dri2! Go figure! ---Gotta go over the fonts wiki entries again...
So the conclusion to the tale is that yes the nvidia driver from extra(repo) works fine with my GeForce. And I must read read read those logs and make sure acpid is installed. Hope this experience helps:)

Should have mentioned that ( 8/24/09)kernel gave message that my bios doesn't support acpi or apm--which it does! And during boot ACPI wasn't mentioned. Don't recall ever having to install acpid in arch. Figured the kernel missed it so I installed acpid. xorg log now shows acpi present and 3D desktop didn't revert to 2D. And the boot sequence includes starting acpi.
Will wait until I get more experience with arch linux before trying aur, though I've used yaourt. More familiar with slackware. Thanx all:)

Install Release Update for Arch?

I have not been using Arch for long. I noticed that there was a 6 month gap between the last two install releases and I was wondering if Arch developers plan to release a 2010.02 this month? I looked around and did not see anything relevant to this topic. I am unable to get 2009.08 installed (USB or CD) to boot on my system to even begin the installer (Yes I filed a bug) and would hope a new install release version will fix whatever is causing it. Anyone know if this month will have a new install release?
*I am very aware that Arch is a rolling release and that the version pertains to the installation media only.*

Tyriel wrote:
EVRAMP wrote:
Tyriel wrote:To be fair, if you are doing a net install there really is no need for a new release. If you do a core install there is nothing pacman -Syu will not fix. The only big difference I personally noticed between the last two versions was the introduction of ext4.
You are right, but you can e.g. buy a new hardware where you cannot get network working and needed drivers are included in later kernel.
That is a very good point and argument for a new release.
One can always install in other ways than with cd, the wiki got a few.

Release name for Arch

I noticed in the newsletter a comment about needing a release name for 0.8. I thought maybe we could use a name with the letters aur in it since one of the major developments was the aur. I did a search for all words with aur in it here:
http://rhymezone.com/r/rhyme.cgi?Word=a … syl&org2=l
What about aurora?
Any other suggestions?

cmp wrote:
Gullible Jones wrote:
Archaeopteryx?
(Getting ready to fly...)
Nahh, that should be the 1.0 release.
i doubt that there will be a 1.0 release. because of the rolling-release system, there will rather be a 0.10 release instead.
there has already been a 0.1 release though
Edit:
rensel wrote:
I would call it "Tundra".
[edit] Btw, seems that "Aurora" was the codename for Delphi 7 and for some Netscape release, it's not that original
I didn't pick it because I thought it was original. I just liked how the first three letters were aur. "Tundra" is not that original either. It is the release name for a truck made by Toyota
Edit again:
I also like Epoch

Why I always have to go back to Arch?

Hi there!
A little story about me. I discovered Linux a long ago (6-7 years ago) while I was interested in computer security and hacking, and got a hint the 1337s uses Linux (or any other unix based system), so I decided to try it (I choose for Linux because it was the most popular *nix based system atm).
My romance with Linux started with Mandrake (today Mandriva) and included distributions like Red hat (it wasn't enterprise oriented that time and AFAIK there were no Fedora yet), and the lovely Gentoo. Its funny tough, Ubuntu wasn't that popular at the time (I even remember how I was on an open lecture in University about open source and Linux, and I got a free CD of Ubuntu and no body heard about it before )
Anyway its stopped there for few years and I got back to Linux about 2 years ago, with Ubuntu on my desktop machine.
My nature is to discover and be different. If everybody uses Ubuntu, I must use something else (even now, I'm on Xubuntu).
This nature lead me to replace Ubuntu with Arch. This was the first time I encountered a very similar distribution to Gentoo, that were optimized to i686 and did not require you to compile every package (for instance I installed Gentoo like 2-3 times from scratch [those were beautiful Friday evenings-nights :romantic-smiley:]) and I remember how compiling Xorg took me hours (and when I say hours I mean hours, something like 6-8 hours).
Then I got back to windows since my old PC died and I got a new one and a laptop.
This leads us to nearly today. Right now I'm using my laptop as my main computer at home and at work (the desktop PC had a motherboard issue that was replaced under warranty. While they were fixing it, I switched to use my laptop).
I started with Ubuntu, simply because it works! But I hated Untiy and I switched to Xubuntu. But you do remember my nature right? This lead me to switching to Arch.
A few days of configurations and I got stable up and running system. Then I came to work and had some issues (especially with keyring) and they blocked me from working, instead I needed to fix them. Then I said "This is it! I need a machine that works!".
I removed Arch, installed Xubuntu and promised my self never to do this pointless Distro_name->Arch->Distro_name path, because well, it ends the same way always.
But you do remember my nature right? So right now I'm standing in front of a decision whether to go back to Arch or not. I must say I really like the rolling release scheme, I'm in love with pacman (its a lot more superior than apt-get or yum IMHO), I like the way you can customize Arch.
But Archs customization is also its weak point, you have to devote hours or reading and configuring to get a stable and running system. And sometimes (especially if you use your laptop for work or for studies or both) you need to have a running system and you don't want to mess with cups just because its 2 A.M. and you need to print your project to submit it in 5 hours (yes you must take care of it at a lot earlier point than 5 hours before deadline )
Ubuntu is nice, but its bloated. It doesn't even give you the option to select what packages you want to install (unlike Fedora for example). Its comes with (IMHO) stupid meta packages like ubuntu-desktop, xubuntu-desktop so its nearly impossible to try gnome3, xfce, openbox on the same system without breaking something.
Ubuntu simply works, but as soon as you need something deeper, you are screwed.
I don't really ask sort of a question here, but I know many of you use Arch (I'm on bbs.archlinux.org doh -.-' ) as your everyday system, so I would like to know how do you handle its configuration, what you do when something breaks and you don't have the time to fix it (you are at work, hitting the deadline of your university project or whatever). Maybe you will be able to convince me why I always want to get back to Arch and what I need to do to choose it as my the one and only distribution. Maybe you will be able to suggest me another distro that fits my needs. Or maybe you want to tell me how sucky I'm.
So simply go on and post, this is why I created this topic.
Thanks you for your time

skwo wrote:
Hi there!
A little story about me. I discovered Linux a long ago (6-7 years ago) while I was interested in computer security and hacking, and got a hint the 1337s uses Linux (or any other unix based system), so I decided to try it (I choose for Linux because it was the most popular *nix based system atm).
My romance with Linux started with Mandrake (today Mandriva) and included distributions like Red hat (it wasn't enterprise oriented that time and AFAIK there were no Fedora yet), and the lovely Gentoo. Its funny tough, Ubuntu wasn't that popular at the time (I even remember how I was on an open lecture in University about open source and Linux, and I got a free CD of Ubuntu and no body heard about it before )
Anyway its stopped there for few years and I got back to Linux about 2 years ago, with Ubuntu on my desktop machine.
My nature is to discover and be different. If everybody uses Ubuntu, I must use something else (even now, I'm on Xubuntu).
This nature lead me to replace Ubuntu with Arch. This was the first time I encountered a very similar distribution to Gentoo, that were optimized to i686 and did not require you to compile every package (for instance I installed Gentoo like 2-3 times from scratch [those were beautiful Friday evenings-nights :romantic-smiley:]) and I remember how compiling Xorg took me hours (and when I say hours I mean hours, something like 6-8 hours).
Then I got back to windows since my old PC died and I got a new one and a laptop.
This leads us to nearly today. Right now I'm using my laptop as my main computer at home and at work (the desktop PC had a motherboard issue that was replaced under warranty. While they were fixing it, I switched to use my laptop).
I started with Ubuntu, simply because it works! But I hated Untiy and I switched to Xubuntu. But you do remember my nature right? This lead me to switching to Arch.
A few days of configurations and I got stable up and running system. Then I came to work and had some issues (especially with keyring) and they blocked me from working, instead I needed to fix them. Then I said "This is it! I need a machine that works!".
I removed Arch, installed Xubuntu and promised my self never to do this pointless Distro_name->Arch->Distro_name path, because well, it ends the same way always.
But you do remember my nature right? So right now I'm standing in front of a decision whether to go back to Arch or not. I must say I really like the rolling release scheme, I'm in love with pacman (its a lot more superior than apt-get or yum IMHO), I like the way you can customize Arch.
tl;dr
But Archs customization is also its weak point, you have to devote hours or reading and configuring to get a stable and running system. And sometimes (especially if you use your laptop for work or for studies or both) you need to have a running system and you don't want to mess with cups just because its 2 A.M. and you need to print your project to submit it in 5 hours (yes you must take care of it at a lot earlier point than 5 hours before deadline )
That is the strong point. Arch's customisation allows me to use it on the craptop and on the desktop. I prefer to see it as an investment of time to get a system I want instead of one a developer thinks I want (don't get me wrong open source devs do an amazing job!)
Ubuntu is nice, but its bloated. It doesn't even give you the option to select what packages you want to install (unlike Fedora for example). Its comes with (IMHO) stupid meta packages like ubuntu-desktop, xubuntu-desktop so its nearly impossible to try gnome3, xfce, openbox on the same system without breaking something.
Ubuntu simply works, but as soon as you need something deeper, you are screwed.
Bloat is the price you pay to have a distro that works out of the box.
I don't really ask sort of a question here, but I know many of you use Arch (I'm on bbs.archlinux.org doh -.-' ) as your everyday system, so I would like to know how do you handle its configuration, what you do when something breaks and you don't have the time to fix it (you are at work, hitting the deadline of your university project or whatever). Maybe you will be able to convince me why I always want to get back to Arch and what I need to do to choose it as my the one and only distribution. Maybe you will be able to suggest me another distro that fits my needs. Or maybe you want to tell me how sucky I'm.
So simply go on and post, this is why I created this topic.
Thanks you for your time
My arch setup is relatively simple, I set cron jobs to back up files with rsync and keep important config files backed up too. I am lucky by the fact I have a laptop I can use if the desktop breaks and vice versa but the sensible solution is to upgrade after work is done.
I keep coming back to arch because of pacman and the AUR. No package manager comes close.

Is Arch the best distro for a friend with a "computer curse?"

Hey, guys! I need some advise. I have a friend who has a "computer curse." It runs in his family as well. Computers break down FAST in his house. He was running Windows initially, but he switched to Linux because he kept getting NASTY viruses. Viruses so nasty, I would often have to use Windows Repair Install. Linux, obviously, eliminated that problem. It created a new problem, though. Linux constantly BREAKS on him! To be more specific, he's been using Debian-based distros. Ubuntu, Linux Mint, etc. All of them have been breaking on him. It's incredibly frustrating.
I've used Arch many times before. I absolutely love it, and I'm beginning to wonder if it would work for him (after I set it up for him, of course). I'm worried, though, about two things directly related - package management, and the system breaking. He ALWAYS finds a way to break the system. ALWAYS. I can lock down package management, but if I do that, he will go months and months without an update; increasing his chances of a breakage if a major security update is needed. Is there a way to balance this out?

I agree, debian stable is the winner choice for him. That's unbreakable without root perms. And I think anyone who uses linux for more then a year, finishes at arch, debian or gentoo (and some insists on slackware). Reason is: full customizability, no company background (mp3 in fedora/suse add repos issue) basic configs didn't variated, gentoo+arch:stable rolling release (innovations 5 years later even with this install, imagine you installed an xp in 2001 and now that install is a win7), debian: stable till die (but out of date)
I killed my first 2 ubuntus one and a half a year ago, and I made panicing kernel half a year ago in gentoo, which even paniced other linux installers if that hard drive was plugged in
If the guy loves innovations, you can set him up arch (or maybe the chakra project) but I basically say, the perfect solution is debian lenny with stable kde3. Old but more stable then anything I've ever seen, and fast (even on an 800Mhz 128 mb ram laptop)
And another advise: don't install him sudo. He may know google too well to use it
Last edited by Vegita (2010-01-02 19:47:03)

I'm going to install Arch this weekend but I have some questions

Currently I use Linux Mint on my primary PC, but I've installed Arch on my older PC at my parents house. I like it a lot and I think I want a distro that is rolling release and also that I build myself (as opposed to installing all of the bloat on Mint). However, I do have some questions; Linux is pretty new to me (been using it since about November) so I don't know exactly how everything works.
1) How do I know what packages I have to have installed for building C/C++ programs? I know in Ubuntu install gcc/g++ by itself doesn't get the job done, you need build-essential to get all of the libraries. Also, what packages do I need for OpenGL and SDL? I'm a CS major and I'm in an OpenGL class so I need to be able to compile OpenGL programs; SDL is for my own purposes, but I'd still like to have it.
2) I use Amarok to transfer songs to my iPod. Amarok 1.48 and libgpod 0.6.0 are in the repos and I know both of those are compatible with my 6th gen. However, I do have a question about transferring album art. On Mint, Amarok transfers the album art as I transfer the album, but someone told me that Amarok doesn't do that by default; there must have been a setting changed somewhere to do that. I looked through Amarok's options and didn't see anything like that... anyone know if I can do that in Arch, and how? There's no real information about Amarok in the wiki.
3) I'm going to use Gnome environment but some KDE apps (like Amarok). Will there be any problems with that that I should know about in advance?
I'll probably have some more questions once I actually install Arch, but that will do for now Those are the important ones.

1) You know by seeing its dependencies on pacman or by the developer documentation. Unlike other distros, you don't need to install dev packages. Example: You have gcc and all the basic gnu tools installed, and you want to compile a program that is not in a arch linux package yet (if theres a package, you can build it using abs / makepkg and pacman will handle the dependencies) and requires, for example, a library called "xyz", you only need to download "xyz" from pacman, and all the dev stuff will come together (like header and etc).
I hope it helps.

Arch & "Rolling Releases"

Similar Messages

Maybe you are looking for