Arch linux install and setup questions

Personally it seems a bit to cumbersome to have to setup xorg etc and is proving to be difficult in getting it to work right. Do they have any arch forks that just auto detect the xorg settings like some of the other distros?
I can't see myself having to setup several of these boxes and doing these install steps one at a time for each one.
It would seem arch could just run a script once installed that would auto sync, download xorg, and setup xorg at the same time.
it would be no different that following each step in the wiki by hand, just via script

itsmeh wrote:Yup it was just an idea , sure people can still do it manual but maybe have something built in to at least take care of the first few steps that most users are going to use
That's the whole point with arch. It's all up to you. Arch doesn't suggest anything for you (exept what you need to get a working linux enviroment).
This is the reason i installed arch. I wanted control over my inviromet and learn more about what's under the hood.

Similar Messages

  • Arch Linux install and management on multiple workstations

    Hello
    I've been asked to install and manage an installation of 20+ computers at my school. They will be dual-boot with Windows and Linux. Even though i've never used Arch, i chose it for the install because i've heard a lot of good about it. I ask for tips on installing and managing such an installation. All the computers and their disks are identical, and i wish all my systems to be identical as well. There is a separate server with LDAP to which i have no root access, but which holds all user accounts.
    Install:
    I probably can suffer installing all the systems manually. I'll dedicate one workstation and create an ftp mirror so that i may do a netinstall on the others -- is this an ok idea? Also, how can i setup hostnames? Can DHCP do this?
    Management:
    The computers will be randomly shut down and powered on as the users please. Perhaps i can boot them all on once a fortnight and use cssh (cluster ssh) to do a system update -- but maybe there is a better way? A local mirror of the packages seems wise. I don't think it's a good idea to create a diskless setup, since if all machines are powered on at once it will significantly decrease boot time and increase network usage, to the annoyance of the users.
    I ask for criticism of those ideas and additional hints.
    PS. I appreciate the joke of asking someone without a Linux installation what the output of a Linux command is:) (when registering to this forum)

    If I was in your position...
    Setup arch on one machine, from scratch to production ready, with the following in mind.
    -Decide what desktop environment and window manager(s) you want installed/want to support.
    -Which applications your users will need
    -Which privileges are required or groups your users will need to be in.
    -Install everything, and note the names of the packages, everything installed after the base install.
    Then make a simple setup sh script that calls pacman to install everything you need and check any settings/group memberships etc, checking the logs afterwords to make sure that there were no issues. You could use this to streamline setup on the rest, or just do what tomk said and clone it.
    I wouldn't wait more than a week between updates. You could probably setup a cron job to run pacman and accept the defaults. You should run this on a test machine before the rest though to insure nothing breaks; that happens occasionally.
    I would probably choose Gnome as the DE personally. Not that it is any better than the others, but it is what I am familiar with, and seemed a bit less cluttered than KDE last time I used it. Less clutter and easily changeable settings means less time fixing users' mistakes.
    You could setup your own local arch repo/mirror and just have the rest of the machines sync from that, it'd save a lot of bandwidth and make updates a bit faster.

  • HT5569 Hi is it possible to install and setup a modem with an iPad only, I would appreciate  an answer to this question thanks

    Hi is it possible to install and setup a modem with an iPad only, I would appreciate  an answer to this question thanks

    The configuration page of my router on Safari (tap to enlarge)
    See IP Adress at the top (center)

  • I must install and setup TDPO on all nodes RAC?

    Hi, everyone
    I have oracle rac 9i four nodes and filesystem archivelogs per hosts. without NFS.
    my question is. Is necesary to backup de archivelogs install and setup TDPO on each node? and by the way. The node_name is tdpo.config it must be the same on all nodes rac.?
    Many Thanks.

    A good place to start is http://www.bombich.com/.

  • Official Arch Linux Install Guide needs edit

    Aloha,
    This line in the Official Arch Linux Install Guide
    Official Arch Linux Install Guide
    dd if=archlinux-XXX.img of=/dev/sdX
    Needs to be changed too
    dd if=archlinux-2010.05-{core|netinstall}-{i686|x86_64|dual}.iso of=/dev/sdx
    Mahalo

    Hum, well I don't know how to write a git patch. I guess it would be a good thing to learn though, but I have to go to bed now.
    Hum..., after further looking at the this forums section I guess I posted in a totally wrong place Live and learn.
    Last edited by hunterthomson (2010-05-26 13:16:47)

  • Computer freezes before BIOS after arch linux install

    This is my first time installing arch linux, I followed the beginners install guide word for word. However, after the install, my computer gets stuck before the BIOS. It identifies the RAM, GPU, and harddrive (all Diagnostic lights are ok on my MOBO). Before the install it booted fine, but now when I boot it up it wont go to the BIOS screen no matter how many times I press delete. However, when I remove the harddrive it boots to the arch linux install screen on my USB drive perfectly.
    I know it is not my hardware because I did many different tests (mem-test, swapped out harddrives, tried 1 stick of RAM in different slots, etc). More importantly, my harddrive is brand new so I know it isn't dead because I could partition it when I was installing arch linux.
    Computer specs:
    AMD Athlon 64 Processor 3700+
    DFI LANPARTY nf4-d
    ATI Radeon x800
    2 GB DDR RAM
    160 GB SSD
    I have no idea what the problem is, I must have screwed something up with the bootloader or something. Any help would be appreciated. Thanks.

    Try clearing the CMOS settings either by removing the motherboard battery for some minutes or by using the relevant jumper on the motherboard. Check the manual of your motherboard for details.
    Last edited by Foucault (2013-07-03 16:19:41)

  • Download install and setup

    first computer and would like to learn how to do this.

    Hello Pat. Im not sure of the computer it was given to me and older one. But Im trying to download install and setup the Adobe flash player to the computer to download music on the computer. Ive done what it said to do but it keeps taking me back to the same site to download. As I stated this is my first computer and I know "NOTHING" about a computer. Im learning this on my own. So Im asking that you my be able to show me how to do this download install and setup.  

  • [Solved] installation and setup question - gummiboot and EFI

    Hi,
    This is the first time i'm installing arch. i hit the below issues. hope someone can help me out. my understanding of uefi is just about an hour worth of reading or less..
    (the wiki didn't mention the kernel requirement when I started installing it... now it does and i updated the beginner's guide wiki post.)
    Problem
    Gummiboot#Installing says that "If you are still running kernel 3.7 or have not booted in EFI mode, creating the boot entry will fail. You should however still be able to boot gummiboot as it copies the binary to the default EFI binary location on your ESP (/boot/EFI/BOOT/BOOTX64.EFI on x64 systems). "
    the gummiboot part fails. gummiboot gives this error at install  saying file system is not FAT EFI, so first I got gummiboot and gummiboot-efi both using pacman. then, I provided the --path to install and now it coughs up because the latest iso isn't 3.8 yet.
    Failed to access EFI variables. Is the "efivarfs" filesystem mounted?
    I can see efivars but this is asking for efivarfs.
    so instead I did,
    # cp /usr/lib/gummiboot/gummibootx64.efi /boot/efi/EFI/gummiboot/gummiboot.efi
    # efibootmgr -c -d /dev/sda -p 1 -w -L "Gummiboot" -l '\EFI\gummiboot\gummiboot.efi'
    how do I set this up? I'm afraid if I overwrite /boot/efi/EFI/BOOT/BOOTX64.EFI  windows might not boot.
    after the installation (of course without gummiboot stuff) now when I reboot, I don't see any gummiboot menu at all because i didn't copy it to EFI/BOOT. it goes straight to windows 8 boot. during boot if I press F12 and see the menu option on thinkpad, I see gummiboot as the first entry(must be because of efibootmgr) followed by windows boot manager. When I select the gummiboot and enter, it just circles back to the same menu. I have setup the loader.conf and arch.conf as described in the wiki.
    how do I set gummiboot to work? i dont mind pressing F12 and then going to gumiboot-> arch until 3.8 is out.
    If this isn't possible, can I install rEFInd without messing gummiboot? looking at the dir structure of FAT EFI I think it should be possible but not 100% sure until i try it out. Also, to get rid of gummiboot, i should delete the directories, uninstall using pacman and get rid of the efibootmgr change I made, right?
    Last edited by lobo2 (2013-03-15 04:07:47)

    srs5694 wrote:
    sidneyk wrote:Maybe not, I use rEFInd and mount ESP on /boot/efi and copy over updated kernels and initramfs. Maybe if you are mounting ESP at /boot the kernel and initramfs get written to both places at once, I don't know. I also don't know if /boot is where gummiboot expects them to be.
    I realize this has been at least partially addressed by others, but I want to make two points very explicit:
    In Linux, partitions (or more precisely, the filesystems that most of them contain) are accessed by mounting them at a mount point (a directory). Thus, when you mount the ESP at /boot, the contents of /boot are the ESP. Put another way, mounting the ESP at /boot means that to read or write the ESP, you read or write files and directories under /boot. You can test this yourself by mounting and unmounting a partition that's not basic to the minute-to-minute functioning of the computer. The ESP will actually work fine for this. Unmount it from /boot or /boot/efi and then issue a mount command to mount it elsewhere (say, "mount /dev/sda1 /mnt"). Then check the files under your temporary mount point, and even copy a file there. You can then unmount it and mount it back where you normally do and look for the file you copied -- it should be there. Play with this until you fully grok it.
    Gummiboot runs before Linux, and it has no conception of Linux mount points. Thus, whether you mount the ESP at /boot, at /boot/efi, at /home/fred/abadmountpointforanesp is irrelevant, so long as the files on the ESP are laid out in the way that gummiboot expects. The trickiest aspect of this for those unfamiliar with the Linux mount system is that you do not include the Linux mount point in filename references for pre-boot programs like gummiboot. That is, if you mount the ESP at /boot and place the kernel at /boot/vmlinuz, then in gummiboot, the kernel would be \vmlinuz.
    Yeah, I figured this out based on the OP's response to that and thinking it through, I just never really thought about it much before now. But I will argue that the mount point /boot and the EFI partition are indeed 2 different locations. I know that once the ESP is mounted at /boot that this distinction disappears until it is umounted and something written to /boot, but while mounted as such, /boot and ESP are virtually the same, at least to the Linux system. I realize the EFI programs are ran before any OS and therefore don't really care about any OS specifics.
    I wonder if the OP has confused the info between gummiboot and rEFInd (possibly from the Beginners Guide) because to me, after reading the gummiboot wiki, it just seems that it would be much simpler to just give gummiboot what it wants, i.e. ESP mounted at /boot, whether it's a separate partion or not. I'll have to take a look again at the Beginners Guide to see again the order of things there, but if it isn't, maybe it should be having the user ensure that pacman has completely updated the new system and then go on to the boot loader specifics. I was just thinking though, that the way the install medium is now, that it was automatically pulling the latest stuff down as part of the install process. I know it's not pulling testing or multilib without user intervention, but the core, stable stuff it is. That seems to be part of the confusion here maybe, 2 different versions of gummiboot with gummiboot changing some things between the 2 and maybe the kernels, too.
    [EDIT] I took another look at the Beginner's Guide and it says that most users will probably want to use the [core], [extra], and [community] repos, which are enabled by default. It goes on to recommend that [multilib] be enabled for users wanting to run 32 bit programs and then reminds you to run #pacman -Sy to update the package lists, but not #pacman -Syu which would apply any changes from the repos since starting the install. Shouldn't matter much, even if the [multilib] repo was enabled because at that point no 32 bit libs have been installed, but it would matter if a user decided to enable the testing repos.
    I wouldn't ordinarily recommend it, but in this case, if the OP is comfortable with the risk, then it might be useful to, at least temporarily, enable the [testing] repo and do #pacman -Syu to pull in the 3.8 kernel. Note that this is likely to also pull in other packages from [testing], but if you haven't installed X yet or any desktop environment, then what is pulled in shouldn't be too overwhelming. He can then immediately go back and disable [testing] if desired. I'm currently using testing without any issues, but there are risks to doing so.
    Last edited by sidneyk (2013-03-14 18:13:43)

  • Need help with arch linux install!

    Hello, let me start off by telling you about the predicament I'm in at the moment. I've been trying to install arch linux on with a cd-drive that has been failing for a little over a year now, so you can already see where my problem starts!  The cd-drive, however will work with such cd's as Windows, Ubuntu, and Slackware, which I'm happy with. Yet, archlinux doesn't work so much.
    And now maybe my solution. I've got a usb cd drive, that I've had laying around that works well, just not well enough for my bios to be able to boot from; which is horrible I know. I've searched the bios for loading a usb cd drive and everything else, including google so I don't think I'm going to get anywhere with that one. Anyways what I'm thinking of doing is that when archlinux is in the terminal right before you type the "/arch/setup" command, I would like to know if there is any command that allows you to change the cd it boots the cd from to go right into the installation. I'm still working on easying my out of the newbie stage of linux at the moment, but I've never run into a command like that, that would help me out. Hopefully there is one though, because I would very much like to archlinux to work, so please respond! Right now I'm running elinks through the terminal right before you type "/arch/setup". Thanks!

    loosec wrote:
    So you get through the boot process just fine then on your old CD drive. The only thing you should have to do after that is to throw your CD into the USB CD drive and proceed with the installation. The installer should then ask you what CD drive you would like to use when you choose install from CD.
    Another way to do it would be to switch cd like above and then manually mount your /dev/[your_usb_cd_here] on /src. The installer then has to be pointed to this place as the source of your installation. To find out what your usb cd drive is called you should unplug it, then replug it and run: dmesg
    The mount command might also need to be shown that this is a cd so:
    mount -t iso9660 /dev/[usb_cd_device_name] /src
    should do the trick.
    If you really want to learn more of what the installer does (and fail installing a few times) you should also check out the command
    /arch/quickinstall
    If your Internet connection is up for it, it seems like you could just choose an FTP install instead.
    Goodluck!
    Once the CD is removed from the drive that it was booted from. the installation will not continue, as the arch install always goes back to the boot drive to continue read the neccessary information for the install. I do not think there is a way to fool the computer into booting from one cd drive and then continueing on with another. To do this you would have to tell the computer that you are switching drives.
    If you can boot up the CD kernel and get to where you type in /arch/setup then the next thing I would od is FTP install. That would limit your CD rom drie usage.
    There is also a way to install arch from within another distro, EVEN ANOTHER LIVECD! <-- this may be your best bet. Check the wiki, there is tons of information there.
    Last edited by rooloo (2008-08-16 12:32:00)

  • [SOLVED] Arch Linux install on UEFI motherboard (Asus UX31A)

    Hello all,
    I am new to Arch Linux (moving over from Mint) and I am having an enormous amount of trouble with booting for UEFI.  I go through all the command line installation steps and get to the step where you install and configure a bootloader.  I have followed everything step-by-step up and cannot seem to get anything to boot, I get the motherboard settings menu instead.  I've tried GRUB, rEFInd, and Gummiboot and all three have not worked for me.  Is there any hope for configuring a bootloader?  I am used to Mint/Ubuntu creating it for me.
    Thanks
    Last edited by somedood (2013-06-09 13:23:51)

    srs5694 wrote:It shouldn't matter from where you launch the EFI shell, although that will affect device names. If you boot a USB flash drive with an EFI shell on it, chances are fs0: will be the USB flash drive itself. Your rEFInd, though, should be installed on your hard disk, and therefore on another device -- probably fs1:, although it could be fs2:, fs3:, or some other number. If the only filesystem you can find is fs0:, then that indicates something is wrong with your hard disk setup -- perhaps your ESP's filesystem is damaged, or maybe there's something wrong with the partition table.
    I followed the steps in the beginners installation guide to create my file system starting with:
    cgdisk /dev/sda1
    I create 4 Linux filesystem partitions with this, then I run:
    mkfs.vfat -32 /dev/sda1
    mkfs.ext4 /dev/sda2
    mkfs.ex4 /dev/sda3
    mkswap /dev/sda4
    I then toggle the boot flag of my FAT32 formatted ESP (/dev/sda1):
    parted
    toggle
    1
    boot
    if it is not turned on according to the output of:
    parted
    print
    After that I mount the drives and turn swap on:
    mount /dev/sda2 /mnt
    mkdir -p /mnt/home
    mount /dev/sda3 /mnt/home
    mkdir -p /mnt/boot/efi
    mount /dev/sda1 /mnt/boot/efi
    swapon /dev/sda4
    I install my base system (without changing /etc/pacman.d/mirrorlist) by running:
    pacstrap -i /mnt base
    Finally I generate my fstab after the Arch installation:
    genfstab -U -p /mnt >> /mnt/etc/fstab
    I then check the fstab and I get the result from my previous post and according to the guide it is ok because everything but my root partition has a 0 or 2.
    somedood wrote:# /dev/sda2
    UUID=264fe719-b816-462d-af5b-1b76c73a875b     /     ext4     rw,relatime,data=ordered     0     1
    # /dev/sda3
    UUID=824c5b36-6bf3-4c66-8878-5c886dadc9dd     /home     ext4     rw,relatime,data=ordered     0     2
    # /dev/sda1
    UUID=49B1-AE7E     /boot/efi     vfat     rw,relatime,fmask=0022, dmask=0022,codepage=437,iocharset=iso8859-1,shortname=mixed,errors=remount-ro     0     2
    # /dev/sda4
    UUID=426143bf-0cb4-4b7e-8868-e82d6fe7bd8b     none     swap     defaults     0     0
    To clarify, sda2 and sda3 are my root and home partitions respectively, sda4 is my swap, and sda1 is my ESP.
    What could have gone wrong here?  This is quite frustrating to me and I really want to use Arch.

  • Why arch linux install media is so big?

    Hi all,
    Yesterday I installed arch linux in my PC. My very first learning/questions regarding arch linux:
    * Why is the install media 650 MB(so big) iso, when it doesnt contain any packages?
    * From the https://wiki.archlinux.org/index.php/Installation_guide page,
       Download
    Download the most recent Arch Linux installation ISO image from the Arch Linux download page: this is a hybrid image that allows booting into an x86_64 or i686 live system, depending on the system's architecture and the user's choice.
    Why the most recent Arch Linux installation ISO? The ISO which is downloaded one year before also will install the latest linux right because everything is downloaded from internet?
    Can anyone please explain what is special about new install media??
    Thanks
    Mario

    Mariappan wrote:
    Download the most recent Arch Linux installation ISO image from the Arch Linux download page: this is a hybrid image that allows booting into an x86_64 or i686 live system, depending on the system's architecture and the user's choice.
    Why the most recent Arch Linux installation ISO? The ISO which is downloaded one year before also will install the latest linux right because everything is downloaded from internet?
    While you are correct that a netinstall will indeed install the latest linux version, it does not mean that an old install medium will boot on the newest hardware, or is up to date with the latest Arch specific packages etc. Using an old installation image can mean that your hardware is not supported, old bugs are included or that you are not at all able to install Arch because of outdated Arch specific packages on the install medium.
    The install medium is merely a specific set of packages combined together in an image, not a special project to install Arch (you can install all packages via pacman). It needs updating, just like a regular Arch install.

  • Arch Linux Partitions And Filesystems

    Hi everyone!
    I'm pretty much new to Linux having used Ubuntu about 2 years back. Recently I've been looking to get a better understanding of the system and wanted to try out a different distro. I searched around and came across Arch. I really liked the principle on which this distribution is based and  have decided to give it a go.
    I have 2 SATA HDDs in my system.
    1. 1TB Seagate --> Windows 7
    2. 160GB Seagate --> Windows XP
    I want to install Arch on the 160GB HDD. I'll remove XP and the entire HDD will be for Arch.
    What I would like to know is the partitions that I should create. Should I go with /boot, /, /var, /usr, /home, /temp? Or lesser number of partitions?
    What sizes should they be?
    And which ones should be primary and which ones extended, if at all? What file systems should I use for the different partitions? (I had been considering ext4,  XFS and Reiser.) It would help if someone would direct me to the appropriate ones for each partition.
    Also, I could not understand the difference between the /usr and /home directories. What are their uses? Where do my programs (obtained through Pacman or otherwise) get installed?
    Any help is greatly appreciated.
    Thank you.

    According to the wiki:
    / (root) The root filesystem is the primary filesystem from which all other filesystems stem; the top of the hierarchy. All files and directories appear under the root directory "/", even if they are stored on different physical devices. The contents of the root filesystem must be adequate to boot, restore, recover, and/or repair the system. Therefore, certain directories under / are not themselves candidates for separate partitions. (See warning below).
    /boot This directory contains the kernel and ramdisk images as well as the bootloader configuration file, and bootloader stages. /boot also stores data that is used before the kernel begins executing userspace programs. This may include saved master boot sectors and sector map files. /boot is essential for booting, but is unique in that it may still be kept on its own separate partition (if required).
    /home User data and user specific configuration files for applications are stored in each user's home directory in a file that starts with the '.' character (a "dot file").
    /usr While root is the primary filesystem, /usr is the secondary hierarchy, for user data, containing the majority of (multi-)user utilities and applications. /usr is shareable, read-only data. This means that /usr shall be shareable between various hosts and must not be written to, except in the case of system update/upgrade. Any information that is host-specific or varies with time is stored elsewhere.
    /tmp directory for programs that require temporary files such as '.lck' files, which can be used to prevent multiple instances of their respective program until a task is completed, at which point the '.lck' file will be removed. Programs must not assume that any files or directories in /tmp are preserved between invocations of the program and files and directories located under /tmp will typically be deleted whenever the system is booted.
    /var contains variable data; spool directories and files, administrative and logging data, pacman's cache, the ABS tree, etc. /var exists in order to make it possible to mount /usr as read-only. Everything that historically went into /usr that is written to during system operation (as opposed to installation and software maintenance) must reside under /var.
    I really don't see the difference between usr and home on a typical Linux install, especially if you're the only person using the PC.
    As for the other storage partition, I am using a primary NTFS partition for storage. I am using NTFS because I have access to the partition inside Windows and Linux, which makes swapping files around easy and I can use all of my Windows files in Linux and vice-versa.

  • Trouble with a wired connection on my fresh Arch Linux Install

    I am a new Arch Linux user. I used to use Ubuntu exclusively, but I'm to the point where I can no longer stand a lot of the issues with the new development and Canonical. Anyways....
    I have a Dell Inspiron 1440 laptop with an RTL8110SC (Module R8169) Ethernet port. (Pardon me if my terminology is wrong; please correct me if it is so I can learn? ) During the install process with the CD that I made, I had no issues connecting to the Internet. However, whenever I boot into my install, there is no connection. I've tried pinging various web pages and nothing will happen. I know that my hardware is compatible, but I do not know what to do next. I've tried everything under the Wiki's Beginner's Guide and the Network Configuration Guide.
    I just realised after I wrote this top portion that it no longer is connecting via CD, either. Pinging is giving me errors about no packets being received.
    Please help me with this matter? I would love to start using Arch Linux!

    Ping is a very basic command that is fundamental to debugging networks at the lowest levels.
    ewaller@odin:~ 1002 %ping 8.8.8.8
    PING 8.8.8.8 (8.8.8.8) 56(84) bytes of data.
    64 bytes from 8.8.8.8: icmp_seq=1 ttl=50 time=42.7 ms
    64 bytes from 8.8.8.8: icmp_seq=2 ttl=50 time=36.7 ms
    64 bytes from 8.8.8.8: icmp_seq=3 ttl=50 time=36.8 ms
    ^C
    --- 8.8.8.8 ping statistics ---
    3 packets transmitted, 3 received, 0% packet loss, time 2003ms
    rtt min/avg/max/mdev = 36.721/38.782/42.768/2.819 ms
    ewaller@odin:~ 1003 %
    What we can tell so far..
    You have an Ethernet Interface.
    You have an Ethernet Address.
    You cannot look up domain names (bad)
    We are trying to see if you can see other computers if you know their address.  This will tell us if it is a DNS or a routing problem
    Last edited by ewaller (2012-10-26 22:58:32)

  • X server won't start (Arch Linux installed on USB key)

    Hi,
    I installed Arch Linux on a USB key (see the original thread: https://bbs.archlinux.org/viewtopic.php?id=185441).
    The problem is that X doesn't start on all machines.
    When starting the system on a machine with NVIDIA GTX 560Ti graphics card:
    - X doesn't start using startx or xinit and there are no log entries in /var/log/Xorg.*.log (as I haven't tried to start X).
    - I'm getting the message "Waiting for X server to begin accepting connections .. .. .. ..".
    - Additionally: The "default terminals tty1/2/3/..." (which I'm using to start X) from have a poor resolution (I think 640x480 pixel).
    When starting the system on a virtual machine or a machine with an ATI Radeon (mobile) graphics card:
    - X starts and runs without any trouble the XFCE desktop environment.
    - Additionally: The default terminals have a proper resolution (I think the maximal resolution of the display).
    What happens if you uninstall nvidia and use nouveau?
    I installed all video drivers recommended here: https://wiki.archlinux.org/index.php/In … eo_drivers
    So I was using nouveau in the first place. I already tried to uninstall the open source driver (xf86-video-nouveau, nouveau-dri) to use the default driver (I think this is xf86-video-vesa) with no effect.
    The poor tty resolution is to be expected if the closed-source nvidia drivers are installed, because they don't support KMS.
    I have never tried to install the proprietary driver (and actually don't want to use the proprietary driver).
    Also have you looked in /etc/X11 ?
    Yes. I don't think that my configuration contains anything preventing X from starting using the NVIDIA graphics card (but I'm not sure what to look for).
    I'm starting X using "startx" or "xinit -- :0 -nolisten tcp vt$XDG_VTNR". Here's my ~/.xinitrc:
    if [ -d /etc/X11/xinit/xinitrc.d ]; then
    for f in /etc/X11/xinit/xinitrc.d/*; do
    [ -x "$f" ] && . "$f"
    done
    unset f
    fi
    exec startxfce4
    Last edited by The Infinity (2014-08-14 21:17:41)

    I figured out where the problem comes from. It is actually the screen (which has a resolution of 1920x1080 pixel). When using a screen (with a smaller resolution) I'm getting a decent resolution in the terminal and - more imporant -  X starts and seems to use the Nouveau driver (which I have installed again). When plugging my normal screen again (while X is running) I'm able to use this screen with the maximum resolution without problems (so only starting with this screen does not work). Now I have a reference point to fix the problem.
    The other thread might be interesting as well.
    Last edited by The Infinity (2014-08-17 02:36:57)

  • No wlan0 interface, fresh arch linux install.

    Hi there,
    I'v installed finally arch linux on my box, and everything is fine, i configured well my system (after reeding the arch wiki installation guide) and now i'm with a # (no gui off course).
    I was trying to set up the wireless interface but when I type ifconfig the wlan0 doesn't show up, only  eth0 and lo
    lspci | grep -i net
    --- ethernet --
    03:00.0 Network controller : Intel Corporation PRO/Wireless 3945ABG [Golan] Network Connection (rev 02)
    uname - a
    Linux hostname 2.6.30-ARCH #1 SMP ...... i686
    ifconfig wlan0 up
    SIOSCSIFFLAGS: No such file or directory
    (this is obvious)
    The strange "thing" is that when i type ifconfig  (on the live cd), I see the wlan0 interface and this doesn't show up on the hdd install.
    What should I do?
    PS: i have only wireless connection, now posting from an macBook
    Last edited by r0b0t (2010-01-09 17:25:08)

    Did you install the firmware package?  It loads automatically during install but only goes to the hard drive during install if you select it in the list of packages.  The package is called 'iwlwifi-3945-ucode'.
    Also, if you do a pacman -Syu, you'll update to the 2.6.32.2 kernel which has a bug.  With some access points, it will disassociate when the 3945 power management puts the card to sleep.  To work around this, as root type 'iwconfig wlan0 power off'.  this should be the default behavior in the next kernel update.  If you stay with 2.6.30 the 3945 should work fine.
    Last edited by brianhanna (2010-01-09 17:54:22)

Maybe you are looking for