CentOS 5 :: Kickstart Installation Hangs At GetNetConfig?
Apr 22, 2011
I am trying to do a kickstart installation of CentOS 5.5 x86 using a static IP on eth1. (The machine has 4 ethernet ports). I set ks, ksdevice, ip, netmask, and noipv6 when the boot disc asks for the init commands.I have similar networking information configured in the kickstart script. I have tried switching the order of the network config lines in the kickstart script to place eth1 at the top. Either way the installation hangs. On tty3 I can see that the command last called is getNetConfig. All of this configuration is using static IPs. I'm not sure why it thinks it has to do anything extra... Is there a problem trying to reuse the same eth1 NIC for the second stage? I have tried using the same and different IPs on the same subnet but nothing changes the outcome. Both the kickstart file and the install tree are on the same subnet, the same server in fact. Both are accessed via FTP but I had this problem with NFS as well.
I developed a bootable Kickstart USB Stick earlier that works flawlessly on a HP7700 computer with a 750 GB drive but when I use the exact same USB Stick on a HP8000 computer with 2 - 2 TB drives, the systems hangs after loading initrd.img
I have searched with my friend Google high and low have been unable to find a solution. I have tried adding acpi=off and maxcpus=1 to the Append line in the syslinux.cfg without success.
As a workaround I created a bootable DVD which I then pointed to the USB Stick to load in the Kickstart configuration but would really prefer to use a single USB stick to perform this task.
Below I have pasted the output of the lshw commands for the HP7700 and HP8000 thinking there must be something in the hardware causing the problem.
what needs to change in the kickstart config to allow this new hardware to work with my bootable Kickstart USB Stick.
LSHW for HP7700 (the pc that works):
stc0025988 description: Low Profile Desktop Computer product: HP Compaq dc7700 Small Form Factor vendor: Hewlett-Packard serial: CAC72905KD width: 32 bits
System gets corrupted when I abort a kickstart installationI have a simple kickstart installation of CentOS 5.5 (same issue on 5.3)The only user interaction is the partitioning screen.if, in my cfg file, I state :-
# Partition clearing information clearpart --all --initlabel or
I put a script into the post installation of kickstart and the log showed that the yum that is started in the script had to wait for yum to finish. I understand I can (and will) put this package in the packages section, but isn't it bad that yum is still finishing up when the post installation is initialized? Has anyone else noticed this?
Does anyone have a ready to deploy kickstart file for creating KVM images? I don't have access to a CentOS machine or a spare computer and would need to create images for a machine on the other side of the Atlantic.I would need something that would create an absolute minimal CentOS installation. I tried to search for something online, but couldn't find anything.
i am trying to install centos 5.5 x86_64 as a guest OS in vmware server 2.0.2 using netinstall iso. Installation runs fine until the point, when it tries to install selinux-policy-targeted-2.4.6-279.el5.noarch, the whole virtual pc hangs at this.any ideas? i tried to google few things about this, but i have found nothing. this has happened 3 times in row, whole virtual pc always hangs at the same package. i dont have any other problems with vmware, gentoo runs and installs fine in it.i would prefer to do installation using netinstall.iso, it would take a lot of time to download all cds or whole dvd and all i require is a very basic set of packages.
I need to set a new centos on an "old" box : PIV 1.8 1.5Gb
The Centos will run on a 40 Gb PATA Hard Drive set as main drive, The Box runs a another SATA Drive with 2 different windows installations.
The installer first hanged on the loading ata_piix driver screen so I restarted with the irqpoll option to complete the wizard. But I'm now stopped on the last X screen : starting installation process please wait.
I have changed the hard drive, the DVD drive, the Centos DVD, tried with a 5.3 and 5.4 with no success.
I'm attempting to install CentOS 5.3 on my IBM IntelliStation Z Pro 6221/37G, powered by some extra equipments as nVidia NV44 GeForce 6200 video card, MV88SX6041 SATA II Controller, 2x1GB DRAM.
The problem: CentOS 5.3 DVD i386 didn't install, the screen becomes unreadable because of vertical bars just after anaconda start running. Choosing at boot time different "linux resolutions" did not help, as well as the "text mode" still useless because Logical Volumes need to be created/handled and the graphical interface is mandatory.
Same result booting CentOS 5.2 Live CD. This sounds really strange to me. I checked iso images against their md5sum and are ok. Centos 5.3 x86_64 does not boot yet :"Your CPU does not support long mode", evem they are 2 x Xeon 2.8GHz / 512KB
I am attempting to upgrade a system from 4.7 to 5.2 using a (now) DVD drive attached to the onboard IDE. Originally I had tried using a remote NFS image and a USB stick but I thought maybe there was a problem with the image. I can get up to the point of the installation of selecting the keyboard for the system and then it freezes and never goes any further. It doesn't appear to be a kernel panic since I can still switch between consoles.
I've got an MSI K9NGM2-FID with 14 drives in it. It serves as a file server for our backup server. It's got a secondary 4 port Silicon Image SII 3114 SATA card using the sata_sil module, and an old IDE Promise FastTrak TX2000. Technically I could have 16 drives but the 750W PS is walking the fine line on tripping it's self-breaker with the 14 drives and 7 fans. I would like to NOT have to disconnect all of this to do the upgrade.
I thought maybe that running the install using the "noprobe" option would help so it didn't detect and load the modules for the Silicon Image or the Promise cards and detect all of the drives but it still gets stuck on the step after selecting the keyboard. The installation info console and the dmesg console don't really provide any useful information. The installation console says:
INFO : moving (1) to step welcome INFO : moving (1) to step language INFO : moving (1) to step keyboard INFO : moving (1) to step findrootparts
And the last lines of the dmesg console says:
<6>device-mapper: multipath: version 1.0.5 loaded <6>device-mapper: multipath round-robin: version 1.0.0 loaded <6>device-mapper: multipath emc: version 0.0.3 loaded
Is there a hidden "debug" option that will turn on a lot of extra logging?
Does anyone know of a way to tell the CentOS installer -not- to use LVM in a kickstart? We've been using a system that lets us define which particular drives to use during the installation as part of our deployment system. This does not work now that LVM is the 'default' in CentOS. I've looked over the options and I see how to FORCE particular LVM configurations, but I see no way to just turn it off.
I created a kickstart file and put it on a floppy. I have installed several times to refine the process and confused. I am not sure if the kickstart file is even being used by grub. I specified
autostep --autoscreenshot The install STILL asks me all those questions I was trying to avoid by doing autostep in the first place, and /root/anaconda-screenshots coes not exist.
I specified linux = hd:fd0:/ks.cfg
as a Grub command line option. I got that line off a forum, so it may be inaccurate. Anyone know if that is correctd? If I screw that up, shouldn't anaconda complain about file-not-found? Are the screenshots only applicable to graphics mode or also to text mode? (I have been using text mode, assuming text file screen dumps would appear in that subdirectory...)
I am trying to kickstart and want in post install to copy some files from a shared directory, to enable passwd less ssh and having same users across the clusters. But the cp does not work, nor does .ssh directory is getting created I have pasted my post install script below.
I'm trying to install CentOS 5.5 from harddrive using a kickstart file. Kickstart file is read correctly, it contains the following 3 lines (+ additional config):
I've downloaded both CentOS-5.5-i386-bin-DVD.iso and CentOS-5.5-i386-bin-1of7.iso, but anaconda (the installer) asks: - What partition and directory on that partition holds the CD (iso9660) images for CentOS? ...
VT3 gives these messages: INFO: partition /dev/sda11 selected INFO: mounting device sda11 for hard drive install INFO: mntloop loop7 on /tmp/loopimage as /tmp/hdimage/repos/CentOS/5.5/isos/i386/CentOS-5.5-i386-bin-1of7.iso fd is 12
I have been using the same kickstart more or less since release 5.2 but it fails with 5.5. It looks like it is good all the way to the final stages. Does anyone know what has changed in anaconda for this release? I think the first boot process has changed as well.
when i use kickstart to install centos from cdrom (i make it myself in my way),i got a %post script problem with the kickstart file. 1.%post script used to copy my own software from cdrom to hard disk.then make install automaitlly with bash script.
the %post script like : %post mkdir -p /myownsoftware cp -r /mnt/myownsoftware/* /myownsoftware cd /myownsoftware
I have downloaded the following kickstart file for installing minimal < 300 MB space centOS 5.2. I have created a Virtual Machine for Linux and attempting to install CentOS.here is my kickstart file:
I am using the "harddrive" option in a kickstart config to have it pick up isolinux files from a USB flash drive. I have been able to get it work by specifying the device name directly, but if I specify a LABEL or UUID, it does not work.Here is what my ks.cfg looks like.
Code: install text harddrive --partition=LABEL="/install" --dir=/ lang en_US.UTF-8
[Code]...
It almost seems like the version of Anaconda in Centos5.4 does not support specifying UUID or LABELs, but I have not been able to confirm that from the Release notes.Appreciate any tips/references/documentation.
we can't get the clients in our lab to do a kickstart install. we're doing the install by booting from the Centos 5.3 net install cd and anaconda starts, but terminates abnormally reporting a SIGSEGV fault. Interestingly, attempts at doing an install from a CD and without the network connection results in this error:
X11TransSocketINETConnect() can't get address for localhost:6001. Temporary failure in name resolution.
I'm trying to dynamically write command section stubs with a pre-script to be included via %include. The simplest of these contains the disk partitioning commands. Following the canonical examples,[URL].. for one, of this found in every source of documentation for RHEL/Fedora variants does not work. Anaconda attempts to prompt for interaction to get the partitioning scheme and a cmdline install stops with "In interactive mode parttype, can't continue". I've stripped things down to the following two cases:
I'm building kickstart files for my various machines.On my xen virtualization servers, I'm trying to get dom0_mem=512M added to the grub kernel line. Unfortunately, the bootloader --append option gets added to the linux kernel line, not the xen kernel line, which doesn't have the desired effect.Is there another way to put this into the kickstart file or should I use sed to put it into grub.conf in my post install section?
Can an unattended Kickstart support both IDE (hda) and SCSI (sda)? The goal is to to create a new virtual machine from scratch. What I have works for Parallels in which a new VM defaults to emulate an IDE hard disk. It does not work for VMware Workstation which defaults to emulate a SCSI disk.
The relevant Kickstart section: bootloader --location=mbr --driveorder=hda --append="rhgb quiet"
For portability reasons; I am building a standalone kickstart ISO; based of Cent5.2. I am to the point where I can load my ks file (linux ks=cdrom:/ks.cfg), it reads it fine; and performs the install as I want.
Where I am having a problem; is a good way to have the install use upgraded RPM's, not the base; specifically a kernel with a few needed tweaks in it; which is packaged in an rpm.
I attempted to place my kernel rpm's into the CentOS directory and rerun creatrepo; but I simply managed to corrupt the base repo on the install media.
I recently set up a kickstart server using Centos 5. I copied all 7 of the centos 5 install cd's to the tree. I made the install cd, it boots fine. I'm using http, when I'm prompted, I put in the web site: 1.1.1.1 and the Centos directory of /network-install/RPM
I get the following error:
Unable to retrieve [URL]
I've been told the double slashes after the ip address is not a problem, and I've tested that through a browser, by browsing the same location. (not sure if that's a valid test, but it did find the directory and display the files.)
we use for the installation of our machines bladelogic. We have different servers. Some servers have only one network interface, but it can be 2, 4 or may be more. There is always one network device for using PXE, but it is not always eth0.Is there any way to run kickstart without the entering of the PXE-Device so, that kickstart checks all the network devices in the system?
I am trying to create a workflow for upgrading various systems using kickstart. I was hoping folks can point me in the right direction.I have a system which already has Centos installed on it. However it is a stripped down version of Centos using a custom kickstart installer. Now I would like to upgrade these systems, using an updated kickstart file spec. I would like to be able to copy over required files into a partition on the system, make a change in the grub.conf, reboot the system and expect the system to use the kickstart file and the iso file located on a partition on the system to self upgrade the entire system.The partition which holds the iso on these systems may be raided, or it may be an LVM partition.
How do I specify in the kickstart file that the location of the iso is on an LVM partition? Is this even supported? I have tried specifiying the disk like so:
< -- isolinux/ks/harddrive.cfg -- > upgrade text harddrive --partition=mapper/VolGroup00-LogVol02 --dir=/isolinux
[code]...
But the installer did not seem to like it. Instead of the iso, can I put the entire tree instead? Would that work?
