From John.Dickinson at nominet.org.uk Thu Dec 1 15:28:33 2005 From: John.Dickinson at nominet.org.uk (John.Dickinson at nominet.org.uk) Date: Thu, 1 Dec 2005 15:28:33 +0000 Subject: non interactive install Message-ID: Hi, I am trying to do a non interactive install of a set of HP BL25p blades using kickstart, pxelinux and NFS. The blades have 4 network interfaces and every time I boot and run the installer I get a message telling me that I have multiple network devices and asking which network interface I wish to do the install through. Is there anyway to get kickstart to stop asking? I just want it to access the nfs share through eth0 using DHCP. When I press OK everything works fine. I have the following kickstart file Generated by Kickstart Configurator #platform=x86, AMD64, or Intel EM64T #System language lang en_GB #Language modules to install langsupport en_GB #System keyboard keyboard uk #System mouse mouse #Sytem timezone timezone Europe/London #Root password rootpw testing #Reboot after installation reboot #Use text mode install text #Install OS instead of upgrade install #Use NFS installation Media nfs --server=A.B.C.D --dir=/var/provisioning/linux/CentOS4.2 #System bootloader configuration bootloader --location=mbr #Clear the Master Boot Record zerombr yes #Partition clearing information clearpart --all --initlabel #Disk partitioning information part /boot --fstype ext3 --size 100 part / --fstype ext3 --size 10240 part swap --size 2048 part /var --fstype ext3 --size 1 --grow #System authorization infomation auth --useshadow --enablemd5 #Network information network --bootproto=dhcp --device=eth0 network --bootproto=dhcp --device=eth1 network --bootproto=dhcp --device=eth2 network --bootproto=dhcp --device=eth3 #Firewall configuration firewall --disabled #Do not configure XWindows skipx #Package install information %packages --resolvedeps Thanks John From clumens at redhat.com Thu Dec 1 15:31:57 2005 From: clumens at redhat.com (Chris Lumens) Date: Thu, 1 Dec 2005 10:31:57 -0500 Subject: non interactive install In-Reply-To: References: Message-ID: <20051201153157.GD18708@exeter.boston.redhat.com> > I am trying to do a non interactive install of a set of HP BL25p blades > using kickstart, pxelinux and NFS. The blades have 4 network interfaces > and every time I boot and run the installer I get a message telling me > that I have multiple network devices and asking which network interface I > wish to do the install through. Is there anyway to get kickstart to stop > asking? I just want it to access the nfs share through eth0 using DHCP. > When I press OK everything works fine. Pass ksdevice=eth0 as a parameter in your bootloader (or, add it to your netboot configs). - Chris From ext.helice.rivera at sncf.fr Thu Dec 1 15:34:45 2005 From: ext.helice.rivera at sncf.fr (EXT / HELICE RIVERA Cedric) Date: Thu, 1 Dec 2005 16:34:45 +0100 Subject: non interactive install Message-ID: Hello, If you have multiple network adapters you have to say which one you want to use in the "bootp/pxe" command line or just comment entries for those you don't need in your "kickstart" file : bootp> rhes3 ks=nfs:xx.xx.xx.xx:/where/you/store/it ksdevice=eth0 I hope this help. ++ C.RIVERA -----Message d'origine----- De : kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] Envoy? : jeudi 1 d?cembre 2005 16:29 ? : kickstart-list at redhat.com Objet : non interactive install Hi, I am trying to do a non interactive install of a set of HP BL25p blades using kickstart, pxelinux and NFS. The blades have 4 network interfaces and every time I boot and run the installer I get a message telling me that I have multiple network devices and asking which network interface I wish to do the install through. Is there anyway to get kickstart to stop asking? I just want it to access the nfs share through eth0 using DHCP. When I press OK everything works fine. I have the following kickstart file Generated by Kickstart Configurator #platform=x86, AMD64, or Intel EM64T #System language lang en_GB #Language modules to install langsupport en_GB #System keyboard keyboard uk #System mouse mouse #Sytem timezone timezone Europe/London #Root password rootpw testing #Reboot after installation reboot #Use text mode install text #Install OS instead of upgrade install #Use NFS installation Media nfs --server=A.B.C.D --dir=/var/provisioning/linux/CentOS4.2 #System bootloader configuration bootloader --location=mbr #Clear the Master Boot Record zerombr yes #Partition clearing information clearpart --all --initlabel #Disk partitioning information part /boot --fstype ext3 --size 100 part / --fstype ext3 --size 10240 part swap --size 2048 part /var --fstype ext3 --size 1 --grow #System authorization infomation auth --useshadow --enablemd5 #Network information network --bootproto=dhcp --device=eth0 network --bootproto=dhcp --device=eth1 network --bootproto=dhcp --device=eth2 network --bootproto=dhcp --device=eth3 #Firewall configuration firewall --disabled #Do not configure XWindows skipx #Package install information %packages --resolvedeps Thanks John _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From John.Dickinson at nominet.org.uk Thu Dec 1 15:40:16 2005 From: John.Dickinson at nominet.org.uk (John.Dickinson at nominet.org.uk) Date: Thu, 1 Dec 2005 15:40:16 +0000 Subject: non interactive install In-Reply-To: <20051201153157.GD18708@exeter.boston.redhat.com> Message-ID: kickstart-list-bounces at redhat.com wrote on 01/12/2005 15:31:57: > > Pass ksdevice=eth0 as a parameter in your bootloader (or, add it to your > netboot configs). > > - Chris > Thanks, that worked. I now feel v stupid - I have spent all day trying various kickstart files :) John From jkeating at j2solutions.net Thu Dec 1 15:46:30 2005 From: jkeating at j2solutions.net (Jesse Keating) Date: Thu, 01 Dec 2005 07:46:30 -0800 Subject: non interactive install In-Reply-To: <20051201153157.GD18708@exeter.boston.redhat.com> References: <20051201153157.GD18708@exeter.boston.redhat.com> Message-ID: <1133451990.22074.19.camel@yoda.loki.me> On Thu, 2005-12-01 at 10:31 -0500, Chris Lumens wrote: > Pass ksdevice=eth0 as a parameter in your bootloader (or, add it to your > netboot configs). There is also ksdevice=link which will use the only nic that has a live link. Sometimes this is a better option, sometimes not (: -- Jesse Keating RHCE (geek.j2solutions.net) Fedora Legacy Team (www.fedoralegacy.org) GPG Public Key (geek.j2solutions.net/jkeating.j2solutions.pub) Was I helpful? Let others know: http://svcs.affero.net/rm.php?r=jkeating From maddingue at free.fr Thu Dec 1 17:59:42 2005 From: maddingue at free.fr (=?iso-8859-1?b?U+liYXN0aWVu?= Aperghis-Tramoni) Date: Thu, 01 Dec 2005 18:59:42 +0100 Subject: Problems installing RHEL3 through PXE+NFS Message-ID: <1133459982.438f3a0e54fd1@imp2-g19.free.fr> Hello all, I am currently working on making a custom install of a RedHat Enterprise Linux 3.0, and am now encountering problem I really don't understand: when I try to install via PXE, it looks like the dependencies have all been scratched, the installer just nistall package in the alphabetical order, and therefore many packages just can't be installed because they miss they prereqs, including the kernel itself. Things are broken enought that the installer doesn't even notice that LILO fails to install the boot sector, while I can see on VT5: Reading boot sector from /dev/hda Fatal: open /dev/hda: No such file or directory I presume this last error is caused by the fact that many core packages are missing. The modifications I made to the installer are pretty small and consist mostly of adding four installation classes which all inherit from the Kickstart class, in order to read their configuration from kickstart files present in the installation tree. The sole method defined in these classes is _init_: def __init__(self, expert): KickstartBase.__init__(self, "/mnt/source/type4i.cfg", 0) self.installType = "install" self.interactive = 1 The really strange thing being that the ISO images created from the same installation tree installs fine. I've just finished a successful install inside Qemu while a PXE install on another machine just failed. Here are the parameters from /tftpboot/linux-install/pxelinux.cfg/default used to boot the PXE installation: label 2 kernel rhel3u5/vmlinuz append initrd=rhel3u5/initrd.img ramdisk_size=10000 method=nfs:server:/rhel3u5/i386 lang=fr_FR.UTF-8 keymap=fr-latin1 ksdevice=link ip=dhcp Has anybody seen something like this before? Thanks in advance -- S?bastien Aperghis-Tramoni Close the world, txEn eht nepO. From teng at dataway.com Thu Dec 1 19:21:36 2005 From: teng at dataway.com (Tedman Eng) Date: Thu, 1 Dec 2005 11:21:36 -0800 Subject: non interactive install Message-ID: <37ED92F9890FAF4BB947613C66FF8B1A08BB2C52@dw-mail.dataway.com> You have 4 "network" clauses. The "network" clause is not to configure the final system, but rather to configure the kickstart environment. If you want to configure additional interfaces for the final system, use the %post section for that. -----Original Message----- From: John.Dickinson at nominet.org.uk [mailto:John.Dickinson at nominet.org.uk] Sent: Thursday, December 01, 2005 7:29 AM To: kickstart-list at redhat.com Subject: non interactive install Hi, I am trying to do a non interactive install of a set of HP BL25p blades using kickstart, pxelinux and NFS. The blades have 4 network interfaces and every time I boot and run the installer I get a message telling me that I have multiple network devices and asking which network interface I wish to do the install through. Is there anyway to get kickstart to stop asking? I just want it to access the nfs share through eth0 using DHCP. When I press OK everything works fine. I have the following kickstart file Generated by Kickstart Configurator #platform=x86, AMD64, or Intel EM64T #System language lang en_GB #Language modules to install langsupport en_GB #System keyboard keyboard uk #System mouse mouse #Sytem timezone timezone Europe/London #Root password rootpw testing #Reboot after installation reboot #Use text mode install text #Install OS instead of upgrade install #Use NFS installation Media nfs --server=A.B.C.D --dir=/var/provisioning/linux/CentOS4.2 #System bootloader configuration bootloader --location=mbr #Clear the Master Boot Record zerombr yes #Partition clearing information clearpart --all --initlabel #Disk partitioning information part /boot --fstype ext3 --size 100 part / --fstype ext3 --size 10240 part swap --size 2048 part /var --fstype ext3 --size 1 --grow #System authorization infomation auth --useshadow --enablemd5 #Network information network --bootproto=dhcp --device=eth0 network --bootproto=dhcp --device=eth1 network --bootproto=dhcp --device=eth2 network --bootproto=dhcp --device=eth3 #Firewall configuration firewall --disabled #Do not configure XWindows skipx #Package install information %packages --resolvedeps Thanks John _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From phr at doc.ic.ac.uk Thu Dec 1 19:31:40 2005 From: phr at doc.ic.ac.uk (Philip Rowlands) Date: Thu, 1 Dec 2005 19:31:40 +0000 (GMT) Subject: Problems installing RHEL3 through PXE+NFS In-Reply-To: <1133459982.438f3a0e54fd1@imp2-g19.free.fr> References: <1133459982.438f3a0e54fd1@imp2-g19.free.fr> Message-ID: On Thu, 1 Dec 2005, S?bastien Aperghis-Tramoni wrote: > I am currently working on making a custom install of a RedHat > Enterprise Linux 3.0, and am now encountering problem I really don't > understand: when I try to install via PXE, it looks like the > dependencies have all been scratched, the installer just nistall > package in the alphabetical order, and therefore many packages just > can't be installed because they miss they prereqs, including the > kernel itself. This question was asked/answered two days ago: https://www.redhat.com/archives/kickstart-list/2005-November/msg00062.html Cheers, Phil From rh-kickstart at matthouse.com Thu Dec 1 21:02:23 2005 From: rh-kickstart at matthouse.com (Matt Sturtz) Date: Thu, 1 Dec 2005 14:02:23 -0700 (MST) Subject: comps RPM Message-ID: <1914.64.207.49.98.1133470943.squirrel@www.ontheside.net> On a somewhat related note (to my previous question), is there a simple procedure documented someplace for rebuilding the comps.rpm file with the new package information? Seems like this should be part of my package update process... Thanks again, -Matt- From skvidal at phy.duke.edu Thu Dec 1 21:04:43 2005 From: skvidal at phy.duke.edu (seth vidal) Date: Thu, 01 Dec 2005 16:04:43 -0500 Subject: comps RPM In-Reply-To: <1914.64.207.49.98.1133470943.squirrel@www.ontheside.net> References: <1914.64.207.49.98.1133470943.squirrel@www.ontheside.net> Message-ID: <1133471083.25713.22.camel@cutter> On Thu, 2005-12-01 at 14:02 -0700, Matt Sturtz wrote: > On a somewhat related note (to my previous question), is there a simple > procedure documented someplace for rebuilding the comps.rpm file with the > new package information? Seems like this should be part of my package > update process... > die, comps, die! -sv From klaus.steden at thomson.net Thu Dec 1 21:07:47 2005 From: klaus.steden at thomson.net (Klaus Steden) Date: Thu, 1 Dec 2005 16:07:47 -0500 Subject: comps RPM In-Reply-To: <1133471083.25713.22.camel@cutter> References: <1914.64.207.49.98.1133470943.squirrel@www.ontheside.net> <1133471083.25713.22.camel@cutter> Message-ID: <20051201210747.GT47794@thomson.net> > > On a somewhat related note (to my previous question), is there a simple > > procedure documented someplace for rebuilding the comps.rpm file with the > > new package information? Seems like this should be part of my package > > update process... > > > I've still to figure out what value this RPM actually offers. What is its history, and what is it supposed to do? Klaus From skvidal at phy.duke.edu Thu Dec 1 21:11:24 2005 From: skvidal at phy.duke.edu (seth vidal) Date: Thu, 01 Dec 2005 16:11:24 -0500 Subject: comps RPM In-Reply-To: <20051201210747.GT47794@thomson.net> References: <1914.64.207.49.98.1133470943.squirrel@www.ontheside.net> <1133471083.25713.22.camel@cutter> <20051201210747.GT47794@thomson.net> Message-ID: <1133471484.25713.27.camel@cutter> On Thu, 2005-12-01 at 16:07 -0500, Klaus Steden wrote: > > > On a somewhat related note (to my previous question), is there a simple > > > procedure documented someplace for rebuilding the comps.rpm file with the > > > new package information? Seems like this should be part of my package > > > update process... > > > > > > I've still to figure out what value this RPM actually offers. What is its > history, and what is it supposed to do? die, comps, die. -sv From tdiehl at rogueind.com Thu Dec 1 21:18:29 2005 From: tdiehl at rogueind.com (Tom Diehl) Date: Thu, 1 Dec 2005 16:18:29 -0500 (EST) Subject: comps RPM In-Reply-To: <1133471484.25713.27.camel@cutter> References: <1914.64.207.49.98.1133470943.squirrel@www.ontheside.net> <1133471083.25713.22.camel@cutter> <20051201210747.GT47794@thomson.net> <1133471484.25713.27.camel@cutter> Message-ID: On Thu, 1 Dec 2005, seth vidal wrote: > On Thu, 2005-12-01 at 16:07 -0500, Klaus Steden wrote: >>>> On a somewhat related note (to my previous question), is there a simple >>>> procedure documented someplace for rebuilding the comps.rpm file with the >>>> new package information? Seems like this should be part of my package >>>> update process... >>>> >>> >> I've still to figure out what value this RPM actually offers. What is its >> history, and what is it supposed to do? > > die, comps, die. Don't hold back now, tell us how you really feel about it!! Seriously though, given that comps is on its last breath, is there a good way to generate a comps like file to use for setting up a yum groups file. I currently use the comps file that comes with a release for a starting point for my yum groups file(s). How are others doing this kind of thing? Regards, Tom From jmclean at parasun.com Thu Dec 1 21:38:41 2005 From: jmclean at parasun.com (John McLean) Date: Thu, 01 Dec 2005 13:38:41 -0800 Subject: Problem with CD based install. Message-ID: <1133473121.13703.82.camel@jdm.softeng.parasun.net> I'm a kickstart newbie and I'm having a problem I haven't been able to find any reference to on the net. Hoping someone might point me in the right direction: 1.) Created cdrom based install cd. 2.) On boot up I get: On vert term 1, gui based "CD not found" "The blah blah CD was not found" "Please insert CD and press OK" "Ok" "Back" On vert term 3 "trying to mount CD device hda" every time you press "OK" The system has booted from my cd and the cd is still in the drive, it never gets ejected or anything. My understanding is that we've gone through the first phase of the install and it's trying to remount the cd to read the ks.cfg file and start loading rpms. Could anyone point me at the document to read to understand what I'm seeing. My ks.cfg is as attached. Thanks in advance. John. -------------- next part -------------- # CRMS 30 anaconda/kickstart configuration - mgmt.test1.cablerocket.net #System language lang en_US #Language modules to install langsupport --default=en_US # Forces a probe of the PCI bus and loads modules for all the devices found if a module is available deviceprobe #System keyboard keyboard us #System mouse mouse none #Sytem timezone # pacific - America/Vancouver or America/Los_Angeles # central - America/Chicago # mountain - America/Denver # eastern - America/New_York timezone PST8PDT #Root password rootpw --iscrypted $1$lhAuR2Z6$OzBr1LRe.SdmvjyJCYckn. #Reboot after installation #reboot #Use text mode install text #Install Red Hat Linux instead of upgrade install #Use CDROM installation media cdrom #url --url http://204.174.16.20/fedora-custom/i386/ #System bootloader configuration bootloader --location=mbr #Clear the Master Boot Record zerombr yes #Partition clearing information clearpart --all --initlabel #Disk partitioning information part /boot --size 100 --ondisk sda part swap --size 2048 --ondisk sda part / --size 1 --grow --ondisk sda part /mnt/data --size 1 --grow --ondisk sdb #System authorization infomation auth --useshadow --enablemd5 #Network information network --bootproto=static --ip=172.19.225.129 --netmask=255.255.0.0 --gateway=172.19.0.1 --nameserver=204.174.18.2 --device=eth0 #network --bootproto=dhcp --device=eth0 #Firewall configuration firewall --disabled #Do not configure XWindows skipx #Package install information %packages comps MAKEDEV SysVinit acl acpid alsa-lib anacron apmd ash aspell aspell-en at atk attr autoconf autofs automake automake14 automake15 automake16 automake17 basesystem bash bc beecrypt bind bind-chroot bind-libs bind-utils binutils bridge-utils byacc bzip2 bzip2-devel bzip2-libs caching-nameserver chkconfig coreutils cpio cpp cracklib cracklib-dicts crontabs ctags curl curl-devel cvs cyrus-sasl cyrus-sasl-devel cyrus-sasl-md5 cyrus-sasl-plain db4 db4-devel db4-utils dev device-mapper devlabel dialog diffstat diffutils dos2unix doxygen dump e2fsprogs e2fsprogs-devel ed elfutils elfutils-libelf ethtool expat expat-devel expect fbset fedora-logos fedora-release file filesystem findutils flex freetype ftp gawk gcc gcc-c++ gcc-g77 gcc-gnat gcc-java gcc34 gcc34-java gdb gdbm gdbm-devel glib glib2 glibc glibc-common glibc-devel glibc-headers glibc-kernheaders gmp gmp-devel gnupg gpm gpm-devel grep groff grub gzip hdparm hesiod hesiod-devel hotplug hwdata indent info initscripts iproute iptables iputils jwhois kbd kernel kernel-utils krb5-devel krb5-libs krbafs krbafs-devel kudzu kudzu-devel less lftp libIDL libacl libart_lgpl libattr libcap libcap-devel libf2c libgcc libgcj libgcj-devel libgcj34 libgcj34-devel libgnat libjpeg libmcrypt libmcrypt-devel libogg libogg-devel libpcap libpng libselinux libstdc++ libstdc++-devel libtermcap libtermcap-devel libtiff libtool libtool-libs libusb libusb-devel libuser libuser-devel libwvstreams libxml2 libxml2-devel libxslt lockdev lockdev-devel logrotate logwatch lrzsz lsof ltrace lvm2 lynx m4 mailcap mailx make man man-pages mdadm mingetty minicom mkinitrd mkisofs mktemp modutils mrtg mt-st mtools mtr nano nc ncurses ncurses-devel #net-snmp #net-snmp-perl #net-snmp-utils net-tools netdump nmap nscd nss_ldap ntp openldap openldap-devel openssh openssh-clients openssh-server openssl openssl-devel #oprofile pam pam-devel pam_krb5 pam_smb passwd patch patchutils pciutils pciutils-devel pcre perl perl-Filter perl-Bit-Vector perl-DBD-Pg perl-DBI perl-Date-Calc perl-DateManip perl-Digest-HMAC perl-Digest-SHA1 perl-Filter perl-Net-DNS perl-TermReadKey perl-URI pkgconfig popt portmap postgresql postgresql-devel postgresql-server postgresql-libs prelink procmail procps psacct psmisc python python-devel quota raidtools rcs rdate readline readline-devel redhat-rpm-config rmt rootfiles rpm rpm-build rpm-devel rrdtool rsh rsync schedutils screen sed sendmail setarch setserial setup shadow-utils shared-mime-info slocate strace stunnel sudo symlinks sysklogd syslinux tar tcl tcp_wrappers tcpdump tcsh telnet termcap texinfo tftp-server time tmpwatch traceroute tzdata unix2dos unzip usbutils usermode utempter util-linux vim-common vim-enhanced vim-minimal vixie-cron wget which words xinetd xorg-x11-libs-data zip zlib zlib-devel %pre %post echo "new installation on `date`" > /root/ks.log #disable unneeded services chkconfig apmd off chkconfig autofs off chkconfig cpuspeed off chkconfig gpm off chkconfig kudzu off chkconfig mdmpd off chkconfig netfs off chkconfig pcmcia off chkconfig portmap off chkconfig rhnsd off chkconfig smartd off #enable misc essential services chkconfig --level 2345 ntpd on chkconfig --level 2345 named on chkconfig --level 2345 sendmail on # bring eth0 up (we need network access to download our rpms) ifup eth0 1>& 2>> /root/ks.log ifconfig 1>& 2>> /root/ks.log route -n 1>& 2>> /root/ks.log # set /etc/sysconfig/network echo -e "NETWORKING=yes\nHOSTNAME=mgmt.test1.cablerocket.net\nGATEWAY=172.19.0.1\n" > /etc/sysconfig/network # the installrpms.sh script relies on the hostname being set properly on the machine hostname mgmt.test1.cablerocket.net sh /stuff_from_make/rpms/installrpms.sh service postgresql stop mv /var/lib/pgsql /mnt/data/ ln -s /mnt/data/pgsql /var/lib/pgsql service postgresql start if [ ! -d /usr/local/cable/data ]; then mkdir /usr/local/cable/data chown cable.cable /usr/local/cable/data fi mv /usr/local/cable/data/ /mnt/data/cable ln -s /mnt/data/cable/ /usr/local/cable/data" 1>& 2>> /root/ks.log fi /usr/bin/eject From rh-kickstart at matthouse.com Thu Dec 1 22:13:03 2005 From: rh-kickstart at matthouse.com (Matt Sturtz) Date: Thu, 1 Dec 2005 15:13:03 -0700 (MST) Subject: comps RPM In-Reply-To: <20051201210747.GT47794@thomson.net> References: <1914.64.207.49.98.1133470943.squirrel@www.ontheside.net><1133471083.25713.22.camel@cutter> <20051201210747.GT47794@thomson.net> Message-ID: <2019.64.207.49.98.1133475183.squirrel@www.ontheside.net> >> On a somewhat related note (to my previous question), is there a >> simple procedure documented someplace for rebuilding the comps.rpm >> file with the new package information? Seems like this should be >> part of my package update process... >> > I've still to figure out what value this RPM actually offers. What is its > history, and what is it supposed to do? I guess I may be speaking of the rpmdb package -- it's the database RPM uses when offering "suggestions" to resolve deps. Clearly it's not an important issue, but I thought it'd still be worth solving... -Matt- From klaus.steden at thomson.net Thu Dec 1 22:15:52 2005 From: klaus.steden at thomson.net (Klaus Steden) Date: Thu, 1 Dec 2005 17:15:52 -0500 Subject: comps RPM In-Reply-To: <1133471484.25713.27.camel@cutter> References: <1914.64.207.49.98.1133470943.squirrel@www.ontheside.net> <1133471083.25713.22.camel@cutter> <20051201210747.GT47794@thomson.net> <1133471484.25713.27.camel@cutter> Message-ID: <20051201221552.GV47794@thomson.net> I could have sworn seth vidal said this Thu, Dec 01, 2005 at 04:11:24PM -0500 ... > > die, comps, die. > -sv > > Tell us how you really feel, Seth. ;> From teng at dataway.com Thu Dec 1 23:38:14 2005 From: teng at dataway.com (Tedman Eng) Date: Thu, 1 Dec 2005 15:38:14 -0800 Subject: comps RPM Message-ID: <37ED92F9890FAF4BB947613C66FF8B1A08BB2C53@dw-mail.dataway.com> The comps.rpm is nothing more than the hdlist and comps.xml rolled into an RPM for installation into /usr/share/comps/ You can download the comps-.src.rpm and look at the spec file if you want to build a new RPM, though anaconda works fine without one. -----Original Message----- From: Matt Sturtz [mailto:rh-kickstart at matthouse.com] Sent: Thursday, December 01, 2005 1:02 PM To: kickstart-list at redhat.com Subject: comps RPM On a somewhat related note (to my previous question), is there a simple procedure documented someplace for rebuilding the comps.rpm file with the new package information? Seems like this should be part of my package update process... Thanks again, -Matt- _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From maddingue at free.fr Fri Dec 2 10:34:42 2005 From: maddingue at free.fr (=?iso-8859-1?b?U+liYXN0aWVu?= Aperghis-Tramoni) Date: Fri, 02 Dec 2005 11:34:42 +0100 Subject: Problems installing RHEL3 through PXE+NFS In-Reply-To: References: <1133459982.438f3a0e54fd1@imp2-g19.free.fr> Message-ID: <1133519682.439023424fc19@imp6-g19.free.fr> Philip Rowlands wrote: > On Thu, 1 Dec 2005, S?bastien Aperghis-Tramoni wrote: > > > I am currently working on making a custom install of a RedHat > > Enterprise Linux 3.0, and am now encountering problem I really don't > > understand: when I try to install via PXE, it looks like the > > dependencies have all been scratched, the installer just nistall > > package in the alphabetical order, and therefore many packages just > > can't be installed because they miss they prereqs, including the > > kernel itself. > > This question was asked/answered two days ago: > https://www.redhat.com/archives/kickstart-list/2005-November/msg00062.html Except that the --productpath option is not recognized by the genhdlist program from Anaconda 9.1.5.8, the version used on RHEL3u5. The commands I currently execute to recreate the installation tree are: /usr/lib/anaconda-runtime/genhdlist $RHDIST chmod 644 $RHDIST/RedHat/base/hdlist{,2} # several commands to rebuild comps.rpm /usr/lib/anaconda-runtime/pkgorder $RHDIST i386 core base acoss \ french-support > $RHDIST/pkgorder.txt cp $RHDIST/pkgorder.txt $RHDIST/pkgorder-bak.txt /usr/lib/anaconda-runtime/buildinstall \ --pkgorder $RHDIST/pkgorder.txt --comp dist-3 \ --version 3ES --product "RedHat Enterprise Linux 3" \ --release "[build A003 - $(date '+%Y-%m-%d %H:%M')]" \ $RHDIST cp /redhat/rhel3u5/pkgorder-bak.txt /redhat/rhel3u5/pkgorder.txt /usr/lib/anaconda-runtime/splittree.py \ --arch=i386 --distdir=$RHDIST \ --total-discs=3 --bin-discs=3 --src-discs=1 \ --pkgorderfile=$RHDIST/pkgorder.txt \ --release-string="[build A003 - $(date '+%Y-%m-%d %H:%M')]" \ --srcdir=$RHDIST/SRPMS rm -f $RHDIST-disc1/RedHat/base/hdlist* /usr/lib/anaconda-runtime/genhdlist --withnumbers \ --fileorder $RHDIST/pkgorder.txt $RHDIST-disc* and after these a few mkisofs to create the corresponding ISO images. Does that mean I need to execute genhdlist just after executing pkgorder? PS: There seems to be a quite long delay between the time I send a mail and the time it hits the list as I haven't received yet any of the two previous mails I sent. -- S?bastien Aperghis-Tramoni Close the world, txEn eht nepO. From cleve.cornelius at kroger.com Fri Dec 2 21:04:13 2005 From: cleve.cornelius at kroger.com (cleve.cornelius at kroger.com) Date: Fri, 2 Dec 2005 16:04:13 -0500 Subject: Cleve Cornelius/MIS/CORP/KrogerCo is out of the office. Message-ID: I will be out of the office starting Fri 12/02/2005 and will not return until Mon 12/05/2005. If this is an urgent matter please contact the NOC and they will page the on call engineer to assist you. From Curtis at GreenKey.net Mon Dec 5 15:26:49 2005 From: Curtis at GreenKey.net (Curtis Doty) Date: Mon, 05 Dec 2005 07:26:49 -0800 Subject: comps RPM In-Reply-To: <2019.64.207.49.98.1133475183.squirrel@www.ontheside.net> References: <1914.64.207.49.98.1133470943.squirrel@www.ontheside.net><1133471083.25713.22.camel@cutter> <20051201210747.GT47794@thomson.net> <2019.64.207.49.98.1133475183.squirrel@www.ontheside.net> Message-ID: <43945C39.7090504@GreenKey.net> Matt Sturtz wrote: >>>On a somewhat related note (to my previous question), is there a >>>simple procedure documented someplace for rebuilding the comps.rpm >>>file with the new package information? Seems like this should be >>>part of my package update process... >>> >>> >>> >>I've still to figure out what value this RPM actually offers. What is its >>history, and what is it supposed to do? >> >> > > >I guess I may be speaking of the rpmdb package -- it's the database RPM >uses when offering "suggestions" to resolve deps. Clearly it's not an >important issue, but I thought it'd still be worth solving... > > I used to keep various rpmdbs around. But then Seth showed me the light in the form of repoquery in the yum-utils package. At least from fc4... ../C -- These are not the droids you're looking for. - Obi-Wan Kenobi -------------- next part -------------- An HTML attachment was scrubbed... URL: From pmatilai at laiskiainen.org Mon Dec 5 16:07:15 2005 From: pmatilai at laiskiainen.org (Panu Matilainen) Date: Mon, 5 Dec 2005 08:07:15 -0800 (PST) Subject: comps RPM In-Reply-To: <43945C39.7090504@GreenKey.net> References: <1914.64.207.49.98.1133470943.squirrel@www.ontheside.net><1133471083.25713.22.camel@cutter> <20051201210747.GT47794@thomson.net> <2019.64.207.49.98.1133475183.squirrel@www.ontheside.net> <43945C39.7090504@GreenKey.net> Message-ID: On Mon, 5 Dec 2005, Curtis Doty wrote: > Matt Sturtz wrote: > >>>> On a somewhat related note (to my previous question), is there a >>>> simple procedure documented someplace for rebuilding the comps.rpm >>>> file with the new package information? Seems like this should be >>>> part of my package update process... >>>> >>>> >>> I've still to figure out what value this RPM actually offers. What is its >>> history, and what is it supposed to do? The only thing that uses comps.rpm is system-config-packages (previously redhat-config-packages), whether you consider that to be of any value is ... well, up to you :) Apt + Synaptic can use it as well through extensions for group operations but whether anybody cares anymore ... I guess not. And the apt+synaptic support doesn't actually need the rpm itself, it just needs a file in format of comps.xml file somewhere on local fs. - Panu - From ben_m_f at yahoo.co.uk Wed Dec 7 15:32:37 2005 From: ben_m_f at yahoo.co.uk (Ben Fitzgerald) Date: Wed, 7 Dec 2005 15:32:37 +0000 Subject: RHN XMLRPC API In-Reply-To: References: Message-ID: <20051207153237.GX11015@bfitzgerald.co.uk> On Mon, Nov 28, 2005 at 10:10:42AM +0100, Christian.Rohrmeier at SCHERING.DE wrote: > > Hi Kickstarters, > > my question relates tangentially to anaconda, so bear with me if you can. > Possibly nobody knows a thing about this, but maybe somebody knows somebody > else who does know. > > I am trying to use the XMLRPC based RHN API to automate the sorting and > downloading of specific typs of updates (there is no other way to get a > list of the security updates, for example, because up2date only gives you > all updates of a particular channel, with security, bugfixes, and feature > enhancements alle mixed together.) Described here: > http://www.redhat.com/docs/manuals/RHNetwork/ref-guide/4.0/ap-rhn-api-access.html > > This relates to my installs, because I would like to have Kickstart apply > all current security updates during a new installation. I realize its not a > Kickstart question, but you guys have such a wealth of knowledge, I'm > hoping somebody out there has experience interfcaing with RHN via the API. > > Many thanks in advance. > > Christian > > PS. I'm trying to use the Frontier XMLRPC module for Perl, as it is > described in the sample code: > http://www.redhat.com/docs/manuals/RHNetwork/ref-guide/4.0/s1-rhn-api-script.html > which I can't get to work. Hi Christian, I have the example script provided by redhat working against our satellite. However, I'm a little underwhelmed by the API. It only provides access to a small subset of the info available in the db. There isn't much info on this so I'm trying to get beyond the RHN pdfs. I've subscribed to rhn-users and rhn-satellite-users so that should be a start. If anyone knows of other detailed sites please do let me know. I'm willing to roll my sleeves up and hack the API to get new methods that do more, if required. Thanks, Ben. -- Registered Linux user number 339435 From James_Martin at ao.uscourts.gov Thu Dec 8 15:51:38 2005 From: James_Martin at ao.uscourts.gov (James_Martin at ao.uscourts.gov) Date: Thu, 8 Dec 2005 10:51:38 -0500 Subject: RHN XMLRPC API In-Reply-To: <20051207153237.GX11015@bfitzgerald.co.uk> Message-ID: I found it much easier to use python than perl against the RHN xmlrpc interface.. here's some simple code that queries your RHN account for systems subscribed, and also lets you delete machines.. I made it because I hate having to go through the GUI to remove a machine from RHN. here's an example use: rhn_tool.py -q -s ---- queries all your subscribed systems rhn_tool.py -s -d -h HOSTNAME --- will delete machine machine HOSTNAME from RHN I've used it successfully many times, but it's not production code, so USE AT YOUR OWN RISK. (In fact the RHN API is not even supported according to Red Hat's documents, so this goes without saying). #! /usr/bin/python import xmlrpclib from optparse import OptionParser def GetOptions(): parser=OptionParser() parser.add_option("-d", "--delete", action="store_true", dest="delete", default=False, help = "Deletes system, group, or channel") parser.add_option("-s", "--system", action="store_true", dest="system", default=False, help="Used when performing operations to machines subscribe to RHN.") parser.add_option("-q", "--query", action="store_true", dest="query", default=False, help="Used in conjuction with -s to show subscribed systems.") parser.add_option("-n", "--name",dest="hostname", help="hostname of machine to perform operation on.", metavar=" hostname") global options (options,args) = parser.parse_args() return options.delete, options.system, options.hostname def getSystemIds(): systems = server.system.list_user_systems(session) return systems def deleteSystem(sid): try: print "attempting to remove SID %s... with hostname of %s" % (sid,options.hostname) delete = server.system.delete_systems(session,sid) print "Deletion of %s successfull." % (options.hostname) except: print "Deletion of %s unsuccessfull." % (options.hostname) host = 'xmlrpc.rhn.redhat.com' username='YOURRHNUSERNAME' password = 'YOURRHNPASSWORD' protocol = 'https' url = "%s://%s/rpc/api" %(protocol,host) server = xmlrpclib.ServerProxy(url) session = server.auth.login(username,password) GetOptions() if options.system: systems = getSystemIds() if options.query: if len(systems) == 0: print "No systems are subscribed to RHN." else: print "These machines are subscribed to RHN\n\n" print "Name: \t\tcheckin: \t\t\tsid: " for vals in systems: print "%s\t\t%s\t\t%s" % (vals['name'],vals['last_checkin'],vals['id']) if options.delete: for vals in systems: if vals['name'] == options.hostname: deleteSystem(vals['id']) If you are looking into getting errata information, check out these API calls: https://rhn.redhat.com/rpc/api/errata/ You might be able to return a hash/dictionary of released errata and parse through the ones you want. James S. Martin, RHCE Contractor Administrative Office of the United States Courts Washington, DC (202) 502-2394 kickstart-list-bounces at redhat.com wrote on 12/07/2005 10:32:37 AM: > On Mon, Nov 28, 2005 at 10:10:42AM +0100, Christian. > Rohrmeier at SCHERING.DE wrote: > > > > Hi Kickstarters, > > > > my question relates tangentially to anaconda, so bear with me if you can. > > Possibly nobody knows a thing about this, but maybe somebody knows somebody > > else who does know. > > > > I am trying to use the XMLRPC based RHN API to automate the sorting and > > downloading of specific typs of updates (there is no other way to get a > > list of the security updates, for example, because up2date only gives you > > all updates of a particular channel, with security, bugfixes, and feature > > enhancements alle mixed together.) Described here: > > http://www.redhat.com/docs/manuals/RHNetwork/ref-guide/4.0/ap-rhn- > api-access.html > > > > This relates to my installs, because I would like to have Kickstart apply > > all current security updates during a new installation. I realize its not a > > Kickstart question, but you guys have such a wealth of knowledge, I'm > > hoping somebody out there has experience interfcaing with RHN via the API. > > > > Many thanks in advance. > > > > Christian > > > > PS. I'm trying to use the Frontier XMLRPC module for Perl, as it is > > described in the sample code: > > http://www.redhat.com/docs/manuals/RHNetwork/ref-guide/4.0/s1-rhn- > api-script.html > > which I can't get to work. > > Hi Christian, > > I have the example script provided by redhat working against our > satellite. > > However, I'm a little underwhelmed by the API. It only provides access > to a small subset of the info available in the db. > > There isn't much info on this so I'm trying to get beyond the RHN pdfs. > > I've subscribed to rhn-users and rhn-satellite-users so that should be a > start. > > If anyone knows of other detailed sites please do let me know. > > I'm willing to roll my sleeves up and hack the API to get new methods > that do more, if required. > > Thanks, > > Ben. > > -- > Registered Linux user number 339435 > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From rh-kickstart at matthouse.com Thu Dec 8 17:45:06 2005 From: rh-kickstart at matthouse.com (Matt Sturtz) Date: Thu, 8 Dec 2005 10:45:06 -0700 (MST) Subject: kickstart timezone -- UTC ?? Message-ID: <2068.64.207.49.98.1134063907.squirrel@www.ontheside.net> Hi all-- Our servers, like many others I would guess, run entirely in UTC... In my ks.cfg I have "timezone --utc UTC". That has worked for a while, but the other day I wanted to do an "interactive" install, to make sure the right partitions were created (or, rather, that the wrong disks were not touched). I got a traceback on the timezone screen. I tried GMT as well. Looking at the code, the timezone of UTC shouldn't work at all, but it has been on non-interactive... I thought of simply adding UTC to the list, but it's a read-only filesystem and would require modifying the installer (as opposed to doing it as part of a %pre). Since the way I'm doing it clearly isn't, what is the proper way to do this? Or is it just a bug? Cheers, -Matt- From brilong at cisco.com Thu Dec 8 19:23:14 2005 From: brilong at cisco.com (Brian Long) Date: Thu, 08 Dec 2005 14:23:14 -0500 Subject: sshd in %pre Message-ID: <1134069794.4372.34.camel@brilong-lnx> Has anyone wrapped this up for RHEL 3 or RHEL 4? http://www.cpqlinux.com/ks-pre-sshd.html I'd love to get something pre-packaged instead of spinning it from scratch. The VNC option does not work when you're installing over a serial redirect (like a Sun V20z) so having ssh up and running for debug purposes would rock. Thanks. /Brian/ -- Brian Long | | | IT Data Center Systems | .|||. .|||. Cisco Linux Developer | ..:|||||||:...:|||||||:.. Phone: (919) 392-7363 | C i s c o S y s t e m s From yhhoo at tpl.com.sg Thu Dec 8 20:01:00 2005 From: yhhoo at tpl.com.sg (yhhoo at tpl.com.sg) Date: Fri, 9 Dec 2005 04:01:00 +0800 Subject: Yee Huei Hoo is out of the office. Message-ID: I will be out of the office starting 12/08/2005 and will not return until 12/14/2005. For urgent matters please contact Kelvin Yang at 62139281 Vincent Tan at 62139144 From prasanna at aol.net Fri Dec 9 06:47:14 2005 From: prasanna at aol.net (Prasanna Ponnada) Date: Fri, 09 Dec 2005 12:17:14 +0530 Subject: Webserver - Redhat/base images not found In-Reply-To: References: Message-ID: <43992872.9010204@aol.net> We setup a web server that serves RedHat Advanced Servers 3.0 and 4.0 for i386 and x86_64. Basically, we copied the contents of CD's to their respective CDs. Eg: redhat advanced server 3 on x86_64 to /redhat/AS3.0/x86_64/ As part of install process, we create kickstart on the fly and the client downloads the required files like vmlinuz, initrd product.img, updates.img and netstg2.img etc from the webserver. Of late, we have been running into a weird problem wherein the path to the above, at time refers to disk1 and at other times refers to the right path in the webserver. Also, although the file exists in a given directory, the server returns a HTTP 404 file not found error. I checked directory permissions on the directory, and they appear to be ok. Please find an extract from our webserver log below: 2005/12/09 06:23:13 10.149.17.213 200 26261 /redhat/AS3.0/x86_64/ks.cfg HTTP/1.1 2005/12/09 06:23:14 10.149.17.213 200 158 /redhat/AS3.0/x86_64/isolinux/vmlinuz HTTP/1.1 2005/12/09 06:23:20 10.149.17.213 200 158 /redhat/AS3.0/x86_64/isolinux/initrd.img HTTP/1.1 2005/12/09 06:25:16 10.149.17.213 404 78 /redhat/AS3.0/x86_64/RedHat/base/updates.img HTTP/1.1 2005/12/09 06:25:17 10.149.17.213 404 78 /redhat/AS3.0/x86_64/disc1/RedHat/base/updates.img HTTP/1.1 2005/12/09 06:25:18 10.149.17.213 200 158 /redhat/AS3.0/x86_64/RedHat/base/product.img HTTP/1.1 2005/12/09 06:25:27 10.149.17.213 404 78 /redhat/AS3.0/x86_64/disc1/RedHat/base/product.img HTTP/1.1 2005/12/09 06:25:28 10.149.17.213 200 159 /redhat/AS3.0/x86_64/RedHat/base/netstg2.img HTTP/1.1 2005/12/09 06:26:36 10.149.17.213 404 78 /redhat/AS3.0/x86_64/disc1/RedHat/base/netstg2.img HTTP/1.1 The other interesting thing is that although I see a HTTP 404 for files like updates.img, the installer does not complain about it and proceeds. However, when it fails to find netstg2.img, it pops up a window on the screen complaining that netstg2.img cannot be found in a given directory although the file exists there. Do I need to run genhdlist or pkgorder once I copy the contents of the CD into the webserver's directory? I thought genhdlist talks only about rpms and not about updates or netstg2. Did anyone run into issues like the above? Any pointers would be greatly appreciated. thanks Prasanna -------------- next part -------------- An HTML attachment was scrubbed... URL: From Christian.Rohrmeier at SCHERING.DE Fri Dec 9 09:31:03 2005 From: Christian.Rohrmeier at SCHERING.DE (Christian.Rohrmeier at SCHERING.DE) Date: Fri, 9 Dec 2005 10:31:03 +0100 Subject: RHN XMLRPC API In-Reply-To: Message-ID: Hi Martin, I finally got XMLRPC working with the Frontier Perl module. The example that is provided by Red Hat has a few minor problems (specifically, you have to use HTTPS and not HTTP... that was the trick). I'm a passable with Perl, but know nothing about Python. Although, Python doesn't look that hard. =) The XMLRPC API is okay. It is decent enough... initially its good enough to let me get a listing of all errata of type "security" for example. Which is what I urgently need. Maybe you can clue me in on the next problem that I have: having identified the specific errata I wish to retrieve, how to I get up2date to download packages/errata from channels that do not match the specific system that I am running up2date from? Lets say my retrieval system is an RHEL 4 WS X86, but I need to download packages/errata for RHEL 3 AS X86_64. Using up2date --channel doesn't seem to work. At the Linux World Expo in Frankfurt a few weeks ago I talked with a rep from RedHat that claimed that it should be possible to download packages for any RHEL release regardless of what type of system the downloading system has. My downloading system has an entitlement for RHEL AS, which means I can see channels for RHEL 2.1, 3, and 4, WS, ES, AS, X86 and X86_64 when I log on to RHN on the web. However, getting up2date to grab anything is difficult. Is up2date the only way to download packages from RHN? Thanks, Christian _________________ Christian Rohrmeier Schering AG Corporate IT - Infrastructure and Services Computer Systems and Operations System Administration - Research and Development Tel +49 30 468 15794 Fax +49 30 468 95794 James_Martin at ao.u scourts.gov Sent by: To kickstart-list-bo Discussion list about Kickstart unces at redhat.com cc 08.12.2005 16:51 Subject Re: RHN XMLRPC API Please respond to Discussion list about Kickstart I found it much easier to use python than perl against the RHN xmlrpc interface.. here's some simple code that queries your RHN account for systems subscribed, and also lets you delete machines.. I made it because I hate having to go through the GUI to remove a machine from RHN. here's an example use: rhn_tool.py -q -s ---- queries all your subscribed systems rhn_tool.py -s -d -h HOSTNAME --- will delete machine machine HOSTNAME from RHN I've used it successfully many times, but it's not production code, so USE AT YOUR OWN RISK. (In fact the RHN API is not even supported according to Red Hat's documents, so this goes without saying). #! /usr/bin/python import xmlrpclib from optparse import OptionParser def GetOptions(): parser=OptionParser() parser.add_option("-d", "--delete", action="store_true", dest="delete", default=False, help = "Deletes system, group, or channel") parser.add_option("-s", "--system", action="store_true", dest="system", default=False, help="Used when performing operations to machines subscribe to RHN.") parser.add_option("-q", "--query", action="store_true", dest="query", default=False, help="Used in conjuction with -s to show subscribed systems.") parser.add_option("-n", "--name",dest="hostname", help="hostname of machine to perform operation on.", metavar=" hostname") global options (options,args) = parser.parse_args() return options.delete, options.system, options.hostname def getSystemIds(): systems = server.system.list_user_systems(session) return systems def deleteSystem(sid): try: print "attempting to remove SID %s... with hostname of %s" % (sid,options.hostname) delete = server.system.delete_systems(session,sid) print "Deletion of %s successfull." % (options.hostname) except: print "Deletion of %s unsuccessfull." % (options.hostname) host = 'xmlrpc.rhn.redhat.com' username='YOURRHNUSERNAME' password = 'YOURRHNPASSWORD' protocol = 'https' url = "%s://%s/rpc/api" %(protocol,host) server = xmlrpclib.ServerProxy(url) session = server.auth.login(username,password) GetOptions() if options.system: systems = getSystemIds() if options.query: if len(systems) == 0: print "No systems are subscribed to RHN." else: print "These machines are subscribed to RHN\n\n" print "Name: \t\tcheckin: \t\t\tsid: " for vals in systems: print "%s\t\t%s\t\t%s" % (vals['name'],vals['last_checkin'],vals['id']) if options.delete: for vals in systems: if vals['name'] == options.hostname: deleteSystem(vals['id']) If you are looking into getting errata information, check out these API calls: https://rhn.redhat.com/rpc/api/errata/ You might be able to return a hash/dictionary of released errata and parse through the ones you want. James S. Martin, RHCE Contractor Administrative Office of the United States Courts Washington, DC (202) 502-2394 kickstart-list-bounces at redhat.com wrote on 12/07/2005 10:32:37 AM: > On Mon, Nov 28, 2005 at 10:10:42AM +0100, Christian. > Rohrmeier at SCHERING.DE wrote: > > > > Hi Kickstarters, > > > > my question relates tangentially to anaconda, so bear with me if you can. > > Possibly nobody knows a thing about this, but maybe somebody knows somebody > > else who does know. > > > > I am trying to use the XMLRPC based RHN API to automate the sorting and > > downloading of specific typs of updates (there is no other way to get a > > list of the security updates, for example, because up2date only gives you > > all updates of a particular channel, with security, bugfixes, and feature > > enhancements alle mixed together.) Described here: > > http://www.redhat.com/docs/manuals/RHNetwork/ref-guide/4.0/ap-rhn- > api-access.html > > > > This relates to my installs, because I would like to have Kickstart apply > > all current security updates during a new installation. I realize its not a > > Kickstart question, but you guys have such a wealth of knowledge, I'm > > hoping somebody out there has experience interfcaing with RHN via the API. > > > > Many thanks in advance. > > > > Christian > > > > PS. I'm trying to use the Frontier XMLRPC module for Perl, as it is > > described in the sample code: > > http://www.redhat.com/docs/manuals/RHNetwork/ref-guide/4.0/s1-rhn- > api-script.html > > which I can't get to work. > > Hi Christian, > > I have the example script provided by redhat working against our > satellite. > > However, I'm a little underwhelmed by the API. It only provides access > to a small subset of the info available in the db. > > There isn't much info on this so I'm trying to get beyond the RHN pdfs. > > I've subscribed to rhn-users and rhn-satellite-users so that should be a > start. > > If anyone knows of other detailed sites please do let me know. > > I'm willing to roll my sleeves up and hack the API to get new methods > that do more, if required. > > Thanks, > > Ben. > > -- > Registered Linux user number 339435 > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From brilong at cisco.com Fri Dec 9 15:18:42 2005 From: brilong at cisco.com (Brian Long) Date: Fri, 09 Dec 2005 10:18:42 -0500 Subject: Ignoring SAN devices during kickstart? In-Reply-To: <20051025175725.GA7985@linuxace.com> References: <1130262766.4050.60.camel@brilong-lnx> <20051025175725.GA7985@linuxace.com> Message-ID: <1134141522.4558.14.camel@brilong-lnx> On Tue, 2005-10-25 at 10:57 -0700, Phil Oester wrote: > On Tue, Oct 25, 2005 at 01:52:46PM -0400, Brian Long wrote: > > latefcload was only implemented in U6 to my knowledge. Also, this just > > means qla2300 will be loaded after mptbase or whatever local SCSI disks > > you have. This is fine when you want to make sure /dev/sda is the local > > SCSI disk; however with HP Proliants, the RAID controller disks show up > > as /dev/cciss/c0d0, c0d0, etc. /dev/sda is the first SAN LUN and I want > > Anaconda to just completely disregard it. > > > > I wonder if Alexander's noprobe option along with "device scsi cciss" or > > something in ks.cfg might work. > > You'd be better off with 'nostorage' on the command line, so that anaconda > will still probe other devices such as network. Reviving an old thread, this works like a charm. I passed "nostorage" to the kernel append args in my pxe config. I added "device scsi cciss" to ks.cfg. Anaconda did not stop and prompt me to reinitialize sda, etc. /Brian/ -- Brian Long | | | IT Data Center Systems | .|||. .|||. Cisco Linux Developer | ..:|||||||:...:|||||||:.. Phone: (919) 392-7363 | C i s c o S y s t e m s From brilong at cisco.com Fri Dec 9 15:20:41 2005 From: brilong at cisco.com (Brian Long) Date: Fri, 09 Dec 2005 10:20:41 -0500 Subject: RHN XMLRPC API In-Reply-To: References: Message-ID: <1134141641.4558.17.camel@brilong-lnx> > Maybe you can clue me in on the next problem that I have: having identified > the specific errata I wish to retrieve, how to I get up2date to download > packages/errata from channels that do not match the specific system that I > am running up2date from? Lets say my retrieval system is an RHEL 4 WS X86, > but I need to download packages/errata for RHEL 3 AS X86_64. Using up2date > --channel doesn't seem to work. Check out this bugzilla: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=162210 I have submitted patches to up2date to do what you require. Red Hat does not plan to accept these patches although many folks have downloaded my changes and successfully implemented them. /Brian/ -- Brian Long | | | IT Data Center Systems | .|||. .|||. Cisco Linux Developer | ..:|||||||:...:|||||||:.. Phone: (919) 392-7363 | C i s c o S y s t e m s From brilong at cisco.com Fri Dec 9 15:25:50 2005 From: brilong at cisco.com (Brian Long) Date: Fri, 09 Dec 2005 10:25:50 -0500 Subject: Webserver - Redhat/base images not found In-Reply-To: <43992872.9010204@aol.net> References: <43992872.9010204@aol.net> Message-ID: <1134141950.4558.23.camel@brilong-lnx> On Fri, 2005-12-09 at 12:17 +0530, Prasanna Ponnada wrote: > > Please find an extract from our webserver log below: > > 2005/12/09 06:23:13 10.149.17.213 200 > 26261 /redhat/AS3.0/x86_64/ks.cfg HTTP/1.1 > 2005/12/09 06:23:14 10.149.17.213 200 > 158 /redhat/AS3.0/x86_64/isolinux/vmlinuz HTTP/1.1 > 2005/12/09 06:23:20 10.149.17.213 200 > 158 /redhat/AS3.0/x86_64/isolinux/initrd.img HTTP/1.1 > 2005/12/09 06:25:16 10.149.17.213 404 > 78 /redhat/AS3.0/x86_64/RedHat/base/updates.img HTTP/1.1 > 2005/12/09 06:25:17 10.149.17.213 404 > 78 /redhat/AS3.0/x86_64/disc1/RedHat/base/updates.img HTTP/1.1 > 2005/12/09 06:25:18 10.149.17.213 200 > 158 /redhat/AS3.0/x86_64/RedHat/base/product.img HTTP/1.1 > 2005/12/09 06:25:27 10.149.17.213 404 > 78 /redhat/AS3.0/x86_64/disc1/RedHat/base/product.img HTTP/1.1 > 2005/12/09 06:25:28 10.149.17.213 200 This is just Anaconda searching for product.img which might contain updates. It's normal for it to not exist :) > 159 /redhat/AS3.0/x86_64/RedHat/base/netstg2.img HTTP/1.1 > 2005/12/09 06:26:36 10.149.17.213 404 > 78 /redhat/AS3.0/x86_64/disc1/RedHat/base/netstg2.img HTTP/1.1 > > > > The other interesting thing is that although I see a HTTP 404 for > files like updates.img, the installer does not complain about it and > proceeds. > However, when it fails to find netstg2.img, it pops up a window on the > screen complaining that netstg2.img cannot be found in a given > directory although the file exists there. netstg2.img is required by Anaconda, so that's why it pops up a window asking about it. > Do I need to run genhdlist or pkgorder once I copy the contents of the > CD into the webserver's directory? Nope. Can you "wget" the same path from an existing machine? > I thought genhdlist talks only about rpms and not about updates or > netstg2. > Did anyone run into issues like the above? Any pointers would be > greatly appreciated. I have not seen this in my experience with HTTP installs. /Brian/ -- Brian Long | | | IT Data Center Systems | .|||. .|||. Cisco Linux Developer | ..:|||||||:...:|||||||:.. Phone: (919) 392-7363 | C i s c o S y s t e m s From prasanna at aol.net Fri Dec 9 15:44:20 2005 From: prasanna at aol.net (Prasanna Ponnada) Date: Fri, 09 Dec 2005 21:14:20 +0530 Subject: Webserver - Redhat/base images not found In-Reply-To: <1134141950.4558.23.camel@brilong-lnx> References: <43992872.9010204@aol.net> <1134141950.4558.23.camel@brilong-lnx> Message-ID: <4399A654.5050802@aol.net> Looks like I am running into something similar to this. https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=138741 I am running U6 of Redhat Advanced Server 3.0 for x86_64. Has anyone experienced this before on U6? Any pointers/help will be greatly appreciated. thanks Prasanna Brian Long wrote: >On Fri, 2005-12-09 at 12:17 +0530, Prasanna Ponnada wrote: > > > >>Please find an extract from our webserver log below: >> >>2005/12/09 06:23:13 10.149.17.213 200 >>26261 /redhat/AS3.0/x86_64/ks.cfg HTTP/1.1 >>2005/12/09 06:23:14 10.149.17.213 200 >>158 /redhat/AS3.0/x86_64/isolinux/vmlinuz HTTP/1.1 >>2005/12/09 06:23:20 10.149.17.213 200 >>158 /redhat/AS3.0/x86_64/isolinux/initrd.img HTTP/1.1 >>2005/12/09 06:25:16 10.149.17.213 404 >>78 /redhat/AS3.0/x86_64/RedHat/base/updates.img HTTP/1.1 >>2005/12/09 06:25:17 10.149.17.213 404 >>78 /redhat/AS3.0/x86_64/disc1/RedHat/base/updates.img HTTP/1.1 >>2005/12/09 06:25:18 10.149.17.213 200 >>158 /redhat/AS3.0/x86_64/RedHat/base/product.img HTTP/1.1 >>2005/12/09 06:25:27 10.149.17.213 404 >>78 /redhat/AS3.0/x86_64/disc1/RedHat/base/product.img HTTP/1.1 >>2005/12/09 06:25:28 10.149.17.213 200 >> >> > >This is just Anaconda searching for product.img which might contain >updates. It's normal for it to not exist :) > > > >> 159 /redhat/AS3.0/x86_64/RedHat/base/netstg2.img HTTP/1.1 >>2005/12/09 06:26:36 10.149.17.213 404 >>78 /redhat/AS3.0/x86_64/disc1/RedHat/base/netstg2.img HTTP/1.1 >> >> >> >>The other interesting thing is that although I see a HTTP 404 for >>files like updates.img, the installer does not complain about it and >>proceeds. >>However, when it fails to find netstg2.img, it pops up a window on the >>screen complaining that netstg2.img cannot be found in a given >>directory although the file exists there. >> >> > >netstg2.img is required by Anaconda, so that's why it pops up a window >asking about it. > > > >>Do I need to run genhdlist or pkgorder once I copy the contents of the >>CD into the webserver's directory? >> >> > >Nope. Can you "wget" the same path from an existing machine? > > > >>I thought genhdlist talks only about rpms and not about updates or >>netstg2. >>Did anyone run into issues like the above? Any pointers would be >>greatly appreciated. >> >> > >I have not seen this in my experience with HTTP installs. > >/Brian/ > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From James_Martin at ao.uscourts.gov Fri Dec 9 17:29:02 2005 From: James_Martin at ao.uscourts.gov (James_Martin at ao.uscourts.gov) Date: Fri, 9 Dec 2005 12:29:02 -0500 Subject: RHN XMLRPC API In-Reply-To: <1134141641.4558.17.camel@brilong-lnx> Message-ID: kickstart-list-bounces at redhat.com wrote on 12/09/2005 10:20:41 AM: > > > Maybe you can clue me in on the next problem that I have: having identified > > the specific errata I wish to retrieve, how to I get up2date to download > > packages/errata from channels that do not match the specific system that I > > am running up2date from? Lets say my retrieval system is an RHEL 4 WS X86, > > but I need to download packages/errata for RHEL 3 AS X86_64. Using up2date > > --channel doesn't seem to work. > > Check out this bugzilla: > https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=162210 > > I have submitted patches to up2date to do what you require. Red Hat > does not plan to accept these patches although many folks have > downloaded my changes and successfully implemented them. > I'm not suprised by this.. It's kind of a loop hole around their licensing.. RHN won't allow you to download packages you don't have entitlements to. If you _DO_ have an entitlement for AS,ES,WS, etc, and would like to use one machine to download packages from each channel, you could simply change the personality of the machine, unsubcribe it from RHN, and re-subscribe. James S. Martin, RHCE Contractor Administrative Office of the United States Courts Washington, DC (202) 502-2394 From smah at vmware.com Mon Dec 12 20:33:20 2005 From: smah at vmware.com (Stephen Mah) Date: Mon, 12 Dec 2005 12:33:20 -0800 Subject: partition syntax Message-ID: <439DDE90.4000903@vmware.com> what's the partitioning syntax to use an existing partition, say a swap partition from redhat 7.2 and reformat it for rhel3? I essentially want to use the same swap partition for both 7.2 and el3. I tried part swap --onpart /dev/sda2 I'm getting unable to open /dev/sda2 for the use of "none" -steve From gkapitany at rogers.com Mon Dec 12 21:12:39 2005 From: gkapitany at rogers.com (GABRIEL KAPITANY) Date: Mon, 12 Dec 2005 16:12:39 -0500 (EST) Subject: pxe install Message-ID: <20051212211240.5574.qmail@web88104.mail.re2.yahoo.com> Hi guys, I'd like to modify an RHEL3 install disk to add a SCSI driver for the pxe install. I'm using a redhat 7.3 documentation and I'm not pretty sure which initrd.img I should modify, the one in images/pxeboot/initrd.img or images/bootdisk.img or both. A diff shows that there are different. Can you point me to some documentation? Thanks, Gabriel -------------- next part -------------- An HTML attachment was scrubbed... URL: From LLavender at kmg.com Mon Dec 12 21:29:10 2005 From: LLavender at kmg.com (Lavender, Leo) Date: Mon, 12 Dec 2005 15:29:10 -0600 Subject: pxe install Message-ID: <044BBA9ACDF3D948BC41F0DD8DFADF314F2524@houexg3.kmg.com> Old procedure http://www.puschitz.com/Kickstart.shtml , but it has a lot. Leo -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of GABRIEL KAPITANY Sent: Monday, December 12, 2005 3:13 PM To: kickstart-list at redhat.com Subject: pxe install Hi guys, I'd like to modify an RHEL3 install disk to add a SCSI driver for the pxe install. I'm using a redhat 7.3 documentation and I'm not pretty sure which initrd.img I should modify, the one in images/pxeboot/initrd.img or images/bootdisk.img or both. A diff shows that there are different. Can you point me to some documentation? Thanks, Gabriel -------------- next part -------------- An HTML attachment was scrubbed... URL: From teng at dataway.com Mon Dec 12 23:24:35 2005 From: teng at dataway.com (Tedman Eng) Date: Mon, 12 Dec 2005 15:24:35 -0800 Subject: partition syntax Message-ID: <37ED92F9890FAF4BB947613C66FF8B1A08BB2C73@dw-mail.dataway.com> Try adding the --noformat in addition to the --onpart > -----Original Message----- > From: Stephen Mah [mailto:smah at vmware.com] > Sent: Monday, December 12, 2005 12:33 PM > To: Discussion list about Kickstart > Subject: partition syntax > > > what's the partitioning syntax to use an existing partition, > say a swap > partition from redhat 7.2 and reformat it for rhel3? > > I essentially want to use the same swap partition for both > 7.2 and el3. > > I tried part swap --onpart /dev/sda2 > > I'm getting unable to open /dev/sda2 for the use of "none" > > -steve > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From alexandre.tournier at jrc.it Tue Dec 13 08:39:59 2005 From: alexandre.tournier at jrc.it (Alex TOURNIER) Date: Tue, 13 Dec 2005 09:39:59 +0100 Subject: RHEL4: cannot kickstart clients with Broadcom DCM5751 Gigabit Message-ID: <439E88DF.801@jrc.it> I have a serious problem with a RHEL4 KS installation. My KS server works well because I can perform installation with no problems on various hardware plaforms. However the latest hardware platform I was delivered (ie Siemens Celsius R630) the installation fails. I am almost sure that the problem is with the NIC (Broadcom NetXtreme DCM 5751 Gigabit ), it looks like the NIC is somehow turned off (even if enabled in the BIOS) , or not working at the right speed and thus the DHCP server is not able to assign the IP. On the DHCP server nothing appears in the log file (as if there was no communication at all). I boot the client from a CD (made with the images available on the 1st CD of RHEL 4 distrib)....I also tried to make another CD including the driver from BROADCOM (bcm5700.ko instead of tg3.ho), but no success. If I disable the Broadcom NIC, and insert a second NIC the KS install works fine using this second NIC!!!! I would be very happy of any advice that could help me in solving the problem, I have many of these PC to be installed, and I need disperatly a KS solution... Thank you in advance. Regards, Alex -- __________________________________________________________________________________________________________ Alexandre TOURNIER System Administrator Joint Research Centre Institute for Environment and Sustainability Global Vegetation Monitoring TP 440 Via Fermi,1 I-21020 ISPRA (VA) ITALY tel: +x-332-785694 fax: +x-332-789073 e-mail: alexandre.tournier at jrc.it web: http://www-gvm.jrc.it/it_support/ FTP: ftp://ftp-gvm.jrc.it "The views expressed are purely those of the writer and may not in any circumstances be regarded as stating an official position of the European Commission." (Disclaimer required under the terms and conditions of use of the internet and electronic mail from Commission equipment) __________________________________________________________________________________________________________ From brilong at cisco.com Tue Dec 13 12:52:25 2005 From: brilong at cisco.com (Brian Long) Date: Tue, 13 Dec 2005 07:52:25 -0500 Subject: RHEL4: cannot kickstart clients with Broadcom DCM5751 Gigabit In-Reply-To: <439E88DF.801@jrc.it> References: <439E88DF.801@jrc.it> Message-ID: <1134478345.4378.29.camel@brilong-lnx> On Tue, 2005-12-13 at 09:39 +0100, Alex TOURNIER wrote: > I have a serious problem with a RHEL4 KS installation. > My KS server works well because I can perform installation with no > problems on various hardware plaforms. > However the latest hardware platform I was delivered (ie Siemens Celsius > R630) the installation fails. > I am almost sure that the problem is with the NIC (Broadcom NetXtreme > DCM 5751 Gigabit ), it looks like the NIC is somehow turned off (even > if enabled in the BIOS) , or not working at the right speed and thus > the DHCP server is not able to assign the IP. On the DHCP server nothing > appears in the log file (as if there was no communication at all). I > boot the client from a CD (made with the images available on the 1st > CD of RHEL 4 distrib)....I also tried to make another CD including the > driver from BROADCOM (bcm5700.ko instead of tg3.ho), but no success. If > I disable the Broadcom NIC, and insert a second NIC the KS install works > fine using this second NIC!!!! > > I would be very happy of any advice that could help me in solving the > problem, I have many of these PC to be installed, and I need disperatly > a KS solution... > Thank you in advance. Alex, We have the HP XW4300 workstation which has BCM5752. I don't know if your issue is the same as ours, but only RHEL 3 U6 and RHEL 4 U2 work. You don't mention which update of RHEL 4 you are kickstarting. Is it U2? If not, please try U2 as it contains an updated tg3 driver for the newest Broadcom chipsets. /Brian/ -- Brian Long | | | IT Data Center Systems | .|||. .|||. Cisco Linux Developer | ..:|||||||:...:|||||||:.. Phone: (919) 392-7363 | C i s c o S y s t e m s From iain at uk.colt.net Tue Dec 13 15:55:33 2005 From: iain at uk.colt.net (Iain Conochie) Date: Tue, 13 Dec 2005 15:55:33 +0000 Subject: RHEL4: cannot kickstart clients with Broadcom DCM5751 Gigabit In-Reply-To: <1134478345.4378.29.camel@brilong-lnx> References: <439E88DF.801@jrc.it> <1134478345.4378.29.camel@brilong-lnx> Message-ID: <439EEEF5.3020501@uk.colt.net> Brian Long wrote: > On Tue, 2005-12-13 at 09:39 +0100, Alex TOURNIER wrote: > >>I have a serious problem with a RHEL4 KS installation. >>My KS server works well because I can perform installation with no >>problems on various hardware plaforms. >>However the latest hardware platform I was delivered (ie Siemens Celsius >>R630) the installation fails. >>I am almost sure that the problem is with the NIC (Broadcom NetXtreme >>DCM 5751 Gigabit ), it looks like the NIC is somehow turned off (even >>if enabled in the BIOS) , or not working at the right speed and thus >>the DHCP server is not able to assign the IP. On the DHCP server nothing >>appears in the log file (as if there was no communication at all). I >>boot the client from a CD (made with the images available on the 1st >>CD of RHEL 4 distrib)....I also tried to make another CD including the >>driver from BROADCOM (bcm5700.ko instead of tg3.ho), but no success. If >>I disable the Broadcom NIC, and insert a second NIC the KS install works >>fine using this second NIC!!!! I have seen a similar issue and i believe that this has been discussed before. However, it goes something like this Once booted, the server will then ask for a DHCP IP address. This has the side effect of resetting the network card and consequently the port on the switch. The DHCP request can time out here as the time that the NIC and the switch take to re-negotiate the link. This may or may not be your issue; i would hazard a guess it is because it does seem to be linked to the Broadcom chipset you describe. Once DHCP fails, do you try again? Does this work the second time? What switch hardware are you using? Cheers Iain >> >>I would be very happy of any advice that could help me in solving the >>problem, I have many of these PC to be installed, and I need disperatly >>a KS solution... >>Thank you in advance. > > > Alex, > > We have the HP XW4300 workstation which has BCM5752. I don't know if > your issue is the same as ours, but only RHEL 3 U6 and RHEL 4 U2 work. > You don't mention which update of RHEL 4 you are kickstarting. Is it > U2? If not, please try U2 as it contains an updated tg3 driver for the > newest Broadcom chipsets. > > /Brian/ > > -- Iain Conochie UNIX System Administrator UK COLT Telecom Units 9-13 Powergate Business Centre, Volt Avenue, London NW10 6PW Tel: +44 (0) 20 7947 1286 Tel: +44 (0) 78 1514 5493 Email: iain.conochie at colt.net www.colt.net From brilong at cisco.com Tue Dec 13 20:17:50 2005 From: brilong at cisco.com (Brian Long) Date: Tue, 13 Dec 2005 15:17:50 -0500 Subject: [Fwd: UP Kernel oops on EM64T with long command-line] Message-ID: <1134505070.4378.62.camel@brilong-lnx> Hello, I sent this to nahant-list earlier but thought some folks on here might have run into my problem. Thanks for any help with this issue. /Brian/ -- Brian Long | | | IT Data Center Systems | .|||. .|||. Cisco Linux Developer | ..:|||||||:...:|||||||:.. Phone: (919) 392-7363 | C i s c o S y s t e m s -------------- next part -------------- An embedded message was scrubbed... From: Brian Long Subject: UP Kernel oops on EM64T with long command-line Date: Tue, 13 Dec 2005 11:25:12 -0500 Size: 5337 URL: From forltran at yahoo.com Wed Dec 14 03:39:29 2005 From: forltran at yahoo.com (Lambert Tran) Date: Tue, 13 Dec 2005 19:39:29 -0800 (PST) Subject: Install exited abnormally... Message-ID: <20051214033929.51132.qmail@web80308.mail.yahoo.com> ... That was the msg when I tried to create a 1 CD installing RH4-ES with a min. kick start. I was pretty much follow Brett Schwarz's procedure. I had rebuilt hdlist[2], had my .discinfo there and when I installed, everything seemed okay, it formatted the drive, transferred the installation package to the harddrive, then suddently, it just exited with the message: Install exited abnormally. It then ummount FS's then the final message: "You may safely reboot your system" [It's like in-your-face from the snake!] Any help? Thanks in advance. -Lambert From error27 at gmail.com Wed Dec 14 07:29:09 2005 From: error27 at gmail.com (Dan Carpenter) Date: Wed, 14 Dec 2005 08:29:09 +0100 Subject: Install exited abnormally... In-Reply-To: <20051214033929.51132.qmail@web80308.mail.yahoo.com> References: <20051214033929.51132.qmail@web80308.mail.yahoo.com> Message-ID: Unfortunately Anaconda failure messages are not normally helpful. Does it still crash if you don't use a kickstart at all but do a completely manual install? > transferred the installation package to the harddrive, I'm not sure what you mean by that. Did it install the RPMs? regards, dan carpenter From alexandre.tournier at jrc.it Wed Dec 14 08:14:24 2005 From: alexandre.tournier at jrc.it (Alex TOURNIER) Date: Wed, 14 Dec 2005 09:14:24 +0100 Subject: RHEL4: cannot kickstart clients with Broadcom DCM5751 Gigabit In-Reply-To: <439EEEF5.3020501@uk.colt.net> References: <439E88DF.801@jrc.it> <1134478345.4378.29.camel@brilong-lnx> <439EEEF5.3020501@uk.colt.net> Message-ID: <439FD460.1070009@jrc.it> Iain Once DHCP fails, if I try gain it fails again..it does not work ...I use a CISCO switch CATALYST 4503. I did try also with RHEL 4 update 2 as adives by Brian, but the behaviour is the same. It works if I insert a 3COM 3C905CX into the host...but it is just a test...i want to use the Broadcom NIC as it's integrated on the mother board and I have a bunch of those to be installed. Do you think that using a simple Hub instead of a Switch could help? Cheers, Alex. > Brian Long wrote: > >> On Tue, 2005-12-13 at 09:39 +0100, Alex TOURNIER wrote: >> >>> I have a serious problem with a RHEL4 KS installation. >>> My KS server works well because I can perform installation with no >>> problems on various hardware plaforms. >>> However the latest hardware platform I was delivered (ie Siemens >>> Celsius R630) the installation fails. >>> I am almost sure that the problem is with the NIC (Broadcom >>> NetXtreme DCM 5751 Gigabit ), it looks like the NIC is somehow >>> turned off (even if enabled in the BIOS) , or not working at the >>> right speed and thus the DHCP server is not able to assign the IP. >>> On the DHCP server nothing appears in the log file (as if there was >>> no communication at all). I boot the client from a CD (made with >>> the images available on the 1st CD of RHEL 4 distrib)....I also >>> tried to make another CD including the driver from BROADCOM >>> (bcm5700.ko instead of tg3.ho), but no success. If I disable the >>> Broadcom NIC, and insert a second NIC the KS install works fine >>> using this second NIC!!!! >> > > I have seen a similar issue and i believe that this has been discussed > before. However, it goes something like this > > Once booted, the server will then ask for a DHCP IP address. This has > the side effect of resetting the network card and consequently the > port on the switch. The DHCP request can time out here as the time > that the NIC and the switch take to re-negotiate the link. > > This may or may not be your issue; i would hazard a guess it is > because it does seem to be linked to the Broadcom chipset you > describe. Once DHCP fails, do you try again? Does this work the second > time? What switch hardware are you using? > > Cheers > > Iain > >>> >>> I would be very happy of any advice that could help me in solving >>> the problem, I have many of these PC to be installed, and I need >>> disperatly a KS solution... >>> Thank you in advance. >> >> >> >> Alex, >> >> We have the HP XW4300 workstation which has BCM5752. I don't know if >> your issue is the same as ours, but only RHEL 3 U6 and RHEL 4 U2 work. >> You don't mention which update of RHEL 4 you are kickstarting. Is it >> U2? If not, please try U2 as it contains an updated tg3 driver for the >> newest Broadcom chipsets. >> >> /Brian/ >> >> > > -- __________________________________________________________________________________________________________ Alexandre TOURNIER System Administrator Joint Research Centre Institute for Environment and Sustainability Global Vegetation Monitoring TP 440 Via Fermi,1 I-21020 ISPRA (VA) ITALY tel: +x-332-785694 fax: +x-332-789073 e-mail: alexandre.tournier at jrc.it web: http://www-gvm.jrc.it/it_support/ FTP: ftp://ftp-gvm.jrc.it "The views expressed are purely those of the writer and may not in any circumstances be regarded as stating an official position of the European Commission." (Disclaimer required under the terms and conditions of use of the internet and electronic mail from Commission equipment) __________________________________________________________________________________________________________ From jkeating at j2solutions.net Wed Dec 14 08:16:06 2005 From: jkeating at j2solutions.net (Jesse Keating) Date: Wed, 14 Dec 2005 00:16:06 -0800 Subject: RHEL4: cannot kickstart clients with Broadcom DCM5751 Gigabit In-Reply-To: <439FD460.1070009@jrc.it> References: <439E88DF.801@jrc.it> <1134478345.4378.29.camel@brilong-lnx> <439EEEF5.3020501@uk.colt.net> <439FD460.1070009@jrc.it> Message-ID: <1134548166.3005.8.camel@yoda.loki.me> On Wed, 2005-12-14 at 09:14 +0100, Alex TOURNIER wrote: > > Once DHCP fails, if I try gain it fails again..it does not work ...I use > a CISCO switch CATALYST 4503. > I did try also with RHEL 4 update 2 as adives by Brian, but the > behaviour is the same. > It works if I insert a 3COM 3C905CX into the host...but it is just a > test...i want to use the Broadcom NIC as it's integrated on the mother > board and I have a bunch of those to be installed. > Do you think that using a simple Hub instead of a Switch could help? Most likely not. I've seen this before, the broadcom just takes too long to get an IPaddress, if it gets one at all. Try setting a static IP address instead of doing DHCP. -- Jesse Keating RHCE (geek.j2solutions.net) Fedora Legacy Team (www.fedoralegacy.org) GPG Public Key (geek.j2solutions.net/jkeating.j2solutions.pub) Was I helpful? Let others know: http://svcs.affero.net/rm.php?r=jkeating From mail2vinodj at yahoo.co.in Wed Dec 14 11:52:15 2005 From: mail2vinodj at yahoo.co.in (vinod kumar) Date: Wed, 14 Dec 2005 11:52:15 +0000 (GMT) Subject: kickstart install without using bootable media Message-ID: <20051214115215.48847.qmail@web8502.mail.in.yahoo.com> hello sir, I have a doubt, whethet i could do the kickstart installation without the bootable media in RedHat Enterprise Linux ES 4.0. and what is the difference between the dhcp and bootp protocol. is that possible. kindly mail me sir, regards vinod kumar j India --------------------------------- Yahoo! India Matrimony: Find your partner now. -------------- next part -------------- An HTML attachment was scrubbed... URL: From i.jochim at querysoft.de Wed Dec 14 13:48:03 2005 From: i.jochim at querysoft.de (Ingo Jochim) Date: Wed, 14 Dec 2005 14:48:03 +0100 Subject: echo and logging in %post Message-ID: <43A02293.5010106@querysoft.de> Hopefully you can help on this. I'm doing a kickstart installation on Fedora Core 4. I like to install some additional packages in the %post part of kickstart. I like to create a new file and want to write something into it. Here is my code: echo $RANDOM > "/root/ID" The result in the file is always fine. But the new file name is strange to me. If I do a "ls -l" I get "ID?" displayed. If I just write the beginning of the file and use the TAB-key then I get "ID^M". What is wrong? And how can I see all the output while the system does all the %post commands? Thank you very much in advance. Ingo From email at jasonkohles.com Wed Dec 14 13:50:46 2005 From: email at jasonkohles.com (Jason Kohles) Date: Wed, 14 Dec 2005 08:50:46 -0500 Subject: echo and logging in %post In-Reply-To: <43A02293.5010106@querysoft.de> References: <43A02293.5010106@querysoft.de> Message-ID: <87f4e9820512140550w14bd7a01w342d4920a8567d68@mail.gmail.com> On 12/14/05, Ingo Jochim wrote: > > Hopefully you can help on this. > > I'm doing a kickstart installation on Fedora Core 4. > I like to install some additional packages in the %post part of kickstart. > > I like to create a new file and want to write something into it. > Here is my code: > echo $RANDOM > "/root/ID" > > The result in the file is always fine. > But the new file name is strange to me. > If I do a "ls -l" I get "ID?" displayed. > If I just write the beginning of the file and use the TAB-key then I get > "ID^M". You are creating your kickstart configuration in a windows/dos text editor, and it has windows line endings, rather than unix ones. Either use a text editor that can save with unix line endings, or use something like the dos2unix command after you create it, or FTP it in text mode to your server before using it. You need to do something to get rid of the extra ^M at the end of every line in your ks.cfg. -- Jason Kohles email at jasonkohles.com - http://www.jasonkohles.com/ "A witty saying proves nothing." -- Voltaire -------------- next part -------------- An HTML attachment was scrubbed... URL: From SSeremeth at anacomp.com Wed Dec 14 13:59:04 2005 From: SSeremeth at anacomp.com (Seremeth, Stephen) Date: Wed, 14 Dec 2005 08:59:04 -0500 Subject: RHEL4: cannot kickstart clients with Broadcom DCM5751 Gigabit Message-ID: <66911C8E1F2DA24EB29DE289297A135ED5F922@usrd104> I apologize if this has already been suggested - I haven't been following the thread that closely, but make sure portfast is enabled on the port on the switch. This will make or break a dhcp install in my experience. Regards, Steve From mmetzner at intergenia.de Wed Dec 14 13:56:19 2005 From: mmetzner at intergenia.de (Marian Marinov) Date: Wed, 14 Dec 2005 15:56:19 +0200 Subject: Detecting hostname Message-ID: <43A02483.3080205@intergenia.de> Hello, I'm using kickstart files for more then 4 months now and I still have one annoing problem. When the installations starts it is trying to find the reverse address of the machine and freezes for 1-2 min. Is there a way to disable this with any boot option change in the initrd? Regards Marian From brilong at cisco.com Wed Dec 14 14:17:00 2005 From: brilong at cisco.com (Brian Long) Date: Wed, 14 Dec 2005 09:17:00 -0500 Subject: Detecting hostname In-Reply-To: <43A02483.3080205@intergenia.de> References: <43A02483.3080205@intergenia.de> Message-ID: <1134569820.4499.14.camel@brilong-lnx> On Wed, 2005-12-14 at 15:56 +0200, Marian Marinov wrote: > Hello, > I'm using kickstart files for more then 4 months now and I still have > one annoing problem. When the installations starts it is trying to find > the reverse address of the machine and freezes for 1-2 min. > Is there a way to disable this with any boot option change in the initrd? If you pass --hostname hostname.domainname in the network line, the reverse lookup will not happen. /Brian/ -- Brian Long | | | IT Data Center Systems | .|||. .|||. Cisco Linux Developer | ..:|||||||:...:|||||||:.. Phone: (919) 392-7363 | C i s c o S y s t e m s From mmetzner at intergenia.de Wed Dec 14 14:32:30 2005 From: mmetzner at intergenia.de (Marian Marinov) Date: Wed, 14 Dec 2005 16:32:30 +0200 Subject: Detecting hostname In-Reply-To: <1134569820.4499.14.camel@brilong-lnx> References: <43A02483.3080205@intergenia.de> <1134569820.4499.14.camel@brilong-lnx> Message-ID: <43A02CFE.3090008@intergenia.de> Brian Long wrote: > On Wed, 2005-12-14 at 15:56 +0200, Marian Marinov wrote: > >>Hello, >>I'm using kickstart files for more then 4 months now and I still have >>one annoing problem. When the installations starts it is trying to find >>the reverse address of the machine and freezes for 1-2 min. >>Is there a way to disable this with any boot option change in the initrd? > > > If you pass --hostname hostname.domainname in the network line, the > reverse lookup will not happen. > > /Brian/ > Many thanks Brian. My installations are now working as a charm. Again Thanks Regards Marian From rspaulding at mail.arc.nasa.gov Wed Dec 14 17:09:27 2005 From: rspaulding at mail.arc.nasa.gov (Ryan C. Spaulding) Date: Wed, 14 Dec 2005 09:09:27 -0800 Subject: partition syntax In-Reply-To: <37ED92F9890FAF4BB947613C66FF8B1A08BB2C73@dw-mail.dataway.com> References: <37ED92F9890FAF4BB947613C66FF8B1A08BB2C73@dw-mail.dataway.com> Message-ID: Not sure if you got your answer or not yet. Below is our kickstart for preserving a previous installation partition: part /boot --fstype ext3 --asprimary --size 128 --onpart sda1 part swap --asprimary --size 1000 --onpart sda3 part / --fstype ext3 --asprimary --size 10000 --onpart sda2 part /home --fstype ext3 --size 40000 --onpart sda5 --noformat part /scratch --fstype ext3 --size 1 --grow --onpart sda6 --noformat You need to have each partition specified or the installation will fail every time. Also do NOT have the following lines in the kickstart: zerombr yes clearpart --all --initlabel Thank you, Ryan On Dec 12, 2005, at 3:24 PM, Tedman Eng wrote: > Try adding the --noformat in addition to the --onpart > >> -----Original Message----- >> From: Stephen Mah [mailto:smah at vmware.com] >> Sent: Monday, December 12, 2005 12:33 PM >> To: Discussion list about Kickstart >> Subject: partition syntax >> >> >> what's the partitioning syntax to use an existing partition, >> say a swap >> partition from redhat 7.2 and reformat it for rhel3? >> >> I essentially want to use the same swap partition for both >> 7.2 and el3. >> >> I tried part swap --onpart /dev/sda2 >> >> I'm getting unable to open /dev/sda2 for the use of "none" >> >> -steve >> >> _______________________________________________ >> Kickstart-list mailing list >> Kickstart-list at redhat.com >> https://www.redhat.com/mailman/listinfo/kickstart-list >> > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From phr at doc.ic.ac.uk Thu Dec 15 01:47:59 2005 From: phr at doc.ic.ac.uk (Philip Rowlands) Date: Thu, 15 Dec 2005 01:47:59 +0000 (GMT) Subject: kickstart install without using bootable media In-Reply-To: <20051214115215.48847.qmail@web8502.mail.in.yahoo.com> References: <20051214115215.48847.qmail@web8502.mail.in.yahoo.com> Message-ID: On Wed, 14 Dec 2005, vinod kumar wrote: > I have a doubt, whethet i could do the kickstart installation without > the bootable media in RedHat Enterprise Linux ES 4.0. It's possible to avoid booting from floppy, CD/DVD-ROM, hard disk, or USB key if you have network boot (PXE) support in the BIOS, or provided by a local alternative (etherboot or Windows RIS disk). > and what is the difference between the dhcp and bootp protocol. At a basic level they both provide support for assinging an IP address to a host over the network. bootp pre-dates DHCP by a good while, and DHCP supports many more options to convey settings other than IP address. Cheers, Phil From i.jochim at querysoft.de Thu Dec 15 10:46:24 2005 From: i.jochim at querysoft.de (Ingo Jochim) Date: Thu, 15 Dec 2005 11:46:24 +0100 Subject: echo and logging in %post In-Reply-To: <87f4e9820512140550w14bd7a01w342d4920a8567d68@mail.gmail.com> References: <43A02293.5010106@querysoft.de> <87f4e9820512140550w14bd7a01w342d4920a8567d68@mail.gmail.com> Message-ID: <43A14980.6090104@querysoft.de> Thank you Jason. You were right. I did create the kickstart script on a linux machine. But then I used WinScp to copy the file to a Windows machine. There I have GuildFTPD running as ftp server. WinScp or GuildFTPD make the changes to my file. Now I started the kickstart script from a floppy and everything works fine. Thank you again. Ingo Jason Kohles wrote: > On 12/14/05, *Ingo Jochim* > wrote: > > Hopefully you can help on this. > > I'm doing a kickstart installation on Fedora Core 4. > I like to install some additional packages in the %post part of > kickstart. > > I like to create a new file and want to write something into it. > Here is my code: > echo $RANDOM > "/root/ID" > > The result in the file is always fine. > But the new file name is strange to me. > If I do a "ls -l" I get "ID?" displayed. > If I just write the beginning of the file and use the TAB-key then > I get > "ID^M". > > > You are creating your kickstart configuration in a windows/dos text > editor, and it has windows line endings, rather than unix ones. > Either use a text editor that can save with unix line endings, or use > something like the dos2unix command after you create it, or FTP it in > text mode to your server before using it. You need to do something to > get rid of the extra ^M at the end of every line in your ks.cfg. > > -- > Jason Kohles > email at jasonkohles.com - > http://www.jasonkohles.com/ > "A witty saying proves nothing." -- Voltaire > >------------------------------------------------------------------------ > >_______________________________________________ >Kickstart-list mailing list >Kickstart-list at redhat.com >https://www.redhat.com/mailman/listinfo/kickstart-list > From rickphilbrick at gmail.com Thu Dec 15 18:49:18 2005 From: rickphilbrick at gmail.com (Rick Philbrick) Date: Thu, 15 Dec 2005 10:49:18 -0800 Subject: bootloader options for SW raid 1 arrangement Message-ID: Hi, I recently did a KS-cfg for a SWraid 1 configuration, RHEL-3 on a Dell 1750 w/ 2 identical SCSI drives. Now when I tested the bootability of both disks only one seemed to work properly. On the disk that didn't I had to rerun grub install to get the mbr written. What option/s should be included in my KS-cfg in order to have either disk bootable? Below are the two that i do use. bootloader --location=mbr zerombr yes RIckp -------------- next part -------------- An HTML attachment was scrubbed... URL: From jpfox17 at gmail.com Thu Dec 15 23:34:57 2005 From: jpfox17 at gmail.com (John Fox) Date: Thu, 15 Dec 2005 18:34:57 -0500 Subject: failure to recognize network config in ks.cfg Message-ID: Hi, I'm running RHEL 4 U1 AS. I got my kickstart working with PXE, TFTP, DHCP, NFS, etc. I'm able to see my kickstart config file and most of the config is picked up during the install, except for the network config. I get prompted to choose which NIC to configure and then I have to specify the configuration. After that, the install goes pretty well, but I need to be able to automate the NIC configuration piece. Here is the network line from my kickstart file (all one line): network --device eth0 --bootproto static --ip 10.1.182.62 --netmask 255.255.255.0 --gateway 10.1.182.1 --nameserver 10.1.180.226 --hostname host1 The machine that is being built is an HP DL380 w/2 GigE NIC cards. Any help would be greatly appreciated. Thanks, John -------------- next part -------------- An HTML attachment was scrubbed... URL: From mike at ftl.com Fri Dec 16 01:50:36 2005 From: mike at ftl.com (Mike Smith) Date: Thu, 15 Dec 2005 17:50:36 -0800 Subject: bootloader options for SW raid 1 arrangement In-Reply-To: References: Message-ID: <43A21D6C.4040003@ftl.com> Rick Philbrick wrote: > > What option/s should be included in my KS-cfg in order to have either > disk bootable? Below are the two that i do use. > bootloader --location=mbr > zerombr yes I use the following in %post grub --batch </dev/null 2>/dev/null root (hd0,0) setup (hd0) root (hd1,0) setup (hd1) quit EOT hope that helps - Mike From maddingue at free.fr Fri Dec 16 11:44:30 2005 From: maddingue at free.fr (=?iso-8859-1?b?U+liYXN0aWVu?= Aperghis-Tramoni) Date: Fri, 16 Dec 2005 12:44:30 +0100 Subject: failure to recognize network config in ks.cfg In-Reply-To: References: Message-ID: <1134733470.43a2a89e9206f@imp6-g19.free.fr> John Fox wrote: > Hi, > > I'm running RHEL 4 U1 AS. I got my kickstart working with PXE, TFTP, DHCP, > NFS, etc. I'm able to see my kickstart config file and most of the config > is picked up during the install, except for the network config. I get > prompted to choose which NIC to configure and then I have to specify the > configuration. After that, the install goes pretty well, but I need to be > able to automate the NIC configuration piece. Here is the network line from > my kickstart file (all one line): > > network --device eth0 --bootproto static --ip 10.1.182.62 --netmask > 255.255.255.0 --gateway 10.1.182.1 --nameserver 10.1.180.226 --hostname > host1 > > The machine that is being built is an HP DL380 w/2 GigE NIC cards. I think someone previously said that this can be done by adding "ksdevice=link" to the append option in your pxelinux.cfg/default. This should autoselect the device physically connected to the network. -- S?bastien Aperghis-Tramoni Close the world, txEn eht nepO. From markryde at gmail.com Sun Dec 18 15:06:04 2005 From: markryde at gmail.com (Mark Ryden) Date: Sun, 18 Dec 2005 17:06:04 +0200 Subject: error in installing FC4 with kickstart: (non existing packeges) Message-ID: Hello, I am trying to use kickstart to install FC4. I did installed FC4 (without kickstart) and took the anaconda-ks.cfg as a basis to use on installation on a new machine (which has the same hw,exactly). I am using http kickstart installation and get many messages like these: ... PACKAGE NAME references package PACKAGE NAME which doesn't exist PACKAGE NAME references package PACKAGE NAME which doesn't exist PACKAGE NAME references package PACKAGE NAME which doesn't exist ... ... and then I see an exception and the installation aborts (with printing trace) googling a bit showed the following thread: http://forums.fedoraforum.org/showthread.php?t=60555 it says there: "...but I was able to cure this crash by removing all of the "-package_name" lines I had in the %packages section. I added "rpm -e package_name" commands in the %post section of the kickstart file instead...that seems to do the trick for me." but this seems to me not so proper way, not to say a bit silly (I did not checked if it works) ; there must be another way. any ideas ? MR From phr at doc.ic.ac.uk Mon Dec 19 13:25:48 2005 From: phr at doc.ic.ac.uk (Philip Rowlands) Date: Mon, 19 Dec 2005 13:25:48 +0000 (GMT) Subject: error in installing FC4 with kickstart: (non existing packeges) In-Reply-To: References: Message-ID: On Sun, 18 Dec 2005, Mark Ryden wrote: > I am trying to use kickstart to install FC4. > I did installed FC4 (without kickstart) and took the anaconda-ks.cfg > as a basis to use on installation on a new machine (which has the same > hw,exactly). Please attach the kickstart file which is causing problems (or at least the %packages section) Did you add/remove/change any *.rpm files from the installation media, and/or run genhdlist? Cheers, Phil From MWeiner at ag.com Mon Dec 19 14:23:13 2005 From: MWeiner at ag.com (MW Mike Weiner (5028)) Date: Mon, 19 Dec 2005 09:23:13 -0500 Subject: Problems doing a ks over NFS Message-ID: <4FD2C985D5E2A642AE25823DFD61C2B0095DA030@orca.agcom.amgreetings.com> Dear list readers - I have been googling and reading, trying to figure this out, but its come time to ask the collective brain trust here for ideas, as none of my efforts have fixed my problems. What i have is about 400 servers that are in need of being kickstarted from RH73 to FC2. The network infrastructure is such that these 400 servers are spread over multiple vlans in order to spread the servers over multiple routers, etc. Normally i would be able to have dhcp forwarding enabled, which MIGHT have solved my problem, but in our environment this would be a huge no-no. So, in an effort to get around that problem, i have a few dhcp-relay servers running, one on each of the subnets (vlans) in this environment So, when kickstarting a box that sits natively on the same subnet as the yum/ks server - this works, nfs is found, mounted, and the kickstart works. Now, for ANY other box, not sitting on the same subnet, this fails, even though i can see the dhcp requests getting there from the relay servers, the nfs mount is not able to be found, nor mounted, so the kickstart fails. By the wway, I have exported the 'repo' I am using for these kickstart installs as follows: I desperately need to get this to work - however, i am limited in what i can 'open' within this infrastructure. Does anyone have any pointers, ideas, places for me to look to try to help debug and figure out JUST WHAT THE HECK is going on here. Thanks in advance Michael Weiner From MWeiner at ag.com Mon Dec 19 14:24:42 2005 From: MWeiner at ag.com (MW Mike Weiner (5028)) Date: Mon, 19 Dec 2005 09:24:42 -0500 Subject: Problems doing a ks over NFS Message-ID: <4FD2C985D5E2A642AE25823DFD61C2B0095DA034@orca.agcom.amgreetings.com> -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of MW Mike Weiner (5028) Sent: Monday, December 19, 2005 9:23 AM To: Discussion list about Kickstart Cc: MW Mike Weiner (5028) Subject: Problems doing a ks over NFS Dear list readers - I have been googling and reading, trying to figure this out, but its come time to ask the collective brain trust here for ideas, as none of my efforts have fixed my problems. What i have is about 400 servers that are in need of being kickstarted from RH73 to FC2. The network infrastructure is such that these 400 servers are spread over multiple vlans in order to spread the servers over multiple routers, etc. Normally i would be able to have dhcp forwarding enabled, which MIGHT have solved my problem, but in our environment this would be a huge no-no. So, in an effort to get around that problem, i have a few dhcp-relay servers running, one on each of the subnets (vlans) in this environment So, when kickstarting a box that sits natively on the same subnet as the yum/ks server - this works, nfs is found, mounted, and the kickstart works. Now, for ANY other box, not sitting on the same subnet, this fails, even though i can see the dhcp requests getting there from the relay servers, the nfs mount is not able to be found, nor mounted, so the kickstart fails. By the wway, I have exported the 'repo' I am using for these kickstart installs as follows: Sorry got cut off, here is my exports: /repo *(ro) /tftpboot *(ro) I desperately need to get this to work - however, i am limited in what i can 'open' within this infrastructure. Does anyone have any pointers, ideas, places for me to look to try to help debug and figure out JUST WHAT THE HECK is going on here. Thanks in advance Michael Weiner _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From phr at doc.ic.ac.uk Mon Dec 19 14:52:45 2005 From: phr at doc.ic.ac.uk (Philip Rowlands) Date: Mon, 19 Dec 2005 14:52:45 +0000 (GMT) Subject: Problems doing a ks over NFS In-Reply-To: <4FD2C985D5E2A642AE25823DFD61C2B0095DA030@orca.agcom.amgreetings.com> References: <4FD2C985D5E2A642AE25823DFD61C2B0095DA030@orca.agcom.amgreetings.com> Message-ID: On Mon, 19 Dec 2005, MW Mike Weiner (5028) wrote: [snip] > The network infrastructure is such that these 400 servers are spread > over multiple vlans in order to spread the servers over multiple > routers, etc. Normally i would be able to have dhcp forwarding > enabled, which MIGHT have solved my problem, but in our environment > this would be a huge no-no. So, in an effort to get around that > problem, i have a few dhcp-relay servers running, one on each of the > subnets (vlans) in this environment So, when kickstarting a box that > sits natively on the same subnet as the yum/ks server - this works, > nfs is found, mounted, and the kickstart works. Now, for ANY other > box, not sitting on the same subnet, this fails, even though i can see > the dhcp requests getting there from the relay servers, the nfs mount > is not able to be found, nor mounted, so the kickstart fails. Are you sure the DHCP relay is working? What happens outside of a kickstart install? Are you sure that NFS is working? What happens outside of a kickstart install? [repeat as necessary; try doing a manual install and note at which point it breaks] Cheers, Phil From MWeiner at ag.com Mon Dec 19 15:00:31 2005 From: MWeiner at ag.com (MW Mike Weiner (5028)) Date: Mon, 19 Dec 2005 10:00:31 -0500 Subject: Problems doing a ks over NFS Message-ID: <4FD2C985D5E2A642AE25823DFD61C2B0095DA092@orca.agcom.amgreetings.com> From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Philip Rowlands [snip] > Are you sure the DHCP relay is working? What happens outside of a kickstart > install? > Are you sure that NFS is working? What happens outside of a kickstart install? > [repeat as necessary; try doing a manual install and note at which point it breaks] Well I would think that since I see messages on the kickstart server FROM the dhcprelay server, that this would imply things are kinda sorta working, yes. As for NFS, yes I know that's working as I can do a 'same vlan' kickstart over NFS and that works fine. I just keep seeing this on the dhcp/kickstart/repo server whenever I attempt to do a box on another subnet: Dec 19 09:22:03 yum1 dhcpd: DHCPDISCOVER from 00:30:48:29:c9:55 via 192.168.20.58: unknown network segment Dec 19 09:22:05 yum1 dhcpd: DHCPDISCOVER from 00:30:48:29:c9:55 via 192.168.20.58: unknown network segment Dec 19 09:22:06 yum1 dhcpd: DHCPINFORM from 192.168.20.160 via eth1: unknown subnet 192.168.20.58 Dec 19 09:22:09 yum1 dhcpd: DHCPDISCOVER from 00:30:48:29:c9:55 via 192.168.20.58: unknown network segment Dec 19 09:22:14 yum1 dhcpd: DHCPINFORM from 192.168.20.160 via eth1: unknown subnet 192.168.20.58 Dec 19 09:22:17 yum1 dhcpd: DHCPDISCOVER from 00:30:48:29:c9:55 via 192.168.20.58: unknown network segment Dec 19 09:22:22 yum1 dhcpd: DHCPINFORM from 192.168.20.160 via eth1: unknown subnet 192.168.20.58 Dec 19 09:22:30 yum1 dhcpd: DHCPINFORM from 192.168.20.160 via eth1: unknown subnet 192.168.20.58 The server 20.58 is a dhcp-relay server. Thanks again for any assistance Michael Weiner From phr at doc.ic.ac.uk Mon Dec 19 15:20:42 2005 From: phr at doc.ic.ac.uk (Philip Rowlands) Date: Mon, 19 Dec 2005 15:20:42 +0000 (GMT) Subject: Problems doing a ks over NFS In-Reply-To: <4FD2C985D5E2A642AE25823DFD61C2B0095DA092@orca.agcom.amgreetings.com> References: <4FD2C985D5E2A642AE25823DFD61C2B0095DA092@orca.agcom.amgreetings.com> Message-ID: On Mon, 19 Dec 2005, MW Mike Weiner (5028) wrote: >> Are you sure the DHCP relay is working? What happens outside of a >> kickstart install? > > Well I would think that since I see messages on the kickstart server > FROM the dhcprelay server, that this would imply things are kinda sorta > working, yes. I'd suggest checking to make sure. > As for NFS, yes I know that's working as I can do a 'same vlan' > kickstart over NFS and that works fine. I'd suggest checking to make sure. In both cases you appear to have given reasons why things should be working through indirect evidence. > I just keep seeing this on the dhcp/kickstart/repo server whenever I > attempt to do a box on another subnet: > > Dec 19 09:22:03 yum1 dhcpd: DHCPDISCOVER from 00:30:48:29:c9:55 via 192.168.20.58: unknown network segment > Dec 19 09:22:05 yum1 dhcpd: DHCPDISCOVER from 00:30:48:29:c9:55 via 192.168.20.58: unknown network segment > Dec 19 09:22:06 yum1 dhcpd: DHCPINFORM from 192.168.20.160 via eth1: unknown subnet 192.168.20.58 This is far from a complete DHCP exchange; the client is not receiving any "offer". Why the "unknown subnet" errors? Cheers, Phil From MWeiner at ag.com Mon Dec 19 15:31:23 2005 From: MWeiner at ag.com (MW Mike Weiner (5028)) Date: Mon, 19 Dec 2005 10:31:23 -0500 Subject: Problems doing a ks over NFS Message-ID: <4FD2C985D5E2A642AE25823DFD61C2B0095DA0D3@orca.agcom.amgreetings.com> -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On >> Are you sure the DHCP relay is working? What happens outside of a >> kickstart install? > > Well I would think that since I see messages on the kickstart server > FROM the dhcprelay server, that this would imply things are kinda > sorta working, yes. I'd suggest checking to make sure. > As for NFS, yes I know that's working as I can do a 'same vlan' > kickstart over NFS and that works fine. I'd suggest checking to make sure. In both cases you appear to have given reasons why things should be working through indirect evidence. > I just keep seeing this on the dhcp/kickstart/repo server whenever I > attempt to do a box on another subnet: > > Dec 19 09:22:03 yum1 dhcpd: DHCPDISCOVER from 00:30:48:29:c9:55 via > 192.168.20.58: unknown network segment Dec 19 09:22:05 yum1 dhcpd: > DHCPDISCOVER from 00:30:48:29:c9:55 via 192.168.20.58: unknown network > segment Dec 19 09:22:06 yum1 dhcpd: DHCPINFORM from 192.168.20.160 via > eth1: unknown subnet 192.168.20.58 -- This is far from a complete DHCP exchange; the client is not receiving any "offer". Why the "unknown subnet" errors? Uhmmm the fact that a kickstart install over NFS works, is direct evidence, not indirect evidence, that the NFS on the same subnet works - I have evidence in the logs that the ks servers recieves an NFS call etc, so THIS WORKS - period. Now, I would agree on the dhcrelay side however, but am unsure how to truly test this - especially when my poor boxes are 1000 miles away. Tho I could take one of the two interfaces and switch it from static to dhcp and temporarily 'break' the box and test it that way. But is there a convenient way to test the communication between dhcrelay and dhcpd server without necessarily having to do that? I pasted the snippet from the log to illustrate the errors that caused me to even begin this thread. If I werent seing those errors, I would imagine things would most likely be working. But they arent. I understand dhcp wont natively work across subnets without permitting dhcp forwarding on the switches that these servers are plumbed into - I cant enable that, so I went for the relay-styled communication to get around that issue. I appreciate your help however Phil!!!!! Michael Weiner From MWeiner at ag.com Mon Dec 19 19:45:46 2005 From: MWeiner at ag.com (MW Mike Weiner (5028)) Date: Mon, 19 Dec 2005 14:45:46 -0500 Subject: Problems doing a ks over NFS Message-ID: <4FD2C985D5E2A642AE25823DFD61C2B0095DA250@orca.agcom.amgreetings.com> Ok, here are some log entries to indicate that the dhcp IS working over the dhcrelays Dec 19 14:57:06 yum1 dhcpd: DHCPDISCOVER from 00:50:da:6e:41:58 via 192.168.31.129 Dec 19 14:57:06 yum1 dhcpd: DHCPOFFER on 192.168.31.106 to 00:50:da:6e:41:58 via 192.168.31.129 Dec 19 14:57:06 yum1 dhcpd: DHCPDISCOVER from 00:50:da:6e:41:58 via 192.168.30.58: network 192.168.30/24: no free leases Dec 19 14:57:06 yum1 dhcpd: DHCPDISCOVER from 00:50:da:6e:41:58 via 192.168.31.129 Dec 19 14:57:06 yum1 dhcpd: DHCPOFFER on 192.168.31.106 to 00:50:da:6e:41:58 via 192.168.31.129 Dec 19 14:57:06 yum1 dhcpd: DHCPREQUEST for 192.168.31.106 (192.168.34.248) from 00:50:da:6e:41:58 via 192.168.31.129 Dec 19 14:57:06 yum1 dhcpd: DHCPACK on 192.168.31.106 to 00:50:da:6e:41:58 via 192.168.31.129 Dec 19 14:57:06 yum1 dhcpd: DHCPREQUEST for 192.168.31.106 (192.168.34.248) from 00:50:da:6e:41:58 via 192.168.30.58: ignored (not authoritative). This indicates that this specific box (31.106) SHOULD work and be able to be kickstarted and install via NFS (providing everything else is ok), as the dhcp issues across the subnets has been resolved (in part) by adding another dhcrelay server onto the 31 subnet rather than leaving it on the 30 subnet, as our masking doesn't necessarily allow the two subnets to talk to one another. So, in short I have the following: 192.168.30.x - dhcrelay server already here 192.168.31.x - dhcrelay server added 192.168.32.x - dhcrelay server already here 192.168.33.x - dhcrelay server *needed* here (though subnet mask allows some cross subnet communication) 192.168.34.x - no dhcrelay server as the dhcpd server sits here (which is also the same box doing the NFS and yum/ks portions of this exercise). I am hoping this will alleviate my headaches. Anything else I may have missed? Anyone? Thanks in advance, and thanks for all the responses to my problem. This e-mail and any attachments may contain confidential information that is legally privileged. The information is solely for the use of the intended recipient(s); any disclosure, copying, distribution, or other use of this information is strictly prohibited. If you have received this e-mail in error, please notify the sender by return e-mail and delete this message. Thank you. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Philip Rowlands Sent: Monday, December 19, 2005 10:21 AM To: Discussion list about Kickstart Subject: RE: Problems doing a ks over NFS On Mon, 19 Dec 2005, MW Mike Weiner (5028) wrote: >> Are you sure the DHCP relay is working? What happens outside of a >> kickstart install? > > Well I would think that since I see messages on the kickstart server > FROM the dhcprelay server, that this would imply things are kinda > sorta working, yes. I'd suggest checking to make sure. > As for NFS, yes I know that's working as I can do a 'same vlan' > kickstart over NFS and that works fine. I'd suggest checking to make sure. In both cases you appear to have given reasons why things should be working through indirect evidence. > I just keep seeing this on the dhcp/kickstart/repo server whenever I > attempt to do a box on another subnet: > > Dec 19 09:22:03 yum1 dhcpd: DHCPDISCOVER from 00:30:48:29:c9:55 via > 192.168.20.58: unknown network segment Dec 19 09:22:05 yum1 dhcpd: > DHCPDISCOVER from 00:30:48:29:c9:55 via 192.168.20.58: unknown network > segment Dec 19 09:22:06 yum1 dhcpd: DHCPINFORM from 192.168.20.160 via > eth1: unknown subnet 192.168.20.58 This is far from a complete DHCP exchange; the client is not receiving any "offer". Why the "unknown subnet" errors? Cheers, Phil _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From Chris.Pierzycki at ge.com Mon Dec 19 23:42:00 2005 From: Chris.Pierzycki at ge.com (Pierzycki, Chris (GE Healthcare, consultant)) Date: Mon, 19 Dec 2005 17:42:00 -0600 Subject: RHEL 4.2 Anaconda problems Message-ID: <9F96C45D9F164446B5ED47EB89786F3B0ACFF9E2@MKEMLVEM04.e2k.ad.ge.com> Hi, I am getting a segfault error message on the F4 screen of the RHEL 4.2 AMD64 installer. Other than that I do not see any other errors. It just bombs out and waits for user input to reboot. The following is a list of modifications: 1. Created DVD-based installation disk with kickstart-based install script. This works without modifications below. 2. Added two packages. One custom package and KSH package taken from Fedora Core 4. 3. Executed genhdlist. Is the procedure for adding packages different for RHEL 4.2 than RHEL 2.1? I noticed this version of RHEL comes with pdksh package, which could conflict with the new ksh package; however, the installation bombs out without the ksh package too. I think perhaps the genhdlist is not working well or there is something else I have to do. Any help greatly appreciated. Thanks, Chris -------------- next part -------------- An HTML attachment was scrubbed... URL: From klaus.steden at thomson.net Tue Dec 20 00:05:40 2005 From: klaus.steden at thomson.net (Klaus Steden) Date: Mon, 19 Dec 2005 19:05:40 -0500 Subject: RHEL 4.2 Anaconda problems In-Reply-To: <9F96C45D9F164446B5ED47EB89786F3B0ACFF9E2@MKEMLVEM04.e2k.ad.ge.com> References: <9F96C45D9F164446B5ED47EB89786F3B0ACFF9E2@MKEMLVEM04.e2k.ad.ge.com> Message-ID: <20051220000540.GP9429@thomson.net> > Hi, > > > > I am getting a segfault error message on the F4 screen of the RHEL 4.2 > AMD64 installer. Other than that I do not see any other errors. It just > bombs out and waits for user input to reboot. The following is a list of > modifications: > > > > 1. Created DVD-based installation disk with kickstart-based install > script. This works without modifications below. > 2. Added two packages. One custom package and KSH package taken from > Fedora Core 4. > 3. Executed genhdlist. > > > > Is the procedure for adding packages different for RHEL 4.2 than RHEL > 2.1? I noticed this version of RHEL comes with pdksh package, which could > conflict with the new ksh package; however, the installation bombs out > without the ksh package too. > > > > I think perhaps the genhdlist is not working well or there is something > else I have to do. Any help greatly appreciated. > Hello Chris, I've run into something like this before ... make sure the packages you add have filenames consistent with their internal package names. Anaconda isn't smart enough to tell that 'foo.rpm' and 'foo-1.2rc.rpm' are the same thing. gedhdlist spits out a ton of stuff while it operates, so make sure it doesn't yield any critical errors. A script posted here may also be useful: http://fedoraproject.org/wiki/AnacondaGenHdList hth, Klaus From jpfox17 at gmail.com Tue Dec 20 00:12:45 2005 From: jpfox17 at gmail.com (John Fox) Date: Mon, 19 Dec 2005 19:12:45 -0500 Subject: failure to recognize network config in ks.cfg In-Reply-To: <1134733470.43a2a89e9206f@imp6-g19.free.fr> References: <1134733470.43a2a89e9206f@imp6-g19.free.fr> Message-ID: I added "ksdevice=link" to the append option in pxelinux.cfg/default and it seemed to help. Now I am not prompted to choose which nic to use. However, before the kickstart config file is found, the install tries to get the nic config from dhcp. This fails and then I'm prompted to enter the network configuration. Are there other options I can append in pxelinux.cfg/default that will tell the client what its network config should be? What I don't understand is why the client is able to get an IP from DHCP to boot off the tftpboot server, but then it is unable to get or keep the IP to configure the nic during the insall. Any ideas? My pxe and dhcp configs are below. Thanks! [root at 192.168.1.61 pxelinux.cfg]# pxeos -l sec-trn-lnx01 Description: sec-trn-lnx01 Protocol: NFS isDiskless: False Server: 192.168.1.61 Location: /kickstart/media/rhel4-u1-as [root at 192.168.1.61 pxelinux.cfg]# pxeboot -l default sec-trn-lnx01 label sec-trn-lnx01 kernel sec-trn-lnx01/vmlinuz append initrd=sec-trn-lnx01/initrd.img ks=nfs: 192.168.1.61:/kickstart/ks-files/192.168.1.62-kickstart.cfgramdisk_size=10000 ksdevice=link [root at 192.168.1.61 pxelinux.cfg]# cat /etc/dhcpd.conf ddns-update-style interim; ignore client-updates; allow booting; allow bootp; subnet 192.168.1.0 netmask 255.255.255.0 { # --- default gateway option routers 192.168.1.1; option subnet-mask 255.255.255.0; option nis-domain "blah.com"; option domain-name "blah.com"; option domain-name-servers blah blah; option time-offset -18000; # Eastern Standard Time option ntp-servers ntp.blah.blah.com; range dynamic-bootp 192.168.1.63 192.168.1.65; default-lease-time 600; max-lease-time 600; host sec-trn-lnx01 { next-server 192.168.1.61; #next-server marvin.redhat.com; hardware ethernet 00:0D:9D:DC:F6:89; fixed-address 192.168.1.62; filename "linux-install/pxelinux.0"; } } On 12/16/05, S?bastien Aperghis-Tramoni wrote: > > John Fox wrote: > > > Hi, > > > > I'm running RHEL 4 U1 AS. I got my kickstart working with PXE, TFTP, > DHCP, > > NFS, etc. I'm able to see my kickstart config file and most of the > config > > is picked up during the install, except for the network config. I get > > prompted to choose which NIC to configure and then I have to specify the > > configuration. After that, the install goes pretty well, but I need to > be > > able to automate the NIC configuration piece. Here is the network line > from > > my kickstart file (all one line): > > > > network --device eth0 --bootproto static --ip 10.1.182.62 --netmask > > 255.255.255.0 --gateway 10.1.182.1 --nameserver 10.1.180.226 --hostname > > host1 > > > > The machine that is being built is an HP DL380 w/2 GigE NIC cards. > > I think someone previously said that this can be done by adding > "ksdevice=link" to the append option in your pxelinux.cfg/default. > This should autoselect the device physically connected to the network. > > -- > S?bastien Aperghis-Tramoni > > Close the world, txEn eht nepO. > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > -------------- next part -------------- An HTML attachment was scrubbed... URL: From Chris.Pierzycki at ge.com Tue Dec 20 00:22:15 2005 From: Chris.Pierzycki at ge.com (Pierzycki, Chris (GE Healthcare, consultant)) Date: Mon, 19 Dec 2005 18:22:15 -0600 Subject: RHEL 4.2 Anaconda problems Message-ID: <9F96C45D9F164446B5ED47EB89786F3B0ACFFA47@MKEMLVEM04.e2k.ad.ge.com> Both my custom package and KSH were freshly built on RHEL 4.2 machine and their file names were not modified. Do I even need to run genhdlist? Chris -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Klaus Steden Sent: Monday, December 19, 2005 6:06 PM To: Discussion list about Kickstart Subject: Re: RHEL 4.2 Anaconda problems > Hi, > > > > I am getting a segfault error message on the F4 screen of the RHEL 4.2 > AMD64 installer. Other than that I do not see any other errors. It just > bombs out and waits for user input to reboot. The following is a list of > modifications: > > > > 1. Created DVD-based installation disk with kickstart-based install > script. This works without modifications below. > 2. Added two packages. One custom package and KSH package taken from > Fedora Core 4. > 3. Executed genhdlist. > > > > Is the procedure for adding packages different for RHEL 4.2 than RHEL > 2.1? I noticed this version of RHEL comes with pdksh package, which could > conflict with the new ksh package; however, the installation bombs out > without the ksh package too. > > > > I think perhaps the genhdlist is not working well or there is something > else I have to do. Any help greatly appreciated. > Hello Chris, I've run into something like this before ... make sure the packages you add have filenames consistent with their internal package names. Anaconda isn't smart enough to tell that 'foo.rpm' and 'foo-1.2rc.rpm' are the same thing. gedhdlist spits out a ton of stuff while it operates, so make sure it doesn't yield any critical errors. A script posted here may also be useful: http://fedoraproject.org/wiki/AnacondaGenHdList hth, Klaus _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From klaus.steden at thomson.net Tue Dec 20 00:42:02 2005 From: klaus.steden at thomson.net (Klaus Steden) Date: Mon, 19 Dec 2005 19:42:02 -0500 Subject: RHEL 4.2 Anaconda problems In-Reply-To: <9F96C45D9F164446B5ED47EB89786F3B0ACFFA47@MKEMLVEM04.e2k.ad.ge.com> References: <9F96C45D9F164446B5ED47EB89786F3B0ACFFA47@MKEMLVEM04.e2k.ad.ge.com> Message-ID: <20051220004202.GR9429@thomson.net> > Both my custom package and KSH were freshly built on RHEL 4.2 machine > and their file names were not modified. > > Do I even need to run genhdlist? > If they're newer versions or replace existing versions, yes. I would, just to be on the safe side. Klaus From phr at doc.ic.ac.uk Tue Dec 20 02:16:04 2005 From: phr at doc.ic.ac.uk (Philip Rowlands) Date: Tue, 20 Dec 2005 02:16:04 +0000 (GMT) Subject: RHEL 4.2 Anaconda problems In-Reply-To: <9F96C45D9F164446B5ED47EB89786F3B0ACFF9E2@MKEMLVEM04.e2k.ad.ge.com> References: <9F96C45D9F164446B5ED47EB89786F3B0ACFF9E2@MKEMLVEM04.e2k.ad.ge.com> Message-ID: Please don't post HTML to mailing lists. On Mon, 19 Dec 2005, Pierzycki, Chris (GE Healthcare, consultant) wrote: > I am getting a segfault error message on the F4 screen of the RHEL 4.2 > AMD64 installer. Segfault, or anaconda/python stack trace? (Does the error appear as a pop-up box within anaconda?) If it's a genuine segfault that's very wrong and should be escalated to RedHat (especially for their RHEL offering). > I think perhaps the genhdlist is not working well or there is > something else I have to do. Any help greatly appreciated. There is a bug (feature?) in recent anacondas, still present it seems, where the package order is not taken into account, and must therefore be generated correctly (with pkgorder). The wiki page which Klaus mentioned has the details of which commands to run. Cheers, Phil From prasanna at aol.net Tue Dec 20 14:24:40 2005 From: prasanna at aol.net (Prasanna Ponnada) Date: Tue, 20 Dec 2005 19:54:40 +0530 Subject: Kickstart file not being found on mirrored disk Message-ID: <43A81428.4000208@aol.net> I have a machine setup that does RAID level 1 mirroring. I create a kickstart config file and modify grub.conf such that it points to my vmlinuz, initrd and kickstart files. My kickstart, vmlinuz and initrd files are in the /boot directory. Filesystem layout: # df Filesystem 1K-blocks Used Available Use% Mounted on /dev/md0 30233808 761376 27936624 3% / none 1990496 0 1990496 0% /dev/shm #cat /etc/fstab /dev/md0 / ext3 defaults 1 1 none /dev/pts devpts gid=5,mode=620 0 0 none /dev/shm tmpfs defaults 0 0 none /proc proc defaults 0 0 none /sys sysfs defaults 0 0 /dev/cciss/c0d0p3 swap swap defaults 0 0 /dev/hda /media/cdrom auto pamconsole,exec,noauto,managed 0 0 /dev/fd0 /media/floppy auto pamconsole,exec,noauto,managed 0 0 vmlinuz and initrd files are detected fine and my machine boots up fine until it fails to find the kickstart file. After that the loader prompts me to input values. Please find the extract from my grub.conf below: /etc/grub.conf entry: -------------------------- root (hd0,0) kernel /boot/vmlinuz-AS4.0-x86_84 ks='hd:md0:/boot/ks.cfg' ro root=/dev/md0 initrd /boot/initrd-AS4.0-x86_64.img Please find the display on tty1 below: * getting kickstart file * getting kickstart file from harddrive * Loading ks from device md0 on path /ks.cfg * getFileFrom BlockDevice(md0, /ks.cfg) * failed to mount /dev/md0: Invalid argument * trying to mount CD device hda I understand that the "partition" information that I am specifying in my path to ks is wrong. However, I tried various other options but cannot seem to get it right at all. Any help/pointers would be greatly appreciated. thanks Prasanna -------------- next part -------------- An HTML attachment was scrubbed... URL: From maddingue at free.fr Tue Dec 20 14:38:47 2005 From: maddingue at free.fr (=?iso-8859-1?b?U+liYXN0aWVu?= Aperghis-Tramoni) Date: Tue, 20 Dec 2005 15:38:47 +0100 Subject: RHEL 4.2 Anaconda problems In-Reply-To: <20051220000540.GP9429@thomson.net> References: <9F96C45D9F164446B5ED47EB89786F3B0ACFF9E2@MKEMLVEM04.e2k.ad.ge.com> <20051220000540.GP9429@thomson.net> Message-ID: <1135089527.43a817778407d@imp5-g19.free.fr> Selon Klaus Steden : > > Hi, > > > > I am getting a segfault error message on the F4 screen of the RHEL 4.2 > > AMD64 installer. Other than that I do not see any other errors. It > just > > bombs out and waits for user input to reboot. The following is a list > of > > modifications: > > > > > > > > 1. Created DVD-based installation disk with kickstart-based install > > script. This works without modifications below. > > 2. Added two packages. One custom package and KSH package taken from > > Fedora Core 4. > > 3. Executed genhdlist. > > > > > > > > Is the procedure for adding packages different for RHEL 4.2 than RHEL > > 2.1? I noticed this version of RHEL comes with pdksh package, which > could > > conflict with the new ksh package; however, the installation bombs out > > without the ksh package too. > > > > > > > > I think perhaps the genhdlist is not working well or there is something > > else I have to do. Any help greatly appreciated. > > > Hello Chris, > > I've run into something like this before ... make sure the packages you add > have filenames consistent with their internal package names. Anaconda isn't > smart enough to tell that 'foo.rpm' and 'foo-1.2rc.rpm' are the same thing. If that may help, I wrote this small Perl script that takes care of renaming RPM files that don't have the expected names (like Sun JDK): ---[begin: pkg-rename.pl]--- #!/usr/bin/perl use strict; use File::Spec; use RPM::Perlonly; # pure Perl clone of RPM::Info die "usage: $0 \n", " check the rpms located in the directory given in argument\n", " and rename them correctly if necessary\n" unless @ARGV and @ARGV==1; my($rpms_dir) = @ARGV; # check that the given arguments exist -d $rpms_dir or die "No such directory: $rpms_dir\n"; opendir(DIR, $rpms_dir) or die "can't read directory '$rpms_dir': $!"; my @dir = grep /\.rpm$/, readdir(DIR); closedir(DIR); for my $file (@dir) { my $path = File::Spec->catfile($rpms_dir, $file); tie my %header, "RPM::Perlonly", $path or next; my $new_name = sprintf "%s-%s-%s.%s.rpm", @header{qw(NAME VERSION RELEASE ARCH)}; if($file ne $new_name) { print " $file --> $new_name\n"; rename($path, File::Spec->catfile($rpms_dir, $new_name)) or warn "can't rename '$file' to '$new_name': $!"; } untie %header; } ---[end]--- -- S?bastien Aperghis-Tramoni Close the world, txEn eht nepO. From maddingue at free.fr Tue Dec 20 14:53:57 2005 From: maddingue at free.fr (=?iso-8859-1?b?U+liYXN0aWVu?= Aperghis-Tramoni) Date: Tue, 20 Dec 2005 15:53:57 +0100 Subject: RHEL 4.2 Anaconda problems In-Reply-To: References: <9F96C45D9F164446B5ED47EB89786F3B0ACFF9E2@MKEMLVEM04.e2k.ad.ge.com> Message-ID: <1135090437.43a81b052a024@imp5-g19.free.fr> Philip Rowlands wrote: > > I think perhaps the genhdlist is not working well or there is > > something else I have to do. Any help greatly appreciated. > > There is a bug (feature?) in recent anacondas, still present it seems, > where the package order is not taken into account, and must therefore be > generated correctly (with pkgorder). The wiki page which Klaus mentioned > has the details of which commands to run. >From what I've learned, there are two tricks. The first one is that you must run genhdlist, then pkgorder, then again genhdlist: $ genhdlist $RHDIST $ pkgorder $RHDIST i386 [list of groups to prioritize] >pkgorder.txt $ genhdlist --fileorder pkgorder.txt $RHDIST The second trick is that if you use buildinstall to regenerate the installation disk images, it itself re-runs pkgorder, hence ignoring the groups you wanted to prioritize. What I do is save a copy of the pkgorder.txt file before running buildinstall, then copy it back. -- S?bastien Aperghis-Tramoni Close the world, txEn eht nepO. From klaus.steden at thomson.net Tue Dec 20 17:40:42 2005 From: klaus.steden at thomson.net (Klaus Steden) Date: Tue, 20 Dec 2005 12:40:42 -0500 Subject: RHEL 4.2 Anaconda problems In-Reply-To: <1135089527.43a817778407d@imp5-g19.free.fr> References: <9F96C45D9F164446B5ED47EB89786F3B0ACFF9E2@MKEMLVEM04.e2k.ad.ge.com> <20051220000540.GP9429@thomson.net> <1135089527.43a817778407d@imp5-g19.free.fr> Message-ID: <20051220174042.GT9429@thomson.net> Yeah, I think that would help a lot ... in the end I ended up grabbing a jpackage'd RPM of Java (rather than the Sun one) because it worked with jedit ... it had the right name, too, which was a nice bonus. :> Sebastien, would this be you would like to post to the Fedora Wiki? I think it would be helpful to other folk out there. Klaus > > If that may help, I wrote this small Perl script that takes care of > renaming RPM files that don't have the expected names (like Sun JDK): > > ---[begin: pkg-rename.pl]--- > #!/usr/bin/perl > use strict; > use File::Spec; > use RPM::Perlonly; # pure Perl clone of RPM::Info > > die "usage: $0 \n", > " check the rpms located in the directory given in argument\n", > " and rename them correctly if necessary\n" > unless @ARGV and @ARGV==1; > > my($rpms_dir) = @ARGV; > > # check that the given arguments exist > -d $rpms_dir or die "No such directory: $rpms_dir\n"; > > opendir(DIR, $rpms_dir) or die "can't read directory '$rpms_dir': $!"; > my @dir = grep /\.rpm$/, readdir(DIR); > closedir(DIR); > > for my $file (@dir) { > my $path = File::Spec->catfile($rpms_dir, $file); > tie my %header, "RPM::Perlonly", $path or next; > my $new_name = sprintf "%s-%s-%s.%s.rpm", @header{qw(NAME VERSION RELEASE > ARCH)}; > if($file ne $new_name) { > print " $file --> $new_name\n"; > rename($path, File::Spec->catfile($rpms_dir, $new_name)) > or warn "can't rename '$file' to '$new_name': $!"; > } > untie %header; > } > ---[end]--- > > -- > S?bastien Aperghis-Tramoni > > Close the world, txEn eht nepO. > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list -- Klaus Steden | Senior Systems Administrator | Technicolor Creative Services | TODO: Toronto | 1) Learn to use my new Unix account. klaus.steden at thomson.net | 2) Learn how to change this list. Phone: (416) 585-9995 | Fax: (416) 364-1585 | From Chris.Pierzycki at ge.com Tue Dec 20 19:22:03 2005 From: Chris.Pierzycki at ge.com (Pierzycki, Chris (GE Healthcare, consultant)) Date: Tue, 20 Dec 2005 13:22:03 -0600 Subject: RHEL 4.2 Anaconda problems Message-ID: <9F96C45D9F164446B5ED47EB89786F3B0AD008AF@MKEMLVEM04.e2k.ad.ge.com> That's what I thought. Both packages are new to RHEL 4.2. I have a feeling the hdlist files get corrupted after running genhdlist. This is AMD64 version of the OS. Chris -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Klaus Steden Sent: Monday, December 19, 2005 6:42 PM To: Discussion list about Kickstart Subject: Re: RHEL 4.2 Anaconda problems > Both my custom package and KSH were freshly built on RHEL 4.2 machine > and their file names were not modified. > > Do I even need to run genhdlist? > If they're newer versions or replace existing versions, yes. I would, just to be on the safe side. Klaus _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From Chris.Pierzycki at ge.com Tue Dec 20 22:11:00 2005 From: Chris.Pierzycki at ge.com (Pierzycki, Chris (GE Healthcare, consultant)) Date: Tue, 20 Dec 2005 16:11:00 -0600 Subject: RHEL 4.2 Anaconda problems Message-ID: <9F96C45D9F164446B5ED47EB89786F3B0AD5F13C@MKEMLVEM04.e2k.ad.ge.com> Thanks Sebastien. This looks like fixed the problem... whatever it was. So in RHEL 4.2 x86_64 you must run pkgorder followed by genhdlist. In RHEL 2.1 you didn't have to run pkgorder to add packages. Chris -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of S?bastien Aperghis-Tramoni Sent: Tuesday, December 20, 2005 8:54 AM To: Discussion list about Kickstart Subject: Re: RHEL 4.2 Anaconda problems Philip Rowlands wrote: > > I think perhaps the genhdlist is not working well or there is > > something else I have to do. Any help greatly appreciated. > > There is a bug (feature?) in recent anacondas, still present it seems, > where the package order is not taken into account, and must therefore be > generated correctly (with pkgorder). The wiki page which Klaus mentioned > has the details of which commands to run. >From what I've learned, there are two tricks. The first one is that you must run genhdlist, then pkgorder, then again genhdlist: $ genhdlist $RHDIST $ pkgorder $RHDIST i386 [list of groups to prioritize] >pkgorder.txt $ genhdlist --fileorder pkgorder.txt $RHDIST The second trick is that if you use buildinstall to regenerate the installation disk images, it itself re-runs pkgorder, hence ignoring the groups you wanted to prioritize. What I do is save a copy of the pkgorder.txt file before running buildinstall, then copy it back. -- S?bastien Aperghis-Tramoni Close the world, txEn eht nepO. _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From iqbal_tannu at yahoo.com Wed Dec 21 18:41:27 2005 From: iqbal_tannu at yahoo.com (Mohammed Iqbal.H) Date: Wed, 21 Dec 2005 10:41:27 -0800 (PST) Subject: kickstart hangs Message-ID: <20051221184127.29531.qmail@web31110.mail.mud.yahoo.com> hi I have noticed a kickstart hangs at the point when the PC being NFS kickstarted is at the "Determining host name and domain" point in the kickstart. please let me know why it hangs.if any one have the kickstart file or document please send it to me. thanks iqbal __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From koreesmith at gmail.com Wed Dec 21 20:23:53 2005 From: koreesmith at gmail.com (Koree A. Smith) Date: Wed, 21 Dec 2005 14:23:53 -0600 Subject: kickstart hangs In-Reply-To: <20051221184127.29531.qmail@web31110.mail.mud.yahoo.com> References: <20051221184127.29531.qmail@web31110.mail.mud.yahoo.com> Message-ID: <676847c60512211223h2fc885cy39aa36f337024cb0@mail.gmail.com> I am getting the same thing. This didn't start until I started kickstarting RHEL4. RHEL3 with an almost identitcal ks.cfg did not behave this way. I don't have a fix, this is just a "me too". Koree On 12/21/05, Mohammed Iqbal.H wrote: > hi > > I have noticed a kickstart hangs at the point when the > PC being NFS kickstarted is at the "Determining host > name and domain" point in the kickstart. > > please let me know why it hangs.if any one have the > kickstart file or document please send it to me. > > thanks > iqbal > > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > -- Koree A. Smith -- koreesmith at gmail.com From drkludge at cox.net Sat Dec 24 03:47:12 2005 From: drkludge at cox.net (Greg Morgan) Date: Fri, 23 Dec 2005 20:47:12 -0700 Subject: kickstart hangs In-Reply-To: <676847c60512211223h2fc885cy39aa36f337024cb0@mail.gmail.com> References: <20051221184127.29531.qmail@web31110.mail.mud.yahoo.com> <676847c60512211223h2fc885cy39aa36f337024cb0@mail.gmail.com> Message-ID: <43ACC4C0.1020407@cox.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Koree A. Smith wrote: > I am getting the same thing. This didn't start until I started > kickstarting RHEL4. RHEL3 with an almost identitcal ks.cfg did not > behave this way. > > I don't have a fix, this is just a "me too". > > Koree > > On 12/21/05, Mohammed Iqbal.H wrote: > >>hi >> >>I have noticed a kickstart hangs at the point when the >>PC being NFS kickstarted is at the "Determining host >>name and domain" point in the kickstart. Fedora Core has done this for awhile. The RHEL3 to RHEL4 sounds like the right time for RHEL. This is a known problem as far as it has been discussed on the mailing list before. One possible answer is located here https://www.redhat.com/archives/kickstart-list/2005-May/msg00092.html . However, I don't have Cisco network equipment. The other more likely reason is located here https://www.redhat.com/archives/kickstart-list/2005-September/msg00038.html . The good news is that it is not the kickstart file. Where I see the problem is when you use a NFS server to install from ISO images an no kickstart file is in use. The second link has a proposed patch too. >> >>please let me know why it hangs.if any one have the >>kickstart file or document please send it to me. >> Greg -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.1 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iD8DBQFDrMTAxyxe5L6mr7IRAo2sAJ4pNtQJDHABm2bMBXvtlYfH8uxpMgCfYplq u+97ppl5N3h7OJsfbb2Iou4= =pA9s -----END PGP SIGNATURE----- From drkludge at cox.net Sat Dec 24 04:06:24 2005 From: drkludge at cox.net (Greg Morgan) Date: Fri, 23 Dec 2005 21:06:24 -0700 Subject: Kickstart file not being found on mirrored disk In-Reply-To: <43A81428.4000208@aol.net> References: <43A81428.4000208@aol.net> Message-ID: <43ACC940.6070107@cox.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Prasanna Ponnada wrote: > I have a machine setup that does RAID level 1 mirroring. > I create a kickstart config file and modify grub.conf such that it > points to my vmlinuz, initrd and kickstart files. > My kickstart, vmlinuz and initrd files are in the /boot directory. Pransanna, Please try removing the /boot from your kickstart path. I believe grub sees your boot partition as / and not /boot during the boot process. Please see this post http://www.redhat.com/archives/fedora-list/2004-May/msg05948.html . Greg > > Filesystem layout: > # df > Filesystem 1K-blocks Used Available Use% > Mounted on > /dev/md0 30233808 761376 27936624 3% / > none 1990496 0 1990496 0% /dev/shm > #cat /etc/fstab > /dev/md0 / ext3 > defaults 1 1 > none /dev/pts devpts > gid=5,mode=620 0 0 > none /dev/shm tmpfs > defaults 0 0 > none /proc proc > defaults 0 0 > none /sys sysfs > defaults 0 0 > /dev/cciss/c0d0p3 swap swap > defaults 0 0 > /dev/hda /media/cdrom auto > pamconsole,exec,noauto,managed 0 0 > /dev/fd0 /media/floppy auto > pamconsole,exec,noauto,managed 0 0 > > vmlinuz and initrd files are detected fine and my machine boots up fine > until it fails to find the kickstart file. > After that the loader prompts me to input values. > > Please find the extract from my grub.conf below: > /etc/grub.conf entry: > -------------------------- > root (hd0,0) > kernel /boot/vmlinuz-AS4.0-x86_84 ks='hd:md0:/boot/ks.cfg' ro root=/dev/md0 > initrd /boot/initrd-AS4.0-x86_64.img > > Please find the display on tty1 below: > * getting kickstart file > * getting kickstart file from harddrive > * Loading ks from device md0 on path /ks.cfg > * getFileFrom BlockDevice(md0, /ks.cfg) > * failed to mount /dev/md0: Invalid argument > * trying to mount CD device hda > > I understand that the "partition" information that I am specifying in my > path to ks is wrong. > However, I tried various other options but cannot seem to get it right > at all. > Any help/pointers would be greatly appreciated. > > thanks > Prasanna -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.1 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iD8DBQFDrMlAxyxe5L6mr7IRAohQAKCZCBjafkX+6fcLxFZkPBmiylBniACghUlJ Q/ihQlQW6OKHXYwBdrFSiIE= =Lwzs -----END PGP SIGNATURE----- From koreesmith at gmail.com Sat Dec 24 14:03:49 2005 From: koreesmith at gmail.com (Koree A. Smith) Date: Sat, 24 Dec 2005 08:03:49 -0600 Subject: kickstart hangs In-Reply-To: <43ACC4C0.1020407@cox.net> References: <20051221184127.29531.qmail@web31110.mail.mud.yahoo.com> <676847c60512211223h2fc885cy39aa36f337024cb0@mail.gmail.com> <43ACC4C0.1020407@cox.net> Message-ID: <676847c60512240603q35140272n7552b415eace16fb@mail.gmail.com> Actually, this is happening to me, and I am not using NFS. Mine is a fully-automated, unattended, http install using a kickstart file. >From what I can tell, it might be some sort of DNS or hostname resolution issue, but I can't figure it out. This wouldn't include any problems with Cisco equipment, because this happens even when I'm running a crossover cable directly between the target server, and the tftp/http server. Koree On 12/23/05, Greg Morgan wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Koree A. Smith wrote: > > I am getting the same thing. This didn't start until I started > > kickstarting RHEL4. RHEL3 with an almost identitcal ks.cfg did not > > behave this way. > > > > I don't have a fix, this is just a "me too". > > > > Koree > > > > On 12/21/05, Mohammed Iqbal.H wrote: > > > >>hi > >> > >>I have noticed a kickstart hangs at the point when the > >>PC being NFS kickstarted is at the "Determining host > >>name and domain" point in the kickstart. > > Fedora Core has done this for awhile. The RHEL3 to RHEL4 sounds like > the right time for RHEL. This is a known problem as far as it has been > discussed on the mailing list before. One possible answer is located > here > https://www.redhat.com/archives/kickstart-list/2005-May/msg00092.html . > However, I don't have Cisco network equipment. The other more likely > reason is located here > https://www.redhat.com/archives/kickstart-list/2005-September/msg00038.html > . The good news is that it is not the kickstart file. Where I see the > problem is when you use a NFS server to install from ISO images an no > kickstart file is in use. The second link has a proposed patch too. > > >> > >>please let me know why it hangs.if any one have the > >>kickstart file or document please send it to me. > >> > > Greg > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.1 (GNU/Linux) > Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org > > iD8DBQFDrMTAxyxe5L6mr7IRAo2sAJ4pNtQJDHABm2bMBXvtlYfH8uxpMgCfYplq > u+97ppl5N3h7OJsfbb2Iou4= > =pA9s > -----END PGP SIGNATURE----- > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > -- Koree A. Smith -- koreesmith at gmail.com From error27 at gmail.com Sat Dec 24 18:44:45 2005 From: error27 at gmail.com (Dan Carpenter) Date: Sat, 24 Dec 2005 10:44:45 -0800 Subject: kickstart hangs In-Reply-To: <676847c60512240603q35140272n7552b415eace16fb@mail.gmail.com> References: <20051221184127.29531.qmail@web31110.mail.mud.yahoo.com> <676847c60512211223h2fc885cy39aa36f337024cb0@mail.gmail.com> <43ACC4C0.1020407@cox.net> <676847c60512240603q35140272n7552b415eace16fb@mail.gmail.com> Message-ID: 2005/12/24, Koree A. Smith : > Actually, this is happening to me, and I am not using NFS. Mine is a > fully-automated, unattended, http install using a kickstart file. > >From what I can tell, it might be some sort of DNS or hostname > resolution issue, but I can't figure it out. This wouldn't include > any problems with Cisco equipment, because this happens even when I'm > running a crossover cable directly between the target server, and the > tftp/http server. > > Koree > What are the last few things that happen on tty3? Could it be a driver/hardware issue with your NIC? regards, dan From koreesmith at gmail.com Sat Dec 24 18:47:01 2005 From: koreesmith at gmail.com (Koree A. Smith) Date: Sat, 24 Dec 2005 12:47:01 -0600 Subject: kickstart hangs In-Reply-To: References: <20051221184127.29531.qmail@web31110.mail.mud.yahoo.com> <676847c60512211223h2fc885cy39aa36f337024cb0@mail.gmail.com> <43ACC4C0.1020407@cox.net> <676847c60512240603q35140272n7552b415eace16fb@mail.gmail.com> Message-ID: <676847c60512241047uff68894hf7e272d979c68d0b@mail.gmail.com> I'm not sure off the top of my head. I'll have to check on Monday when I return to the office. However, eventually, everything works. It just sits at determining hostname and domain for ~=10 minutes before continuing. Koree On 12/24/05, Dan Carpenter wrote: > 2005/12/24, Koree A. Smith : > > Actually, this is happening to me, and I am not using NFS. Mine is a > > fully-automated, unattended, http install using a kickstart file. > > >From what I can tell, it might be some sort of DNS or hostname > > resolution issue, but I can't figure it out. This wouldn't include > > any problems with Cisco equipment, because this happens even when I'm > > running a crossover cable directly between the target server, and the > > tftp/http server. > > > > Koree > > > > What are the last few things that happen on tty3? Could it be a > driver/hardware issue with your NIC? > > regards, > dan > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > -- Koree A. Smith -- koreesmith at gmail.com From brilong at cisco.com Tue Dec 27 15:16:04 2005 From: brilong at cisco.com (Brian Long) Date: Tue, 27 Dec 2005 10:16:04 -0500 Subject: kickstart hangs In-Reply-To: <676847c60512241047uff68894hf7e272d979c68d0b@mail.gmail.com> References: <20051221184127.29531.qmail@web31110.mail.mud.yahoo.com> <676847c60512211223h2fc885cy39aa36f337024cb0@mail.gmail.com> <43ACC4C0.1020407@cox.net> <676847c60512240603q35140272n7552b415eace16fb@mail.gmail.com> <676847c60512241047uff68894hf7e272d979c68d0b@mail.gmail.com> Message-ID: <1135696564.4461.17.camel@brilong-lnx> On Sat, 2005-12-24 at 12:47 -0600, Koree A. Smith wrote: > I'm not sure off the top of my head. I'll have to check on Monday > when I return to the office. > However, eventually, everything works. It just sits at determining > hostname and domain for ~=10 minutes before continuing. If you specify --hostname on the network line of ks.cfg, Anaconda skips the reverse lookup completely. /Brian/ -- Brian Long | | | IT Data Center Systems | .|||. .|||. Cisco Linux Developer | ..:|||||||:...:|||||||:.. Phone: (919) 392-7363 | C i s c o S y s t e m s From koreesmith at gmail.com Tue Dec 27 15:29:01 2005 From: koreesmith at gmail.com (Koree A. Smith) Date: Tue, 27 Dec 2005 09:29:01 -0600 Subject: kickstart hangs In-Reply-To: <1135696564.4461.17.camel@brilong-lnx> References: <20051221184127.29531.qmail@web31110.mail.mud.yahoo.com> <676847c60512211223h2fc885cy39aa36f337024cb0@mail.gmail.com> <43ACC4C0.1020407@cox.net> <676847c60512240603q35140272n7552b415eace16fb@mail.gmail.com> <676847c60512241047uff68894hf7e272d979c68d0b@mail.gmail.com> <1135696564.4461.17.camel@brilong-lnx> Message-ID: <676847c60512270729q4f2c599ct19af53b0d90c9ea8@mail.gmail.com> That didn't seem to affect it. The last line in the messages on tty3 is: * waiting for link * 0 seconds After sitting in this state for about 10 minutes, it goes on. I added the --hostname option to the line in the ks.cfg. *shrug* Koree On 12/27/05, Brian Long wrote: > On Sat, 2005-12-24 at 12:47 -0600, Koree A. Smith wrote: > > I'm not sure off the top of my head. I'll have to check on Monday > > when I return to the office. > > However, eventually, everything works. It just sits at determining > > hostname and domain for ~=10 minutes before continuing. > > If you specify --hostname on the network line of ks.cfg, > Anaconda skips the reverse lookup completely. > > /Brian/ > > -- > Brian Long | | | > IT Data Center Systems | .|||. .|||. > Cisco Linux Developer | ..:|||||||:...:|||||||:.. > Phone: (919) 392-7363 | C i s c o S y s t e m s > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > -- Koree A. Smith -- koreesmith at gmail.com From koreesmith at gmail.com Tue Dec 27 15:35:22 2005 From: koreesmith at gmail.com (Koree A. Smith) Date: Tue, 27 Dec 2005 09:35:22 -0600 Subject: kickstart hangs In-Reply-To: <676847c60512270729q4f2c599ct19af53b0d90c9ea8@mail.gmail.com> References: <20051221184127.29531.qmail@web31110.mail.mud.yahoo.com> <676847c60512211223h2fc885cy39aa36f337024cb0@mail.gmail.com> <43ACC4C0.1020407@cox.net> <676847c60512240603q35140272n7552b415eace16fb@mail.gmail.com> <676847c60512241047uff68894hf7e272d979c68d0b@mail.gmail.com> <1135696564.4461.17.camel@brilong-lnx> <676847c60512270729q4f2c599ct19af53b0d90c9ea8@mail.gmail.com> Message-ID: <676847c60512270735k5740a70ct8ba17e60eee3e50c@mail.gmail.com> It does say "reverse lookup failed" when it goes on... Koree On 12/27/05, Koree A. Smith wrote: > That didn't seem to affect it. The last line in the messages on tty3 is: > > * waiting for link > * 0 seconds > > After sitting in this state for about 10 minutes, it goes on. I added > the --hostname option to the line in the ks.cfg. > > *shrug* > > Koree > > On 12/27/05, Brian Long wrote: > > On Sat, 2005-12-24 at 12:47 -0600, Koree A. Smith wrote: > > > I'm not sure off the top of my head. I'll have to check on Monday > > > when I return to the office. > > > However, eventually, everything works. It just sits at determining > > > hostname and domain for ~=10 minutes before continuing. > > > > If you specify --hostname on the network line of ks.cfg, > > Anaconda skips the reverse lookup completely. > > > > /Brian/ > > > > -- > > Brian Long | | | > > IT Data Center Systems | .|||. .|||. > > Cisco Linux Developer | ..:|||||||:...:|||||||:.. > > Phone: (919) 392-7363 | C i s c o S y s t e m s > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > > > > > -- > Koree A. Smith > -- > koreesmith at gmail.com > -- Koree A. Smith -- koreesmith at gmail.com From bill at wiliweld.com Tue Dec 27 18:07:20 2005 From: bill at wiliweld.com (Bill Schoolcraft) Date: Tue, 27 Dec 2005 10:07:20 -0800 (PST) Subject: Kickstart from eth1 (eth0 socket bad) Message-ID: <20051227100342.U31984@bsd.billschoolcraft.com> Hello, I have a series of servers that all kick from eth0, we have one server that has a busted socket in the back and we use eth1 for use, not a big deal but when I went to kickstart this box from eth1 the install stopped at the ipaddr setup screen. Nowhere in the ksfile of any of the other boxes is eth0 specified in the "network" stanza. (question) Can I implicitly state eth1 somewhere in the ksfile for this one server? Thanks -- Bill Schoolcraft PO Box 210076 San Francisco, CA 94121 "UNIX, A Way of Life." http://wiliweld.com From Elizabeth.Brosch at thomson.com Tue Dec 27 18:09:12 2005 From: Elizabeth.Brosch at thomson.com (Elizabeth.Brosch at thomson.com) Date: Tue, 27 Dec 2005 13:09:12 -0500 Subject: Kickstart from eth1 (eth0 socket bad) Message-ID: <2213CAA79DA9094AB52517FDADEF07FA021E50F2@tshuspaphimbx01.ERF.THOMSON.COM> I use this in my ks.cfg file: network --bootproto dhcp --device eth0 This works very well for me. Liz Brosch TSH System Services Philadelphia, PA 19104 office: (215) 386-0100 x1144 cell: (267) 784-9166 -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Bill Schoolcraft Sent: Tuesday, December 27, 2005 1:07 PM To: Kickstart Subject: Kickstart from eth1 (eth0 socket bad) Hello, I have a series of servers that all kick from eth0, we have one server that has a busted socket in the back and we use eth1 for use, not a big deal but when I went to kickstart this box from eth1 the install stopped at the ipaddr setup screen. Nowhere in the ksfile of any of the other boxes is eth0 specified in the "network" stanza. (question) Can I implicitly state eth1 somewhere in the ksfile for this one server? Thanks -- Bill Schoolcraft PO Box 210076 San Francisco, CA 94121 "UNIX, A Way of Life." http://wiliweld.com _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From bill at wiliweld.com Tue Dec 27 18:48:58 2005 From: bill at wiliweld.com (Bill Schoolcraft) Date: Tue, 27 Dec 2005 10:48:58 -0800 (PST) Subject: Kickstart from eth1 (eth0 socket bad) In-Reply-To: <2213CAA79DA9094AB52517FDADEF07FA021E50F2@tshuspaphimbx01.ERF.THOMSON.COM> References: <2213CAA79DA9094AB52517FDADEF07FA021E50F2@tshuspaphimbx01.ERF.THOMSON.COM> Message-ID: <20051227103938.V31984@bsd.billschoolcraft.com> At Tue, 27 Dec 2005 it looks like Elizabeth.Brosch at thomson.com composed: > I use this in my ks.cfg file: > > network --bootproto dhcp --device eth0 > > This works very well for me. Thanks Liz, The files generated by our setup takes a hostname arg, OS version arg, partition scheme arg (etc) then checks /etc/hosts and /etc/ethers and then , among other lines in the host-specific ksfile, has this stanza: I think I can enter it after the 'bootproto' option, thanks. ##################################################### # Network Settings network --bootproto static --ip 192.168.1.11 netmask 255.255.255.0 --gateway 192.168.1.254 --nameserver 192.168.1.253 auth --useshadow --enablemd5 firewall --disabled ##################################################### > > Liz Brosch > TSH System Services > Philadelphia, PA 19104 > office: (215) 386-0100 x1144 > cell: (267) 784-9166 > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Bill Schoolcraft > Sent: Tuesday, December 27, 2005 1:07 PM > To: Kickstart > Subject: Kickstart from eth1 (eth0 socket bad) > > Hello, > > I have a series of servers that all kick from eth0, we have one > server that has a busted socket in the back and we use eth1 for > use, not a big deal but when I went to kickstart this box from > eth1 the install stopped at the ipaddr setup screen. > > Nowhere in the ksfile of any of the other boxes is eth0 > specified in the "network" stanza. > > (question) > Can I implicitly state eth1 somewhere in the ksfile for this one > server? > > Thanks > > -- Bill Schoolcraft PO Box 210076 San Francisco, CA 94121 "UNIX, A Way of Life." http://billschoolcraft.com From Elizabeth.Brosch at thomson.com Tue Dec 27 18:51:32 2005 From: Elizabeth.Brosch at thomson.com (Elizabeth.Brosch at thomson.com) Date: Tue, 27 Dec 2005 13:51:32 -0500 Subject: Kickstart from eth1 (eth0 socket bad) Message-ID: <2213CAA79DA9094AB52517FDADEF07FA021E50F3@tshuspaphimbx01.ERF.THOMSON.COM> Yes just add it somewhere as part of the network definition. Thanks. Liz Brosch TSH System Services Philadelphia, PA 19104 office: (215) 386-0100 x1144 cell: (267) 784-9166 -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Bill Schoolcraft Sent: Tuesday, December 27, 2005 1:49 PM To: Discussion list about Kickstart Subject: RE: Kickstart from eth1 (eth0 socket bad) At Tue, 27 Dec 2005 it looks like Elizabeth.Brosch at thomson.com composed: > I use this in my ks.cfg file: > > network --bootproto dhcp --device eth0 > > This works very well for me. Thanks Liz, The files generated by our setup takes a hostname arg, OS version arg, partition scheme arg (etc) then checks /etc/hosts and /etc/ethers and then , among other lines in the host-specific ksfile, has this stanza: I think I can enter it after the 'bootproto' option, thanks. ##################################################### # Network Settings network --bootproto static --ip 192.168.1.11 netmask 255.255.255.0 --gateway 192.168.1.254 --nameserver 192.168.1.253 auth --useshadow --enablemd5 firewall --disabled ##################################################### > > Liz Brosch > TSH System Services > Philadelphia, PA 19104 > office: (215) 386-0100 x1144 > cell: (267) 784-9166 > > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Bill Schoolcraft > Sent: Tuesday, December 27, 2005 1:07 PM > To: Kickstart > Subject: Kickstart from eth1 (eth0 socket bad) > > Hello, > > I have a series of servers that all kick from eth0, we have one > server that has a busted socket in the back and we use eth1 for > use, not a big deal but when I went to kickstart this box from > eth1 the install stopped at the ipaddr setup screen. > > Nowhere in the ksfile of any of the other boxes is eth0 > specified in the "network" stanza. > > (question) > Can I implicitly state eth1 somewhere in the ksfile for this one > server? > > Thanks > > -- Bill Schoolcraft PO Box 210076 San Francisco, CA 94121 "UNIX, A Way of Life." http://billschoolcraft.com _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From bill at wiliweld.com Tue Dec 27 19:31:25 2005 From: bill at wiliweld.com (Bill Schoolcraft) Date: Tue, 27 Dec 2005 11:31:25 -0800 (PST) Subject: Kickstart from eth1 (eth0 socket bad) In-Reply-To: <2213CAA79DA9094AB52517FDADEF07FA021E50F3@tshuspaphimbx01.ERF.THOMSON.COM> References: <2213CAA79DA9094AB52517FDADEF07FA021E50F3@tshuspaphimbx01.ERF.THOMSON.COM> Message-ID: <20051227112747.S31984@bsd.billschoolcraft.com> At Tue, 27 Dec 2005 it looks like Elizabeth.Brosch at thomson.com composed: > Yes just add it somewhere as part of the network definition. > Liz, I just went and tried it and the "Red Hat" install gui (standard one shown during "ip/hostname" query, still had "eth0" in the titlebar of what it was seaching for. I was then defautled to the ipaddr setup screen again. Is there something hardcoded into the kickstart process where it defaults to eth0 despite what is specified in the ksfile? It made no attempts to try to reach eth1, just gave me the setup screen. Thanks > -----Original Message----- > From: kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Bill Schoolcraft > Sent: Tuesday, December 27, 2005 1:49 PM > To: Discussion list about Kickstart > Subject: RE: Kickstart from eth1 (eth0 socket bad) > > At Tue, 27 Dec 2005 it looks like Elizabeth.Brosch at thomson.com composed: > >> I use this in my ks.cfg file: >> >> network --bootproto dhcp --device eth0 >> >> This works very well for me. > > Thanks Liz, > > The files generated by our setup takes a hostname arg, OS > version arg, partition scheme arg (etc) then checks > /etc/hosts and /etc/ethers and then , among other lines in the > host-specific ksfile, has this stanza: > > I think I can enter it after the 'bootproto' option, thanks. > > ##################################################### > > # Network Settings > network --bootproto static --ip 192.168.1.11 netmask > 255.255.255.0 --gateway 192.168.1.254 --nameserver 192.168.1.253 > auth --useshadow --enablemd5 firewall --disabled > > ##################################################### > > >> >> Liz Brosch >> TSH System Services >> Philadelphia, PA 19104 >> office: (215) 386-0100 x1144 >> cell: (267) 784-9166 >> >> -----Original Message----- >> From: kickstart-list-bounces at redhat.com >> [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Bill > Schoolcraft >> Sent: Tuesday, December 27, 2005 1:07 PM >> To: Kickstart >> Subject: Kickstart from eth1 (eth0 socket bad) >> >> Hello, >> >> I have a series of servers that all kick from eth0, we have one >> server that has a busted socket in the back and we use eth1 for >> use, not a big deal but when I went to kickstart this box from >> eth1 the install stopped at the ipaddr setup screen. >> >> Nowhere in the ksfile of any of the other boxes is eth0 >> specified in the "network" stanza. >> >> (question) >> Can I implicitly state eth1 somewhere in the ksfile for this one >> server? >> >> Thanks >> >> > > -- Bill Schoolcraft PO Box 210076 San Francisco, CA 94121 "UNIX, A Way of Life." http://billschoolcraft.com From Elizabeth.Brosch at thomson.com Tue Dec 27 19:41:00 2005 From: Elizabeth.Brosch at thomson.com (Elizabeth.Brosch at thomson.com) Date: Tue, 27 Dec 2005 14:41:00 -0500 Subject: Kickstart from eth1 (eth0 socket bad) Message-ID: <2213CAA79DA9094AB52517FDADEF07FA021E50F4@tshuspaphimbx01.ERF.THOMSON.COM> I was having a problem initially with Kickstart. I had eth0 hardcoded in my ks.cfg file and also in the pxelinux config file. But every time I attempted my install I would get stopped at the screen for the host info. I had multiple NICs and this caused some confusion. What I considered eth0 was actually eth3. PXE worked just find but Anaconda seen my eth0 as eth3! I found this out by booting from CD and using linux rescue. You can search the archives for my email thread regarding this: October 2005 archives Subject is: Kicstart fails Thanks. Liz Brosch TSH System Services Philadelphia, PA 19104 office: (215) 386-0100 x1144 cell: (267) 784-9166 From MWeiner at ag.com Tue Dec 27 19:42:26 2005 From: MWeiner at ag.com (MW Mike Weiner (5028)) Date: Tue, 27 Dec 2005 14:42:26 -0500 Subject: Kickstart across subnets Message-ID: <4FD2C985D5E2A642AE25823DFD61C2B0097236D5@orca.agcom.amgreetings.com> Dear list readers - I am having some major headaches trying to get this to work in my production environment. My "design" is kickstart off a cd, to do an install across NFS. The network is basically comprised of 4 vlans (192.168.30.0/24,192.168.31.0/24,192.168.32.0/24,192.168.34.0/24) via nfs - and i realize you can dhcp across subnets without some other 'magic' so, i have a dhcrelay running on each of the subnets. Now, for some reason, the 3 subnets that are not native to that of the yum server arent working. The subnet the kickstart box is on works, or at least 'seems' to work and the install goes fairly smoothly. I am unfortunately 1000miles away from these servers so its hard to tell whats actually going on, but i created a bootable kickstart cd for the technicians working in our datacenter, and whenever they use it on those 3 subnets that fail - i see NOTHING in any logs on any of the dhcrelay servers nor dhcpd server, as if the request is not even getting relayed. And i really cant tell what is actually going on here when it fails. I created a ks boot cd, with all the required files (i.e. vmlinuz, initrd.img, isolinux.bin, etc) and the key line in my isolinux.cfg is as follows: label linux kernel vmlinuz append ksdevice=eth1 console=tty0 load_ramdisk=1 prompt_ramdisk=0 initrd=initrd.img network ip=dhcp ks=nfs:192.168.34.248:/tftpboot/ks-fc2.cfg selinux=0 which seems to work, anyone see an issue with this? Once, booted, it SHOULD go out and try to grab in IP address from either a dhcrelay or the main dhcpd server (depending on the subnet). Once it has an IP address, it is supposed to continue on and mount the nfs share containing the installation tree (/repo/fedora/linux/core/2/i386/os) and complete in roughly 10 minutes. Below is my ks.cfg file (also called ks-fc2.cfg on the tftpboot server). I am sure others have done cross subnet kickstarts, though poorly documented, i was wondering if someone may have sometime to help me debug this. I have to complete these 400+ kickstarts by the end of January 2006. I appreciate ANY pointers and assistance you may be able to provide. Michael Weiner # Kickstart file for Automated Fedora Core 2 installs via NFS install nfs --server=192.168.34.248 --dir=/repo/fedora/linux/core/2/i386/os lang en_US.UTF-8 langsupport --default en_US.UTF-8 en_US.UTF-8 keyboard us skipx network --device eth1 --bootproto dhcp rootpw --iscrypted $1$.pio2FpF$qdZBN7oU0oOG6R3oV6y5X. firewall --disabled selinux --disabled authconfig --enableshadow --enablemd5 timezone America/New_York bootloader --location=mbr --append rhgb quiet # The following is the partition information clearpart --linux part /boot --fstype ext3 --size=101 part /disk2 --fstype ext3 --size=12289 part / --fstype ext3 --size=12289 part swap --size=2048 part /weblog --fstype ext3 --size=1 --grow %packages --resolvedeps @ web-server @ mail-server @ dns-server @ dialup @ network-server @ sql-server @ editors @ admin-tools @ system-tools @ news-server @ smb-server @ authoring-and-publishing @ sound-and-video @ server-cfg @ graphics @ ftp-server @ development-tools @ engineering-and-scientific @ text-internet kernel-smp grub e2fsprogs -cups -isdn -pcmcia %post # run rpm error report rpm -Va --nofiles --nomd5 >> /tmp/rpm-problems.txt # Update thyself yum -y update # The Last Thing happening in Kickstart: # get a file via TFTP which indicates we are done. echo "get Kickstart_end" | /usr/bin/tftp 192.168.96.11 From bill at wiliweld.com Tue Dec 27 20:00:54 2005 From: bill at wiliweld.com (Bill Schoolcraft) Date: Tue, 27 Dec 2005 12:00:54 -0800 (PST) Subject: Kickstart from eth1 (eth0 socket bad) In-Reply-To: <2213CAA79DA9094AB52517FDADEF07FA021E50F4@tshuspaphimbx01.ERF.THOMSON.COM> References: <2213CAA79DA9094AB52517FDADEF07FA021E50F4@tshuspaphimbx01.ERF.THOMSON.COM> Message-ID: <20051227115814.I31984@bsd.billschoolcraft.com> At Tue, 27 Dec 2005 it looks like Elizabeth.Brosch at thomson.com composed: > I was having a problem initially with Kickstart. I had eth0 hardcoded > in my ks.cfg file and also in the pxelinux config file. But every time > I attempted my install I would get stopped at the screen for the host > info. > > I had multiple NICs and this caused some confusion. What I considered > eth0 was actually eth3. PXE worked just find but Anaconda seen my eth0 > as eth3! I found this out by booting from CD and using linux rescue. > You can search the archives for my email thread regarding this: > > October 2005 archives > Subject is: Kicstart fails > Ahh, I had not thought of the "prelinux" config file. I was not the admin who came in to setup this multi os, multi partition profile, multi package profile system. I will dig for that file and see if I can do a "one-off" install with the orig file saved and edits done to the "prelinux" conf file. Thank You Liz. > Thanks. > > Liz Brosch > TSH System Services > Philadelphia, PA 19104 > office: (215) 386-0100 x1144 > cell: (267) 784-9166 > > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > -- Bill Schoolcraft PO Box 210076 San Francisco, CA 94121 "UNIX, A Way of Life." http://billschoolcraft.com From sjbrowne at bluebottle.com Tue Dec 27 21:26:21 2005 From: sjbrowne at bluebottle.com (Stuart J. Browne) Date: Wed, 28 Dec 2005 08:26:21 +1100 Subject: Kickstart across subnets References: <4FD2C985D5E2A642AE25823DFD61C2B0097236D5@orca.agcom.amgreetings.com> Message-ID: <00bc01c60b2c$3040af90$271aa8c0@bluebottle.com> Dumb questions, just for clarification. The machines running 'dhcprelay' in each of the subnets can see the other subnets without any issue? I'm assuming you've got shell access to those machines, have you run any packet sniffers across the interface when they're attemping a rebuild, looking for dhcp requests? ----- Original Message ----- From: "MW Mike Weiner (5028)" To: "Discussion list about Kickstart" Cc: "MW Mike Weiner (5028)" Sent: Wednesday, December 28, 2005 6:42 AM Subject: Kickstart across subnets > Dear list readers - > > I am having some major headaches trying to get this to work in my > production environment. My "design" is kickstart off a cd, to do an > install across NFS. The network is basically comprised of 4 vlans > (192.168.30.0/24,192.168.31.0/24,192.168.32.0/24,192.168.34.0/24) via > nfs - and i realize you can dhcp across subnets without some other > 'magic' so, i have a dhcrelay running on each of the subnets. Now, for > some reason, the 3 subnets that are not native to that of the yum server > arent working. The subnet the kickstart box is on works, or at least > 'seems' to work and the install goes fairly smoothly. I am unfortunately > 1000miles away from these servers so its hard to tell whats actually > going on, but i created a bootable kickstart cd for the technicians > working in our datacenter, and whenever they use it on those 3 subnets > that fail - i see NOTHING in any logs on any of the dhcrelay servers nor > dhcpd server, as if the request is not even getting relayed. And i > really cant tell what is actually going on here when it fails. > > I created a ks boot cd, with all the required files (i.e. vmlinuz, > initrd.img, isolinux.bin, etc) and the key line in my isolinux.cfg is as > follows: > > label linux > kernel vmlinuz > append ksdevice=eth1 console=tty0 load_ramdisk=1 prompt_ramdisk=0 > initrd=initrd.img network ip=dhcp > ks=nfs:192.168.34.248:/tftpboot/ks-fc2.cfg selinux=0 > > which seems to work, anyone see an issue with this? Once, booted, it > SHOULD go out and try to grab in IP address from either a dhcrelay or > the main dhcpd server (depending on the subnet). Once it has an IP > address, it is supposed to continue on and mount the nfs share > containing the installation tree (/repo/fedora/linux/core/2/i386/os) and > complete in roughly 10 minutes. Below is my ks.cfg file (also called > ks-fc2.cfg on the tftpboot server). > > I am sure others have done cross subnet kickstarts, though poorly > documented, i was wondering if someone may have sometime to help me > debug this. I have to complete these 400+ kickstarts by the end of > January 2006. > > I appreciate ANY pointers and assistance you may be able to provide. > > Michael Weiner > > # Kickstart file for Automated Fedora Core 2 installs via NFS > install > nfs --server=192.168.34.248 --dir=/repo/fedora/linux/core/2/i386/os > lang en_US.UTF-8 > langsupport --default en_US.UTF-8 en_US.UTF-8 > keyboard us > skipx > network --device eth1 --bootproto dhcp > rootpw --iscrypted $1$.pio2FpF$qdZBN7oU0oOG6R3oV6y5X. > firewall --disabled > selinux --disabled > authconfig --enableshadow --enablemd5 > timezone America/New_York > bootloader --location=mbr --append rhgb quiet > > # The following is the partition information > clearpart --linux > part /boot --fstype ext3 --size=101 > part /disk2 --fstype ext3 --size=12289 > part / --fstype ext3 --size=12289 > part swap --size=2048 > part /weblog --fstype ext3 --size=1 --grow > > %packages --resolvedeps > @ web-server > @ mail-server > @ dns-server > @ dialup > @ network-server > @ sql-server > @ editors > @ admin-tools > @ system-tools > @ news-server > @ smb-server > @ authoring-and-publishing > @ sound-and-video > @ server-cfg > @ graphics > @ ftp-server > @ development-tools > @ engineering-and-scientific > @ text-internet > kernel-smp > grub > e2fsprogs > -cups > -isdn > -pcmcia > > %post > # run rpm error report > rpm -Va --nofiles --nomd5 >> /tmp/rpm-problems.txt > > # Update thyself > yum -y update > > # The Last Thing happening in Kickstart: > # get a file via TFTP which indicates we are done. > echo "get Kickstart_end" | /usr/bin/tftp 192.168.96.11 > > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From MWeiner at ag.com Tue Dec 27 22:32:24 2005 From: MWeiner at ag.com (MW Mike Weiner (5028)) Date: Tue, 27 Dec 2005 17:32:24 -0500 Subject: Kickstart across subnets Message-ID: <4FD2C985D5E2A642AE25823DFD61C2B009723730@orca.agcom.amgreetings.com> NOT a dumb question at all, yes each of the dhcrelay servers can 'see' the main dhcp server, in fact when testing this setup, I would see the DHCPREQUEST on the dhcp server coming from the dhcrelay server - when it worked that is. I tested this by changing one of the interfaces on a server on the same network as the dhcrelay to bootproto=dhcp and then ifup'd the device, and after a few moments the interface would be up with the correct IP, aqnd logs on the dhc server indicated the communication that had occurred (DHCPDISCOVERY - DHCPREQUEST - DHCPACK) I need to do some tcpdumps, like you said, this might help a bit. Thanks for the response. This e-mail and any attachments may contain confidential information that is legally privileged. The information is solely for the use of the intended recipient(s); any disclosure, copying, distribution, or other use of this information is strictly prohibited. If you have received this e-mail in error, please notify the sender by return e-mail and delete this message. Thank you. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Stuart J. Browne Sent: Tuesday, December 27, 2005 4:26 PM To: Discussion list about Kickstart Subject: Re: Kickstart across subnets Dumb questions, just for clarification. The machines running 'dhcprelay' in each of the subnets can see the other subnets without any issue? I'm assuming you've got shell access to those machines, have you run any packet sniffers across the interface when they're attemping a rebuild, looking for dhcp requests? ----- Original Message ----- From: "MW Mike Weiner (5028)" To: "Discussion list about Kickstart" Cc: "MW Mike Weiner (5028)" Sent: Wednesday, December 28, 2005 6:42 AM Subject: Kickstart across subnets > Dear list readers - > > I am having some major headaches trying to get this to work in my > production environment. My "design" is kickstart off a cd, to do an > install across NFS. The network is basically comprised of 4 vlans > (192.168.30.0/24,192.168.31.0/24,192.168.32.0/24,192.168.34.0/24) via > nfs - and i realize you can dhcp across subnets without some other > 'magic' so, i have a dhcrelay running on each of the subnets. Now, for > some reason, the 3 subnets that are not native to that of the yum server > arent working. The subnet the kickstart box is on works, or at least > 'seems' to work and the install goes fairly smoothly. I am unfortunately > 1000miles away from these servers so its hard to tell whats actually > going on, but i created a bootable kickstart cd for the technicians > working in our datacenter, and whenever they use it on those 3 subnets > that fail - i see NOTHING in any logs on any of the dhcrelay servers nor > dhcpd server, as if the request is not even getting relayed. And i > really cant tell what is actually going on here when it fails. > > I created a ks boot cd, with all the required files (i.e. vmlinuz, > initrd.img, isolinux.bin, etc) and the key line in my isolinux.cfg is as > follows: > > label linux > kernel vmlinuz > append ksdevice=eth1 console=tty0 load_ramdisk=1 prompt_ramdisk=0 > initrd=initrd.img network ip=dhcp > ks=nfs:192.168.34.248:/tftpboot/ks-fc2.cfg selinux=0 > > which seems to work, anyone see an issue with this? Once, booted, it > SHOULD go out and try to grab in IP address from either a dhcrelay or > the main dhcpd server (depending on the subnet). Once it has an IP > address, it is supposed to continue on and mount the nfs share > containing the installation tree (/repo/fedora/linux/core/2/i386/os) and > complete in roughly 10 minutes. Below is my ks.cfg file (also called > ks-fc2.cfg on the tftpboot server). > > I am sure others have done cross subnet kickstarts, though poorly > documented, i was wondering if someone may have sometime to help me > debug this. I have to complete these 400+ kickstarts by the end of > January 2006. > > I appreciate ANY pointers and assistance you may be able to provide. > > Michael Weiner > > # Kickstart file for Automated Fedora Core 2 installs via NFS > install > nfs --server=192.168.34.248 --dir=/repo/fedora/linux/core/2/i386/os > lang en_US.UTF-8 > langsupport --default en_US.UTF-8 en_US.UTF-8 > keyboard us > skipx > network --device eth1 --bootproto dhcp > rootpw --iscrypted $1$.pio2FpF$qdZBN7oU0oOG6R3oV6y5X. > firewall --disabled > selinux --disabled > authconfig --enableshadow --enablemd5 > timezone America/New_York > bootloader --location=mbr --append rhgb quiet > > # The following is the partition information > clearpart --linux > part /boot --fstype ext3 --size=101 > part /disk2 --fstype ext3 --size=12289 > part / --fstype ext3 --size=12289 > part swap --size=2048 > part /weblog --fstype ext3 --size=1 --grow > > %packages --resolvedeps > @ web-server > @ mail-server > @ dns-server > @ dialup > @ network-server > @ sql-server > @ editors > @ admin-tools > @ system-tools > @ news-server > @ smb-server > @ authoring-and-publishing > @ sound-and-video > @ server-cfg > @ graphics > @ ftp-server > @ development-tools > @ engineering-and-scientific > @ text-internet > kernel-smp > grub > e2fsprogs > -cups > -isdn > -pcmcia > > %post > # run rpm error report > rpm -Va --nofiles --nomd5 >> /tmp/rpm-problems.txt > > # Update thyself > yum -y update > > # The Last Thing happening in Kickstart: > # get a file via TFTP which indicates we are done. > echo "get Kickstart_end" | /usr/bin/tftp 192.168.96.11 > > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From teng at dataway.com Tue Dec 27 23:14:20 2005 From: teng at dataway.com (Tedman Eng) Date: Tue, 27 Dec 2005 15:14:20 -0800 Subject: Kickstart from eth1 (eth0 socket bad) Message-ID: <37ED92F9890FAF4BB947613C66FF8B1A08BB2C99@dw-mail.dataway.com> There are 2 places to you need to specify eth0/1: In the kernel line: linux initrd=initrd.img ks=http://some.server/ks.cfg ip=1.2.3.4 netmask=255.255.255.0 gateway=1.2.3.4 dns=1.2.3.4 ksdevice=eth0 AND in the ks.cfg file: network --bootproto=static --ip=1.2.3.4 --netmask=255.255.255.0 --gateway=1.2.3.4 --nameserver=1.2.3.4 --device=eth0 If you're booting from CDROM or Floppy, edit isolinux.cfg or syslinux.cfg. If you're booting from network, you'll need to edit pxelinix.cfg I believe your confusion is because you're editing the ks.cfg file only. But without using the proper interface, your computer cannot even reach the ks.cfg file in the first place. > -----Original Message----- > From: Bill Schoolcraft [mailto:bill at wiliweld.com] > Sent: Tuesday, December 27, 2005 12:01 PM > To: Discussion list about Kickstart > Subject: RE: Kickstart from eth1 (eth0 socket bad) > > > At Tue, 27 Dec 2005 it looks like > Elizabeth.Brosch at thomson.com composed: > > > I was having a problem initially with Kickstart. I had > eth0 hardcoded > > in my ks.cfg file and also in the pxelinux config file. > But every time > > I attempted my install I would get stopped at the screen > for the host > > info. > > > > I had multiple NICs and this caused some confusion. What I > considered > > eth0 was actually eth3. PXE worked just find but Anaconda > seen my eth0 > > as eth3! I found this out by booting from CD and using > linux rescue. > > You can search the archives for my email thread regarding this: > > > > October 2005 archives > > Subject is: Kicstart fails > > > > Ahh, I had not thought of the "prelinux" config file. > > I was not the admin who came in to setup this multi os, multi > partition profile, multi package profile system. I will dig for > that file and see if I can do a "one-off" install with the orig > file saved and edits done to the "prelinux" conf file. > > Thank You Liz. > > > > Thanks. > > > > Liz Brosch > > TSH System Services > > Philadelphia, PA 19104 > > office: (215) 386-0100 x1144 > > cell: (267) 784-9166 > > > > > > > > _______________________________________________ > > Kickstart-list mailing list > > Kickstart-list at redhat.com > > https://www.redhat.com/mailman/listinfo/kickstart-list > > > > -- > Bill Schoolcraft > PO Box 210076 > San Francisco, CA 94121 > "UNIX, A Way of Life." > http://billschoolcraft.com > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list > From phr at doc.ic.ac.uk Wed Dec 28 03:55:09 2005 From: phr at doc.ic.ac.uk (Philip Rowlands) Date: Wed, 28 Dec 2005 03:55:09 +0000 (GMT) Subject: Kickstart across subnets In-Reply-To: <4FD2C985D5E2A642AE25823DFD61C2B0097236D5@orca.agcom.amgreetings.com> References: <4FD2C985D5E2A642AE25823DFD61C2B0097236D5@orca.agcom.amgreetings.com> Message-ID: On Tue, 27 Dec 2005, MW Mike Weiner (5028) wrote: > I created a ks boot cd, with all the required files (i.e. vmlinuz, > initrd.img, isolinux.bin, etc) and the key line in my isolinux.cfg is as > follows: > > label linux > kernel vmlinuz > append ksdevice=eth1 console=tty0 load_ramdisk=1 prompt_ramdisk=0 > initrd=initrd.img network ip=dhcp > ks=nfs:192.168.34.248:/tftpboot/ks-fc2.cfg selinux=0 Why the console, load_ramdisk, prompt_ramdisk, and network arguments? I'm not sure these are necessary, and they potentially cloud the problem. Is it possible that the network switches for the problematic machines are taking some time to bring up the port? This can happen with high(er) end hardware and the Spanning Tree Protocol. anaconda's undocumented "linksleep" setting can help here. > I have to complete these 400+ kickstarts by the end of January 2006. Good luck! Cheers, Phil From MWeiner at ag.com Wed Dec 28 13:34:16 2005 From: MWeiner at ag.com (MW Mike Weiner (5028)) Date: Wed, 28 Dec 2005 08:34:16 -0500 Subject: Kickstart across subnets Message-ID: <4FD2C985D5E2A642AE25823DFD61C2B009723759@orca.agcom.amgreetings.com> Philip - Good question, no real reason other than this was an old ks.cfg modified for this project - I will clean it up, though I doubt that has mich impact. Thanks!! Michael This e-mail and any attachments may contain confidential information that is legally privileged. The information is solely for the use of the intended recipient(s); any disclosure, copying, distribution, or other use of this information is strictly prohibited. If you have received this e-mail in error, please notify the sender by return e-mail and delete this message. Thank you. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Philip Rowlands Sent: Tuesday, December 27, 2005 10:55 PM To: Discussion list about Kickstart Subject: Re: Kickstart across subnets On Tue, 27 Dec 2005, MW Mike Weiner (5028) wrote: > I created a ks boot cd, with all the required files (i.e. vmlinuz, > initrd.img, isolinux.bin, etc) and the key line in my isolinux.cfg is > as > follows: > > label linux > kernel vmlinuz > append ksdevice=eth1 console=tty0 load_ramdisk=1 prompt_ramdisk=0 > initrd=initrd.img network ip=dhcp > ks=nfs:192.168.34.248:/tftpboot/ks-fc2.cfg selinux=0 Why the console, load_ramdisk, prompt_ramdisk, and network arguments? I'm not sure these are necessary, and they potentially cloud the problem. Is it possible that the network switches for the problematic machines are taking some time to bring up the port? This can happen with high(er) end hardware and the Spanning Tree Protocol. anaconda's undocumented "linksleep" setting can help here. > I have to complete these 400+ kickstarts by the end of January 2006. Good luck! Cheers, Phil _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From LLavender at kmg.com Wed Dec 28 14:12:14 2005 From: LLavender at kmg.com (Lavender, Leo) Date: Wed, 28 Dec 2005 08:12:14 -0600 Subject: Kickstart across subnets Message-ID: <044BBA9ACDF3D948BC41F0DD8DFADF314F2546@houexg3.kmg.com> Michael, So dhcp ain't happening at all? It's gotta supply a gateway so you can see the kickstart on a different subnet, right? Leo -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of MW Mike Weiner (5028) Sent: Wednesday, December 28, 2005 7:34 AM To: Discussion list about Kickstart Subject: RE: Kickstart across subnets Philip - Good question, no real reason other than this was an old ks.cfg modified for this project - I will clean it up, though I doubt that has mich impact. Thanks!! Michael This e-mail and any attachments may contain confidential information that is legally privileged. The information is solely for the use of the intended recipient(s); any disclosure, copying, distribution, or other use of this information is strictly prohibited. If you have received this e-mail in error, please notify the sender by return e-mail and delete this message. Thank you. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Philip Rowlands Sent: Tuesday, December 27, 2005 10:55 PM To: Discussion list about Kickstart Subject: Re: Kickstart across subnets On Tue, 27 Dec 2005, MW Mike Weiner (5028) wrote: > I created a ks boot cd, with all the required files (i.e. vmlinuz, > initrd.img, isolinux.bin, etc) and the key line in my isolinux.cfg is > as > follows: > > label linux > kernel vmlinuz > append ksdevice=eth1 console=tty0 load_ramdisk=1 prompt_ramdisk=0 > initrd=initrd.img network ip=dhcp > ks=nfs:192.168.34.248:/tftpboot/ks-fc2.cfg selinux=0 Why the console, load_ramdisk, prompt_ramdisk, and network arguments? I'm not sure these are necessary, and they potentially cloud the problem. Is it possible that the network switches for the problematic machines are taking some time to bring up the port? This can happen with high(er) end hardware and the Spanning Tree Protocol. anaconda's undocumented "linksleep" setting can help here. > I have to complete these 400+ kickstarts by the end of January 2006. Good luck! Cheers, Phil _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From MWeiner at ag.com Wed Dec 28 17:14:42 2005 From: MWeiner at ag.com (MW Mike Weiner (5028)) Date: Wed, 28 Dec 2005 12:14:42 -0500 Subject: Kickstart across subnets Message-ID: <4FD2C985D5E2A642AE25823DFD61C2B0097237BA@orca.agcom.amgreetings.com> Leo - That's correct, I have that specified in the dhcpd.conf file as follows # dhcpd.conf # ddns-update-style ad-hoc; not authoritative; subnet 192.168.30.0 netmask 255.255.254.0 { option routers 192.168.30.1; } subnet 192.168.32.0 netmask 255.255.255.0 { option routers 192.168.32.1; } subnet 192.168.34.0 netmask 255.255.254.0 { option routers 192.168.35.254; } group { deny unknown-clients; allow booting; allow bootp; option ip-forwarding false; option mask-supplier false; option root-path "/tftpboot"; default-lease-time 259200; max-lease-time 259200; option domain-name-servers 192.168.30.25, 192.168.30.26; option domain-name "americangreetings.com"; option vendor-class-identifier "PXEClient"; option vendor-encapsulated-options 09:0f:80:00:0c:4e:65:74:77:6f:72:6b:20:62:6f:6f:74:0a:07:00:50 :72:6f:6d:70:74:06:01:02:08:03:80:00:00:47:04:80:00:00:00:ff; filename "pxelinux.0"; # filename "ks-fc2.cfg"; next-server 192.168.34.248; get-lease-hostnames true; host gate-ag250 { hardware ethernet 00:04:75:CB:63:1F; option host-name "ag250"; fixed-address 192.168.30.58; filename "pxelinux.0"; next-server 192.168.34.248; } ... Thank you for your response Michael Weiner This e-mail and any attachments may contain confidential information that is legally privileged. The information is solely for the use of the intended recipient(s); any disclosure, copying, distribution, or other use of this information is strictly prohibited. If you have received this e-mail in error, please notify the sender by return e-mail and delete this message. Thank you. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Lavender, Leo Sent: Wednesday, December 28, 2005 9:12 AM To: Discussion list about Kickstart Subject: RE: Kickstart across subnets Michael, So dhcp ain't happening at all? It's gotta supply a gateway so you can see the kickstart on a different subnet, right? Leo -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of MW Mike Weiner (5028) Sent: Wednesday, December 28, 2005 7:34 AM To: Discussion list about Kickstart Subject: RE: Kickstart across subnets Philip - Good question, no real reason other than this was an old ks.cfg modified for this project - I will clean it up, though I doubt that has mich impact. Thanks!! Michael This e-mail and any attachments may contain confidential information that is legally privileged. The information is solely for the use of the intended recipient(s); any disclosure, copying, distribution, or other use of this information is strictly prohibited. If you have received this e-mail in error, please notify the sender by return e-mail and delete this message. Thank you. -----Original Message----- From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Philip Rowlands Sent: Tuesday, December 27, 2005 10:55 PM To: Discussion list about Kickstart Subject: Re: Kickstart across subnets On Tue, 27 Dec 2005, MW Mike Weiner (5028) wrote: > I created a ks boot cd, with all the required files (i.e. vmlinuz, > initrd.img, isolinux.bin, etc) and the key line in my isolinux.cfg is > as > follows: > > label linux > kernel vmlinuz > append ksdevice=eth1 console=tty0 load_ramdisk=1 prompt_ramdisk=0 > initrd=initrd.img network ip=dhcp > ks=nfs:192.168.34.248:/tftpboot/ks-fc2.cfg selinux=0 Why the console, load_ramdisk, prompt_ramdisk, and network arguments? I'm not sure these are necessary, and they potentially cloud the problem. Is it possible that the network switches for the problematic machines are taking some time to bring up the port? This can happen with high(er) end hardware and the Spanning Tree Protocol. anaconda's undocumented "linksleep" setting can help here. > I have to complete these 400+ kickstarts by the end of January 2006. Good luck! Cheers, Phil _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list