Re: [opensuse] remote and diskless install - mkinitrd
Paul Neuwirth wrote:
right, my problem therewith is, that after installation of packages yast reboots the system (i am not sitting all the time im front...). How to avoid this, so that i can run mkinitrd manually.
better of cause would be, if yast would call mkinitrd correctly.
I think it's actually mkinitrd that is the problem - it should work out that the root filesystem is on nfs, and therefore include network and nfs support. I guess your initrds are stored on your tftp-server, so whatever is built locally isn't important? /Per -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
initrds are symlinks to the filesystem which is exportet with nfs: /srv/tftp/tftpboot/delta_initrd: symbolic link to `/exports/delta/boot/initrd' hmm, maybe this matters in /var/log/YaST2/y2logmkinitrd: /lib/mkinitrd/setup/11-storage.sh: line 283: /etc/fstab: No such file or directory there is actually no /etc/fstab and mkinitrd thinks root was on a tmpfs?: **************************** * WARNING * No fsck for your rootfs * could be found. * This might be bad! * Please install: fsck.tmpfs **************************** On Feb 26 2014 11:10, Per Jessen wrote:
Date: Wed, 26 Feb 2014 11:10:54 +0100 From: Per Jessen <per@computer.org> To: Paul Neuwirth <mail@paul-neuwirth.de> Cc: opensuse@opensuse.org Subject: Re: [opensuse] remote and diskless install - mkinitrd
Paul Neuwirth wrote:
right, my problem therewith is, that after installation of packages yast reboots the system (i am not sitting all the time im front...). How to avoid this, so that i can run mkinitrd manually.
better of cause would be, if yast would call mkinitrd correctly.
I think it's actually mkinitrd that is the problem - it should work out that the root filesystem is on nfs, and therefore include network and nfs support.
I guess your initrds are stored on your tftp-server, so whatever is built locally isn't important?
/Per -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Paul Neuwirth wrote:
initrds are symlinks to the filesystem which is exportet with nfs: /srv/tftp/tftpboot/delta_initrd: symbolic link to `/exports/delta/boot/initrd'
Aha, thanks. Well, the usual way to influence how the initrd is built is to edit "/etc/sysconfig/kernel" and set the appropriate parameters. This won't work for you, afaict. I think there is a way to make the installation pause before the reboot, but I don't remember how. Check the opensuse linuxrc arguments.
hmm, maybe this matters in /var/log/YaST2/y2logmkinitrd: /lib/mkinitrd/setup/11-storage.sh: line 283: /etc/fstab: No such file or directory
there is actually no /etc/fstab
That's odd, YaST should be creating one, even if it is superfluous.
and mkinitrd thinks root was on a tmpfs?: **************************** * WARNING * No fsck for your rootfs * could be found. * This might be bad! * Please install: fsck.tmpfs ****************************
I seem to remember something similar with my systems that have nfsroot. I only have 3-4, and I had to stop the installation process to rerun the mkinitrd manually. /Per -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Thanks, that was a very helpfull hint. Will try WaitReboot=0 . But don't know if there's no timeout or no reboot. But I found several other usefull parameters, as linemode, loghost, startshell (this maybe will also prevent reboot) - the corresponding dialog is "Start shell before and after YaST?" will try this evening. On Feb 26 2014 11:28, Per Jessen wrote:
Date: Wed, 26 Feb 2014 11:28:46 +0100 From: Per Jessen <per@computer.org> To: Paul Neuwirth <pane@paneologe.ch> Cc: Paul Neuwirth <mail@paul-neuwirth.de>, opensuse@opensuse.org Subject: Re: [opensuse] remote and diskless install - mkinitrd
Paul Neuwirth wrote:
initrds are symlinks to the filesystem which is exportet with nfs: /srv/tftp/tftpboot/delta_initrd: symbolic link to `/exports/delta/boot/initrd'
Aha, thanks.
Well, the usual way to influence how the initrd is built is to edit "/etc/sysconfig/kernel" and set the appropriate parameters. This won't work for you, afaict. I think there is a way to make the installation pause before the reboot, but I don't remember how. Check the opensuse linuxrc arguments.
hmm, maybe this matters in /var/log/YaST2/y2logmkinitrd: /lib/mkinitrd/setup/11-storage.sh: line 283: /etc/fstab: No such file or directory
there is actually no /etc/fstab
That's odd, YaST should be creating one, even if it is superfluous.
and mkinitrd thinks root was on a tmpfs?: **************************** * WARNING * No fsck for your rootfs * could be found. * This might be bad! * Please install: fsck.tmpfs ****************************
I seem to remember something similar with my systems that have nfsroot. I only have 3-4, and I had to stop the installation process to rerun the mkinitrd manually.
/Per -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
created a working initrd. now i am stuck again: booting into the initrd: err, enp9s0: failed to llokup hostname via DNS: Temporary failure in name resolution at the same time i get the following on the server: IPv4: martian destination 127.0.0.1 from 172.18.0.23, dev eth0 ? in dhcpcd.conf: option domain-name-servers 172.18.0.1; NFS is not being mounted... On Feb 26 2014 12:47, Paul Neuwirth wrote:
Date: Wed, 26 Feb 2014 12:47:04 +0100 (CET) From: Paul Neuwirth <pane@paneologe.ch> To: Per Jessen <per@computer.org> Cc: opensuse@opensuse.org Subject: Re: [opensuse] remote and diskless install - mkinitrd
Thanks, that was a very helpfull hint. Will try WaitReboot=0 . But don't know if there's no timeout or no reboot. But I found several other usefull parameters, as linemode, loghost, startshell (this maybe will also prevent reboot) - the corresponding dialog is "Start shell before and after YaST?"
will try this evening.
On Feb 26 2014 11:28, Per Jessen wrote:
Date: Wed, 26 Feb 2014 11:28:46 +0100 From: Per Jessen <per@computer.org> To: Paul Neuwirth <pane@paneologe.ch> Cc: Paul Neuwirth <mail@paul-neuwirth.de>, opensuse@opensuse.org Subject: Re: [opensuse] remote and diskless install - mkinitrd
Paul Neuwirth wrote:
initrds are symlinks to the filesystem which is exportet with nfs: /srv/tftp/tftpboot/delta_initrd: symbolic link to `/exports/delta/boot/initrd'
Aha, thanks.
Well, the usual way to influence how the initrd is built is to edit "/etc/sysconfig/kernel" and set the appropriate parameters. This won't work for you, afaict. I think there is a way to make the installation pause before the reboot, but I don't remember how. Check the opensuse linuxrc arguments.
hmm, maybe this matters in /var/log/YaST2/y2logmkinitrd: /lib/mkinitrd/setup/11-storage.sh: line 283: /etc/fstab: No such file or directory
there is actually no /etc/fstab
That's odd, YaST should be creating one, even if it is superfluous.
and mkinitrd thinks root was on a tmpfs?: **************************** * WARNING * No fsck for your rootfs * could be found. * This might be bad! * Please install: fsck.tmpfs ****************************
I seem to remember something similar with my systems that have nfsroot. I only have 3-4, and I had to stop the installation process to rerun the mkinitrd manually.
/Per -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
В Wed, 26 Feb 2014 15:38:31 +0100 (CET) Paul Neuwirth <pane@paneologe.ch> пишет:
created a working initrd. now i am stuck again: booting into the initrd: err, enp9s0: failed to llokup hostname via DNS: Temporary failure in name resolution
at the same time i get the following on the server: IPv4: martian destination 127.0.0.1 from 172.18.0.23, dev eth0
?
You may be interested in http://www.unixresources.net/linux/lf/57/archive/00/00/05/66/56676.html -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Paul Neuwirth wrote:
created a working initrd. now i am stuck again: booting into the initrd: err, enp9s0: failed to llokup hostname via DNS: Temporary failure in name resolution
Either a network failure or a problem with /etc/resolv.conf.
at the same time i get the following on the server: IPv4: martian destination 127.0.0.1 from 172.18.0.23, dev eth0
I guess 172.18.0.23 is the address handed out by DHCP. If you boot a to-be-installed system with "usessh=1 sshpassword=paul", you can check your network config before you start YaST. /Per -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 2/26/2014 12:23 PM, Per Jessen wrote:
at the same time i get the following on the server: IPv4: martian destination 127.0.0.1 from 172.18.0.23, dev eth0
I guess 172.18.0.23 is the address handed out by DHCP.
But even that wouldn't account for a packet destined for 127.0.0.1 being sent/received via 172.18.0.23. Sounds like the lo interface failed to come up, or there is a hosts file entry listing localhost as something other than 127.0.0.1. -- _____________________________________ ---This space for rent--- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
John Andersen wrote:
On 2/26/2014 12:23 PM, Per Jessen wrote:
at the same time i get the following on the server: IPv4: martian destination 127.0.0.1 from 172.18.0.23, dev eth0
I guess 172.18.0.23 is the address handed out by DHCP.
But even that wouldn't account for a packet destined for 127.0.0.1 being sent/received via 172.18.0.23.
No, I have no idea where that came from. -- Per Jessen, Zürich (4.5°C) http://www.dns24.ch/ - free dynamic DNS, made in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
That's right. dhcpcd is called before interface lo is set up (this is going to be done at the end of 12-network.sh) and performs a dns lookup of the own host (23.0.18.172.in-addr.arpa) - but dhcpcd is launched with "dhcpcd -R -Y -N", so resolving will not succeed at this point. I don't think that this is the real problem. Actually think the initrd is kind of broken: the relevant part of run_all.sh: if [ "$interface" -o "$dhcp" -o "$ip" -o "$nfsaddrs" -o "$drvlink" ]; then modules="af_packet $bonding_module" [ "$debug" ] && echo running 12-network.sh source boot/12-network.sh [ "$modules" ] && load_modules fi [ "$debug" ] && echo preping 21-devinit_done.sh [ "$debug" ] && echo running 21-devinit_done.sh source boot/21-devinit_done.sh [ "$modules" ] && load_modules [ "$debug" ] && echo preping 21-nfs.sh [ "$rootfstype" ] || rootfstype='nfs' if [ "$rootfstype" = "nfs" -o "$need_nfs" ]; then [ "$debug" ] && echo load nfs modules modules="nfs nfsv2 nfsv3 nfsv4" [ "$debug" ] && echo running 21-nfs.sh source boot/21-nfs.sh [ "$modules" ] && load_modules fi [ "$debug" ] && echo preping 81-mount.sh [ "$rootdev" ] || rootdev='172.18.0.1:/exports/delta' if [ ! "$root_already_mounted" ]; then [ "$debug" ] && echo running 81-mount.sh source boot/81-mount.sh [ "$modules" ] && load_modules fi [ "$debug" ] && echo preping 82-remount.sh [ "$debug" ] && echo running 82-remount.sh source boot/82-remount.sh [ "$modules" ] && load_modules reading the output is difficult, kernel debug and script debug messages overlap... the last lines I see on my serial console: -the dns lookup error from dhcpcd "running 21-devNFS: Registering the id_resolver key type Key type id_resolver registered Key type id_legacy registered Give root password for maintenance (or type Control-D to continue):" after pressing CTRL-D it continues with 81-mount.sh will have to find out where exactly this happens.. continuing this weekend.. any possibility to get the output more verbose (which shell script steps being executed?) without putting echo's everywhere? thanks On Feb 26 2014 13:09, John Andersen wrote:
Date: Wed, 26 Feb 2014 13:09:32 -0800 From: John Andersen <jsamyth@gmail.com> To: opensuse@opensuse.org Subject: Re: [opensuse] remote and diskless install - mkinitrd
On 2/26/2014 12:23 PM, Per Jessen wrote:
at the same time i get the following on the server: IPv4: martian destination 127.0.0.1 from 172.18.0.23, dev eth0
I guess 172.18.0.23 is the address handed out by DHCP.
But even that wouldn't account for a packet destined for 127.0.0.1 being sent/received via 172.18.0.23.
Sounds like the lo interface failed to come up, or there is a hosts file entry listing localhost as something other than 127.0.0.1.
-- _____________________________________ ---This space for rent--- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Paul Neuwirth wrote:
That's right. dhcpcd is called before interface lo is set up (this is going to be done at the end of 12-network.sh) and performs a dns lookup of the own host (23.0.18.172.in-addr.arpa) - but dhcpcd is launched with "dhcpcd -R -Y -N", so resolving will not succeed at this point.
I don't think that this is the real problem. Actually think the initrd is kind of broken:
the relevant part of run_all.sh: if [ "$interface" -o "$dhcp" -o "$ip" -o "$nfsaddrs" -o "$drvlink" ]; then modules="af_packet $bonding_module" [ "$debug" ] && echo running 12-network.sh source boot/12-network.sh [ "$modules" ] && load_modules fi [ "$debug" ] && echo preping 21-devinit_done.sh [ "$debug" ] && echo running 21-devinit_done.sh source boot/21-devinit_done.sh [ "$modules" ] && load_modules [ "$debug" ] && echo preping 21-nfs.sh [ "$rootfstype" ] || rootfstype='nfs' if [ "$rootfstype" = "nfs" -o "$need_nfs" ]; then [ "$debug" ] && echo load nfs modules modules="nfs nfsv2 nfsv3 nfsv4" [ "$debug" ] && echo running 21-nfs.sh source boot/21-nfs.sh [ "$modules" ] && load_modules fi [ "$debug" ] && echo preping 81-mount.sh [ "$rootdev" ] || rootdev='172.18.0.1:/exports/delta' if [ ! "$root_already_mounted" ]; then [ "$debug" ] && echo running 81-mount.sh source boot/81-mount.sh [ "$modules" ] && load_modules fi [ "$debug" ] && echo preping 82-remount.sh
[ "$debug" ] && echo running 82-remount.sh source boot/82-remount.sh [ "$modules" ] && load_modules
reading the output is difficult, kernel debug and script debug messages overlap... the last lines I see on my serial console: -the dns lookup error from dhcpcd "running 21-devNFS: Registering the id_resolver key type Key type id_resolver registered Key type id_legacy registered Give root password for maintenance (or type Control-D to continue):"
after pressing CTRL-D it continues with 81-mount.sh
will have to find out where exactly this happens.. continuing this weekend.. any possibility to get the output more verbose (which shell script steps being executed?) without putting echo's everywhere?
Use debug=1, but you have already, right? Apart from mkinitrd not properly recognizing that root in on NFS, and a recent bug https://bugzilla.novell.com/show_bug.cgi?id=861218, I have not had any real issues with booting from nfsroot. Uh, I did at some point have to force NFS v3 though - can't remember why. -- Per Jessen, Zürich (3.8°C) http://www.dns24.ch/ - your free DNS host, made in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
This bug could be exactly my problem. A diff from version 12.3/13.1 boot-storage.sh shows the difference... maybe i try with a newly created initrd. meanwhile i changed/added following to boot params.. nfsroot=172.18.0.1:/exports/delta,nfsvers=3 ipv6.disable=1 debug=1 On Feb 27 2014 09:01, Per Jessen wrote:
Date: Thu, 27 Feb 2014 09:01:33 +0100 From: Per Jessen <per@computer.org> To: opensuse@opensuse.org Subject: Re: [opensuse] remote and diskless install - mkinitrd
Paul Neuwirth wrote:
That's right. dhcpcd is called before interface lo is set up (this is going to be done at the end of 12-network.sh) and performs a dns lookup of the own host (23.0.18.172.in-addr.arpa) - but dhcpcd is launched with "dhcpcd -R -Y -N", so resolving will not succeed at this point.
I don't think that this is the real problem. Actually think the initrd is kind of broken:
the relevant part of run_all.sh: if [ "$interface" -o "$dhcp" -o "$ip" -o "$nfsaddrs" -o "$drvlink" ]; then modules="af_packet $bonding_module" [ "$debug" ] && echo running 12-network.sh source boot/12-network.sh [ "$modules" ] && load_modules fi [ "$debug" ] && echo preping 21-devinit_done.sh [ "$debug" ] && echo running 21-devinit_done.sh source boot/21-devinit_done.sh [ "$modules" ] && load_modules [ "$debug" ] && echo preping 21-nfs.sh [ "$rootfstype" ] || rootfstype='nfs' if [ "$rootfstype" = "nfs" -o "$need_nfs" ]; then [ "$debug" ] && echo load nfs modules modules="nfs nfsv2 nfsv3 nfsv4" [ "$debug" ] && echo running 21-nfs.sh source boot/21-nfs.sh [ "$modules" ] && load_modules fi [ "$debug" ] && echo preping 81-mount.sh [ "$rootdev" ] || rootdev='172.18.0.1:/exports/delta' if [ ! "$root_already_mounted" ]; then [ "$debug" ] && echo running 81-mount.sh source boot/81-mount.sh [ "$modules" ] && load_modules fi [ "$debug" ] && echo preping 82-remount.sh
[ "$debug" ] && echo running 82-remount.sh source boot/82-remount.sh [ "$modules" ] && load_modules
reading the output is difficult, kernel debug and script debug messages overlap... the last lines I see on my serial console: -the dns lookup error from dhcpcd "running 21-devNFS: Registering the id_resolver key type Key type id_resolver registered Key type id_legacy registered Give root password for maintenance (or type Control-D to continue):"
after pressing CTRL-D it continues with 81-mount.sh
will have to find out where exactly this happens.. continuing this weekend.. any possibility to get the output more verbose (which shell script steps being executed?) without putting echo's everywhere?
Use debug=1, but you have already, right?
Apart from mkinitrd not properly recognizing that root in on NFS, and a recent bug https://bugzilla.novell.com/show_bug.cgi?id=861218, I have not had any real issues with booting from nfsroot. Uh, I did at some point have to force NFS v3 though - can't remember why.
-- Per Jessen, Zürich (3.8°C) http://www.dns24.ch/ - your free DNS host, made in Switzerland.
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
I managed to connect a VGA to the client. now i see some more output (why not on the serial console??? lost hours for that..): following is the real error: mount -o rw,nolock -t nfs 172.18.0.1:/exports/delta,nfsvers=3 /root mount.nfs: no mount point provided searching using google for that error: zero results. only thing maybe usefull is the source code.. don't know what is wrong.. On Feb 27 2014 10:16, Paul Neuwirth wrote:
Date: Thu, 27 Feb 2014 10:16:44 +0100 (CET) From: Paul Neuwirth <pane@paneologe.ch> To: Per Jessen <per@computer.org> Cc: opensuse@opensuse.org Subject: Re: [opensuse] remote and diskless install - mkinitrd
This bug could be exactly my problem. A diff from version 12.3/13.1 boot-storage.sh shows the difference... maybe i try with a newly created initrd.
meanwhile i changed/added following to boot params.. nfsroot=172.18.0.1:/exports/delta,nfsvers=3 ipv6.disable=1 debug=1
On Feb 27 2014 09:01, Per Jessen wrote:
Date: Thu, 27 Feb 2014 09:01:33 +0100 From: Per Jessen <per@computer.org> To: opensuse@opensuse.org Subject: Re: [opensuse] remote and diskless install - mkinitrd
Paul Neuwirth wrote:
That's right. dhcpcd is called before interface lo is set up (this is going to be done at the end of 12-network.sh) and performs a dns lookup of the own host (23.0.18.172.in-addr.arpa) - but dhcpcd is launched with "dhcpcd -R -Y -N", so resolving will not succeed at this point.
I don't think that this is the real problem. Actually think the initrd is kind of broken:
the relevant part of run_all.sh: if [ "$interface" -o "$dhcp" -o "$ip" -o "$nfsaddrs" -o "$drvlink" ]; then modules="af_packet $bonding_module" [ "$debug" ] && echo running 12-network.sh source boot/12-network.sh [ "$modules" ] && load_modules fi [ "$debug" ] && echo preping 21-devinit_done.sh [ "$debug" ] && echo running 21-devinit_done.sh source boot/21-devinit_done.sh [ "$modules" ] && load_modules [ "$debug" ] && echo preping 21-nfs.sh [ "$rootfstype" ] || rootfstype='nfs' if [ "$rootfstype" = "nfs" -o "$need_nfs" ]; then [ "$debug" ] && echo load nfs modules modules="nfs nfsv2 nfsv3 nfsv4" [ "$debug" ] && echo running 21-nfs.sh source boot/21-nfs.sh [ "$modules" ] && load_modules fi [ "$debug" ] && echo preping 81-mount.sh [ "$rootdev" ] || rootdev='172.18.0.1:/exports/delta' if [ ! "$root_already_mounted" ]; then [ "$debug" ] && echo running 81-mount.sh source boot/81-mount.sh [ "$modules" ] && load_modules fi [ "$debug" ] && echo preping 82-remount.sh
[ "$debug" ] && echo running 82-remount.sh source boot/82-remount.sh [ "$modules" ] && load_modules
reading the output is difficult, kernel debug and script debug messages overlap... the last lines I see on my serial console: -the dns lookup error from dhcpcd "running 21-devNFS: Registering the id_resolver key type Key type id_resolver registered Key type id_legacy registered Give root password for maintenance (or type Control-D to continue):"
after pressing CTRL-D it continues with 81-mount.sh
will have to find out where exactly this happens.. continuing this weekend.. any possibility to get the output more verbose (which shell script steps being executed?) without putting echo's everywhere?
Use debug=1, but you have already, right?
Apart from mkinitrd not properly recognizing that root in on NFS, and a recent bug https://bugzilla.novell.com/show_bug.cgi?id=861218, I have not had any real issues with booting from nfsroot. Uh, I did at some point have to force NFS v3 though - can't remember why.
-- Per Jessen, Zürich (3.8°C) http://www.dns24.ch/ - your free DNS host, made in Switzerland.
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Paul Neuwirth wrote:
I managed to connect a VGA to the client. now i see some more output (why not on the serial console??? lost hours for that..):
Yeah, a serisl console is a must in this context. I've always got minicom running somewhere :-)
following is the real error: mount -o rw,nolock -t nfs 172.18.0.1:/exports/delta,nfsvers=3 /root mount.nfs: no mount point provided
Should that not be: mount -o rw,nolock,v3 -t nfs 172.18.0.1:/exports/delta /root This is from one of my PXE configs: append initrd=stork1/initrd-3.2.0-2-desktop root=/dev/nfs rw nfsroot=192.168.2.170:/var/pifs/stork1/root rootflags=hard,intr,v3,wsize=8192,rsize=8192 /Per -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
got it working. mentioned bug exists, using root=/dev/nfs leads into mount -t nfs /dev/nfs /root ... so this works now for me: append initrd=delta_initrd nosplash textonly showopts root=172.18.0.1:/exports/delta rw nfsroot=172.18.0.1:/exports/delta nfsvers=3 ipv6.disable=1 debug=1 console=ttyS0,115200 btw. I use DEC VT520 thanks for all your help. On Feb 28 2014 15:11, Per Jessen wrote:
Date: Fri, 28 Feb 2014 15:11:17 +0100 From: Per Jessen <per@computer.org> To: Paul Neuwirth <mail@paul-neuwirth.de> Cc: opensuse@opensuse.org Subject: Re: [opensuse] remote and diskless install - mkinitrd
Paul Neuwirth wrote:
I managed to connect a VGA to the client. now i see some more output (why not on the serial console??? lost hours for that..):
Yeah, a serisl console is a must in this context. I've always got minicom running somewhere :-)
following is the real error: mount -o rw,nolock -t nfs 172.18.0.1:/exports/delta,nfsvers=3 /root mount.nfs: no mount point provided
Should that not be: mount -o rw,nolock,v3 -t nfs 172.18.0.1:/exports/delta /root
This is from one of my PXE configs:
append initrd=stork1/initrd-3.2.0-2-desktop root=/dev/nfs rw nfsroot=192.168.2.170:/var/pifs/stork1/root rootflags=hard,intr,v3,wsize=8192,rsize=8192
/Per -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Paul Neuwirth wrote:
following is the real error: mount -o rw,nolock -t nfs 172.18.0.1:/exports/delta,nfsvers=3 /root mount.nfs: no mount point provided
.... it doesn't seem to like your "/root", does it exist? other possibility, looks like either your "device" (which doesn't look alot like a device, so maybe mount is confused) or your mount dir are being ignored. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Linda Walsh wrote:
Paul Neuwirth wrote:
following is the real error: mount -o rw,nolock -t nfs 172.18.0.1:/exports/delta,nfsvers=3 /root mount.nfs: no mount point provided
.... it doesn't seem to like your "/root", does it exist?
It's https://bugzilla.novell.com/show_bug.cgi?id=861218 -- Per Jessen, Zürich (3.4°C) http://www.dns24.ch/ - free dynamic DNS, made in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
could reproduce all "problems" on a different system. i could report or contribute to 3 bugs regarding installation of opensuse 13.1 on nfs: -no fstab being created -mkinitrd doesn't include nfs modules -initrd dislikes root=/dev/nfs will do that as soon as I have some time left... On Mar 1 2014 09:08, Per Jessen wrote:
Date: Sat, 01 Mar 2014 09:08:53 +0100 From: Per Jessen <per@computer.org> To: opensuse@opensuse.org Subject: Re: [opensuse] Re: remote and diskless install - mkinitrd
Linda Walsh wrote:
Paul Neuwirth wrote:
following is the real error: mount -o rw,nolock -t nfs 172.18.0.1:/exports/delta,nfsvers=3 /root mount.nfs: no mount point provided
.... it doesn't seem to like your "/root", does it exist?
It's https://bugzilla.novell.com/show_bug.cgi?id=861218
-- Per Jessen, Zürich (3.4°C) http://www.dns24.ch/ - free dynamic DNS, made in Switzerland.
-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
participants (6)
-
Andrey Borzenkov -
John Andersen -
Linda Walsh -
Paul Neuwirth -
Paul Neuwirth -
Per Jessen