[opensuse] 13.1 fails to boot with kernel 3.11.6-4
Today I updated from 12.3 to 13.1 using the DVD from box but I cannot boot using the default 13.1 kernel. Here are the details: 12.3 was installed on /dev/sdb6. I cloned it to /dev/sdb7 as a backup and ran the update on /dev/sdb6. The update appeared to complete successfully, but on booting it immediately went to emergency mode. The log showed several failures: Failed to insert module 'autofs4' Failed to open /dev/autofs: No such file or directory Failed to initialize automounter FATAL: Could not load /lib/modules/3.11.6-4-desktop/modules.dep: No such file or directory as well as several mount failures with these messages: Dependency failed for Local File Systems mount process exited, code=exited status=21 The log also showed that the system root was mounted to /dev/sdb7 instead of /dev/sdb6. I don't understand why this happened, but it explains why /lib/modules/3.11.6-4-desktop/modules.dep could not be found, since that exists only on /dev/sdb6. But from the Grub2 menu I can select the cloned 12.3 on /dev/sdb7, and it boots fine (with kernel 3.7.10-1.16). And if I select, from the Advanced options section of the Grub2 entry for 13.1, kernel 3.7.10-1.16, which the update left installed on /dev/sdb6 (as well as adding the newer kernel), then I can boot 13.1 -- but, again, only with the 12.3 kernel. So why is /dev/sdb7 mounted as root when I try to boot the default 13.1 kernel, and what do I need to fix that and use the newer kernel? Thanks for any help. Steve Berman -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On Thu, Nov 21, 2013 at 11:08 PM, Stephen Berman
Today I updated from 12.3 to 13.1 using the DVD from box but I cannot boot using the default 13.1 kernel. Here are the details: 12.3 was installed on /dev/sdb6. I cloned it to /dev/sdb7 as a backup and ran the update on /dev/sdb6. The update appeared to complete
About those /dev/sdXY names: I havent seen those in ages with my setups of openSuSE. Isnt this supposed to be the newer way to write device and storage names e.g. /dev/disk/by-id/...... and similar? Maybe dev/sdXY isnt supported any more everywhere in all parts of the scripts, update parts and so on? Good luck. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On Thu, 21 Nov 2013 23:23:42 +0100 cagsm
On Thu, Nov 21, 2013 at 11:08 PM, Stephen Berman
wrote: Today I updated from 12.3 to 13.1 using the DVD from box but I cannot boot using the default 13.1 kernel. Here are the details: 12.3 was installed on /dev/sdb6. I cloned it to /dev/sdb7 as a backup and ran the update on /dev/sdb6. The update appeared to complete
About those /dev/sdXY names: I havent seen those in ages with my setups of openSuSE. Isnt this supposed to be the newer way to write device and storage names e.g. /dev/disk/by-id/...... and similar? Maybe dev/sdXY isnt supported any more everywhere in all parts of the scripts, update parts and so on? Good luck.
I used /dev/sdb6 and /dev/sdb7 in my OP for simplicity and brevity; the actual entries in /etc/fstab and the system log are /dev/disk/by-id/...-part6 and /dev/disk/by-id/...-part7. However, /etc/mtab does contain /dev/sdb6 and /dev/sdb7 instead. Is this unexpected (and if so, how do I change it)? Steve Berman -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 2013-11-21 23:55 (GMT+0100) Stephen Berman composed:
On Thu, 21 Nov 2013 23:23:42 +0100 cagsm wrote:
Stephen Berman wrote:
Today I updated from 12.3 to 13.1 using the DVD from box but I cannot boot using the default 13.1 kernel. Here are the details: 12.3 was installed on /dev/sdb6. I cloned it to /dev/sdb7 as a backup and ran the update on /dev/sdb6. The update appeared to complete
About those /dev/sdXY names: I havent seen those in ages with my setups of openSuSE. Isnt this supposed to be the newer way to write device and storage names e.g. /dev/disk/by-id/...... and similar? Maybe dev/sdXY isnt supported any more everywhere in all parts of the scripts, update parts and so on? Good luck.
I used /dev/sdb6 and /dev/sdb7 in my OP for simplicity and brevity; the actual entries in /etc/fstab and the system log are /dev/disk/by-id/...-part6 and /dev/disk/by-id/...-part7. However, /etc/mtab does contain /dev/sdb6 and /dev/sdb7 instead. Is this unexpected (and if so, how do I change it)?
What does can't boot mean exactly? Is the Grub menu working OK? Is the prior kernel there and working OK? Failsafe (nomodeset) boot stanza is no help? Are the boot menu entries on both sda6 and sda7 correct? Ditto fstabs? Did you assign new UUID to either sda6 or sda7 to ensure no trouble due to duplicates? Do both sda6 and sda7 share the same volume label? Any difference if you use device name or by-label instead of by-id in the Grub stanza or fstab? -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation) Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 2013-11-22 00:37, Felix Miata wrote:
Did you assign new UUID to either sda6 or sda7 to ensure no trouble due to duplicates?
Ah! I vote for this one. :-) -- Cheers / Saludos, Carlos E. R. (from 12.3 x86_64 "Dartmouth" at Telcontar)
On Thu, 21 Nov 2013 18:37:50 -0500 Felix Miata
On 2013-11-21 23:55 (GMT+0100) Stephen Berman composed:
On Thu, 21 Nov 2013 23:23:42 +0100 cagsm wrote:
Stephen Berman wrote:
Today I updated from 12.3 to 13.1 using the DVD from box but I cannot boot using the default 13.1 kernel. Here are the details: 12.3 was installed on /dev/sdb6. I cloned it to /dev/sdb7 as a backup and ran the update on /dev/sdb6. The update appeared to complete
About those /dev/sdXY names: I havent seen those in ages with my setups of openSuSE. Isnt this supposed to be the newer way to write device and storage names e.g. /dev/disk/by-id/...... and similar? Maybe dev/sdXY isnt supported any more everywhere in all parts of the scripts, update parts and so on? Good luck.
I used /dev/sdb6 and /dev/sdb7 in my OP for simplicity and brevity; the actual entries in /etc/fstab and the system log are /dev/disk/by-id/...-part6 and /dev/disk/by-id/...-part7. However, /etc/mtab does contain /dev/sdb6 and /dev/sdb7 instead. Is this unexpected (and if so, how do I change it)?
What does can't boot mean exactly?
The system doesn't come up normally. As I wrote in my OP: "on booting it immediately went to emergency mode"
Is the Grub menu working OK?
I'm not sure just what you mean; all the entries appear to be there and I can select them all. Or do you mean something else?
Is the prior kernel there and working OK?
Yes, when booting either 12.3 or 13.1 (as I also previously noted).
Failsafe (nomodeset) boot stanza is no help?
No, same thing happens (emergency mode).
Are the boot menu entries on both sda6 and sda7 correct?
That's sdb6 and sdb7, but AFAIK yes. In /boot of the former grub.cfg currently contains this: And in /boot of /dev/sdb7 grub.cfg currently contains this:
Ditto fstabs?
How can I ascertain if they are correct? Both contained the mountpoint entries I assigned using /dev/disk/by-id/... (Currently they differ, because I've tried other options, noted below.)
Did you assign new UUID to either sda6 or sda7 to ensure no trouble due to duplicates?
I didn't, and perhaps this is what caused the problems. (I did read the warning about persistent device names in the Start-Up manual, but didn't appreciate its importance; I've often installed different versions of openSUSE in different partitions of the same harddisk, and never had problems -- though those were fresh installations, not updates, which I now did for the first time.) After getting these problems, I changed the fstab options for these partitions to mount by UUID, but it didn't help; nor did mounting by device name or path. In fact, trying these resulted in even weirder behavior: I couldn't boot 13.1 at all anymore, even with the 12.3 kernel, but I could boot 12.3 with the 13.1 kernel (!?). Currently, I'm in 13.1 with the 12.3 kernel, with sdb6 (the root file system) mounted by UUID and sdb7 mounted by ID. But I still can't boot the 13.1 kernel on sdb6. Is there any way to repair this without reinstalling?
Do both sda6 and sda7 share the same volume label?
Neither partition has a volume label.
Any difference if you use device name or by-label instead of by-id in the Grub stanza or fstab?
As mentioned, using by name mounting in fstab didn't help; I'll try by label the next time I'm ready to reboot. How do I try these in Grub? Steve Berman
Le 22/11/2013 11:59, Stephen Berman a écrit :
On Thu, 21 Nov 2013 18:37:50 -0500 Felix Miata
wrote:
Is the Grub menu working OK?
I'm not sure just what you mean; all the entries appear to be there and I can select them all. Or do you mean something else?
in grub with Esc and in grub2 with "e" you can open an edit menu, where you can try to fix the boot options going to emergency mode is often because the root partition is incorrect hope this help :-) jdd -- http://www.dodin.org -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On Fri, 22 Nov 2013 12:43:56 +0100 jdd
Le 22/11/2013 11:59, Stephen Berman a écrit :
On Thu, 21 Nov 2013 18:37:50 -0500 Felix Miata
wrote: Is the Grub menu working OK?
I'm not sure just what you mean; all the entries appear to be there and I can select them all. Or do you mean something else?
in grub with Esc and in grub2 with "e" you can open an edit menu, where you can try to fix the boot options
I know about "e", but I don't know what to change to fix the problem...
going to emergency mode is often because the root partition is incorrect
The Grub menu entry shows the correct root partition (what I've been calling /dev/sdb6), but still when I try to boot with the 13.1 kernel (which I installed in that partition), the root system is mounted to /dev/sdb7 (the partition containing the clone of 12.3). Steve Berman -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 11/22/2013 01:43 PM, jdd wrote:
Le 22/11/2013 11:59, Stephen Berman a écrit :
On Thu, 21 Nov 2013 18:37:50 -0500 Felix Miata
wrote: Is the Grub menu working OK?
I'm not sure just what you mean; all the entries appear to be there and I can select them all. Or do you mean something else?
in grub with Esc and in grub2 with "e" you can open an edit menu, where you can try to fix the boot options
going to emergency mode is often because the root partition is incorrect
hope this help :-) jdd
after moving to 13,1 : the only way that i can boot is by command line grub , like: [my root is on /dev/sdb] 1. grub : setup (hd1,0) 2. kernel /boot/vmlinuz 3. initrd /boot/initrd 4. root (hd1,0) ............. regards -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 2013-11-22 11:59 (GMT+0100) Stephen Berman composed:
Ditto fstabs?
How can I ascertain if they are correct? Both contained the mountpoint entries I assigned using /dev/disk/by-id/... (Currently they differ, because I've tried other options, noted below.)
Start with output of 'mount | grep sdb' to ascertain which partition is your mounted root partition. Try to match it up to output of /proc/cmdline in rebuilding fstab on whichever partition actually contains the 13.1 kernel's modules in /lib/modules/3.11.6-4 and the kernel and initrd in /boot. That will probably be the partition where the rest of 13.1 actually got installed. Using /dev/sdb6 & /dev/sdb7 in fstab will make your job easier at least until this is all sorted through. I always use LABEL= for both Grub root= entries and fstab entries whether I'm using clones or not, and ensure to change appropriate labels when creating clones, and I change at least one UUID among partitions involved in cloning. Before your next boot, ensure sdb6 and sdb7 have different UUIDs by giving at least one of them a new one if both still match.
Did you assign new UUID to either sda6 or sda7 to ensure no trouble due to duplicates?
I didn't, and perhaps this is what caused the problems. (I did read the warning about persistent device names in the Start-Up manual, but didn't appreciate its importance; I've often installed different versions of openSUSE in different partitions of the same harddisk, and never had problems -- though those were fresh installations, not updates, which I now did for the first time.) After getting these problems, I changed the fstab options for these partitions to mount by UUID, but it didn't help; nor did mounting by device name or path. In fact, trying these resulted in even weirder behavior: I couldn't boot 13.1 at all anymore, even with the 12.3 kernel, but I could boot 12.3 with the 13.1 kernel (!?). Currently, I'm in 13.1 with the 12.3 kernel, with sdb6 (the root file system) mounted by UUID and sdb7 mounted by ID. But I still can't boot the 13.1 kernel on sdb6. Is there any way to repair this without reinstalling?
Yes, but I'm not sure the odds are in your favor. It does appear the installation process was fouled by the existence of two partitions with identical UUIDs.
Do both sda6 and sda7 share the same volume label?
Neither partition has a volume label.
It could help to assign one to each.
Any difference if you use device name or by-label instead of by-id in the Grub stanza or fstab?
As mentioned, using by name mounting in fstab didn't help; I'll try by label the next time I'm ready to reboot. How do I try these in Grub?
I only use Grub Legacy, so can't give a specific answer without looking up how Grub2 works. Likely simplest initially will be to edit the stanza at each boot to use an appropriate root=/dev/sdb# until you know fstab matches the home of 13.1, have 13.1 booting on the 13.1 kernel, and mounts are as they need to be. Once it's sorted out, rebuild the Grub2 menu normally. -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation) Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On Fri, 22 Nov 2013 07:22:34 -0500 Felix Miata
On 2013-11-22 11:59 (GMT+0100) Stephen Berman composed:
Before your next boot, ensure sdb6 and sdb7 have different UUIDs by giving at least one of them a new one if both still match.
Did you assign new UUID to either sda6 or sda7 to ensure no trouble due to duplicates?
I didn't, and perhaps this is what caused the problems.[...] Is there any way to repair this without reinstalling?
Yes, but I'm not sure the odds are in your favor. It does appear the installation process was fouled by the existence of two partitions with identical UUIDs.
That was indeed what happened, due to cloning the 12.3 partition. But after assigning a new UUID to it and, following your suggestion, giving both partitions labels, and then rebuilding grub.cfg, I can now boot 13.1 with its default kernel. This was quite an educational experience for me. Thanks for your help! Steve Berman -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 2013-11-22 11:59 (GMT+0100) Stephen Berman composed:
Ditto fstabs?
How can I ascertain if they are correct? Both contained the mountpoint entries I assigned using /dev/disk/by-id/... (Currently they differ, because I've tried other options, noted below.)
Start with output of 'mount | grep sdb' to ascertain which partition is your mounted root partition. Try to match it up to output of /proc/cmdline in rebuilding fstab on whichever partition actually contains the 13.1 kernel's modules in /lib/modules/3.11.6-4 and the kernel and initrd in /boot. That will probably be the partition where the rest of 13.1 actually got installed. Using /dev/sdb6 & /dev/sdb7 in fstab will make your job easier at least until this is all sorted through. I always use LABEL= for both Grub root= entries and fstab entries whether I'm using clones or not, and ensure to change appropriate labels when creating clones, and I change at least one UUID among partitions involved in cloning. Before your next boot, ensure sdb6 and sdb7 have different UUIDs by giving at least one of them a new one if both still match.
Did you assign new UUID to either sda6 or sda7 to ensure no trouble due to duplicates?
I didn't, and perhaps this is what caused the problems. (I did read the warning about persistent device names in the Start-Up manual, but didn't appreciate its importance; I've often installed different versions of openSUSE in different partitions of the same harddisk, and never had problems -- though those were fresh installations, not updates, which I now did for the first time.) After getting these problems, I changed the fstab options for these partitions to mount by UUID, but it didn't help; nor did mounting by device name or path. In fact, trying these resulted in even weirder behavior: I couldn't boot 13.1 at all anymore, even with the 12.3 kernel, but I could boot 12.3 with the 13.1 kernel (!?). Currently, I'm in 13.1 with the 12.3 kernel, with sdb6 (the root file system) mounted by UUID and sdb7 mounted by ID. But I still can't boot the 13.1 kernel on sdb6. Is there any way to repair this without reinstalling?
Yes, but I'm not sure the odds are in your favor. It does appear the installation process was fouled by the existence of two partitions with identical UUIDs.
Do both sda6 and sda7 share the same volume label?
Neither partition has a volume label.
It could help to assign one to each.
Any difference if you use device name or by-label instead of by-id in the Grub stanza or fstab?
As mentioned, using by name mounting in fstab didn't help; I'll try by label the next time I'm ready to reboot. How do I try these in Grub?
I only use Grub Legacy, so can't give a specific answer without looking up how Grub2 works. Likely simplest initially will be to edit the stanza at each boot to use an appropriate root=/dev/sdb# until you know fstab matches the home of 13.1, have 13.1 booting on the 13.1 kernel, and mounts are as they need to be. Once it's sorted out, rebuild the Grub2 menu normally. -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation) Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
* Stephen Berman
The system doesn't come up normally. As I wrote in my OP: "on booting it immediately went to emergency mode"
I see this on one system, *but* if I am patient and wait for a minute or two, the system continues and finishes booting. try it :^) mutch remaining quotes removed! -- (paka)Patrick Shanahan Plainfield, Indiana, USA @ptilopteri http://en.opensuse.org openSUSE Community Member facebook/ptilopteri http://wahoo.no-ip.org Photo Album: http://wahoo.no-ip.org/gallery2 Registered Linux User #207535 @ http://linuxcounter.net -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
participants (7)
-
cagsm
-
Carlos E. R.
-
ellanios82
-
Felix Miata
-
jdd
-
Patrick Shanahan
-
Stephen Berman