Hello, I just cloned a sata ssd to a nvme disk of same size on an other computer. Windows boots, but openSUSE don't (I have Tumbleweed and 15.3 on this disk). I get to the grub menu but stop soon after What could be the simpler way to fix this? try from grub console or recovery disk? I suspect either an UID or device name problem in fstab thanks jdd -- mon serveur usenet: dodin.fr.nf c'est quoi, usenet? http://www.dodin.org/wiki/pmwiki.php?n=Usenet.Usenet
Hello, In the Message; Subject : go from sata to nvme Message-ID : <4a946c90-d45b-bc90-b2e8-61aa30aa86ed@dodin.org> Date & Time: Thu, 6 Apr 2023 13:03:45 +0200 [jdd] == "jdd@dodin.org" <jdd@dodin.org> has written: jdd> Hello, jdd> I just cloned a sata ssd to a nvme disk of same size on an other jdd> computer. jdd> Windows boots, but openSUSE don't (I have Tumbleweed and 15.3 on jdd> this disk). jdd> I get to the grub menu but stop soon after jdd> What could be the simpler way to fix this? How about changing the boot options as follows? 1. remove quiet splash 2. add nvme_load=yes Regards. --- ┏━━┓彡 Masaru Nomiya mail-to: nomiya @ lake.dti.ne.jp ┃＼／彡 ┗━━┛ "A society bound by e-mail and mobile phones deprives us of the freedom to face ourselves and indulge our fantasies." -- Michael Crichton (Speech in Japan) --
Le 06/04/2023 à 13:20, Masaru Nomiya a écrit :
2. add
nvme_load=yes
it's much better, but not enouth. Before it stopped just after the ACPI error (pretty common and not harmfull) now it goes up to a lot of usb config lines. I will try per one thanks jdd -- mon serveur usenet: dodin.fr.nf c'est quoi, usenet? http://www.dodin.org/wiki/pmwiki.php?n=Usenet.Usenet
On 2023-04-06 14:46, jdd@dodin.org wrote:
Le 06/04/2023 à 13:20, Masaru Nomiya a écrit :
2. add
nvme_load=yes
it's much better, but not enouth. Before it stopped just after the ACPI error (pretty common and not harmfull)
now it goes up to a lot of usb config lines.
Have you tried a full power off? Unplug the mains. Sometimes in my computer some USB device, or perhaps the USB hardware in the computer, act up. There are a lot of errors in the log, and system "walks" slowly. The solution is to power off fully. Orderly if possible, but in my case it can take minutes, or not work at all. -- Cheers / Saludos, Carlos E. R. (from 15.4 x86_64 at Telcontar)
Le 06/04/2023 à 14:53, Carlos E. R. a écrit :
Have you tried a full power off? Unplug the mains.
it's a laptop :-( jdd -- mon serveur usenet: dodin.fr.nf c'est quoi, usenet? http://www.dodin.org/wiki/pmwiki.php?n=Usenet.Usenet
On 2023-04-06 15:50, jdd@dodin.org wrote:
Le 06/04/2023 à 14:53, Carlos E. R. a écrit :
Have you tried a full power off? Unplug the mains.
it's a laptop :-(
Well, the button should be enough. Heh, my new laptop has a reset button under it, to be used with a straightened clip. -- Cheers / Saludos, Carlos E. R. (from 15.4 x86_64 at Telcontar)
jdd@dodin.org composed on 2023-04-06 14:46 (UTC+0200):
nvme_load=yes
it's much better, but not enouth. Before it stopped just after the ACPI error (pretty common and not harmfull)
now it goes up to a lot of usb config lines.
OP of this thread never arrived, so details are missing. This may be same problem I reported @ https://bugzilla.opensuse.org/show_bug.cgi?id=1205261#c4 rd.hostonly=0 along with already present noresume followed by rebuilding initrds then removing rd.hostonly=0 was fix for me. -- Evolution as taught in public schools is, like religion, based on faith, not based on science. Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata
On 2023-04-06 17:47, Felix Miata wrote:
jdd@dodin.org composed on 2023-04-06 14:46 (UTC+0200):
nvme_load=yes
it's much better, but not enouth. Before it stopped just after the ACPI error (pretty common and not harmfull)
now it goes up to a lot of usb config lines.
OP of this thread never arrived, so details are missing.
Interestingly, I'm also missing at least two of his posts. I have them on another account. -- Cheers / Saludos, Carlos E. R. (from 15.4 x86_64 at Telcontar)
We can stop this thread. I have on this disk two openSUSE installs, including one of 15.3, now deprecated. So I could use this place to make a fresh install of Tumbleweed. From this I could go more easily (than with rescue disk) to the faulty cloned install. I could see that many files on /sbin have a zero length, meaning that the clone is faulty. So, end of the problem, not fixable. I just have to reinstall the special software. just a guess: no problem on *data* partitions, so I think btrfs subvolumes are not correctly managed with my (some month old) clonezilla install. jdd -- mon serveur usenet: dodin.fr.nf c'est quoi, usenet? http://www.dodin.org/wiki/pmwiki.php?n=Usenet.Usenet
On 2023-04-06 23:01, jdd@dodin.org wrote:
We can stop this thread.
I have on this disk two openSUSE installs, including one of 15.3, now deprecated.
So I could use this place to make a fresh install of Tumbleweed.
From this I could go more easily (than with rescue disk) to the faulty cloned install.
I could see that many files on /sbin have a zero length, meaning that the clone is faulty.
So, end of the problem, not fixable.
I just have to reinstall the special software.
just a guess: no problem on *data* partitions, so I think btrfs subvolumes are not correctly managed with my (some month old) clonezilla install.
We could investigate why clonezilla failed to restore. Or failed to backup. Is it an image backup as done by dd? Is it a file by file backup, similar to rsync? Is it a "smart" backup? with some kind of smartness that claims to understand btrfs, an image backup that skips empty sectors and has other features? Do you still have the original disk? -- Cheers / Saludos, Carlos E. R. (from 15.4 x86_64 at Telcontar)
Le 06/04/2023 à 23:42, Carlos E. R. a écrit :
We could investigate why clonezilla failed to restore. Or failed to backup.
may be, but it's hard
Is it a "smart" backup? with some kind of smartness that claims to understand btrfs, an image backup that skips empty sectors and has other features?
yes, clonezilla do a smart work, copying only the used parts of the disk
Do you still have the original disk?
more or less: The new computer is not yet in use (I have to add some hardware) and I still use the old one jdd -- mon serveur usenet: dodin.fr.nf c'est quoi, usenet? http://www.dodin.org/wiki/pmwiki.php?n=Usenet.Usenet
On 2023-04-07 08:04, jdd@dodin.org wrote:
Le 06/04/2023 à 23:42, Carlos E. R. a écrit :
We could investigate why clonezilla failed to restore. Or failed to backup.
may be, but it's hard
Is it a "smart" backup? with some kind of smartness that claims to understand btrfs, an image backup that skips empty sectors and has other features?
yes, clonezilla do a smart work, copying only the used parts of the disk
Ok, then that is the culprit. That smartness is dumb.
Do you still have the original disk?
more or less: The new computer is not yet in use (I have to add some hardware) and I still use the old one
Ok, can you tell clonezilla to use plain dd for the imaging, without using its smartness? Then do another backup and try again. If not, you need to use dd yourself. I can share the way I do it, with compression. -- Cheers / Saludos, Carlos E. R. (from 15.4 x86_64 at Telcontar)
Le 07/04/2023 à 11:50, Carlos E. R. a écrit :
Ok, can you tell clonezilla to use plain dd for the imaging, without using its smartness?
I don't think so and anyway it's not necessary
If not, you need to use dd yourself.
sure
I can share the way I do it, with compression.
compression is not necessary I will try this. The relevant partition is only 100Gb. But I have to boot an other distro than the present one (in use :-() - reason why I have several :-) may be long, even creating ext3 inodes takes ages on this usb3 pen :-( jdd -- mon serveur usenet: dodin.fr.nf c'est quoi, usenet? http://www.dodin.org/wiki/pmwiki.php?n=Usenet.Usenet
On 2023-04-07 12:28, jdd@dodin.org wrote:
Le 07/04/2023 à 11:50, Carlos E. R. a écrit :
Ok, can you tell clonezilla to use plain dd for the imaging, without using its smartness?
I don't think so and anyway it's not necessary
If not, you need to use dd yourself.
sure
I can share the way I do it, with compression.
compression is not necessary
I will try this. The relevant partition is only 100Gb. But I have to boot an other distro than the present one (in use :-() - reason why I have several :-)
may be long, even creating ext3 inodes takes ages on this usb3 pen :-(
You can use the XFCE Rescue image on USB stick, it is real good for these things. <http://download.opensuse.org/distribution/leap/15.4/live/openSUSE-Leap-15.4-Rescue-CD-x86_64-Media.iso> -- Cheers / Saludos, Carlos E. R. (from 15.4 x86_64 at Telcontar)
Le 07/04/2023 à 08:04, jdd@dodin.org a écrit :
more or less: The new computer is not yet in use (I have to add some hardware) and I still use the old one
well, I give up... even after dd from the original, I can't make it to boot given I have a working computer now, I won't spend more time debugging thanks everyone for your help jdd -- mon serveur usenet: dodin.fr.nf c'est quoi, usenet? http://www.dodin.org/wiki/pmwiki.php?n=Usenet.Usenet
jdd@dodin.org wrote:
Hello, I just cloned a sata ssd to a nvme disk of same size on an other computer.
Windows boots, but openSUSE don't (I have Tumbleweed and 15.3 on this disk).
I get to the grub menu but stop soon after
What could be the simpler way to fix this? try from grub console or recovery disk?
I suspect either an UID or device name problem in fstab
Agree, that is what it sounds like - you can make it boot from a different root device from the command line, but if you have e.g. home on a separate partition, you have to fix the fstab. Boot a rescue system, mount the partition and fix the fstab. Fix the initrd at the same time. -- Per Jessen, Zürich (11.9°C) Member, openSUSE Heroes (2016 - present) We're hiring - https://en.opensuse.org/openSUSE:Heroes
On 2023-04-06 13:40, Per Jessen wrote:
jdd@dodin.org wrote:
Hello, I just cloned a sata ssd to a nvme disk of same size on an other computer.
Windows boots, but openSUSE don't (I have Tumbleweed and 15.3 on this disk).
I get to the grub menu but stop soon after
What could be the simpler way to fix this? try from grub console or recovery disk?
I suspect either an UID or device name problem in fstab
Agree, that is what it sounds like - you can make it boot from a different root device from the command line, but if you have e.g. home on a separate partition, you have to fix the fstab.
Boot a rescue system, mount the partition and fix the fstab. Fix the initrd at the same time.
Ha, that would be "mkinitrd" not that long ago :-D Now: dracut -f --regenerate-all --logfile /var/log/YaST2/mkinitrd.log update-bootloader --refresh -- Cheers / Saludos, Carlos E. R. (from 15.4 x86_64 at Telcontar)
Le 06/04/2023 à 13:40, Per Jessen a écrit :
Boot a rescue system, mount the partition and fix the fstab. Fix the initrd at the same time.
I did that pretty often, but right now failing. I boot a rescue image, mount the root partition, and can access it. Then I mount --bind sys, dev and proc and chroot in /mnt. but then I was used to fire yast to make the hard work like editing fstab and updating the boot system. but here, no yast!! All the yast in /sbin or /usr/sbin are ??? (orphaned links? on the original disk they are there) I made the clone with clonezilla, may be a bug in it? I tried to fix fstab by hand, but with all the subvols it's hard I have till tomorrow to try booting this new computer, I ordered some more hardware for it. If not successful, I will reinstall, but I have some extra repos that are not that easy to cope with :-( the old laptop was intel i7 sata, the new is amd rysen nvme... thanks jdd -- mon serveur usenet: dodin.fr.nf c'est quoi, usenet? http://www.dodin.org/wiki/pmwiki.php?n=Usenet.Usenet
jdd@dodin.org wrote:
Le 06/04/2023 à 13:40, Per Jessen a écrit :
Boot a rescue system, mount the partition and fix the fstab. Fix the initrd at the same time.
I did that pretty often, but right now failing. I boot a rescue image, mount the root partition, and can access it. Then I mount --bind sys, dev and proc and chroot in /mnt.
Sounds good. I usually also bind-mount /run, but I don't actually know if it's needed.
but then I was used to fire yast to make the hard work like editing fstab and updating the boot system. but here, no yast!! All the yast in /sbin or /usr/sbin are ??? (orphaned links? on the original disk they are there)
Hmm, maybe your copying/cloning didn't go quite right? -- Per Jessen, Zürich (12.8°C) Member, openSUSE Heroes (2016 - present) We're hiring - https://en.opensuse.org/openSUSE:Heroes
jdd@dodin.org wrote:
Hello, I just cloned a sata ssd to a nvme disk of same size on an other computer.
Windows boots, but openSUSE don't (I have Tumbleweed and 15.3 on this disk).
Once did that, too, moving a btrfs RAID1 from two SATAs to two NVMEs. (moved on the running system...) Reboot failed, because I forgot to re-do the initrds with NVME (module) support...
Pit Suetterlin composed on 2023-04-07 21:40 (UTC+0100):
jdd wrote:
I just cloned a sata ssd to a nvme disk of same size on an other computer.
Windows boots, but openSUSE don't (I have Tumbleweed and 15.3 on this disk).
Once did that, too, moving a btrfs RAID1 from two SATAs to two NVMEs. (moved on the running system...) Reboot failed, because I forgot to re-do the initrds with NVME (module) support...
This may be same problem I reported @ https://bugzilla.opensuse.org/show_bug.cgi?id=1205261#c4 rd.hostonly=0 along with already present noresume followed by rebuilding initrds then removing rd.hostonly=0 was fix for me. -- Evolution as taught in public schools is, like religion, based on faith, not based on science. Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata
participants (6)
-
Carlos E. R. -
Felix Miata -
jdd＠dodin.org -
Masaru Nomiya -
Per Jessen -
Pit Suetterlin