home and backup where LUCS encripted with the same passphrase. The installation asked for this passphrase, but could mount only /backup (It event offered to reformat the home partition, which I rejected. After the installation success the home partition was still engaged somehow and I could not do with it nothing more:
# p3=/dev/nvme0n1p3
# udisksctl unlock -b $p3
Passphrase:
Error unlocking /dev/nvme0n1p3: GDBus.Error:org.freedesktop.UDisks2.Error.Failed: Device /dev/nvme0n1p3 is already unlocked as /dev/dm-1
Since “/dev/mapper/myhome” exists, that seems to have been done. Check if there is an appropriate entry in “/etc/crypttab”. Maybe post the content of “/etc/crypttab”).
Let me remark too, that /backup and /home decription was always released together, so - this is my speculation - if /backup is mounted properly, than /home must be involved somehow too.
I see no inconsistences. /etc/crypttab says about it which partitions are encripted and fstab says, what is currently mounted.
Well, the installation made “home” a part of the second partition, so I used “/h” for my efforts to mount the third partition.
With the question for passphrase on the graphical screen another question appears in the terxt mode reached with CTRL ALT F1. Possibly one ist for my /backup and another one for my /home . It is well possible, that such a situation appeared while installing the system. Then one question has been answered and the other not. Nevertheless the /home partition is beeing bound somehow - Could this be the explanation ?
What do you mean by re-install?
Were you expecting to re-use your encrypted partitions?
If so, I wouldn’t expect that to be possible, you’d have to decrypt your partitions before setting up encryption again after the re-install.
Unless someone comes up with something I wouldn’t expect…
Another question is, what the following message appearing on the text screen while booting means:
cr_nvme-Samsung_SSD_970_EVO_2TB_S464NB0M401155J
request region for resource [mem 0x00c5fffc-0x00c5ffff]
4.040724] KVM disabled by BIOS
which is shown the logfile as
Found device /dev/nvme0n1p3.
Starting Cryptography Setup for cr_nvme-Samsung_SSD_970_EVO_2TB_S464NB0M401155J-part3...
Found device /dev/disk/by-uuid/C9B2-DCAF.
Found device ST31000528AS 4.
Starting Cryptography Setup for cr-auto-2...
Started Forward Password Requests to Plymouth.
Found device ST31000528AS 2.
Preferably, do that just after reboot, and before you have done anything that might affect the crypto handling.
With the question for passphrase on the graphical screen another question appears in the terxt mode reached with CTRL ALT F1. Possibly one ist for my /backup and another one for my /home . It is well possible, that such a situation appeared while installing the system. Then one question has been answered and the other not. Nevertheless the /home partition is beeing bound somehow - Could this be the explanation ?
The installer asks separately for each encrypted partition. But, for normal boots, if both partitions use the same pass phrase, then you are normally only asked once.
The aim, for the moment, is to try to understand what is the situation with crypto just after booting.
Yes I installed LEAP 15.1 once more leaving my user data in /home and /backup partition as they were before. That means root partition has been formated once more. Ussually the SuSE installation script asks for a passphrase and manages everything else alone. Now for me it seems as if it has managed LUKS encription for one partition only, but the other is not quite left alone for a manual handling.
Those might be three independent messages (not related to one another).
I used to regularly see the “KVM disabled” message, until I enabled it in BIOS settings – I think BIOS setting showed it as CPU features. I often see a “request region for resource” message, and I ignore it. I think it’s is a debug message for kernel maintainers.
What to do now ? I ussually tried to mount /home with yast->partitioner
And there is only root and backup mounted
...
/dev/nvme0n1p2 on /root type btrfs (rw,relatime,ssd,space_cache,subvolid=265,subvol=/@/root)
/dev/nvme0n1p2 on /opt type btrfs (rw,relatime,ssd,space_cache,subvolid=266,subvol=/@/opt)
/dev/mapper/cr-auto-2 on /backup type ext4 (rw,relatime,data=ordered)
tmpfs on /run/user/1000 type tmpfs (rw,nosuid,nodev,relatime,size=3276596k,mode=700,uid=1000,gid=100)
...
# mount /dev/dm-1 /h
mount: /h: wrong file system, wrong option, wrong superblok on /dev/mapper/cr_nvme-Samsung_SSD_970_EVO_2TB_S464NB0M401155J-part3, lack of code page or a help program or another error
Nearly each filesystem can be repaired. I assume /dev/dm-1 is now a raw partition without any LUKS and the task is to repair it as a filesystem, am I right ?
# lsblk -o name,mountpoint,size,type,ro,label,uuid
...
nvme0n1 1,8T disk 0
...
└─nvme0n1p3
1,8T part 0 dc3763d9-fce5-4537-81a6-967e83a93e82
└─cr_nvme-Samsung_SSD_970_EVO_2TB_S464NB0M401155J-part3
1,8T crypt 0
