[SOLVED] upgrade issues

D

Deleted member 33567

Guest
Code: 
Setting up proxmox-ve (5.1-42) ...
Processing triggers for initramfs-tools (0.130) ...
update-initramfs: Generating /boot/initrd.img-4.13.16-2-pve
cryptsetup: WARNING: failed to detect canonical device of /dev/sda1
cryptsetup: WARNING: could not determine root device from /etc/fstab

Any idea why this occurred its first time i see this, runned upgrade on 3 nodes only one reported this.
 
UHL-Services said:
Code: 
Setting up proxmox-ve (5.1-42) ...
Processing triggers for initramfs-tools (0.130) ...
update-initramfs: Generating /boot/initrd.img-4.13.16-2-pve
cryptsetup: WARNING: failed to detect canonical device of /dev/sda1
cryptsetup: WARNING: could not determine root device from /etc/fstab

Any idea why this occurred its first time i see this, runned upgrade on 3 nodes only one reported this.
Click to expand...

check out /usr/share/initramfs-tools/hooks/cryptroot ;)

the message says that the hook script was not able to get the underlying device for your / mountpoint from /etc/fstab.
 
@
fabian said:
check out /usr/share/initramfs-tools/hooks/cryptroot ;)

the message says that the hook script was not able to get the underlying device for your / mountpoint from /etc/fstab.
Click to expand...


Yes, and I would appreciate when this will be taken seriously, hence this behavior is NOT expected and NOT normal.

What should I debug to determine what caused this on a perfect working node, after a simple upgrade.

This is my fstab on node2

Code: 
# <file system> <mount point>   <type>  <options>       <dump>  <pass>
/dev/sda1       /       ext3    errors=remount-ro       0       1
/dev/sda2       swap    swap    defaults        0       0
/dev/pve/data   /var/lib/vz     ext3    defaults        1       2
#/dev/node02-sata2tb/lvol0 /2tb  ext4    defaults       1       2
proc            /proc   proc    defaults        0       0
sysfs           /sys    sysfs   defaults        0       0

This is node3

Code: 
# <file system> <mount point>   <type>  <options>       <dump>  <pass>
/dev/sda2       /       ext3    errors=remount-ro       0       1
/dev/sda3       swap    swap    defaults        0       0
/dev/pve/data   /var/lib/vz     ext3    defaults        1       2
proc            /proc   proc    defaults        0       0
sysfs           /sys    sysfs   defaults        0       0

And only 1 reported the issue mentioned above. May I know why?
 
UHL-Services said:
@



Yes, and I would appreciate when this will be taken seriously, hence this behavior is NOT expected and NOT normal.
Click to expand...

the initramfs scripts are not shipped by PVE, but by Debian.

What should I debug to determine what caused this on a perfect working node, after a simple upgrade.

This is my fstab on node2

Code: 
# <file system> <mount point>   <type>  <options>       <dump>  <pass>
/dev/sda1       /       ext3    errors=remount-ro       0       1
/dev/sda2       swap    swap    defaults        0       0
/dev/pve/data   /var/lib/vz     ext3    defaults        1       2
#/dev/node02-sata2tb/lvol0 /2tb  ext4    defaults       1       2
proc            /proc   proc    defaults        0       0
sysfs           /sys    sysfs   defaults        0       0

This is node3

Code: 
# <file system> <mount point>   <type>  <options>       <dump>  <pass>
/dev/sda2       /       ext3    errors=remount-ro       0       1
/dev/sda3       swap    swap    defaults        0       0
/dev/pve/data   /var/lib/vz     ext3    defaults        1       2
proc            /proc   proc    defaults        0       0
sysfs           /sys    sysfs   defaults        0       0

And only 1 reported the issue mentioned above. May I know why?
Click to expand...

look at the cryptroot hook script, check the contents of /etc/crypttab on both nodes - this is not PVE related at all.. are you even using a LUKS-encrypted device?
 
In my knowledge none of the nodes have any of the disks encrypted. And hence not related to pve, why it occurs on one node only... and the kernel is the pve kernel that run the initframes.
 
UHL-Services said:
In my knowledge none of the nodes have any of the disks encrypted. And hence not related to pve, why it occurs on one node only... and the kernel is the pve kernel that run the initframes.
Click to expand...

I don't know why it occurs on one node only (I gave you some pointers how you could find out though).

the initramfs-tools package is not maintained by Proxmox, it is just the regeneration that is triggered by the pve-kernel-* packages. if the generated initramfs works, the warning is likely benign and can be ignored.
 
To be honest after seeing this error, I think the only way to test initframfs it is to reboot the server, and the things reporting main HDD of proxmox node got me wondering if I should reboot in first place.
 
You must log in or register to reply here.
Top Bottom