I have a very specific and strange problem and ita kinda urgent to solve it in a timely manner. Yesterday I encountered a kernel panic on my proxmox server
looking at server management logs apparently there was a power outage and ups was out of service for maintenance
booting to the default kernel:
cannot open "scripts/zfs" no such file or directory
kernel panic "not syncing attempted to kill init"
I was able to boot from recovery mode in grub to a previous kernel then after checking system journal I deleted the problem a malfunctioning virtual network card causing VMs to not start I removed the ovsitch, did an apt update and fix broken packages and install the new available kernel. After reboot I still get the kernel panic, and because I updated grub the previous kernel which was working is no longer available. Reinstalling system is out of the question so I need help to repair the kernel.
this is what I've tried
So far I imported the zfs pool to the live os I can see and mount some of the guest vm disks but I proxmox "/" (root directory) is not mounted.
I did "zpool import rpool" then 2 directories appeared rpool/DATA and rpool/ROOT/pve-1 but they are both empty.
with lsblk I can see most of the guest VMs partitions but not proxmox's.
also after importing the zfs pool I did "mdadm --assemble --scan" two additional partitions appeared and I dont know what they are ( the reason I assembled raid arrays after importing zfs is because before importing the rpool mdadm wont recognize any raid arrays)
its worth mentioning that I have selinux on permissive mode.
A clean install is my last option so before is there anything I can try? if there is nothing to do, moving forward I need some guidance on exporting my data to a clean install
system configuration:
software RAID 5
filesystem ZFS
storage : rpool zfs / rpool.
looking at server management logs apparently there was a power outage and ups was out of service for maintenance
booting to the default kernel:
cannot open "scripts/zfs" no such file or directory
kernel panic "not syncing attempted to kill init"
I was able to boot from recovery mode in grub to a previous kernel then after checking system journal I deleted the problem a malfunctioning virtual network card causing VMs to not start I removed the ovsitch, did an apt update and fix broken packages and install the new available kernel. After reboot I still get the kernel panic, and because I updated grub the previous kernel which was working is no longer available. Reinstalling system is out of the question so I need help to repair the kernel.
this is what I've tried
So far I imported the zfs pool to the live os I can see and mount some of the guest vm disks but I proxmox "/" (root directory) is not mounted.
I did "zpool import rpool" then 2 directories appeared rpool/DATA and rpool/ROOT/pve-1 but they are both empty.
with lsblk I can see most of the guest VMs partitions but not proxmox's.
also after importing the zfs pool I did "mdadm --assemble --scan" two additional partitions appeared and I dont know what they are ( the reason I assembled raid arrays after importing zfs is because before importing the rpool mdadm wont recognize any raid arrays)
its worth mentioning that I have selinux on permissive mode.
A clean install is my last option so before is there anything I can try? if there is nothing to do, moving forward I need some guidance on exporting my data to a clean install
system configuration:
software RAID 5
filesystem ZFS
storage : rpool zfs / rpool.
Last edited: