I ran
To link the modules for the host to a container.
This produced the line in /etc/pve/lxc/100.conf od
Which caused an error in startup:
And searching did not help until I thought about it:
I had to modify the line to include the /usr in both place:
The question is, is this a bug in pct, or in the config of proxmox? I feel like the pct tool should not produce bad config on a vanilla system.
Also posting for future me when I hit this problem again and don't remember what I did to fix it.
Bash:
pct set 100 --mp0 /lib/modules/$(uname -r),mp=/lib/modules/$(uname -r),ro=1
To link the modules for the host to a container.
This produced the line in /etc/pve/lxc/100.conf od
Code:
mp0: /lib/modules/6.17.4-1-pve,mp=/lib/modules/6.17.4-1-pve,ro=1
Which caused an error in startup:
Code:
../src/lxc/utils.c:run_buffer:560 - Script exec /usr/share/lxc/hooks/lxc-pve-prestart-hook 100 lxc pre-start produced output: symlink encountered at: //lib
And searching did not help until I thought about it:
Code:
ls -l /
...
lrwxrwxrwx 1 root root 7 May 12 2025 lib -> usr/lib
lrwxrwxrwx 1 root root 9 May 12 2025 lib64 -> usr/lib64
I had to modify the line to include the /usr in both place:
Code:
mp0: /usr/lib/modules/6.17.4-1-pve,mp=/usr/lib/modules/6.17.4-1-pve,ro=1
The question is, is this a bug in pct, or in the config of proxmox? I feel like the pct tool should not produce bad config on a vanilla system.
Also posting for future me when I hit this problem again and don't remember what I did to fix it.