Debian 12 upgrade to kernel 6.1.0-34 & network interface issue

yoursunny

@Fritz said:
Any idea to prevent this issue?

Just updated mine and not dare to restart now.

Mentally strong people configure networking the Debian way and confidently restart their server.

https://www.freedesktop.org/software/systemd/man/latest/systemd-networkd.service.html

Carlin0

@david said:
I had this issue today with my vps at: Virtono, HostSailor, and HostDare.

@virtono @HostSailor @hostdare

david

@Fritz said:
Any idea to prevent this issue?

Just updated mine and not dare to restart now.

Is your ethernet device named eth0 right now? If it's already using a different name and working, then it's probably fine.

You can check ifconfig to see what it's using.

Otherwise, check /etc/default/grub to make sure the file is there and it has these options in the cmd line:

GRUB_CMDLINE_LINUX="net.ifnames=0 biosdevname=0"

If it's already there, it should be OK to reboot. If not, edit the file to include those options and then do an "update-grub"

If the file is missing, maybe there's a grub.ucf-dist file you can copy, and verify it's set correctly, then do an "update-grub"

Then it should be OK to reboot.

TimboJones

I upgraded two proxmox servers and had network issues for both. One was related to Realtek 5Gbps nic driver and pve-headers and the other looks like crash and reboot, several times. /shrug

Fritz

@david said:

@Fritz said:
Any idea to prevent this issue?

Just updated mine and not dare to restart now.

Is your ethernet device named eth0 right now? If it's already using a different name and working, then it's probably fine.

You can check ifconfig to see what it's using.

Otherwise, check /etc/default/grub to make sure the file is there and it has these options in the cmd line:

GRUB_CMDLINE_LINUX="net.ifnames=0 biosdevname=0"

If it's already there, it should be OK to reboot. If not, edit the file to include those options and then do an "update-grub"

If the file is missing, maybe there's a grub.ucf-dist file you can copy, and verify it's set correctly, then do an "update-grub"

Then it should be OK to reboot.

Here is mine:

GRUB_DEFAULT=0                                                    GRUB_TIMEOUT=5                                                    GRUB_DISTRIBUTOR=`lsb_release -i -s 2> /dev/null || echo Debian`
GRUB_CMDLINE_LINUX_DEFAULT=""
GRUB_CMDLINE_LINUX="console=tty0 console=ttyS0,115200 earlyprintk>GRUB_TERMINAL="console serial"
GRUB_SERIAL_COMMAND="serial --speed=115200"

and unfortunately it uses eth0.

david

@Fritz said: Here is mine:

and unfortunately it uses eth0.

You can try inserting the "net.ifnames=0 biosdevname=0" in GRUB_CMDLINE_LINUX and running "update-grub" before rebooting.

Or you can just make sure vnc is working, and if it has a problem, you can make that change over vnc and reboot again.

I had a similiar GRUB_CMDLINE_LINUX on my vps at Host-C that uses eth0, and I didn't have a problem with that one, but it was using a slightly different "-cloud" kernel "6.1.0-34-cloud-amd64".

GRUB_CMDLINE_LINUX="console=tty0 console=ttyS0,115200 earlyprintk=ttyS0,115200 consoleblank=0"

w3e4r5

thanks for ur hint...