Error: Nexthop has invalid gateway

[The Real Bev]

I just installed slack 15 on the BMAX B1 Pro. Everything worked. Then
I did slackpkg update and slackpkg upgrade-all and now I have no
ethernet. When I try to run rc.inet1 I get "Error: Nexthop has invalid
gateway" and no eth0. As best I can figure, Nexthop has something to do
with postfix, which isn't running. Any help will be appreciated!

[The Real Bev]

Post by The Real Bev
I just installed slack 15 on the BMAX B1 Pro. Everything worked. Then
I did slackpkg update and slackpkg upgrade-all and now I have no
ethernet. When I try to run rc.inet1 I get "Error: Nexthop has invalid
gateway" and no eth0. As best I can figure, Nexthop has something to do
with postfix, which isn't running. Any help will be appreciated!

I concluded that the problem was that after updating the kernel I left
elilo as it was. I repeated the complete installation/upgrade process,
but this time I ran eliloconfig before rebooting. Now it doesn't finish
booting == "No kernel modules found for linux 5.15.145" -- and the
system is locked up.

What should I have done when the kernel was updated? Is there any way I
can recover without going through the whole process again?

[Henrik Carlqvist]

Post by The Real Bev
What should I have done when the kernel was updated? Is there any way I
can recover without going through the whole process again?

From the ChangeLog.txt:

+--------------------------+
Tue Dec 26 00:20:26 UTC 2023
patches/packages/kernel-firmware-20231222_a7dee43-noarch-1.txz: Upgraded.
Updated to the latest kernel firmware.
patches/packages/linux-5.15.145/*: Upgraded.
These updates fix various bugs and security issues.
Thanks to jwoithe for the PCI fix!
Be sure to upgrade your initrd after upgrading the kernel packages.
If you use lilo to boot your machine, be sure lilo.conf points to the
correct kernel and initrd and run lilo as root to update the bootloader.
If you use elilo to boot your machine, you should run eliloconfig to
copy the kernel and initrd to the EFI System Partition.
For more information, see:
Fixed in 5.15.140:
https://www.cve.org/CVERecord?id=CVE-2023-46862
Fixed in 5.15.141:
https://www.cve.org/CVERecord?id=CVE-2023-6121
(* Security fix *)
+--------------------------+

The fact that you now do not have any eth0 is most likely because no
kernel module has been loaded for your network card.

If you are running a generic kernel with an initrd instead of the huge
kernel you do not only have to worry about the modules installed on your
root file system, you also have your initrd to worry about.

Did you upgrade to the kernel-modules-5.15.145-*-1.txz package?
Did you rebuild any initrd if you are booting with a generic kernel?

It is possible to boot from the Slackware installation media and point
its kernel to your root partition as described in the boot message:

"In a pinch, you can boot your system from here with a command like: ..."

regards Henrik

[The Real Bev]

Post by Henrik Carlqvist

Post by The Real Bev
What should I have done when the kernel was updated? Is there any way I
can recover without going through the whole process again?

+--------------------------+
Tue Dec 26 00:20:26 UTC 2023
patches/packages/kernel-firmware-20231222_a7dee43-noarch-1.txz: Upgraded.
Updated to the latest kernel firmware.
patches/packages/linux-5.15.145/*: Upgraded.
These updates fix various bugs and security issues.
Thanks to jwoithe for the PCI fix!
Be sure to upgrade your initrd after upgrading the kernel packages.
If you use lilo to boot your machine, be sure lilo.conf points to the
correct kernel and initrd and run lilo as root to update the bootloader.
If you use elilo to boot your machine, you should run eliloconfig to
copy the kernel and initrd to the EFI System Partition.
https://www.cve.org/CVERecord?id=CVE-2023-46862
https://www.cve.org/CVERecord?id=CVE-2023-6121
(* Security fix *)
+--------------------------+
The fact that you now do not have any eth0 is most likely because no
kernel module has been loaded for your network card.
If you are running a generic kernel with an initrd instead of the huge
kernel you do not only have to worry about the modules installed on your
root file system, you also have your initrd to worry about.
Did you upgrade to the kernel-modules-5.15.145-*-1.txz package?
Did you rebuild any initrd if you are booting with a generic kernel?
It is possible to boot from the Slackware installation media and point
"In a pinch, you can boot your system from here with a command like: ..."

The problem seems to be that after I upgraded I had to upgrade elilo,
but I didn't know that and it's trying to boot with the old kernel,
which is not there. I can't boot from the install because the install
boots 5.15.19 and the updated kernel is 5.15.145.

I completely went through a new install, after which I installed grub
with grub-mkconfig to write in /boot/grub/grub.cfg. Then I rebooted and
made sure that grub worked. Then I redid the update, updating
everything to the latest 15. And now it's back to running with elilo.
When I installed grub, elilo should have been out of the picture.

[root]

Post by The Real Bev
I completely went through a new install, after which I installed grub
with grub-mkconfig to write in /boot/grub/grub.cfg. Then I rebooted and
made sure that grub worked. Then I redid the update, updating
everything to the latest 15. And now it's back to running with elilo.
When I installed grub, elilo should have been out of the picture.

Your problem came from the grub-mkconfig command which asked
to load the old kernel.

[root]

Post by The Real Bev

Post by The Real Bev
I just installed slack 15 on the BMAX B1 Pro. Everything worked. Then
I did slackpkg update and slackpkg upgrade-all and now I have no
ethernet. When I try to run rc.inet1 I get "Error: Nexthop has invalid
gateway" and no eth0. As best I can figure, Nexthop has something to do
with postfix, which isn't running. Any help will be appreciated!

I concluded that the problem was that after updating the kernel I left
elilo as it was. I repeated the complete installation/upgrade process,
but this time I ran eliloconfig before rebooting. Now it doesn't finish
booting == "No kernel modules found for linux 5.15.145" -- and the
system is locked up.
What should I have done when the kernel was updated? Is there any way I
can recover without going through the whole process again?

Go through the install process again. This time, make a boot usb.
Do not install elilo (EVER).

After the installation, reboot the machine using the boot usb.

run grub-mkconfig >/boot/grub/grub.cfg
edit that file to delete menu entries that do not include
those you want. In particular, change entries that look like
this: linux /boot/vmlinuz-...... and delete the stuff
like 5.15.59, or whatever. The only boot kernel entries
should be /boot/vmlinuz

Run grub-install XXX

where XXX is the device that contains the slackware installation.
For example, if the installation went to mmcblk0p5, the grub
command would be:
grub-install /dev/mmcblk0

[Petri Kaukasoina]

Post by root
run grub-mkconfig >/boot/grub/grub.cfg
edit that file to delete menu entries that do not include
those you want. In particular, change entries that look like
this: linux /boot/vmlinuz-...... and delete the stuff
like 5.15.59, or whatever. The only boot kernel entries
should be /boot/vmlinuz

Or edit /etc/grub.d/10_linux first. Search for the first line with word
'vmlinuz'. It's this:

for i in /boot/vmlinuz-* /vmlinuz-* /boot/kernel-* ; do

Change it to:

for i in /boot/vmlinuz ; do

After that change grub-mkconfig won't create versioned kernel entries.

[root]

Post by Petri Kaukasoina

Post by root
run grub-mkconfig >/boot/grub/grub.cfg
edit that file to delete menu entries that do not include
those you want. In particular, change entries that look like
this: linux /boot/vmlinuz-...... and delete the stuff
like 5.15.59, or whatever. The only boot kernel entries
should be /boot/vmlinuz

Or edit /etc/grub.d/10_linux first. Search for the first line with word
for i in /boot/vmlinuz-* /vmlinuz-* /boot/kernel-* ; do
for i in /boot/vmlinuz ; do
After that change grub-mkconfig won't create versioned kernel entries.

Entries such as that must only be for Slackware, and I never would
have guessed that a special section would be provided. That's ectremely
handy, Thanks.

[Marco Moock]

Post by The Real Bev
and no eth0.

In such cases, run
lspci -nnk
to see if a kernel module is loaded and to see the device ID.

Also check dmesg for messages about that module.