interface configuration being ignored

[zeegull]

UPDATE:

I found the problem to be caused by incorrect permission (no execution permission) for /usr/share/ifupdown2/__main__.py

I made it executable as root and it loaded interface configuration. I'm not sure if it is a bug I triggered by doing the upgrade.

Permission after the fix


Hi,

I have a 3 node pve setup with no-subscription. One of the nodes suddenly lost its network configuration when I ran the following

apt update
apt upgrade -y
reboot

after reboot, the interface file is intact, but when I try to do

systemctl restart networking

it is unusually fast and the node is no longer accessible. upon closer look, all nics are configured to manual. The interfaces file is not being parsed and the systemctl status networking showed initialization being successful without any error (and active). any modification in /etc/network/interfaces appear to have no effect when I run systemctl restart networking.

The following is my config on another node, where all nodes are identical

# network interface settings; autogenerated
# Please do NOT modify this file directly, unless you know what
# you're doing.
#
# If you want to manage parts of the network configuration manually,
# please utilize the 'source' or 'source-directory' directives to do
# so.
# PVE will preserve these directives, but will NOT read its network
# configuration from sourced files, so do not attempt to move any of
# the PVE managed interfaces into external files!

auto lo
iface lo inet loopback

auto eno1
iface eno1 inet manual

iface eno3 inet manual

iface eno4 inet manual

auto eno2
iface eno2 inet manual

auto enp59s0
iface enp59s0 inet manual

auto enp59s0d1
iface enp59s0d1 inet manual

auto bond0
iface bond0 inet manual
        bond-slaves eno1 eno2
        bond-miimon 100
        bond-mode 802.3ad
        bond-xmit-hash-policy layer2+3

auto bond1
iface bond1 inet static
        address 10.1.1.1/24
        bond-slaves enp59s0 enp59s0d1
        bond-miimon 100
        bond-mode 802.3ad
        bond-xmit-hash-policy layer2+3
#cluster network

auto vmbr0
iface vmbr0 inet static
        address 192.168.101.1/16
        gateway 192.168.200.1
        bridge-ports bond0
        bridge-stp off
        bridge-fd 0
        bridge-vlan-aware yes
        bridge-vids 2-4094

auto vmbr0.30
iface vmbr0.30 inet static
        address 10.2.2.1/24
#ceph public network

auto vmbr0.20
iface vmbr0.20 inet static
        address 10.3.3.1/24
#cluster comm

source /etc/network/interfaces.d/*

 

[gbilbatua]

Same problem. I've had my server running for a year without any issues. A couple of days ago, after I upgraded from kernel 6.8.8-1-pve to 6.8.8-2-pve, the system refused to boot.

I downgraded to a previous KERNEL so that the system could boot but all the nics remain down, refuse to start and it's no longer possible to reach the web interface. Besides, the existing CT's and VM's keep stopped.

Now I can only interface with the server by using a monitor and a keyboard.

Running journalctl -xe I get among other information:
start failed: QEMU exited with code 1
bridge 'vmbr1' does not exist, and the same for the other vmbrXX

The following setup is the one I've had running.



This is my HW:



I hope someone can help. I'm blocked. (Hopefully I'm running a backup server)

Thanx.

 

[zeegull]

Same problem. I've had my server running for a year without any issues. A couple of days ago, after I upgraded from kernel 6.8.8-1-pve to 6.8.8-2-pve, the system refused to boot.

I downgraded to a previous KERNEL so that the system could boot but all the nics remain down, refuse to start and it's no longer possible to reach the web interface. Besides, the existing CT's and VM's keep stopped.

Now I can only interface with the server by using a monitor and a keyboard.

Running journalctl -xe I get among other information:
start failed: QEMU exited with code 1
bridge 'vmbr1' does not exist, and the same for the other vmbrXX

The following setup is the one I've had running.

View attachment 71975

This is my HW:

View attachment 71976

I hope someone can help. I'm blocked. (Hopefully I'm running a backup server)

Thanx.

I suspect it is similar to my case, check permission for files inside /usr/share/ifupdown2/ and try to use ifup manually

 

[shanreich]

I downgraded to a previous KERNEL so that the system could boot but all the nics remain down, refuse to start and it's no longer possible to reach the web interface. Besides, the existing CT's and VM's keep stopped.

How does the output of ip a look like?
Can you post the output of ifreload -ad ?

You also have the same subnet configured twice (10.71.87.0/24) - which shouldn't be the cause for that issue - but nevertheless something you should look at after you fixed your network devices not coming up.

 

[gbilbatua]

Thank you zeegull and shanreich for your response!!!!!!!

I tried zeegull's suggestion and the problem with the netword is gone. The ugo x permit was missing for /usr/share/__main__.py.

On the other hand, I found out that the /boot drive was full. I purged some old kernels and I could upgrade the system and now I'm able to boot it with the newest kernel version.

 

[gbilbatua]

Sorry, the right path is /usr/share/ifupdown2/__main__.py