[SOLVED] Proxmox 8.1 Kernel 6.5.11-4 network issue.

[anoo222]

Hi all,

After updating proxmox to kernel 6.5.11-4, i have network issues. The host is booting, but i can't connect to it, neither can i ping my router from the host itself.

If i boot with kernel 6.2.16-19, my host is behaving normally.

Below an attached screenshot with more information. Any suggestions?

Thanks in advance!

 

[cheiss]

Hi,

can you please post the output of ip address, and possible the output of dmesg -H (as file)?

Since it does say bridge port enp1s0 does not exist, either the interface gets renamed differently under 6.5 or the driver does not load correctly.

 

[anoo222]

Hi, sure!

I will get my host from basement again and reboot it with kernel 6.5 with a screen and keyboard attached and run both commands.

I remember from ip a that there was an ip (192.168.1.36) assigned, dont know if this could be of any help.

Could u please tell me how i could provide u with dmesg as a file since i only have acces to the host with kernel 6.5 with a screen and keyboard, i can't transfer the file over network since i have no connection, and currently don't find a spare usb . Are there any other methods u are aware of?

Thanks!

 

[anoo222]

Hi again cheiss,

I've managed to borrow a usb from my neighbor.

The output of ip address;
(I've removed ipv6 address and 63 entries from containers & vm's, let me know please if u would need them)

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host noprefixroute valid_lft forever preferred_lft forever 3: vmbr0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000 link/ether aa:d8:f4:a3:f3:a4 brd ff:ff:ff:ff:ff:ff inet 192.168.1.36/24 scope global vmbr0 valid_lft forever preferred_lft forever inet6 xxxx/64 scope link valid_lft forever preferred_lft forever

As for the output of dmesg -H, i'm not sure if it could contain any sensitive information, therefore i'll send it to u over private message.

Thanks

 

[ravage2k]

There is a problem with the latest pve-firmware package (3.9-1).
The network driver cannot be loaded.
It affects all kernel versions. I fixed it by downgrading the package version.

apt install pve-firmware=3.8-5
update-initramfs -u

If you booted another kernel you may have to specify the initramfs to update with the -k parameter instead of -u

1b:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller (rev 15)

 

[anoo222]

There is a problem with the latest pve-firmware package (3.9-1).
The network driver cannot be loaded.
It affects all kernel versions. I fixed it by downgrading the package version.

apt install pve-firmware=3.8-5
update-initramfs -u

If you booted another kernel you may have to specify the initramfs to update with the -k parameter instead of -u

1b:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller (rev 15)

Hi, i just tried this method with no succes.

I'll reproduce my steps, maybe i did something wrong;

I've booted my host with kernel 6.2.16-19, to downgrade pve-firmware to 3.8-5 (Since i have no internet connection with kernel 6.5).
After downgrading, i used the command update-initramfs -u -k 6.5.11-4-pve.
I've rebooted with kernel 6.5, and still had the same issue, host is booting, but no connection possible & not able to ping router.
I've checked with apt list --installed | grep pve-firmware, and the version was indeed 3.8-5.
I've reverted those steps and for now i will pin kernel 6.2 with proxmox-boot-tool.

 

[anoo222]

Hi,

I think i can pinpoint the cause of this issue, i first looked if maybe the interface gets named differently as told by @cheiss, but the interface isn't even created.

I remember installing r8168-dkms, after the host regullary freezing after updating to 8 (https://forum.proxmox.com/threads/updated-from-7-4-to-8-since-regulary-freezing.130259/#post-571294).

lspci on 6.2 kernel

01:00.0 Ethernet controller [0200]: Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller [10ec:8168] (rev 15)
        Subsystem: Dell RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller [1028:085c]
        Kernel driver in use: r8168
        Kernel modules: r8168

lspci on 6.5 kernel

01:00.0 Ethernet controller [0200]: Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller [10ec:8168] (rev 15)
        Subsystem: Dell RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller [1028:085c]       
        Kernel modules: r8168

Note that on 6.5 kernel, there is no mention of kernel driver in use.

I've tried to modprobe r8168, but it was not there. r8169 however was there, so i tested with r8169.
After that i could ping my home router, but no internet connection.

Now i think booting 6.5 kernel & installing r8168-dkms maybe will fix this issue, but i'm not sure if this is the correct way.
If someone more knowledgeable would be able to guide me the correct way to reslove this issue, it would be highly appreciated.
I like trying things on vm's & containers, but on the host itself i want to make sure i don't break anything.

Thanks!

 

[anoo222]

Currently solved.

As suggested here: https://forum.proxmox.com/threads/a...3-network-not-work-anymore.136831/post-609005, I removed r8168-dkms package & the configuration file from modprobe.d.

Currently running 6.5 with network, using r8169 kernel driver.
Hopefully the host won't freeze regulary as it did from updating to 7.4 to 8, why i initially needed to install r8168-dkms and blacklist r8169.

 

[strausmann]

Currently solved.

As suggested here: https://forum.proxmox.com/threads/a...3-network-not-work-anymore.136831/post-609005, I removed r8168-dkms package & the configuration file from modprobe.d.

Currently running 6.5 with network, using r8169 kernel driver.
Hopefully the host won't freeze regulary as it did from updating to 7.4 to 8, why i initially needed to install r8168-dkms and blacklist r8169.

for me the same ... Dell OptiPlex 3070 r8168-dkms removed

 

[wolpi]

Alternatively, you can use the new version of the r8168-dkms package. Since version 8.051.02-3, Kernel 6.5 is supported by the package. You only need to update dkms as well as the new version requires dkms >= 3.0.11. Also see my post HERE.

 

[FishBoy]

Currently solved.

As suggested here: https://forum.proxmox.com/threads/a...3-network-not-work-anymore.136831/post-609005, I removed r8168-dkms package & the configuration file from modprobe.d.

Currently running 6.5 with network, using r8169 kernel driver.
Hopefully the host won't freeze regulary as it did from updating to 7.4 to 8, why i initially needed to install r8168-dkms and blacklist r8169.

Can you give an update concerning the freezes? Are they reoccuring for you is it working on the r8169 driver? Curious to know as I am still holding the update back for this reason.

 

[anoo222]

Since updating, my node is on 24/7 without a freeze. I can almost confidently say it is working on the r8169 driver.

 

[Johan-P]

I got similar problem when I read this. However, I do not use Dell Optiplex, so I am not sure the source of problem.
I tried to reboot several times, it is useless....
I use Proxmox VE 8.1.4
The problem happened when tried to update. The update is success. Whenever I tried to reboot, I saw on the server monitor. it works. However, I cannot connect from the web. the network failed.
Is there anyone can help?

 

[anoo222]

Looks similar to the issue i had a while back, could u check with lspci what network controller u have on your node?

 

[Johan-P]

Thank you for your fast response. Give a minute, I will check with the lspci

 

[Johan-P]

Looks similar to the issue i had a while back, could u check with lspci what network controller u have on your node?

Here is the network Controller that I collect with lspci command (attached here). basically there are 3 slots with Intel network X722

 

[anoo222]

Hmm seems like you don't have the same network controller as i did when i had this issue, so i can't be of much help here unfortunatly.

Hopefully someone more knowledgable can assist u here.

What i would do currently in your case is check with lspci aswell the kernel module in use on your network controller. (Don't know exact command as im on my phone), then use 'proxmox-boot-tool kernel pin xxxx...' to revert to the previous kernel which worked before u updated, and check again kernel module in use on your network controller.

If u compare these, possibly this may point u to the issue, and someone else maybe can assist u then on how to resolve it.

U could probably temporarly use the previous kernel till u know the cause & solution.