vmbr0 entered blocking state, then disabled, then blocking, then forwarding

[vescovialex]

Hello,

I've just purchased a used server Dell R730 and fresh installed on 2 SSD 2TB in Raid1 zfs Proxmox.

From the first start of the Virtual Environment I have this "issue" with the NIC.
This messages appears:

device eno1 entered promiiscuous mode
vmbr0: port 1(eno1) entered blocking state
vmbr0: port 1(eno1) entered disabled state
vmbr0: port 1(eno1) entered forwarding state

I've tried to change eno, to eno2, eno3 but the same message is coming out.

Effectively I'm not having issue with the network (or maybe I lost a ping when this message appear, who knows), this message is annoying and I don't understand why is going out.

The frequency of this message is totally random.

I have some Proxmox installation but this is the first time I've receive this messages.

Thanks

 

[floh8]

show /etc/interfaces?

 

[anktarius]

Hi,
I have the same issue here.

i installed proxmox on my NUC 11 two month ago.
My setup is simple :
1 NIC with this config :

auto lo
iface lo inet loopback

iface enp88s0 inet manual

auto vmbr0
iface vmbr0 inet static
        address 192.168.10.6/24
        gateway 192.168.10.1
        bridge-ports enp88s0
        bridge-stp off
        bridge-fd 0
        bridge-vlan-aware yes
        bridge-vids 2-4094

Few weeks ago i had a weird behavior (network instability due to a VM acting as dhcp server hosted on proxmox).
Today the same issue again ; Some client lost their connectivity due to lack of DHCP answer, proxmox is up and running (can ping), vm behind don't have any warning message.
I tried to reboot the VM etc. same thing until i reboot completely the proxmox server, i still don't understand why i have this (can be after few days or weeks), i had this on 7.3 and 7.4 (two different NUC)

Please find below the only line that appearing everytime i have this issue :

Apr  2 00:50:25 nuc11 kernel: [624233.977334] x86/split lock detection: #AC: CPU 0/KVM/1069 took a split_lock trap at address: 0xfffff8074401e643
Apr  2 01:00:25 nuc11 kernel: [624834.025564] x86/split lock detection: #AC: CPU 0/KVM/1069 took a split_lock trap at address: 0xfffff8074401e643
Apr  2 01:02:25 nuc11 kernel: [624954.027989] x86/split lock detection: #AC: CPU 0/KVM/1069 took a split_lock trap at address: 0xfffff8074401e643
.........
Apr  7 10:12:53 nuc11 kernel: [1089987.046048] x86/split lock detection: #AC: CPU 0/KVM/1069 took a split_lock trap at address: 0xfffff8074401e643
Apr  7 11:46:53 nuc11 kernel: [1095627.455498] x86/split lock detection: #AC: CPU 0/KVM/1069 took a split_lock trap at address: 0xfffff8074401e643
Apr  7 11:52:53 nuc11 kernel: [1095987.477464] igc 0000:58:00.0 enp88s0: NIC Link is Down
Apr  7 11:52:53 nuc11 kernel: [1095987.477493] vmbr0: port 1(enp88s0) entered disabled state
Apr  7 11:56:47 nuc11 kernel: [1096220.987351] nfs: server 192.168.2.13 not responding, timed out
Apr  7 11:59:47 nuc11 kernel: [1096401.017130] nfs: server 192.168.2.13 not responding, timed out
Apr  7 11:59:52 nuc11 kernel: [1096406.141064] nfs: server 192.168.2.13 not responding, timed out
Apr  7 12:00:23 nuc11 kernel: [1096436.830444] igc 0000:58:00.0 enp88s0: NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX
Apr  7 12:00:23 nuc11 kernel: [1096436.830709] vmbr0: port 1(enp88s0) entered blocking state
Apr  7 12:00:23 nuc11 kernel: [1096436.830714] vmbr0: port 1(enp88s0) entered forwarding state
Apr  7 12:06:53 nuc11 kernel: [1096827.529892] x86/split lock detection: #AC: CPU 0/KVM/1069 took a split_lock trap at address: 0xfffff8074401e643
Apr  7 12:20:53 nuc11 kernel: [1097667.577582] x86/split lock detection: #AC: CPU 0/KVM/1069 took a split_lock trap at address: 0xfffff8074401e643
Apr  7 12:37:13 nuc11 kernel: [1098647.569034] device tap100i1 entered promiscuous mode
Apr  7 12:37:13 nuc11 kernel: [1098647.608016] vmbr0: port 11(fwpr100p1) entered blocking state
Apr  7 12:37:13 nuc11 kernel: [1098647.608019] vmbr0: port 11(fwpr100p1) entered disabled state
Apr  7 12:37:13 nuc11 kernel: [1098647.608071] device fwpr100p1 entered promiscuous mode
Apr  7 12:37:13 nuc11 kernel: [1098647.608445] vmbr0: port 11(fwpr100p1) entered blocking state
Apr  7 12:37:13 nuc11 kernel: [1098647.608446] vmbr0: port 11(fwpr100p1) entered forwarding state
Apr  7 12:37:13 nuc11 kernel: [1098647.625312] fwbr100i1: port 1(fwln100i1) entered blocking state
Apr  7 12:37:13 nuc11 kernel: [1098647.625315] fwbr100i1: port 1(fwln100i1) entered disabled state
Apr  7 12:37:13 nuc11 kernel: [1098647.625364] device fwln100i1 entered promiscuous mode
Apr  7 12:37:13 nuc11 kernel: [1098647.625389] fwbr100i1: port 1(fwln100i1) entered blocking state
Apr  7 12:37:13 nuc11 kernel: [1098647.625390] fwbr100i1: port 1(fwln100i1) entered forwarding state

I am not able to understand the logs neither explain them (forwarding, blocking etc point me to the network with spanning-tree but it can be that as nothing was done on the network and no L2 issue on anything else)
If someone has an idea about what can be the root cause, i am interested, i have already checked the network (switch, etc.) so i think the issue is ont the server side

 

[floh8]

according to the PVE Docu u should configure your Host IP different. https://pve.proxmox.com/pve-docs/pve-admin-guide.html#sysadmin_network_configuration
A vlan-aware bridge tagged the untagged packets of the trunk by default to the vlan 1.

 

[anktarius]

according to the PVE Docu u should configure your Host IP different. https://pve.proxmox.com/pve-docs/pve-admin-guide.html#sysadmin_network_configuration
A vlan-aware bridge tagged the untagged packets of the trunk by default to the vlan 1.

You mean that i should configure this way ?

auto lo
iface lo inet loopback

iface eno1 inet manual


auto vmbr0.5
iface vmbr0.5 inet static
        address  10.10.10.2/24
        gateway  10.10.10.1

auto vmbr0
iface vmbr0 inet manual
        bridge-ports eno1
        bridge-stp off
        bridge-fd 0
        bridge-vlan-aware yes
        bridge-vids 2-4094

I am using the vlan 1 / native vlan for the management (bad but this is at home) so it is working but it is not best practice.
Do you think it can explain something about the logs ?
Thanks anyway for pointing me this !

 

[floh8]

yes, that i mean, but u use a different port naming.
u can configure a other native vlan with "bridge-pvid x".
maybe a loop is the reason but i dont understand why.

 

[anktarius]

yes, that i mean, but u use a different port naming.
u can configure a other native vlan with "bridge-pvid x".
maybe a loop is the reason but i dont understand why.

I have updated the config of my 2 NUCS ;

auto lo
iface lo inet loopback

iface enp88s0 inet manual

auto vmbr0
iface vmbr0 inet manual
        bridge-ports enp88s0
        bridge-stp off
        bridge-fd 0
        bridge-vlan-aware yes
        bridge-vids 2-4094

auto vmbr0.10
iface vmbr0.10 inet static
        address 192.168.10.6/24
        gateway 192.168.10.1
#Management

I have removed the native vlan as i don't need it in fact, so now everything is "proper"
Let's see if it's fix the issue

 

[hvisage]

vmbr0: port 1(eno1) entered blocking state
vmbr0: port 1(eno1) entered disabled state
vmbr0: port 1(eno1) entered forwarding state

That is the normal state that a bridge interface goes through (especially when spanning trees are configured) as it starts up.
Blocking state: don't forward any traffic to prevent loops
Forwarding: This interface doesn't make a loop, or the spanning tree coose it to forward traffic, so enable forwarding.
IF and only IF: you are sure this interface in not in any way possible to create a loop (ie. a host interface) you can turn off STP or make the times shorter than the normal 30 odd seconds that STP takes to converge