[SOLVED] Issues with interface bonding

[aljaxus]

Hi, I'm having some issues with interface bonding. I have four interfaces on my server (running Proxmox VE 6.3-3).
The network config summary is as follows;
- vm1 -> vmbr0
- vm2 -> vmbr0
- vm3 -> vmbr0
- vm4 -> vmbr0
- vm5 -> vmbr0
- vmbr0 -> bond0
- bond0 -> enp4s0f0 enp4s0f1

enp* are physical interfaces.

I have tested with all four network interfaces bound to bond0 (enp3s0f0 enp3s0f1 enp4s0f0 enp4s0f1), but that doesn't work either.
All interfaces are "active" and have "autostart" enabled.

Spoiler: /etc/network/interfaces

root@apollo:~# cat /etc/network/interfaces.new
# 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 enp3s0f0
iface enp3s0f0 inet manual

auto enp3s0f1
iface enp3s0f1 inet manual

auto enp4s0f0
iface enp4s0f0 inet manual

auto enp4s0f1
iface enp4s0f1 inet manual

auto bond0
iface bond0 inet manual
        bond-slaves enp4s0f0 enp4s0f1
        bond-miimon 100
        bond-mode balance-rr

auto vmbr0
iface vmbr0 inet static
        address 192.168.1.100/24
        gateway 192.168.1.1
        bridge-ports bond0
        bridge-stp off
        bridge-fd 0

All VMs can reach the "outside" internet, but they can't always reach each other.
Tracepath does finish from given VM to another one, but with ~2-3 second delay.
When a VM is unable to reach another one, it's unable to ping or reach it by any means, while the tracepath still does finish.

As soon as I switch the vmbr0 slaves from bond0 to any physical interface (eg. enp3s0f0) the connectivity is perfectly fine.
The switch that all four interfaces are connected to is a "HPE V1910-24G Switch JE006A" (in case anyone knows that these have any issues.. ?). Operating as a "dumb" switch.

 

[mira]

Try active-backup instead of balance-rr, or LACP if both the NICs and the switch support it.

 

[aljaxus]

Try active-backup instead of balance-rr, or LACP if both the NICs and the switch support it.

Thanks! Switching to "LCAP" option worked like a charm.