Dropped network packets from VM but host OK.

[gob]

I have a single host PVE version 8.1.4 with a pretty basic out of the box configuration
I have
2 debian 12 LXC containers
1 debian 12 VM
1 Windows 2019 Server VM

The PVE host itself has no problems accessing the outside world and shows consistent ping latency with now packet loss. However the linux CT and VM guests display packetloss when using vmbr0 and the same NIC as the host..
However the Windows VM on vmbr0 shows no packet loss.

Pinging the PVE host IP from the guest shows no dropped packets.
However the below is from pinging the network gateway from the guest:
ping 192.168.29.1
PING 192.168.29.1 (192.168.29.1) 56(84) bytes of data.
64 bytes from 192.168.29.1: icmp_seq=7 ttl=64 time=0.188 ms
64 bytes from 192.168.29.1: icmp_seq=8 ttl=64 time=0.175 ms
64 bytes from 192.168.29.1: icmp_seq=9 ttl=64 time=0.168 ms
64 bytes from 192.168.29.1: icmp_seq=10 ttl=64 time=0.180 ms
64 bytes from 192.168.29.1: icmp_seq=11 ttl=64 time=0.183 ms
64 bytes from 192.168.29.1: icmp_seq=12 ttl=64 time=0.190 ms
64 bytes from 192.168.29.1: icmp_seq=13 ttl=64 time=0.174 ms
64 bytes from 192.168.29.1: icmp_seq=14 ttl=64 time=0.186 ms
64 bytes from 192.168.29.1: icmp_seq=15 ttl=64 time=0.179 ms
64 bytes from 192.168.29.1: icmp_seq=16 ttl=64 time=0.195 ms
64 bytes from 192.168.29.1: icmp_seq=44 ttl=64 time=0.182 ms
64 bytes from 192.168.29.1: icmp_seq=80 ttl=64 time=0.193 ms
64 bytes from 192.168.29.1: icmp_seq=81 ttl=64 time=0.151 ms
64 bytes from 192.168.29.1: icmp_seq=82 ttl=64 time=0.233 ms
64 bytes from 192.168.29.1: icmp_seq=83 ttl=64 time=0.144 ms

Note how the first 6 pings are dropped and then seq 17 to 43 and 45 to 79 etc.
The same ping from the host is fine.
The system is under no heavy load - < 3% CPU and 10% RAM
Any suggestions on how to troubleshoot this further?

Thanks

 

[Guillaume Delanoy]

Hello,

Please post the content of /etc/network/interfaces of your PVE server, as well as network configuration (ip a + ip r) of all his hosts.

Kind regards,

GD

 

[gob]

Hi @GuillaumeDelaney

I'm not sure exactly what you mean by (ip a + ip r) but here's the basic network info.

PVE Host:

auto lo
iface lo inet loopback

iface eno1 inet manual

iface eno3 inet manual

iface eno4 inet manual

iface eno2 inet manual

auto vmbr0
iface vmbr0 inet static
address 192.168.29.20/23
gateway 192.168.29.1
bridge-ports eno1
bridge-stp off
bridge-fd 0

auto vmbr1
iface vmbr1 inet static
address 10.10.29.20/24
bridge-ports eno4
bridge-stp off
bridge-fd 0
#DMZ

Debian 12 LXC:

auto lo
iface lo inet loopback

auto eth0
iface eth0 inet static
address 192.168.29.21/23
gateway 192.168.29.1





Windows 2019 Server VM

 

[Guillaume Delanoy]

Hello,

"ip a" is the short order for "ip address show", and
"ip r" is the short order for "ip route show".

Simply type them in a terminal to get interesting network related infos.

But it's okay, since the print screens and configs you sent also give us that information

In your IP configuration, IP ranges given to each interface and to each VM are overlapping each other. Moreover, some have been given the same IP address.
vmbr0 : address 192.168.29.20/23
vmbr1 : address 192.168.29.20/24 same address, and the subnet is overlapping with vmbr0's.

Also, your Windows VM has the same address as your eth0 PVE interface.

Avoid overlapping, solve your IP address conflicts and you should solve your dropped packets issue .

Kind regards,


GD

 

[gob]

Thanks GD

Really don't know how I missed that IP clash. I appreciate the fresh pair of eyes on it.

Oh, the two VMBR's are on different subnets... one is 192.168.29.20/23 the other is 10,10,29.20/24

Cheers!