two independent installs but one has networking trouble

[fixjunk]

I am stumped and I have come here for help!

I have two proxmox installs that are pretty similar.

pve1: pve-manager/8.0.4/d258a813cfa6b390 (running kernel: 6.2.16-19-pve)
pve1 is also running LMDE6 as a workstation and a bunch of lxc and vm stuff.

pve2: pve-manager/8.0.4/d258a813cfa6b390 (running kernel: 6.2.16-19-pve)
pve2 has just two things running on it: a debian lxc for pihole and a FreeBSD vm for opnsense and that's it.
opnsense is set up for verizon fios IPV6 and my pihole(s) have 2600: addresses. so does the LAN interface. the WAN interface does not, only an fe80 link local (this is a verizon thing)

for some reason pve2 wants to do things over IPV6 and fails. it resolves an address and then fails to contact it. for example, this just hangs on PVE2:

root@pve2:~ $ ping -6 google.com
PING google.com(wv-in-f139.1e100.net (2607:f8b0:4004:c1b::8b)) 56 data bytes

note that PVE2 CAN reliably ping opnsense over ipv6.

pve1 realizes that IPV6 is not completely available and uses IPV4 just fine or errors out:

root@pve2:~ $ ping google.com -c 2
PING google.com (142.251.163.100) 56(84) bytes of data.
64 bytes from wv-in-f100.1e100.net (142.251.163.100): icmp_seq=1 ttl=60 time=3.67 ms
64 bytes from wv-in-f100.1e100.net (142.251.163.100): icmp_seq=2 ttl=60 time=3.80 ms

root@pve1:~ $ ping -6 google.com
ping: connect: Network is unreachable

this isn't typically a problem except when I try to do things with the web interface at https://pve2.localdomain.com:8006 (which is a legit domain with valid cert):

• logging in with openid fails because pve2 can't reach https://accounts.google.com
• downloading templates fails because pve2 can't reach http:download.proxmox.com
• other stuff

It used to work. possibly before I enabled IPV6 in opnsense on pve2.


how can I fix it?

 

[fixjunk]

I think I've resolved this.

I had an unused onboard lan that was bridged to the VM and getting IPV6 address. I disabled it and things started working properly.