Multi VLAN Setup driving me crazy

[SofaKingBoring]

Hello erveryone!

I really need some help...
I'm neither a professional administrator nor a sophisticated amateur, but I have my Proxmox Hypervisor running with guests in 5 different VLANs and also docker running on a host with containers in different VLANs on the same host.

But my first TrueNAS Scale setup is driving me nuts, just with the basic setup!
I have spent several days on this with little success, so I hope someone can point me to my idiocy:
I'd like to start simple with 2 VLANs: 101 (Server/Management) and 102 (Client/AD).

I failed to get even the simplest setup running until I stumbled accross THIS post and the corresponding VISIO.
This was more or less the opposite of what I was trying before (phys. interface -> bridge -> VLAN(s) how I'm used to from proxmox).

VLANs have parent interface eno1, br102 has member vlan102
This works so far, WebGUI is bound to 10.12.1.202.
I have a Test VM running, connected to br102 with static IP that's working as well.
Because my Servers/VMs generally reside in vlan101 (same as the WebGUI) I want to move my test VM to vlan101, which would require a br101 with vlan101 as a member.
As soon as I do this, I lose access to the GUI, while the rest continues to work, as far I could see.

I'm either doing something fundamentally wrong, or it's just a tiny problem with the GUI configuration.
Either way I'm lost.

Thanks and Regards
Chris

 

[sretalla]

If you're assigning a bridge to the interface where TrueNAS is running (and in general if you want broadcast to work properly) you need to assign the IP address of TrueNAS to the bridge, not the VLAN.

 

[SofaKingBoring]

After some hiccups with my test VM that actually worked!
I have VMs running in both VLANs now and GUI works as well.
Thank you very much.

It took me so much time, that I was not thinking clearly anymore, although I'm pretty sure I tried it like that before (of course.)
Anyway thanks again.

 

[sretalla]

Great that it worked for you.

Just to have a good record of the thing that works (as your original post contains a version of it that's a bit incorrect:

Physical NIC(s) -> LAGG (If you're doing that) -> VLAN (if you're doing that) -> Bridge (IP address of TrueNAS goes here if you're using one for this connection) -> VMs or services attached to the bridge.

 

[timtaler]

Hey guys,

I´m struggeling with the same problem and also tried a lot. My setup ist
TrueNAS Scale 22.12.1, migrated from TrueNAS Core at a SuperMicro Superserver

Physical NIC enp1s0f0
-> vlan1 -> br1 (with TrueNAS Scale IP 10.0.11.11/26)
-> vlan701 -> br701 (VMs, no TrueNAS IP anymore, 10.0.11.83/28)
-> vlan799 -> br799 (TrueNAS DHCP of a different LAN, Samba Share, reachable from it)

According to this thread, there is no untagged data at NIC enp1s0f0 anymore.

My Problem resides at vlan701/br701: There are a couple of VMs and Apps running for testing purposes, meanwhile. Each of this servicves can talk to each other, also to a TrueNAS IP (which was 10.0.11.82 for testing).

BUT: its not possible to talk to the router outside at 10.0.11.81 (VLAN701), no way in any tested constellation.

If I switch a VM to br1 (and changing the IP inside, of course), everything works well. So the VMs are not the problem. Also virtio vs. et1000 arent to point.
If there is a TrueNAS IP at this bridge701 or not, isnt also a problem. Also after each network configuration test, a reboot was done.
All the Linux known config files and tools are useless, because TrueNAS seems to do this completely different.

Hopefully, you have some suggestions or ideas to prevent me jumping out of the window...

Kind regards,
Tim