pfSense not enabling port

georgelza

pfSense is on first topton, others wil run Proxmox, already have one... need to order the balance and the aggregation switch which are the 8 additional ports.

trying to get this one working, onto the switch... implying the card/SFP+'s etc all works together.

G

Gblenn

@georgelza Thing is though, pfsense runs on freebsd, not Linux, and what you are seeing here is likely a driver issue related to freebsd.

Proxmox on the other hand is Debian which will have different drivers, and Unifi switches are also Linux based. So it may be so that you can use your Dell modules in all your other machines, not just the one running pfsense.

But the only way to find out is to test it... Plug a module into the Proxmox boxe, connect it to the Unifi switch and you will see immediately if it works or not...

stephenw10

@georgelza said in pfSense not enabling port:

you forgetting i'm re-assigning the main lan interface... i have one lan interface over which i run multiple vlans, this is the link between me... wifi into core switch or hard wired into core switch.

Can you not add one of the other igc NICs as a management interface at least temporarily?

georgelza

@Gblenn

Hehehe
Exactly what I started cabling this morning.
. Will advise ltr.

Curious, what dif would a DAC cable make as that’s just everything pre packaged, what I’m getting from our problem is pfSense not liking the Dell/EMC SFP+ itself.

G

georgelza

@stephenw10
I have a spare usw flex mini. Going to see if I can configure a management network using that directly into the pfSense. As a backup/backdoor
G

Gblenn

@georgelza said in pfSense not enabling port:

@Gblenn

Hehehe
Exactly what I started cabling this morning.
. Will advise ltr.

Curious, what dif would a DAC cable make as that’s just everything pre packaged, what I’m getting from our problem is pfSense not liking the Dell/EMC SFP+ itself.

G

There is actually a bit of software (firmware) in all those modules (transceivers). And it is this firmware may which is giving you the incompatibility issue here. So either you have to solve it in pfsense, with a driver that works with the DELL module. Or replace the module with something different.

And no it doesn't have to be a DAC cable... a fiber module from a different vendor may also work. It's just that DAC's are typically cheaper...

Gblenn

@georgelza said in pfSense not enabling port:

@stephenw10
I have a spare usw flex mini. Going to see if I can configure a management network using that directly into the pfSense. As a backup/backdoor
G

Or do what I suggested, use the ix0/1 as the testing connections towards your USW instead. Keep the working stuff as is and don't start reassigning interfaces until things are working with the 10G ports.

georgelza

@Gblenn

Need to figure out how to get this done... as there is allot of common bits here.
and sharing...

I got a 2nd topton with Proxmox on it... patched that into the Unifi Pro Max, SFP2.
configure Proxmox to use the fiber port ix0 as a 2nd bridge. the port is alive, i can ping the port from local. but i can't ping out, which tells me something is wrong more somewhere... the port is active as far as proxmox is concerned, if I can get this working then at least I know the hw is compatible on both sides... aka (problem sits inside pfSense).

I have ordered a DAC cable and a 2nd SFP+, different brand, allot cheaper than these enterprise level Dell/EMC's.

G

Gblenn

@georgelza said in pfSense not enabling port:

@Gblenn

Need to figure out how to get this done... as there is allot of common bits here.
and sharing...

I got a 2nd topton with Proxmox on it... patched that into the Unifi Pro Max, SFP2.
configure Proxmox to use the fiber port ix0 as a 2nd bridge. the port is alive, i can ping the port from local. but i can't ping out

So the bridge you created, how are you using it for thist testing / pinging?
Like, do you have a VM running on Proxmox that has the SFP assigned to it?

Here is what it looks like from a machine of mine, where enp9s0 is the motherboard NIC and enp10s0f0, f1 are the two ports on my X520 card, which I have assigned vmbr1 and 2 respectively.

So if you SSH in to Proxmox and do 'ip a', do you see the connected NIC reading something like this:

Key here is UP, meaning that my port has linked up with something, my switch in this case.

From the Proxmox host perspective I only have an IP assigned to vmbr0, which is where I access the Proxmox host interface (UI and SSH). When I start a VM which has vmbr1 assigned however, I will be able to see the IP from within the VM, and ping from it...

I have ordered a DAC cable and a 2nd SFP+, different brand, allot cheaper than these enterprise level Dell/EMC's.

G

Yes there are plenty available, although sometimes a good idea to check the compatibility list, or from someone who has tested already.

georgelza

@Gblenn said in pfSense not enabling port:

Hi
What I did was click on the pmox1 and click on shell

As per suggestion, see below. Looks good, vmbr30 which sows up

root@pmox1:~# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host noprefixroute 
       valid_lft forever preferred_lft forever
2: enp2s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq master vmbr0 state UP group default qlen 1000
    link/ether a8:b8:e0:02:a3:71 brd ff:ff:ff:ff:ff:ff
3: enp3s0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
    link/ether a8:b8:e0:02:a3:72 brd ff:ff:ff:ff:ff:ff
4: enp5s0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
    link/ether a8:b8:e0:02:a3:73 brd ff:ff:ff:ff:ff:ff
5: enp6s0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
    link/ether a8:b8:e0:02:a3:74 brd ff:ff:ff:ff:ff:ff
6: enp4s0f0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq master vmbr30 state DOWN group default qlen 1000
    link/ether a8:b8:e0:05:f0:91 brd ff:ff:ff:ff:ff:ff
7: enp4s0f1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
    link/ether a8:b8:e0:05:f0:92 brd ff:ff:ff:ff:ff:ff
8: vmbr0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether a8:b8:e0:02:a3:71 brd ff:ff:ff:ff:ff:ff
    inet 172.16.10.51/24 scope global vmbr0
       valid_lft forever preferred_lft forever
    inet6 fe80::aab8:e0ff:fe02:a371/64 scope link 
       valid_lft forever preferred_lft forever
10: vmbr30: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether a8:b8:e0:05:f0:91 brd ff:ff:ff:ff:ff:ff
    inet 172.16.30.11/24 scope global vmbr30
       valid_lft forever preferred_lft forever
    inet6 fe80::aab8:e0ff:fe05:f091/64 scope link 
       valid_lft forever preferred_lft forever

If I ping 172.16.30.1 however. also if i ping from my laptop to the 172.16.30.11 ip which is suppose to be assigned on the pmox1 host it fails.

root@pmox1:~# ping 172.16.30.1
PING 172.16.30.1 (172.16.30.1) 56(84) bytes of data.
From 172.16.30.11 icmp_seq=4 Destination Host Unreachable
From 172.16.30.11 icmp_seq=5 Destination Host Unreachable
From 172.16.30.11 icmp_seq=6 Destination Host Unreachable
From 172.16.30.11 icmp_seq=9 Destination Host Unreachable
From 172.16.30.11 icmp_seq=10 Destination Host Unreachable
From 172.16.30.11 icmp_seq=11 Destination Host Unreachable
^C
--- 172.16.30.1 ping statistics ---

Gblenn

@georgelza Ok but the Proxmox host, vmbr0 and the vmbr30 are on different subnets. So unless you have rules set up to allow them to communicate with each other, they can't.

So either you need to put vmbr30 into the same subnet as vmbr0, or make sure it is possible to communicate between the 172.16.10 and 172.16.30 subnets...

That said, since it is clearly saying it is UP, and it is also getting an IP, my guess it is working fine here.
So the card and the module are ok to use with Linux (Proxmox at least). And it is likley only with pfsense (freebsd) that you will have an issue, which you will be able to solve when the DAC and/or new module arrives.

stephenw10

You are seeing the replies from 172.16.30.11 which implies the pmox1 is using it. Which we know it is.

Host unreachable implies it cannot ARP for the address so a layer2 failure.

georgelza

@Gblenn Let me go check...
I know my MBP that sit on 172.16.20.29 has full access to everything on 172.16.10.0 ... need to confirm I have a rule that allows similar to 172.16.30.0
sure i did check, but lets recheck/verify ;)
G

Gblenn

@stephenw10 Hmm, I missed that it was actually two different IP's there. So I guess the ping was meant to have been to 172.168.30.11? Where the .1 belongs to pfsense VLAN where that .11 IP was actually handed out.

Anyway, I think the DELL module is working on Linux and the DAC and new Fiber module will fix it.

stephenw10

Well it looks like 172.16.30.11 to 172.16.30.1 to me. And it's failing which implies it cannot ARP for it inside the same subnet.

georgelza

@stephenw10

that last rule is never used as as it already allowed by second.

georgelza

@Gblenn ye... but at the moment, going from the topton hosting pmox through fiber onto unifi sfp+ port 1, and then via my 2.5GbE Cat 6 uplink to pfSense is failing.

vLan30 with 172.16.30.1 lives on igb1.30 at the moment.

G

georgelza

Gblenn

@georgelza said in pfSense not enabling port:

As per suggestion, see below. Looks good, vmbr30 which sows up

root@pmox1:~# ip a
6: enp4s0f0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq master vmbr30 state DOWN group default qlen 1000
link/ether a8:b8:e0:05:f0:91 brd ff:ff:ff:ff:ff:ff
7: enp4s0f1: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
link/ether a8:b8:e0:05:f0:92 brd ff:ff:ff:ff:ff:ff
10: vmbr30: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
link/ether a8:b8:e0:05:f0:91 brd ff:ff:ff:ff:ff:ff
inet 172.16.30.11/24 scope global vmbr30
valid_lft forever preferred_lft forever
inet6 fe80::aab8:e0ff:fe05:f091/64 scope link
valid_lft forever preferred_lft forever

Hmm, perhaps it isn't working after all. Just noticed that you have the NO-CARRIER notification = No cable connected?! As well as it is not enabled as in "noop state DOWN group".

Further on the vmbr30 there is no LOWER_UP as in the physical connection at the link layer is not there?? Compare vmbr0 to vmbr30:

vmbr0: <BROADCAST,MULTICAST,UP,LOWER_UP>
vmbr30: <BROADCAST,MULTICAST,UP>

What does it look like in the Proxmox UI, for the host and the listings under network?

georgelza

@Gblenn see above.

The UI implies it's up...

guess i need to say i wait... this is rightly not a netgate problem... well until i install the DAC cable from the Topton running pfSense into y core switch.

This atm is more unifi/Topton comm...

netgate/pfsense related though, at the moment it's not allowing me to specify a default gw or simply a gw to use for the 172.16.30.0 network.

i can ping from the pmox my 172.16.10.1 gw, but that's going via the 2.5GbE copper link, to switch and onwards to pfSense to the igc0 port

G