SOLVED Network: Everything I try fail

[otpi]

TrueNAS-SCALE-22.02-RC.1-2

I am trying to create a bridge for VMs to LAN. But whenever I "test" network changes I am kicked out of the GUI. After a while it usually reverts (but I have been forced to reboot a couple of times as well)

I have 4 1G ports. TrueNAS GUI, smb, nfs are all on eno1 (static .11). While my VM macvtap0 (static .21) is on eno4 (which is .120, dynamic provided by dhcp).

Furthermore: When I reboot the system, and it is back to defaults (no bridges). The VM behind macvtap0 has a static .21 ip. But .21 can no longer ping .11 (which is used for an NFS share to the VM), this fixes itself after a while... But I've observed that macvtap0 does not get an IPV4 address, only ipv6.

Well. This may be all good tomorrow when the next RC is out...

 

[morganL]

Its worthwhile checking the default gateway IP address..... if you lose connectivity without confirming it will revirt.

 

[otpi]

This?

root@truenas[~]# ip route show default via 10.0.0.1 dev eno1 10.0.0.0/24 dev eno1 proto kernel scope link src 10.0.0.11 10.0.0.0/24 dev eno4 proto kernel scope link src 10.0.0.120

I have pfSense on .1

Any change I make in GUI "network" fails at "test", even setting a static IP.

I tried using shell, just for fun

ip link add br0 type bridge
ip link set eno4 master br0

As long as eno4 is not "busy" this apparently works (ip -details a), but it wont show in the GUI or be persistent (which is intended?)

Edit: I just did a "change" and lost connection to the web interface on .11. It is also not responding on .120. However: I can ping the VM on .21

Code:

PING 10.0.0.1 (10.0.0.1) 56(84) bytes of data.
64 bytes from 10.0.0.1: icmp_seq=1 ttl=64 time=0.221 ms
64 bytes from 10.0.0.1: icmp_seq=2 ttl=64 time=0.100 ms

--- 10.0.0.1 ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 13ms
rtt min/avg/max/mdev = 0.100/0.160/0.221/0.061 ms

PING 10.0.0.11 (10.0.0.11) 56(84) bytes of data.

--- 10.0.0.11 ping statistics ---
5 packets transmitted, 0 received, 100% packet loss, time 85ms

PING 10.0.0.21 (10.0.0.21) 56(84) bytes of data.
64 bytes from 10.0.0.21: icmp_seq=1 ttl=64 time=0.438 ms
64 bytes from 10.0.0.21: icmp_seq=2 ttl=64 time=0.463 ms
64 bytes from 10.0.0.21: icmp_seq=3 ttl=64 time=0.489 ms

--- 10.0.0.21 ping statistics ---
3 packets transmitted, 3 received, 0% packet loss, time 53ms
rtt min/avg/max/mdev = 0.438/0.463/0.489/0.027 ms

 

[otpi]

Further details: The web interface is not responding.

Remote into server:
ip route show is the same as before.
ip -details a | grep eno shows the correct IP on all enos.
service middlewared restart does something. But GUI is still not responding.

 

[otpi]

@morganL I had high hopes for RC2, but still the same behavior. Tried using GUI to create br0, no members just name: br0 and random Alias IP 192.168.1.169. Disconnected from GUI and reconnects after ~60 s, with no change.

What now? I just wnat to create a bridge

 

[morganL]

I can't see what is different from normal.
Could you review your sequence with the docs: https://www.truenas.com/docs/scale/virtualization/accessingnasfromvm/
If you can't see anything different, then please "report-a bug".
I'd suggest writing it up in a sequence where it's easy to replicate the whole issue.

 

[engineerdj]

I can't see what is different from normal.
Could you review your sequence with the docs: https://www.truenas.com/docs/scale/virtualization/accessingnasfromvm/
If you can't see anything different, then please "report-a bug".
I'd suggest writing it up in a sequence where it's easy to replicate the whole issue.

@morganL This is where I have the issue with the steps outlined in the doc:

Click Add under IP Addresses and enter the active interface’s IP and subnet mask.

When I do that, it returns that the IP address is already in use. In the docs, it seems like the active interface enp7s0 IP address just jumps to the br0 interface. On the following screenshot you see the active interface IP, the VM IP, and an IPv6 IP. Some other steps I've seen call to delete the active interface IP (enp7s0) port but the GUI doesn't allow moving past the IP in use error.

 

[otpi]

@engineerdj

I had .11 reserved/static for this NIC in pfSense, but afaik Truenas eno0 was 'dhcp' and thus lost 'route' whenever I tried to test.

These steps worked for me:

1. In the network gui i selected eno1, uncheck DHCP, add 10.0.0.11. Test and confirm changes.
2. create bridge br0 add member ensf0 (sfp+). I chose DHCP for bridge and used pfsense to reserve .13.

I then changed VM NICs to use br0. Had to reboot for this to take effect. macvtaps disappeared, and VMs are given IPs by pfsense dhcp, which I can reserve or not as needed.

What I do not understand is why DHCP is disabled for NICs and not just eno1? But I dont need more physical connections yet, and I can always set IP manually if needed, so I left it as is.

 

[engineerdj]

@otpi I have a similar setup.

1. TrueNAS Web UI is set to .27 on enp7s0. DHCP unchecked, and statically set.
2. Create bridge, add member enp7s0 from above, DHCP checked, and then I try to add the "the active interface's IP and subnet mask"

This is how I interpreted the doc. https://www.truenas.com/docs/scale/virtualization/accessingnasfromvm/

This is where it fails the first time, returning that the IP address is already in use. If I leave the IP address blank and only use DHCP, then in the test process, connection to the web UI is lost. This is the second failure, and the only resolution is to delete br0 from the ipmi. This should make sense because the TrueNAS Web UI and the bridge can't(?) share the same IP.

I'm wondering if I repeat step 2, but rather than adding "the active interface's IP and subnet mask" .27 I add .28 instead. Then the enp7s0 would have it's own IP .27 for the TrueNAS Web UI and the bridge br0 would have its own IP .28 bound to the enp7s0 interface. Then I could reconfigure the VM to use br0 instead. Thinking reads as sounds, but I would have to test.

 

[engineerdj]

Although, the way the doc reads, the br0 interface should take "the active interface's IP and subnet mask" from the enp7s0 interface on config and test and that should function correctly with the bridge and the TrueNAS Web UI. This is where my confusion starts because my interpretation of the doc steps gives me different results.

 

[marrbacca]

Hello,
I recognize this is an old post, but I'm having the exact same issue as the OP. I can see the thread marked as solved, but I do not see any solution in posts.
My setup is a little bit different I have 2 interfaces but currently using only one: "enp7s0".
I have several apps installed all using this interface + I have a VM on the same interface which I would like to have access to TrueNas. The VM can see other ips in the network (pings them ok), but do not see the TrueNas IP.

I have followed https://www.truenas.com/docs/scale/virtualization/accessingnasfromvm/ but it look like I'm missing something, as doing the steps from the guide makes my TrueNas gui not accessible (along with all the apps installed). It automatically reverts back to correct settings after set time, but I'm unable to make it work.

What is the accepted solution here?

 

[otpi]

Afaik: What solved my issues was to set the default route under Network/Global Configuration. I.e. open "global configuration settings", type in the IP you want to use, e.g. 10.0.0.11 in my case, the same it already received/used from dhcp.

Network |

Network

www.truenas.com