Messages

I think the Hardened BSD drivers are not just optimized for the NICs of R4S.

I'm back to using OpenWRT snapshot and I am able to get 937 Mbps download without Software Flow Offloading. Note that I also tried the official firmware FriendlyWRT, and some other custom WRTs version and all are able to get past 900+ Mbps.

The drivers will get better overtime, when I switch to a lower speed plan I'll use opnsense.

I have a 1 Gigabit internet, unfortunately I can only get 500 mbps download speed. R4S has a Realtek nic and custom nic which is pci-e based.

I finally made it work. The mac address of LAN is 00:00:00... so you won't be able to connect to the admin page and get a connection if you have: R4S -> a switch or a router -> your pc. You must connect your pc directly to R4S and update the mac address of LAN.

Do you mean it doesn't boot at all?  Coz r4s have 2 types of ram used so maybe it have different uboot for it. 

I have R4S with 4 GB ram. It doesn't boot at all. How long until it boots btw? I read that I need to wait for a long time but I already did this. Do I also need to connect R4S directly to my PC and login then edit the mac address?

Can anyone build and share the R4S image from https://github.com/yrzr/opnsense-tools/tree/22.1/r4s?

The June 12 image here doesn't work on first boot. Did anyone got it working including internet connection on the WAN interface?

https://personalbsd.org/?page_id=2

I know the mac address of LAN needs to be configured first so you have to connect directly to the pc. I haven't tried it though. Is there a working build that has a working connection on first boot?

When I tried to delete the LAN interface (eth1), add it back and configure it with ipv6, I no longer lose ipv6 connectivity.

I removed the 0.0.0.0 from the Endpoint.

I finally know what's the problem. It's my WG0 interface. For some reason it was using igb0 which is the WAN interface. I'm using PPPoE so there was another interface created igb0 pppoe which is used by WAN. What's weird is there's no wg0:000:000 assignment to choose from the dropdown. So I deleted the WG0 interface, reinstalled WireGuard and enabled it and did the setup properly again. This time wg0:000:000 can be selected.

[Local]

In the wg server config don't enter a DNS server and restart wg.

This didn't fix the issue.

It looks like it's the server instance that's causing this. I enabled WireGuard but disabled the server instance, it didn't affect the LAN internet. When I enabled the server instance that's when the issue happens.

The server instance setup is standard:
Local

Code Select Expand

Listen Port: 51820
DNS Server: 192.168.30.1
Tunnel Address: 10.11.0.1/24
Peers: <see Endpoint below>

Endpoint

Code Select Expand


Allowed IPs: 0.0.0.0, 10.11.0.2/32


Again, I have Unbound setup with DNSCrypt-Proxy.

Yeah the VPN subnet is 10.0.0.1/24 while LAN is 192.168.0.1/24

I followed this to setup WireGuard: https://docs.opnsense.org/manual/how-tos/wireguard-client.html
It's working fine when I connected my phone as a client and I can access my network, my phone also has internet.

Other computers (non WireGuard clients) that are connected to LAN unfortunately don't have internet access. I have to disable Wireguard VPN to bring back the internet connection. It looks like a DNS resolution issue as indicated in my browser. Unbound by default is listening to all outgoing interfaces so this shouldn't be a problem? Btw, I also use DNSCrypt-Proxy so Unbound serves as a forwarder. Any idea on how to solve this?

As discussed via pm, this has something to do with

Code Select Expand


kernel pflog0: promiscuous mode enabled
kernel pflog0: promiscuous mode disabled


It looks like a software issue. I'll try to use my Edge Router x and Hex-S, see if I still lose ipv6.
Other users seem to experience the same: https://forum.opnsense.org/index.php?topic=22086.0

I also get this and I lose ipv6 connectivity.

Ok after I set the ipv6 type in WAN to none, saving, then set it again to DHCPV6 and save, I got my ipv6 address back. Then I lose it again after a few minutes.

I connect to my ISP via IPoE. PPoE doesn't make sense and it must be a timing issue earlier. Now I lost my ipv6 address again, even after restaring the radvd it didn't fix it.

I get this from the logs:

Code Select Expand

/usr/local/opnsense/scripts/dhcp/prefixes.php: The command '/sbin/route add -inet6 '<an ipv6 address:/62>' ''' returned exit code '71', the output was 'route: : Name does not resolve'