Menu

Show posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Show posts Menu

Messages - mag

#1
Until now, I had the WebGUI bound (listen interface) to the LAN interface only (I don't want to bind it to the WAN interface + there's already another service (HAProxy) bound to the same port there).

Even though the LAN interface uses fully static configuration (as mentioned at https://docs.opnsense.org/manual/settingsmenu.html#listen-interfaces), the WebGUI now fails to automatically start after updating from 24.1.8 to 24.1.10_8 (it can be still started manually).

Therefore I went the way of adding a custom loopback interface (device lo1, interface opt1, IP 127.0.0.10) to which I bound the WebGUI and then added a NAT port forwarding rule (along with the associated filter rule that allows the related traffic) for the LAN interface that makes the WebGUI accessible from LAN just like before.

This works fine, however the problem is with accessing the WebGUI from VPN (IPsec road-warrior VPN clients connected to the OPNsense host). I also added the same NAT port forwarding rule (along with the associated filter rule) for the IPsec interface, but VPN clients still can't access the WebGUI.

When enabling logging on these firewall rules I can see that the traffic passes these rules as expected (same as in the LAN case).
I also tried packet capture, but it doesn't capture the traffic that's subject to the NAT port forwarding in any useful way, so it didn't really help (I can only see the initial SYN packet coming from the VPN client side - at the IPSec enc0 device).

Now I'm quite out of ideas what could be wrong and what to try. Perhaps routing issue (with traffic coming back to the VPN client) or something else?
Does anyone have any ideas or tips what to look at or try?
#2
The 1st set of numbers (0/90064/-1/-1/120071) is timing information (TR/Tw/Tc/Tr/Ta).
The 2nd set of numbers (5/3/0/0/3) are connection counters (actconn/feconn/beconn/srv_conn/retries).

What's also important is the termination state just before the 2nd set of numbers (sC--) which would explain why the connection was closed.

sC-- means:

s: the server-side timeout expired while waiting for the server to send or receive data.
C: the proxy was waiting for the CONNECTION to establish on the server. The server might at most have noticed a connection attempt.

aka
sC: The "timeout connect" stroke before a connection to the server could complete. When this happens in HTTP mode, the status code is likely a 503 or 504 here.

More information (and description of each of those fields):
https://www.haproxy.com/blog/introduction-to-haproxy-logging/
http://docs.haproxy.org/2.4/configuration.html#8.2.3
http://docs.haproxy.org/2.4/configuration.html#8.4
http://docs.haproxy.org/2.4/configuration.html#8.5
#3
opnsense-revert -r 21.7.5 haproxy22 in the shell should do it.
I'm also having some (different, increase of dropped connections) issues with HAProxy after updating to 21.7.6 so considering the same.

As for the request size limit, it should be tied to the tune.bufsize parameter (Services: HAProxy: Settings:  Buffer size):
http://cbonte.github.io/haproxy-dconv/2.2/configuration.html#tune.bufsize
But it didn't change between these HAProxy versions so the actual reason behind your problem might be different.