@tjabas said in need some help assigning ports in new router:

i was trying to do so by assigning igc2 as the same start ip as igc1(192.168.1.1) but it wasnt possible so i made it 192.168.2.1, but i cant get no internet connection in that port.

Unlike the main LAN, you need to provide rules to allow the traffic. Here's what I have for my guest WiFi. It's on a VLAN, but it works the same way. It allows connecting only to the Internet and pinging the interfauce.

b2263027-ed7b-4ba3-876d-a871ead6e324-image.png

@Troutpocket So based on what I have recently done i think you will need a radius proxy.
To add some color, I recently set up Cisco DUO 2FA. Logging into my pfsense using LDAP or OpenVPN using LDAP auth, admins and end-users will get a DUO Push notification. Works brilliantly.
In order to get this accomplished I needed to have a working LDAP server that everyone initially can reach and auth to. Once that is working I set up a DUO Proxy server that also listens for LDAP requests. Now you point all your LDAP configuration to the DUO Ldap you set up and when any LDAP request get sent to this proxy server, the proxy talks to the LDAP server confirms that auth is good then talks to the DUO service to have a push notification sent to the end-users phone.
I bring all that up to say i think you require a proxy as well. You need to have something talking radius and all radius requets get sent to it and the proxy will turn around and send it to Azure.

I found this while searching.

https://wiki.freeradius.org/config/Proxy

@vahidmoghadam

You'd do it with a shell script. Here's one I wrote years ago on Linux:

#! /bin/sh
while [ 1 ]
do
ping 99.246.124.1 -c 1 || date >> ~/log;sleep 50
done

The commands I run on failure are date and sleep for 50 seconds.

Apologies for the thread necro, but I figured I should give a final update.

Was all ready with Wireshark, and waiting for the problem to happen again - but it didn't. And four months later, it still hasn't happened. Everyone seems to be working fine. And I still had no idea what the problem was, or why it suddenly vanished.

It could be sunspots for all I know. Thanks for all the advice and suggestions. 😃

Both of those crashes are different but appear to be in PPPoE and netgraph.

They don't quite line up with other similar recent issues either.

First panic backtrace:

db:0:kdb.enter.default> bt Tracing pid 0 tid 100007 td 0xfffffe0012e2c720 kdb_enter() at kdb_enter+0x32/frame 0xfffffe001099f960 vpanic() at vpanic+0x183/frame 0xfffffe001099f9b0 panic() at panic+0x43/frame 0xfffffe001099fa10 pppoe_addsession() at pppoe_addsession+0xb1/frame 0xfffffe001099fa40 ng_pppoe_rcvdata_ether() at ng_pppoe_rcvdata_ether+0x8ba/frame 0xfffffe001099fad0 ng_apply_item() at ng_apply_item+0x2de/frame 0xfffffe001099fb70 ng_snd_item() at ng_snd_item+0x2e0/frame 0xfffffe001099fbb0 ether_demux() at ether_demux+0x1a4/frame 0xfffffe001099fbe0 ether_nh_input() at ether_nh_input+0x349/frame 0xfffffe001099fc30 netisr_dispatch_src() at netisr_dispatch_src+0xad/frame 0xfffffe001099fc90 ether_input() at ether_input+0x99/frame 0xfffffe001099fcf0 iflib_rxeof() at iflib_rxeof+0xe59/frame 0xfffffe001099fe00 _task_fn_rx() at _task_fn_rx+0x7a/frame 0xfffffe001099fe40 gtaskqueue_run_locked() at gtaskqueue_run_locked+0xa7/frame 0xfffffe001099fec0 gtaskqueue_thread_loop() at gtaskqueue_thread_loop+0xc2/frame 0xfffffe001099fef0 fork_exit() at fork_exit+0x80/frame 0xfffffe001099ff30 fork_trampoline() at fork_trampoline+0xe/frame 0xfffffe001099ff30

First panic message buffer:

<5>ng_pppoe[15]: no matching session panic: Bad list head 0xfffffe0096ade060 first->prev != head cpuid = 0 time = 1685171251 KDB: enter: panic

Second panic backtrace:

Tracing pid 13 tid 100046 td 0xfffffe0012ec4020 kdb_enter() at kdb_enter+0x32/frame 0xfffffe00858c9cf0 vpanic() at vpanic+0x183/frame 0xfffffe00858c9d40 panic() at panic+0x43/frame 0xfffffe00858c9da0 ng_destroy_hook() at ng_destroy_hook+0x309/frame 0xfffffe00858c9de0 ng_apply_item() at ng_apply_item+0x97/frame 0xfffffe00858c9e80 ngthread() at ngthread+0x26b/frame 0xfffffe00858c9ef0 fork_exit() at fork_exit+0x80/frame 0xfffffe00858c9f30 fork_trampoline() at fork_trampoline+0xe/frame 0xfffffe00858c9f30

Second panic message buffer:

panic: refcount 0xfffff800188fd2a8 wraparound cpuid = 3 time = 1685245381 KDB: enter: panic

Ah, OK. Interesting. I'm not aware of anything specific that would have affected that but it could well have been an unintended effect of some other change. That would be unlikely to have been tested since it's considered an invalid setup.

I would open bug report with the specific findings you have there.
https://redmine.pfsense.org/

IMO it should be valid since PPP is point-to-point and can work in a setup like that. It may just need to be excluded from a check somewhere.

Steve

@hoandco
I don't think so.

I can hardly follow your comments. To get clear, pleas post the "DNS Server Settings" from System > General Setup.

And what do you get exactly, when doing a lookup for google.com in Diagnostics > DNS Lookup?

Do you have the DNS Resolver enabled on pfSense?

@Bummer : did your import config file to 64bit version wokred?

@Gektor the patch in that thread should fix it.

@Lavarocker said in OpenVPN Issues Anyone After Upgrading to 23.05?:

Re: To 23.05 or not ? that is the question :)

Has anyone tested OpenVPN after upgrading to 23.05? The last official update broke it.

Thanks

I have issue with my OpenVPN Server and client. I was able to get OpenVPN Server working again by disable it and then re-enable it. As for my 2 OpenVPN clients, both show are connected but the gateway show offline/packetloss. I have rebuild it again since I can't fine any issue. My suggest is that there likely some new field being added to OpenVPN which causing issue on my end.

@NollipfSense said in Lock Top navigation...?:

@furom You could restart webGUI at the console or reboot.

Cool! I was logged out and when logging back in it stuck! Thanks :)

@TheRedHerring Ah but devices go around DNS with DOH.Browsers use it by default. I’ve one device that requires it.

This is an overly detailed guide but may help with blocking other DNS if that’s your goal:
https://jpgpi250.github.io/piholemanual/doc/Block%20DOH%20with%20pfsense.pdf

After running around in circles for almost a week, trying all suggestions on the forum I finally got it working. The solution? Using the console and

13) Update from console

I would go with a small unit that comes with 3 LAN ports and having an looking eye forward it should
maybe coming by site with 3 x 2.5 GBit/s ports.

WAN /LAN /DMZ / (WLAN)

CPU 4C/4T w/ 4 GB RAM
CPU 4C/8T w/ 8 GB RAM

Or if the RAM module will be able to change
it may be the best option! (SO-DIMM)

If you need a fall back line (3G/LTE) it should comes sorted with an SIM socket and/or
miniPCIe or M.2 socket inside.

If you want to use the captive portal with
FreeRadius it should be also sorted with a
miniPCIe or M.2 for a wireless card (WiFi)

It should be able to connect to the console
in any kind. Serial DB9-RS232 or RJ45 CON
or USB CON port. For solving out any problems.

Netgate itself are offering smaller units in their
shop. Perhaps there is something for your needs
in? Have look please: Netgate Shop

What is your entire WAN speed?
What packages will you run?
Do you using PPPoE?

@Gektor Sounds similar but different situation. New thread? Check the logs to see why PHP is crashing.

Have you tried to set the latency to a different mode?

Quick update:
I have decided to try reinstalling pfsense on ZFS instead of UFS. I backed up all my configs first, did a clean install with 1 striped disk and set zfs copies=2. It hasn't been long since I restored everything but pfsense has been running for 16 hours now without any crashes so it's looking promising so far.

I will post another update with how it's running after a few days or when it crashes again.

@Sorjal said in Errors that occurred during upgrades to 23.01 and 23.05 RC:

I try to switch away from 192.x.x.x

A good idea in most cases. 👍