Can you help me with the procedure? No.  It has nothing to do with pfSense.  While I have compiled squid from source before on Ubuntu, that was a few years ago.  Perhaps I was unclear.  When I said you would need to spin up your own server, I was talking about a Linux box for instance, not another pfSense.  Compile squid form source on a Linux box and then use that squid. Is that could be a security issue? Nothing to do with security. A better solution might be to allow this specific user to go around the proxy. As the squid page notes, this has only been tested up to 32K and your URL is 36K, so it may not even work.

By default, you cannot connect to WebGUI via WAN, only LAN.  If you want to enable WebGUI from WAN, see this: https://doc.pfsense.org/index.php/How_can_I_access_the_webGUI_from_the_WAN Also note you should edit your WAN interface (Interfaces - WAN) and uncheck the Block private networks and loopback addresses option.

Overkill.

Just as an update, if you check the ticket we just merged a fix from PiBa. Thanks!

once you rules load correctly it should be there. 5,000,000 is a bit drastic.. 400,000 should be more than enough and is going to be the new default. You should always be able to cycle if antilock is enabled disabled under System / Advanced / Admin Access

If your device has integrated graphics and a BIOS setting to specify how much system memory will be dedicated to the graphic card you can set this to the smallest available amount.

you using xn interfaces in Xen?  I don't think they support vlans..

While Ciscos are great (probably SG300 or 350 series?) they aren't cheap. Some users here like these D-Link DGS-1100-08 GBit switches. Compared to the Cisco SG300/350 they are smart switches "only" and not fully managed.

In System->Advanced->Networking, There is a section, which name is Network Interfaces.There are five options,  Device polling,  Hardware Checksum Offloading,  Hardware TCP Segmentation Offloading,  Hardware Large Receive Offloading and  ARP Handling. All of them are not checked. Yes all are blocks !

Of course I can do cu command when LTE isn't connected or pfsense is trying to reconnect (moment when log screaming 'the modem is not responding to AT commands' then I can do cu on cuaU0.0). Edit or resave the ppp setup do nothing. Only solution is reboot pfsense… Very strange because a lot of people posts this config as working. Tommorow I will try downgrade pfsense to 3.5 version. Martin

Use of syslog has been long standing best practice when you need long term storage of logs from pretty much any sort of device.. What good is the logs in your SSD, that is in the device when the device shits the bed or the disk shits the bed and you loose your logs..  If the logs are so important that you want them and lots of them then send them to your syslog server, or multiples of them.  All of which are more likely to have redundancy and backup of the logs, etc. So yes the correct solution to the question - I need lots of LOGS.. Well then send them to your syslog server(s) that have lots of space ;)

Sounds good. I'll forget about them and disable them in the bios. I just figured it would be nice to utilize the onboard ones and then have the pcie slot open for future expansion of who knows what or to use pcie NIC as a backup when/if the realtek ones died.

Hi Stephenw10, Sure thing. Will do it for sure! Appreciate the message. Fabricio.

Thank you for sharing. Had the same problem with a very slow dns and changing the default. But in the dns custom config I put: server: include: /var/unbound/pfb_dnsbl.*conf server: private-domain: "plex.direct" and I don't get the plex server running out of the network like before without dnsbl… any advice?

I'm on Virgin in the UK and can confirm, with the SuperHub in modem mode, you'll get a Public IP address on the WAN interface on PfSense. The 192.168.100.1 address can also be used to look at the modem stats and to toggle modem mode back off again should you wish.