Subcategories

  • Discussions about packages which handle caching and proxy functions such as squid, lightsquid, squidGuard, etc.

    4k Topics
    21k Posts
    E
    I even tried deleting and creating a new certificate. Any suggestions?
  • Discussions about packages whose functions are Intrusion Detection and Intrusion Prevention such as snort, suricata, etc.

    2k Topics
    16k Posts
    bmeeksB
    @NRgia said in Suricata on Pfsense: Your suggestion is preferred, but from I understood from you, nobody is interested or have the knowledge. Again thank you, for updating this package over the years. I'm sure there is someone here on the forum using the package that has the knowledge to maintain it. Another option if IDS/IPS is critical is to use the Linux package on a separate virtual machine or hardware appliance. Inline IPS performance would actually be very good using a Linux box (or even a FreeBSD box) with two separate NICs and configure true netmap hardware-to-hardware mode. That is many times more performant than the hardware-to-host mode that is required when using netmap within pfSense. Of course using a separate box would mean no GUI, but that's how the vast majority of the world uses Suricata already (without a GUI).
  • Discussions about packages that handle bandwidth and network traffic monitoring functions such as bandwidtd, ntopng, etc.

    572 Topics
    3k Posts
    keyserK
    @Antibiotic No it’s not possible with NtopNG as it is not a Netflow collector. You need nProbe for that which will “translate” recieved netflows into flows that NtopNG understands and can visualize (with very very little detail might I add as Netflows has no additonal information apart from sender/reciever and volume). The NtopNG package and the product in general is more geared towards visualising and recording traffic details from actual packet captures. This contains MUCH more metadata about the sessions than netflows (DNS names, protocol information and myriads of other things). But pffSense Plus has a builtin Netflow exporter if you have an external netflow collector on hand.
  • Discussions about the pfBlockerNG package

    3k Topics
    20k Posts
    W
    I want to share my experience and, hopefully, help you with your problem. For a long time, I also faced the fact that sites loaded incredibly slowly due to domain blocking via pfblockerNG together with my local Pi-hole DNS servers. And, frankly speaking, I was already starting to think that I would have to put up with this forever. But after a little experimentation, I managed to significantly speed up page loading! The solution turned out to be simple, but brilliant - configure pfblockerNG and Pi-hole so that they return "empty" responses instead of waiting for a timeout. Thanks to this, the browser stopped constantly trying to load blocked domains, and the network speed increased noticeably. What is especially nice is that this solution worked on all devices in my network, including mobile phones and tablets, without the need to install any extensions for Firefox or Chrome. As a result, my system works smoothly: Squid caches, SSL is filtered, ClamAV checks, and sites open instantly. Setting this up was a bit of a pain, but the results exceeded all expectations - finally a feeling of complete control and stability of the network!
  • Discussions about Network UPS Tools and APCUPSD packages for pfSense

    101 Topics
    2k Posts
    dennypageD
    @jhg said in NUT fails to start after 2.7.2 -> 2.8.0 upgrade: Interesting. I would have thought the initial reboot, which occurred as part of the upgrade, would have done the trick, but it took a second reboot, just now, to get things working. Glad you have it sorted. There was no difference in the output of usbconfig show_ifdrv at any point -- before or after unplugging/replugging the USB cable, nor after rebooting. ... Question: What would tell me whether or not a driver was loaded? If there were an attached driver, it should have shown up with the show_ifdrv command. If you use the command and look at the other usb devices, I think they will show attached drivers. I don't expect to see a driver attached to the ups, because there is a quirk that tells the OS to ignore that device (and not attach a driver). Look for idVendor and idProduct in the above output. The Vendor ID for your device is 0764, which corresponds to Cyber Power Systems, and the Product ID for your device is 0601, which is registered as "PR1500LCDRT2U UPS" (don't sweat an exact match for the name). You can see the quirk with the following command: [25.07-RC][root@fw]/root: usbconfig dump_device_quirks | grep 0764 VID=0x0764 PID=0x0005 REVLO=0x0000 REVHI=0xffff QUIRK=UQ_HID_IGNORE VID=0x0764 PID=0x0501 REVLO=0x0000 REVHI=0xffff QUIRK=UQ_HID_IGNORE VID=0x0764 PID=0x0601 REVLO=0x0000 REVHI=0xffff QUIRK=UQ_HID_IGNORE [25.07-RC][root@fw]/root: Your device is third on the list. The HID_IGNORE quirk says to ignore the device and not attach a driver. @jhg said in NUT fails to start after 2.7.2 -> 2.8.0 upgrade: You might consider adding this resolution to the release notes for 2.8. LOL... sorry, I don't have input to the release notes (I don't work here). While I wrote and maintain various packages, including NUT, I'm still just a volunteer. Most packages are actually written by volunteers.
  • Discussions about the ACME / Let’s Encrypt package for pfSense

    501 Topics
    3k Posts
    A
    Hi, Please help to forward / report the bugs in ACME 1.0 package. Thanks.
  • Discussions about the FRR Dynamic Routing package on pfSense

    295 Topics
    1k Posts
    J
    Anyone else happen to notice that when configuring BFD, if you create a peer and select a profile - after save, re-edit the peer and the Profile is not represented. It appears as "None". You have to check the raw config to determine if the profile was actually assigned to the peer. This is on 2.8.1 (all packages up to date as of the date/time of this post). UPDATE: if re-edit and save (without re-configuring the profile none to what you want) - the save will strip the profile from the peer.
  • Discussions about the Tailscale package

    91 Topics
    611 Posts
    T
    Hi All, I use HAProxy to redirect to a range of https internal resources, this works really well at the moment through the WAN where I have source limits set up, and I can connect to the internal resources from limited external IP Addresses. Given I have tailscale I would like to basically be able to put custom dns entries in to point these hostnames to my pfsense tailscale IP4 address (100.89.148.118) but I am not having any luck getting this working. At the moment, I am just trying to connect to HAProxy using https://100.89.148.118 but it is getting blocked by the firewall. Sep 11 11:55:58 tailscale0 Default deny rule IPv4 (1000000103) 100.89.148.10:53148 100.89.148.118:443 TCP:S I have tried with and without NAT redirecting internally to 127.0.0.1, and I also have rules set up to allow any traffic to and from my tailnets (defined in an alias) but I still keep getting these connections from my other tailscale machines being blocked on the pfsense machine. Can someone give me some pointers on what I am missing because I can see the requests are coming through to the pfsense machine, and in theory the rules should allow it through but I cant see why they don't. I do have tailscale ACL in place, but clearly that is not an issue as the requests are making it through to the firewall. 0/0 B IPv4+6 TCP/UDP TailNets * TailNets * * none Allow across Tailnets 0/0 B IPv4+6 TCP/UDP * * * 443 (HTTPS) * none Allow Tailscale IP4 I also tried adding a EasyRule but because the tailscale0 interface doesn't exist in pfsense it throws an error and won't let me add that rule. Appreciate any help or tips, Cheers.
  • Discussions about WireGuard

    701 Topics
    4k Posts
    QuantumParadoxQ
    I am on PFsense Plus 25.07.1 and I am trying to setup my VPN's wireguard and at first it worked now it will not. Once I set up WireGuard for the first time, it all worked. I could toggle on and off the WireGuard and everything would work as it should, so I made a backup of the system. A few days later, after I rebooted PFsense, the writeguard came on but it disabled the Unbound DNS. and when I went to enable it, I still would not get any traffic. Once I disable Wireguard, I'll get internet again. I went and reinstalled the backup and same thing, it does not work. The VPN I am using is TORguard, and I had the techs from TORguard remotely into my machine to set it up, and they have the same issue. they can ping their VPN traffic out and they can Ping my IPS traffic but there is a bug with switching between the two. Can anyone on here help me with this?
  • Squid ldap auth slow

    Locked
    2
    0 Votes
    2 Posts
    2k Views
    M
    Try the package forum…
  • Firewall issue

    Locked
    5
    0 Votes
    5 Posts
    2k Views
    C
    Moving this to the packages board as it's squid related.
  • Pfflowd stopped status

    Locked
    10
    0 Votes
    10 Posts
    8k Views
    B
    Yeah, it does have a dependency on the state table.  It could be the case that the way my default rules are they just lend the,selves to the pfflowd process.
  • How can i add manual ACL in SquidGuard

    Locked
    2
    0 Votes
    2 Posts
    3k Views
    D
    In current version only modify '/usr/local/pkg/squidguard_configurator.inc' file. Current version not supported Users option, but in near time mabe added
  • Need snort, kinda screwed. Plz help

    Locked
    2
    0 Votes
    2 Posts
    2k Views
    ?
    The snort package is no longer available for pfSense.  I would encourage you to create a separate snort box and use that in conjunction with your firewall.
  • SNORT in pfSense1.2

    Locked
    7
    0 Votes
    7 Posts
    4k Views
    ?
    Yes, it is recommended that you consider starting a bounty project for the package.
  • 0 Votes
    3 Posts
    7k Views
    C
    To Fix Squid add this to the /boot/loader.conf kern.ipc.nmbclusters=32768 kern.maxfiles=65536 kern.maxfilesperproc=32768 net.inet.ip.portrange.last=65535 or just delete it and replace with autoboot_delay="1" #kern.ipc.nmbclusters="0" hint.apic.0.disabled=1 kern.hz=100 #for squid kern.ipc.nmbclusters="32768" kern.maxfiles="65536" kern.maxfilesperproc="32768" net.inet.ip.portrange.last="65535" you might ask why squid is so slow? its because default configuration of pfsense is router not as a server thats why kern.ipc.nmbclusters="0" <- is set to zero. if you just simply remove this squid will be just fine. but to tune the squid i add this kern.ipc.nmbclusters=32768 kern.maxfiles=65536 kern.maxfilesperproc=32768 net.inet.ip.portrange.last=65535 i just figure out why squid is slow. but i don't like the binary package of squid. i'll be using the squid HEAD bec of the store_rewrite feature for caching youtubes videos and other video files and mp3.
  • Enabling squid proxy slows down speedtest.net speed test

    Locked
    8
    0 Votes
    8 Posts
    13k Views
    C
    To Fix Squid add this to the /boot/loader.conf kern.ipc.nmbclusters=32768 kern.maxfiles=65536 kern.maxfilesperproc=32768 net.inet.ip.portrange.last=65535 or just delete it and replace with autoboot_delay="1" #kern.ipc.nmbclusters="0" hint.apic.0.disabled=1 kern.hz=100 #for squid kern.ipc.nmbclusters="32768" kern.maxfiles="65536" kern.maxfilesperproc="32768" net.inet.ip.portrange.last="65535" you might ask why squid is so slow? its because default configuration of pfsense is router not as a server thats why kern.ipc.nmbclusters="0" <- is set to zero. if you just simply remove this squid will be just fine. but to tune the squid i add this kern.ipc.nmbclusters: 32768 kern.maxfiles=65536 kern.maxfilesperproc=32768 net.inet.ip.portrange.last: 65535 i just figure out why squid is slow. but i don't like the binary package of squid. i'll be using the squid HEAD bec of the store_rewrite feature for caching youtubes videos and other video files and mp3.
  • Major slowdown through squid

    Locked
    9
    0 Votes
    9 Posts
    12k Views
    C
    To Fix Squid add this to the /boot/loader.conf kern.ipc.nmbclusters=32768 kern.maxfiles=65536 kern.maxfilesperproc=32768 net.inet.ip.portrange.last=65535 or just delete it and replace with autoboot_delay="1" #kern.ipc.nmbclusters="0" hint.apic.0.disabled=1 kern.hz=100 #for squid kern.ipc.nmbclusters="32768" kern.maxfiles="65536" kern.maxfilesperproc="32768" net.inet.ip.portrange.last="65535" you might ask why squid is so slow? its because default configuration of pfsense is router not as a server thats why kern.ipc.nmbclusters="0" <- is set to zero. if you just simply remove this squid will be just fine. but to tune the squid i add this kern.ipc.nmbclusters: 32768 kern.maxfiles=65536 kern.maxfilesperproc=32768 net.inet.ip.portrange.last: 65535 i just figure out why squid is slow. but i don't like the binary package of squid. i'll be using the squid HEAD bec of the store_rewrite feature for caching youtubes videos and other video files and mp3.
  • Stunnel from external IP

    Locked
    2
    0 Votes
    2 Posts
    4k Views
    P
    Fixed it; I moved Stunnel to the host Windows machine running 443, 993 and 465 and pfsense is NATting. I suspect Stunnel on PFSense is broken.
  • [squid] transparent proxy - don't proxy a specific client?

    Locked
    2
    0 Votes
    2 Posts
    3k Views
    V
    Guess I was too quick to post, but this may help others anyway.  First you need to uncheck "Allow users on interface" which automatically creates an allow for all clients in local network.  Since you want to deny an ip you need to put the deny before the allow, which doesn't happen automatically.  In the custom section add the following (where 192.168.0.0 is your own local network). acl ps3 src 192.168.0.105/32; acl localnet src 192.168.0.0/24; http_access deny ps3; http_access allow localnet Again, the order of the http_access lines are important.  Now the "ps3" acl won't get proxied.
  • Squid on Multi-WAN

    Locked
    8
    0 Votes
    8 Posts
    6k Views
    ?
    Which is exactly what I said 3 posts ago.
  • Problem posting in FF3 via Squid to forums

    Locked
    5
    0 Votes
    5 Posts
    3k Views
    F
    Thanks for trying, it's when I enter an article, and post it it fails. I can browse without problems. I've tried excempting the domain from caching, but that didn't help. I wonder if it can be something in their end  ??? I've tried it with two different pfsenses now, they both excibit the same problem…
  • Snort stops working

    Locked
    8
    0 Votes
    8 Posts
    3k Views
    S
    I just removed the package.  It needs a new maintainer.
  • Freeradius and dd-wrt

    Locked
    8
    0 Votes
    8 Posts
    15k Views
    ?
    Thanks for your answer, I made the test with only one AP, and use the same shared key. Do you have any experience with a configuration like this? First I try with hotspotlogin.cgi, but it show a 500 internal server error, so I tried with hotspotlogin.php, it seems to be working, but after some time, it shows the login screen again, and didn't pass the authorization to the AP. My knowledge of pfesense and dd-wrt is not to much, so I'm stuck on this and don't see any solution for this configuration, any help or hint is wellcome. Thanks
  • Squid 2.7

    Locked
    3
    0 Votes
    3 Posts
    2k Views
    J
    @mhab12: The best way might be to post a small bounty in the bounty forum.  I know the devs are busy on 1.2.1 and 1.3 right now, perhaps ermal or dvserg could help us out…  Exactly what bugfixes / features were you looking for? Thanks Mhab12 !! I would like the new store URL rewriter
  • Imspector log to mysql

    Locked
    1
    0 Votes
    1 Posts
    2k Views
    No one has replied
  • [squid] Can't see local websites

    Locked
    5
    0 Votes
    5 Posts
    2k Views
    S
    Alrite, thank you for your advices ;)
  • LightSquid : number of connections per domain

    Locked
    4
    0 Votes
    4 Posts
    2k Views
    X
    OK, so i understood… We can't have the number of visits (i think it was possible to say : "such user had visited this website 3 times today"... ) per site but instead we have something like all "connections" to the website due to pictures,JS,etc loadings... Thank you for the time answer !! (I hadn't notice the "Megabytes" unit !). Thank you for your answers  :) Sincerely.
  • NMAP-how reliable is it?

    Locked
    2
    0 Votes
    2 Posts
    3k Views
    C
    With only one open port and everything else filtered, you can't get a fully accurate OS detection. The MAC address is key here - Askey Computer is the manufacturer of the device. From the looks of the products on their website, it's most likely a wireless access point. http://www.askey.com.tw Further complicating OS detection is the fact that the open port is likely not open on the device itself I would guess, it's probably a port forward. The most likely situation is this is a wireless AP using NAT to connect multiple devices. If DHCP was showing it inactive, there's a good chance the person who installed it took whatever lease they were initially assigned and statically configured it.
Copyright 2025 Rubicon Communications LLC (Netgate). All rights reserved.