Through trial and error, I decided the best way is to set the filtering on high and expect to investigate false positives for a few weeks. Snort enables you to easily allow exceptions at a granular level directly from the alerts page. Anything with a '1' is suspicious. If it involves port 80 and, for example comes from Google or Akamai, it's probably OK and you can put it through there. But look into it a little first. I have found the need to suppress full rules not needed as badly with this approach.

I'm seeing a lot of this now as well.  Seems to all be originating from one machine running Storjshare. I tried increasing the Flow Memory Cap, but so far that hasn't accomplished anything. It's at 256MB at the moment. edit I tried restarting Suricata on the LAN interface, and now it refuses to start even after resetting things back the way they were.  :o edit2 Needed to remove commas from the byte sizes in flow and stream memory cap.  pfSense should automatically parse this or at least return an error if you attempt to provide invalid input parameters. edit3 I'm not certain whether it was necessary to completely restart Suricata on the interface for the setting to take effect, so it's possibly that my current 512MB setting is way more than necessary, but it's stopped the error messages, so I'm going to leave it alone for now. edit4 Nope.  Up to 1GB and still receiving this error.  I don't think increasing the flow memory cap is a solution. edit5 I noticed that there's a separate Flow/State Memcap setting under LAN App Parsers -> DNS App-Layer Parser Settings.  The default is only 512KB.  I upped it to 1MB and reset the Flow Memory Cap setting to its default.

@coffeecup25: @bimmerdriver: So, the snort service stopped again.  Am I the only one this is happening to? https://forum.pfsense.org/index.php?topic=130993.msg723503#msg723503 No. I also complained. The thread this link is in mentioned a possible third reason it fails to start properly. To me, it looks like a buggy upgrade as it worked great before updating pfSense and snort to the newest versions. Right now, I have snort disabled. I will enable it when I see a new package update for it being made available. Getting the exact behavior here on my APU2C4 since the upgrade.

Were you able to get this working ever? I only get a sample log like such to my syslog server from using the barnyard2 May 31 01:42:38 pfsense.rando.local nginx: 10.0.0.3 - - [31/May/2017:01:42:38 +0000] "GET /css/pfSense.css HTTP/1.1" 200 7239 "https://10.0.0.1/snort/snort_barnyard.php?id=0" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.36" I don't actually get the snort alerts…if I turn it to log to the pfsense system log, it works fine but I want it to be a separate log.

I used to have an "SuricataWhitelist" alias containing hosts (also alliasses). Now it's type is "networks" (old hosts aliasses are still there). So this has changed. Maybe this is causing the blocks since the last version?

nevermind, solution and explanation can be found here: https://forum.pfsense.org/index.php?action=profile;area=showposts;u=17262

There are issues with Inline mode and many NICs.  The links provided to another thread here on the forum provides the evidence.  The problems with Inline IPS mode are related to compatibility issues between various NIC drivers and the Netmap module.  This is complicated even further on pfSense because some things done to help with limiters seem to have a negative impact on Netmap.  In short, if you have problems with Inline mode it is almost certainly due to something with your specific NIC driver and Netmap. There is a known issue with traffic shaping on pfSense and Netmap (that's the limiter thing mentioned above).  Those two absolutely don't play well together at this point.  Things will eventually improve as Netmap bugs are ironed out.  Until then, you may have to be content with Legacy Mode blocking.  Usually em drivers are OK, so do you by chance have a traffic shaper enabled?  If so, try disabling it and see if Inline mode works then. Bill

Either of these two methods would be correct for syntax – #DHCP Guest Range suppress gen_id 0, sig_id 0, track by_src, ip 192.168.20.10/28 #DHCP Guest Range suppress gen_id 1, sig_id 0, track by_src, ip 192.168.20.10/28 The first will apply to all Generator IDs and SIDs, while the latter will apply only to generic rules and not to other preprocessors. Exactly how are you assigning the Suppress List you created to the interface?  I assume you know that after creating a manual Suppress List, you then have to go to the INTERFACE SETTINGS tab for the interface where you want to use it and assign it to that interface.  You do that by selecting the named list in the drop-down box in the Suppression List section.  Save the change and then restart Snort on that interface.  Simply creating the list on the SUPPRESS tab is only 50% of the work. Another possibility is you have multiple Snort instances running on the same interface.  Run this CLI command to make sure only a single Snort instance is running on each configured interface: ps -ax |grep snort Bill

Probably false positives.  There have been some reports of flakiness with the TLS decoder rules in Suricata of late.  There is a post on the Suricata Redmine site about some other TLS issues. Bill

Looks ilke a syntax error in one of the Emerging Threats rules. Bill

@hescominsoon: I currently run pfsense 2.3.4 under hyper-v.  The issue i am experiencing with suricata is when it gets an alert triggered instead of blocking the host that causes the alert it kills the entire interface.  I have read how inline is a bit buggy…is this another inline bug for suricata? It's not necessarily a Suricata bug.  Inline mode is entirely dependent on Netmap for operation, and Netmap in turn is totally dependent on 100% support from the NIC driver.  There are only a tiny handful of NIC drivers that fully support Netmap on FreeBSD.  From your experience, it seems the Hyper-V NIC drivers are not on that list.  Netmap inserts itself between the NIC and the rest of the operating system.  Nothing moves from the Ethernet wire into pfSense (or from pfSense into the Ethernet wire) without going through the Netmap layer.  The NIC driver has to understand how to talk to Netmap.  Any inconsistencies in how the NIC driver interracts with Netmap will cause problems with Suricata inline mode. Bill

The pull request for the GUI package update to support Snort 2.9.9.0 update has been posted.  Here is a link:  https://github.com/pfsense/FreeBSD-ports/pull/361. This should get merged into 2.4-BETA very quickly, and then appear a bit later for the 2.3.4 RELEASE and 2.3.5 snapshot builds. Bill

@tortue: Only path forward I can think of at the moment is a log rule and handling the log parsing and dedupe individually and using a FW alias for that deduped file. Correct, that would be a way to do it, but it is a bit labor intensive to develop.  I would challenge if the effort was worth the gain, though.  The firewall is already going to drop inbound traffic that is unsolicited or that does not match an open port.  Good operating practice says to keep all Internet-facing public services (web, email, etc.) well patched and hardened.  Having an automated system that locks out an IP due to it attempting to connect to a single non-existent port could be an issue for a commercial enterprise.  Suppose the user who is also a customer of yours just fat-fingered the hostname or accidentally used your hostname when he was trying to connect to another service (RDP perhaps) when he meant to use another.  Now his IP will be blocked from any contact with your domain, including your web site. Of course if you are strictly a home user, then none of the above matters to you.  But at the same time, as a home user you likely have no public-facing services (and thus open ports), so why bother? Bill

When you select and enable an IPS Policy, that overrides manual selection of Snort VRT rules – hence the red mouse cursor.  Those checkboxes are disabled by design when using an IPS Policy.  This is because the chosen policy is determining which of those grayed-out categories and rules to use. Bill

This is a bug within the Suricata binary itself.  It has been reported to Suricata upstream and will be fixed in the next release, I think.  It is only a warning and will not affect operation. Some changes were made to a section of code in the 3.2.x series, but one line got missed when those changes were made.  You can find the bug report and details on the Suricata Redmine site. Bill

Thank you that's very good to know for the future!

Currently the GUI package builds a snort.conf file configured for alerting and capturing packets related to those alerts.  It was never intended to use Snort like Wireshark on pfSense, so that capability is not part of the standard snort.conf generation. Bill

thanks Bill you tha MAN