This is fixed now

Yeah, right. I disabled snort interface and all stuff works. But I need ISP on my network, what should I do to do? Where we can see what is a problem with snort? Last update makes it broken, if I  don't wrong. UPDATE: I turned on snort after two days and everything works fine… sometimes. Sometimes works and after few minutes stopped, Permission denied or domain no exist (or something like this). Maybe it is some kind of cache error with snort? UPDATE2: It looks like this http://wiki.squid-cache.org/SquidFaq/TroubleShooting#FATAL:Failed_to_make_swap_directory.2Fvar.2Fspool.2Fcache:_.2813.29_Permission_denied in squid FAQ, but I have cache directory and owner is correctly, user: squid | group: proxy.

If it's transparent then yes it should just pick up all port 80 traffic.  It should just work.  You could shell in and take a look at /var/squid/logs/cache.log and see if there are any errors.  I would also recommend removing the squid package and then reinstalling it.  That some times works for these weird glitches.

First pointer would be to start reading this forum.  Every other post is pretty much exactly like yours.  From there you can either start using search engines to show you step by step what you want to do, or you start trying it yourself and then asking specific questions here if you run into problems.  But I can assure you that squid + squidguard can filter HTTPS.

@doktornotor: Regardless, HAProxy can run on any port, and multiple ports, and in general I don't get what's the question here…. Sorry, I was trying to answer what appears to be two separate issues (Teamviewer WOL and HAProxy using any ports other than 80/443) in one thread; I thought they were related

Hi everyone I've similiar problems when activating transparent proxy setting in squid3 on pfsense 2.3.2 amd64. The forwarding of port 80 to the squid interface wont work. If i manually add them in the Browsersetting its working. So squid is running. I've tried adding a Portforwarding Rule but with the same result. Connection got terminated. The only difference is I don't even use a bridge  :o

@doktornotor, thank you so much, looks like issue resolved after changing to immediate mode.

You can use wpad or https ssl for the solution https://doc.pfsense.org/index.php/WPAD_Autoconfigure_for_Squid https://forum.pfsense.org/index.php?topic=123874.0

@doktornotor: Try with 2.3.3 snapshots perhaps.. thank you for the recommendation. Unfortunately, i am still pulling my hair out on this. It seems like another issue is at work here, since i am experiencing delays, to a lesser degree, on a clear connect as well (no bumping). With no HTTPS, the browser shows "Waiting for proxy tunnel". This only seems to happen on my test pfsense (which is chained to my main PFSENSE/firewall). I will look at testing this in another way and report back if a solution is found

And how should I use it?  (within pfsense)

Yeah, so use LDAP -> problem solved. If you invent your own authentication methods non-existent in the pfSense package, you'll need to pick up the pieces yourself.

Ah that explains it indeed, i should probably check and prevent that combination. Or try and 'fix' the ipfw rule by resolving the backserver ips.. Still tricky if it changes though which haproxy itself could do with the dnsresolvers setting.

If you want the logs elsewhere, do it the other way round – like, SCP/SFTP the logs from pfSense to your NAS via cron.

yes what's ur opinion in this issue

In general, yes, you would be a whole lot better off dropping the nanobsd altogether.  And no, I don't think it's a mirror down. Can't write 1440 bytes to /var/db/clamav... suggests the ramdisk is full or whatever other fluke related to nanobsd is happening.

Yeah, obvious hint here would be getting rid of the redundant router behind pfSense that's producing double NAT, requires static routes configured on pfSense and generally is a royal PITA.