When the portal page doesn't come up it is almost always:
HTTPS set as browser's home page (go to a regular http site - just type 10.10.10.10 in the browser or something)
DNS (usually static DNS servers in the client even though they're DHCP)
Find out what port your portal is on and try connecting directly to that. (eg http://192.168.10.1:8002)
Hello Gertjan!
Thanks for the reply.
with squid non transparent I am sure downloading continues after disconnecting via CP. In transparent squid everything is ok.
I am using pfsense 2.0.3 only because in this version I can use non transparent squid with CP login option as using non transport squid I can block https traffic and also able to cache https traffic.
@ermal:
You are sure this is still true on 2.2?
Not sure if MAC passthrough is still slow when new users join and there are already thousands in the list, but just allowing them through, the database appears to be cleared on reboot.
When you upload a customised CP login page it does get backed up as part of the XML config file (I've done it so I know it works). The pictures ought to be included as well, assuming you uploaded them via the 'file upload' section of the captive portal configuration pages.
Yes but its easier for me to add a whole subnet in captiveportal.inc since I have no control what IPs the servers will get. I only supply a pre configured pfsense.
The symptom is the same, the crash is quite a bit different from the looks of your crash report. 2.2 release is now out, upgrading to it is your next best step.
It is impossible infeasible to successfully captive portal HTTPS connections unless you control all your prospective client devices and can pre-install trusted root certificates. It is better to just drop TCP/443 traffic until the portal session has been negotiated via TCP/80.