Figured it out, will need to do some re-coding

I see this post seems to be related to my cause. Very helpful info.  I can see from his code that int= is the interval rate. I do see the authentication problem is more work than implementing the graph. https://forum.pfsense.org/index.php?topic=43852 sample code: http://pf.sense.ip/_graph.php?ifnum=wan&ifname=WAN&timeint=10

Noticed the widget breaking a couple of days ago. Gonna try out your solution.

I can confirm that turning off device polling fixed it, thermal widget workes as expected now.  Explains why loading a config showed the issue, device polling is off be default, config kept turning it back on. Also a positive, my CPU would hover idle right around 13-15% load, all the time when device polling was on.  It is now idle at 0% with device polling off. Weird though, device polling didn't have an issue running in 2.1.x…

Was affected by the bad SSL certificate problem described in the sticky section. Had to first use a différent IP on the console and then revert it back to the original one through the web GUI. Issues my certificate through the Windows CA of the company. Are there special requirements for the SSL certificate?

Saw this : Widget "Captive Portal Status" small bug ? Another one line visual bug  ;)

yeah. If there is a way to delegate access to Squidguard only it would be really useful for us

Apparently waiting 18 hours has fixed this issue for now. Not sure what happened, but every change I make is taking now.

@cmb: Yeah deleting the leases file there is fine. I was curious how it got to be so big, but guess it's too late to ask that now. :) Hmmm… could be bigger I guess? Found a ~23MB dhcpd6.leases file here on a full install. (About a year old, DHCPv6 not used any more on that box.)

For the record - Never thought he was crazy.  Just couldn't locate the issue myself.

I am using ESXi 5.5 with VMXNET3 adapters. Turning adapters on of off worked good with version 2.1.x I use this function to identify the correct (ESXi virtual) adapter within pfSense. Maybe this also helps: When I install pfSense there is an page where you could "auto detected" the NIC's, this is also not working! :o

@Rewt0r: Not sure what has broken it but if it's in the Dev build it will soon be rolled out to the standard version, hence my post here. You're reporting it to the wrong place if that's the case. What we have works in every stable Chrome version ever released, Opera, Safari, Firefox, and IE. If it's broken in Chrome dev it's something Google needs to fix.

The key part there seems to be "Too many open files in system". What has a slew of files open is the question. Check output of 'fstat' for what all is open.

From a new pfSense user, many thanks for this theme it's more acceptable to my eyes than the inbuilt themes. Well done. :)

Oh man! Thank you guys! I have OpenVPN set to port 443/tcp, so no wonder the web configurator cannot bind to that port if it's too slow. So yes, I will bind the web configurator to a different port then. Thanks!

/conf/config.xml

Here's how I'm doing it… services->reverse proxy general/web servers tab should be straightforward.  Jump in at mappings tab - you can create a mapping group like this: .sub1.domain.com.$ .sub2.domain.com.$ .sub3.domain.com.$ .domain.com/URI1.$ .domain.com/URI2.$ .domain.com/URI3.$ .domain.com.*$ I've also had success playing with the Redirects tab rules, but that's been really unreliable.  I'll change something completely unrelated and the rules on that tab break. anyway once you're passed to the apache server on your internal network, you can do yet more with redirects and RewriteRules.

Will this be fixed or? I am running Supermicro server with no vlans. 2.1.5 - very annoying :)