Interfaces: WAN is where both the assigned static IP address and the gateway IP address are configured.

The CIDR network (mask bits) must also be set correctly.

OK, this is a serious and confirmed PEBKAC.  ::)

This is the completed system…...

20150430_173515.jpg
20150430_173515.jpg_thumb

Looks like I had a couple of issues.
1:  Putty wasn't showing me the right stuff and even when it did it wouldn't let me type.
2:  While the 64-bit archive downloaded for the Tx site ine and unpacked fine it appears to have been corrupt.  I downloaded it from the Ky mirror and it loads fine.
3:  I forgot about the whole assigning NICs thing before it would load.  DOH!  Using Hyperterminal I was able to get in and assign them.

I had the same issue after updating from 2.1.5 to 2.2.2.

The dashboard show the OS updated, the widgets had a new look but the system never really rebooted.  The old kernel was still being used.  Uptime was the same:
–------------------------------------------------------------------------------------------------------------------------------------------
Version 2.2.2-RELEASE (i386) built on Mon Apr 13 20:10:33 CDT 2015
FreeBSD REDACTED 8.3-RELEASE-p16 FreeBSD 8.3-RELEASE-p16 #0: Mon Aug 25 08:25:41 EDT 2014 root@pf2_1_1_i386.pfsense.org:/usr/obj.i386/usr/pfSensesrc/src/sys/pfSense_SMP.8 i386

You are on the latest version.
Platform pfSense
CPU Type Intel(R) Pentium(R) D CPU 3.00GHz
Uptime 238 Days 23 Hours 24 Minutes 25 Seconds

I tried multiple times rebooting via the webGUI it didn't work.  I ended up having to push the power button on the server to get a clean shutdown and reboot.  Then the Uptime looked fine.  Error message went away.  After login, pfSense said that it was regenerating SSH keys.  This procedure worked for two server updates.

You can do a few things.

1. Set up your workstation to be behind that vm in general. I do something similar with virtualbox on one of my work computers, and I have virtualbox using my NIC as the WAN for pfsense, and I have another virtual NIC that is 'LAN' or equivalent that my laptop uses as if it was it's only network interface. The pfsense VM's wan gets whatever is at the physical ethernet port of the machine, and then my laptop gets it's route through a virtual NIC and is behind pfsense.

Yes, pfsense has to be booted before packets flow at all, so that may not be ideal. Works great for me, pfsense boots fast enough it makes almost no difference. Once you do this, you can usually pass whatever captive portal exists because you are natted behind the MAC and IP they are seeing.

2. Create a script that uses cURL to do a POST or whatever action is normally taken to get you into the system. For example, pfsense captive portal just wants a POST to a certain URL (when it's configured with no auth), and so you could do something like

curl --data "" http://firewall.ip:port/

At that point, you figure out how long before they idle you off, and schedule your script to run accordingly.

3. Clone the MAC address of a device that already passed the hotel's auth

This is a little trickier, I use an old smartphone for this. Basically, use the mobile device to pass the portal, then make your pfsense WAN interface bound to that MAC. I've done this many many times over the years with varying success.

I have a friend who runs an ISP.  He provides Internet to a condo building in Downtown Atlanta somewhere.  One of the issues he has run it to (and since solved) is end users plugging in their consumer router/firewalls "backwards"  with the LAN side connected to the WAN.  The built-in, on by default, DHCP server on these devices would take down the whole building.

So, to make a point…yes you only want one DHCP server enabled or your network will not work right, at all.  Basically you'll DOS yourself.

Modern switch software allows you to lock down which port DHCP requests can be replied from to guard against this kind of problem.

Upgrade are generally fine, but depending on how old the hardware is (especially the disk), it may be a good time to reinstall.

Reinstall+restore should work identically to an upgrade except for a few bits that may be left over during an upgrade. Usually if it doesn't appear to work properly it's due to some difference in the image or configuration's console settings, interfaces, etc.

Looks like you submitted the same question via the support portal this morning and someone responded similarly. Probably best to continue there if you'd like to pursue figuring out why the reinstall didn't work as expected.

Weekend? This takes about 10 minutes, reinstall and restore the configuration backup.

Turned out to be pretty simple. 
IPCop hosts table extracts easily into a comma delimited format and that can be converted into the XML of the XML layout of the backup file, that can then be restored.

(I don't know how to format code on this board, so forgive me..)
Simple batch job to do the conversion, then paste the output in between the appropriate sections of the backup file, then restore.

I think the biggest issue here is that I couldn't find anything on this because I couldn't figure out appropriate keywords for searches.
I know someone else has had this issue before and already solved it…

#!/bin/ksh -x
INFILE=$1
OUTFILE=$2
for LINE in cat $INFILE
do
IP=echo $LINE | cut -f1 -d,
HOST=echo $LINE | cut -f2 -d,
DOMAIN=echo $LINE | cut -f3 -d,

print "\t<hosts>" >> $OUTFILE
print "\t\t<host>$HOST</host>" >> $OUTFILE
print "\t\t<domain>$DOMAIN</domain>" >> $OUTFILE
print "\t\t<ip>$IP</ip>" >>$OUTFILE
print "\t\t<descr>" >> $OUTFILE
print "\t\t<aliases>" >> $OUTFILE
print "\t</aliases></descr></hosts>" >> $OUTFILE
done

Downgraded to 2.2.1 and no sync as well,

Will downgrade to 2.2 and test

I'd just like to give an update to my problem as I've finally able to install 2.2 onto my system. I had tinkered with the SCSI drives, cables, assortment of system tunables, boot loader configs, etc but nothing helped. I was able to reinstall 2.1.5 just fine, but somehow when it comes to 2.2 the system would get stuck at that dreaded "bsdlabel" command.

Then one fine morning when the system booted up with the 2.2. Install CD, I noticed something odd with how 2.2 reported the transfer rates of the two SCSI disks attached to the Adaptec controller. It reported the SCSI disks as having a transfer rate of about 11MB/s. I can't remember how it was reported under 2.1.5, but I have a SCSI-2 disk there, which should give me a transfer rate of 20MB/s. But I also knew pfSense loaded the correct disk driver (ahc) for it. And so, I went to the SCSI BIOS and had a look at what I could change so that pfSense 'detects' the appropriate transfer rate.

Under the Adaptec SCSI BIOS, the "Initiate Wide Negotiation" was set at a default of "Yes" and at the default transfer rate of 40MB/s. I disabled the setting by selecting "No" and with that the maximum default transfer rate went down to 20MB/s (the 40MB/s option was no longer there).

With the above setting in the SCSI BIOS, the installation was no longer stuck at bsdlabel (in fact, the command just did it's thing within a second as it should) and I was able to complete the installation of 2.2 and restored the config.

Hope this help someone with a similar setup and issue.

No idea what the problem was.

I reinstalled 2.2.1 and restored a config backup so now I'm back to square one.

I think I'll refrain from upgrading to 2.2.2 for a bit.

Hello again,
Well, a big THANK YOU for the replies. Disabled serial port 1 and 2 in BIOS and have rebooted a few times from both slice one and slice two. The only thing now is to re-do all the settings but that's a trade off I'm happy to pay. Thanks again.
Marked as resolved.

I finally got around to trying this today.
It booted and ran 2.1.5 just fine, and I had it auto-update to 2.2.2.  Upon rebooting, it had the same ATA problem and didn't boot.

Back to the CF card with M0n0wall for now…

They will hang onto the existing state. Did you make sure the checkbox under advanced, misc 'State Killing on Gateway Failure' is unchecked?
Also make sure the provider is allowing the phones to register from the WAN2 IP.