Should have probably stated I'm using i386 embedded (nanobsd) on a Firebox x700.

Using virtual adapters the detection code may not be able to see the change correctly. Just select the interfaces manually instead of using the auto-detect. I imagine they will be shown as em0 and em1. This shouldn't affect the normal operating of the NICs. Steve

not sure how your email got buried in my inbox (sales emails only come to me until we get our ticket system upgraded in a couple weeks) but I just responded to you. Definitely something we can help you with, can have that entire config done within a day of purchase.

Configure your access point to assign the two ssids to two VLANs. Then setup VLAN interfaces on pfSense to connect to it. That way the 'internal' and 'guest' traffic will arrive on separate interfaces so you can have different rules and CP etc. Steve

The nanobsd images are meant to be written to flash media, such as a CF card. PfSense will then boot directly from that. No hard drive needed. The memstick images are used to create a bootable USB stick that you can use to install pfSense to the hard drive. See: http://www.pfsense.org/index.php?option=com_content&task=view&id=43&Itemid=44 Steve

Any particular reason you want to downgrade ?

Thank you for your answer. For the moment, they were in raid mode, than i changed them to ahci and set up pfSense on one single disk. The problem is that i would like to ensure that if one disk fails the firewall could work further without downtime… do that work fine with gmirror?

@johnpoz: That thread is about nat reflection - which no where in your post did you state where you were trying to access these sites from.  You would have to assume from outside your firewall - since the normal operation in accessing a site while on that network would be to access it via its common IP on your own network. Also you state that wan admin access is "default" - that is not the case.  You can not access admin gui of pfsense from wan in a default state of pfsense. You would have to configure firewall rules to allow access to web gui/ssh from wan - which is not a recommended configuration. Glad you got what you wanted sorted, but you really need to accurately convey your issue and your setup if don't want people guessing what your question/problem is. You know what, you are absolutely right! I was a bit short on time to get this sorted out and the fact that my problem description sucked completely did not help anyone… Question 1: I got slightly confused as I tried to access the wan from within lan and never thought about the firewall accepting it as I was testing from inside. Question 2: I was trying to see my sites behind opt1 from LAN and got the NAT reflection issue. Stress sometimes work for you, sometimes against! Thanks and sorry for wasting your time! Ozz.

@Paul47: I was just wondering about the construction, "/dev/ufs/pfsense0". I'm unfamiliar with that. I was expecting something like "/dev/da0s1a", that's all. It's a GEOM label, so we can refer to the slice by name instead of by its physical designation. So we don't have to worry about the disk driver/device name at all, we use our own label for the filesystem. @Paul47: I know about the read-only business in nanobsd, except for a couple details: I'm guessing a restore operation automatically mounts the cf partition rw? It had to have, otherwise my restore wouldn't have worked. I'm a little shakey on where the gui automatically mounts it rw, and where I have to do it manually. I will dig into that some more. It handles that automatically for changes made in the GUI. If you want to manually make changes at the shell or over scp, etc, then you'll need to switch it manually. @Paul47: As to clicking on the logo, that makes little sense. What's wrong with a nice, obvious "cancel" button? Then people don't have to field so many newbie questions in the forum.  :) I added some text to the wizard to state that clicking the logo gets out of it.

@stephenw10: @Nachtfalke: Perhaps they are updating all sites and mirros for pfsense 2.0.3 !? Positive thinking. Nice. ;) Steve pfsense.org and packages are up again but no pfsense 2.0.3  ;)

@Wamba: Thank you Sir! Toggling with hard drive options in BIOS solved the problem! I ran into the same issue last night after a mobo/cpu/ram upgrade.  I had to change hard drive access from AHCI to IDE to allow installation. Switching back to AHCI after install causes pfSense not to boot.

There's this: http://doc.pfsense.org/index.php/Automatically_Restore_During_Install Not completely unattended though. Steve

Thanks :):) It`s work :):)

Thanks for all the feedback guys,  I have to admit I have found about 3 different ways to do things and as you say sometimes one way is better depending on what you want to do. I have to admit though, this is one very very good product that I simply can not fault.  Well that's not entirely true, some of the documentation is a little vague in places but then again with so much flexibility that's to be expected.

@joako: I'm facing similar issues attempting to run SquidGuard on embedded (NanoBSD.) The /var partition is not large enough to store the URL blacklist download, then the extracted files, etc. Simply put is there any way to increase the size of the /var partition? Then whatever issues exist with files missing after a reboot can be solved with the package itself. Well, I've recently addressed this question, e.g. NanoBSD + Squid, in this forum. Maybe, this thread is helpful: http://forum.pfsense.org/index.php/topic,59932.msg322453.html

Well you should be able to find them on your network just do a nbtstat -A to the Ips and that should give you the hostname. They are clearly clients, looking for those devices.  I guest printers from the MFC in their name. As to such problems on a guest vlan - their broadcasts would be limited to your guest vlan, so not sure how it would be a problem? As to periodically?  That 192.168.0.175 box really wants to find that printer, the is broadcasting every half second or so for it ;) If I were you I would spend some time watching your broadcast traffic and cleaning up - if they are looking for something that does not exist fix the client so he has the new name, or remove the old device from his setup.  if your seeing broadcast for stuff that answers - look to why it got all the way to broadcasting for it, broadcasting is normally last method. Should look to host, lmhost first then dns, then if node type is hybrid it would do lookup to wins if your running it then finally broadcast for the name.  So fix your name resolution, are clients/printers not registering themselves in dns?  Are they dhcp or static - pretty much any device on your network that would ever be accessed should be in dns.  And while  your at it get your reverse zones setup as well, so you can just query PTR for 192.168.0.175 and know what its hostname is, etc.

remove the package and then from the shell run pkg_add -r libxml2 Or backup your config, edit out the bad package, reinstall from an iso, then restore your config

Hi Jimp, I configured the following system tunables: kern.ipc.nmbclusters=131072 hw.igb.num_queues=1 hw.em.num_queues=1 hw.igb.fc_setting=0 hw.em.fc_setting=0 mbuf usage sometimes is 19000+ or 20000, average of 20000 my wan download is at 10mbps while my upload is at 1mbps

I have a 4GB on order…...... Will keep you posted on how I get on..... Once again.. thanks for your assistance Regards Martin

^ seems like a long way to go about it, cloning Why not just take a snapshot of your pfsense VM, and then do your update - if everything goes good your on the new machine in a simple 1 step update.  If goes wrong, just rollback to your snap Cloning and then bringing up new vm seems like more work to me ;)