That’s a very helpful diagnostic, and I will no longer suspect the Embedded edition! Given your input Phil, and now that I have an alternative pfSense appliance at my client site, I am going to completely re-install pfSense on my first device to see if the PPPoE problem is magically fixed by a fresh re-install.  I anticipate that a re-install will not fix the problem, but I want to put a nail in that coffin because it may be suggested or requested by a pfSense developer. Out of curiosity Phil, what is the make of the WAN interface on your problematic unit (e.g., Intel, Broadcom, etc.)?  I may be overlooking this information in your PPP log.

Rebooted modem, everything seems to be working now.

Thanks. Looking for the info in the wrong place… I will scribble that link into my pfsense book for when I forget about it next time.  ::)

@stephenw10: This looks to have been fixed this morning. The updaters URL now contains the correct images.  :) Yes, the images should all be OK now. cmb made some changes and put the correct images in place.

Issue has been resolved: https://github.com/pfsense/pfsense/commit/7b15d22967a9f9fefe7b8b11fa2d68c762c55219 Applied it via System Patches and rebooted, problem gone. Thanks to the pfSense team!

Thank you Jimp

The RoadWarrior connections will, but the site-site connections absolutely need it to be able to complete their connections. Note that I'm assuming (what a terrible thing to do  ::) ) that the site-site conx's are all pfSense based clients, manually configured.  The RoadWarriors are installed via the appropriate Client Export Utility entry. With those caveats, I believe everything I've stated to this point describes the operation of OpenVPN under pfSense (I'm certainly willing to be proven wrong of course).  ;)

@dgcom: I perfectly understand implications of this particular issue, and yes - it is not just a matter of replacing openssl executable… What I am saying is that recompiling everything is not very efficient. But, I guess, you know your product :) … @dgcom: shouldn't build system be smart enough to recompile only if dependencies changed? the build system is, but the system is packaged in a way that is a forklift upgrade on every update. We're exploring ways to update differently (such that updates are more like freebsd-update), but even here there are impacts that you might not imagine. PBIs are dead after 2.2, btw. @dgcom: I, personally, do not run anything, based on recent versions of openssl - except pfSense. this is probably mistake at this point.

Thanks Jason

Nikkon, Not wishing to be hurtful but whenever I ask for help on a forum (quite a lot on pfSense actually) I always report back with my results so others might learn and those who helped me might feel a bit more appreciated. I would hope that you could do the same. I know I would appreciate knowing what you did.

Thanks! I had searched but didn't find that particular post. I also had a different solution. I searched http://www.filewatcher.com for the string: "/ports/i386/packages-8.3-release/Latest/perl.tbz" this found all the mirrors and copies out there. Now, most of these links (like 90%) are dead, however by using a firefox plugin called DownloadThemAll, i told it to start a download of all the sites at once. Then I watched for which sites actually started a download, and i chose the one thar was the fastest and went there. Sure enough that site had a functioning copy if the 8.3 package tree. So i installed that and a Compat6xxx package.

Glad you got it back up and running. pfSense is one amazing piece of effective code  :D

Many thanks…that got me out of package Jail...not sure yet what it was hung on, and so far everything looks fine...will dig some more later when I have time... -ftrain

Thanks a lot!

Yes, just installed the nano update over the vga one. Now I get boot messages but no console. Thanks for the fix.

Gertjan, you helped me on another thread. Much appreciate. Give your Windows 7 each a "fixed IP" (no DHCP). Now that is an interesting idea. Give each of my computers and end devices a static IP. That would eliminate the need for NAT on the Windows computers and surrender it to pfSense where I think NAT, etc. belongs. Is that what you are saying? Some background on my "present" (definitely will change) setup: WAN through pfSense Network Appliance out single LAN NIC to Cisco SG300 10 port Managed Switch out to each computer and end device. Have 2 more NIC's in the pfSense box (not yet being used).

You could try adding the myk driver but the msk driver is newer and far more likely to work IMHO. Since the msk driver is built into the kernel I have no idea which driver would end up attaching to the card. The PCI Express error is likely not causing any problems as you've found. I have spent some time trying to stop that happening and did succeed eventually but it made no difference at all. Support for MSI in the current msk(4) driver can be buggy. You should add: hw.msk.msi_disable=1 to the file /boot/loader.conf.local.  Create that file if it doesn't exist yet. The symptoms you describe seem far more likely to be a problem with the express card hardware to me. I would guess it's being put into some power saving mode from which it's not recovering correctly. There may be options in either the BIOS or the express-card driver to restrict standby power modes. Steve

If your problem is that the drive is no longer at the logical location it installed to, because you installed on a different machine for example, then enter the new correct location at the mountroot> prompt. Then once the box has booted edit the fstab. Steve

Yes it's still possible for it to be the BIO. Depending on how the BIOS handles booting from USB devices it may not like the multiple slices that exist on NanoBSD