Ok that would make sense :) But the 4 Gigabyte in note refers to the amount of data that will be transfered until the counter overflows. It's not like gigabyte stands next to the error-counter itself.

Got it working RC2 Fri Dec 12 16:00:58 EST 2008. ryts

thanks for your help. I noticed ntpd is usable about 10-15 mins after a reboot of my pfsense box but your right, its not showing in the services as running but using the ps command shows it is there.

I am getting the same problem running 1.2.1 RC2. They are both stuck in this state. "dhcp0"  recover  2008/12/11 13:44:22  unknown-state  2008/12/11 13:44:22 Neither will answer DHCP requests in this state. Both of the firewalls are synced on time and I have a firewall rule on the LAN interface right now that just allows everything from anywhere and still no luck. I did use the right IP addresses in the Failover peer section.

I run 1.2.1 rc2  on FX5621 and I'm getting these watchdog timeouts too. They start on ET0 but can then be seen on the other RL interfaces as well. As seen by peterclo these occur after about 2days of uptime. But from time to time they occur immediately on bootup. For my setup I only needed 2 interfaces (LAN/WAN) for a dedicated transparent squid box on one WAN of my main firewall (also 1.2.1, but not FX5621 hardware). Now a few days after switching to the 100Mbit RL interfaces only, I have not seen any watchdog timeouts. So it looks like there may be a problem with the ET0/1 driver/config on 1.2.1 when running fx5621 hardware. Well that makes two of us, hard to believe there are just two of us running 1.2.1 rc2 on FX5621, is nobody else seeing this? Nick.

http://blog.pfsense.org/?p=284 Firewalling for OpenVPN is in 2.0

The problem is ntop - an update with ntop installed knocks out RRD and requires an update without ntop to get it going again. Re-installing ntop causes the same symptoms, with syslog entries as posted above. ryts

And we're good to go! For those interested, in the second thread fredde posted (http://forum.pfsense.org/index.php/topic,11594.15.html) AudiAddicts posts are what worked for me (and here they are): Problem fixed with the help of dvserg!! (THANKS!!) He asked me to post the instructions Remove package from GUI Go to the firewall shell and type pkg_info If lightsquid is listed, use PKG_delete to remove it (pkg_delete lightsquid-1.7.1_1) Reinstall from GUI now and it should be fixed! Uninstall like explained above and then reinstall the package in the gui then type in the shell : ln -s /usr/local/bin/perl /usr/bin/perl Make sure the log setting is switched on for squid and set to the standard dir and then go to lightsquid and refresh/update settings and it should work.

Is the WAN interface slowness fixed in the lastest build?

It only looked like it. Actually it's the same. I'm going back to freebsd 6 based pfsense.

This is a default installation with changes only to local lan ip-range. One WAN connection with DHCP. One LAN connection with DHCP 192.168.10.*. No aditional rules added. ISP:s do not recognize this DNS problem as a problem of them. When changing to laptop directly connected to WAN modem, DNS is correctly set.

this is specific to the 1.2 load balancing with slbd. If you are not using load balancer pools it will perfectly fine and you will not see any flapping. The RRD graphs have nothing to do with this.

I would love to leave it in auto, but it will not work unless manually set. It could be an issue with the NIC driver, but there are issues with Intel NIC drivers also, and they are onboard nics. :( I will switch back to 1.2 Release and see if the problem persists.

Well, I changed the firewall rules to conservative without deleting the other rules and it fixed it. I guess I'll have to read up on that option Thanks for all your HELP!  ;D

Tried a firmware upgrade, it says latest version dec 05, downloads, it says updating, then the fw reboots, when clicking on fw update again, it shows that it's not been updated. Another test regarding the alias settings, I add an ip to the alias list (to allow IP access to a certain server) and this doesn;t work until I reboot several times.

http://forum.pfsense.org/index.php/topic,9414.0.html

It just started happening with the 1.2.1 release, so maybe some caching did get turned on during the change. The values change right away with 1.2.

@MattMeyer: 1.  The file size seems too big.  If I install a fresh image using the ISO, manually install the tools package, then zip up the vmdk, the file size is just around 50M or the same as the ISO.  This is what I was expecting, but the standard virtual appliance is over 3x larger.  I cannot figure out why this is. I wondered the same, there is only about 100 MB of data on the vmdk, but even after a defrag it's significantly larger than that.  Wrote it off, as I don't have time to chase that one down right now.