I managed to solve the problem by changing the configuration of the network addressing. [image: 264mmw5.jpg] Now i have Internet connection in my test platform.

My guess would be that pfsense uses a combination of time/packets (pps) and packet size (64-1500) to calculate speed (Mb/s). So it's possible there were a few large packet counts (maybe jumbo frames) for a few milliseconds and it screwed up the math? This is a complete BS answer I'm just guessing.

Well that Intel page also conflicts with the Dell specs page: http://www.dell.com/us/dfb/p/poweredge-r200/pd#TechSpec That also states ECC RAM and Xeon X3200 series CPUs. The chipset used supports ECC RAM: http://ark.intel.com/products/34387/Intel-3200-Memory-Controller The ECC line in the CPU spec seems irrelevant. The X3230 doesn't bother to list it. Steve

@ghackett: Namely, I have a domain name via dyndns, lets call it mydomain.com and I would like all traffic addressed to specific subdomains to be forwarded to different servers within my NAT… i.e. in-bound traffic to webhost1.mydomain.com would route to 192.168.0.2 and traffic to webhost2.mydomain.com would route to 192.168.0.3, etc. As I recall, the paid version of dyndns allows you to do redirection as well. What you can then do is to add a redirect Webhop for the individual sub-domains with individual ports. eg. webhost1.mydomain.com will redirect to <your wan="" ip="">:22 webhost2.mydomain.com will redirect to <your wan="" ip="">:23 And you'll use pfSense to port forward <your wan="" ip="">:22 to the internal server that hosts webhost1.mydomain.com And port forward <your wan="" ip="">:23 to the internal server that hosts webhost2.mydomain.com</your></your></your></your>

So when you say 'it's not VLAN'd off yet', you mean that everything is in the same network space, correct? The LAN and the Wireless network would be the same network then, so you would need a way to intercept and log packets from just the AP's. Is it possible? Yea… With Squid most likely... but it will be a pain. Better to just enable Squid globally, and give DHCP reservations to your LAN computers so you can differentiate between wired/wireless hosts.

As a side note, Im looking at about 10% of the speed currently I had a few weeks ago….

So, I guess it is worth mentioning that the LAN clients really can't access the server by host name locally as a limitation of the software they are connecting to.  It expects all clients on the 'base url', which I have to configure as the external fqdn.  I've yet to think of a way that I can have the services both internally and externally accessible at the same time, which is a problem other users have found.  This is not a pfsense issue. I think this one is unsolvable for what I need… at first I was hoping it would be as simple as some firewall rules, but I don't think it will work out. Thanks anyway to those who gave it a look.

@stephenw10: Watch the RRD WAN quality graphs to get some idea of your ping times and packet loss rate. In System: Routing: click the 'e' to edit your default gateway. Click advanced to access the apinger parameters. Try increasing them if they are below values you regularly see on your WAN. Alternatively you can disable apinger altogether as a test by checking 'Disable Gateway Monitoring'. Steve I've experiencing the same issue with my cable modem connection over the past few months. I ended up increasing the gateway monitoring parameters that Stephen is talking about. Once in a while I'll see some entries in the GW log but I can live with it, before it was 4-5 times a day. On another note, apinger has become really sensitive since 2.1. Before I changed my parameters, a speedtest within my provider networks would trigger an alarm. Stephen

Hello again, new problem. i have activated public 3g IP from my Carrier, and im trying To get it running. it says i should use the apn internet4 but it just doesnt work, it shows the green Arrow on wan that the link is up but no IP adress in the textbox under it. the modem is connected judging by the constant lightblue light. When i connect the modem to my computers usb port and enter apn internet4 it works, i get public ip address. any ideas? EDIT: Ok so now i it works, i just restarted the router a few times and removed the modem and put it back and nowi got public ip on my 3g connection! :)

#NSA #BUMP

@stephenw10: There is now package for apcupsd which is especially for this. There is no need to connect more than one machine to the UPS. Whichever machine is connected runs a daemon that broadcasts messages over the network to inform other machines, running the client, of the UPS status. I've not tried the pfSense package, it was only released a few days ago, so I don't know it's capabilities. It will be able to run in server mode in which case you would connect the UPS to the pfSense box and then configure the NAS to look at it. The NAS is probably running apcupsd underneath but I have no idea if it's able to be configured to allow a client on pfSense to lok at it. Try it and see. Steve Thanks for the suggestion Steve  ;D The Synology is running some (I don't know which one yet) server, so I would need a tool that gets the signal from the Synology. I already found the thread of the apcupsd package, so I will ask there too. [image: ups004.jpg] [image: ups004.jpg_thumb]

i have just started experiencing this problem. though it seems to be occuring with add-on intel cards see also http://forum.pfsense.org/index.php/topic,70832.0.html

Or better yet, once it's solved post a summary of what the problem was and how it was fixed, for future reference.

Well its a 16-bit EtherExpress 10/100, so its not cardbus (32-bit). USB adapter was recognized for its chipset (dmesg showed it and the system reacted when I unplugged/plugged it in).  But apparently no native drivers. Flipping the interfaces from wan/lan to lan/wan had no effect, other than the collisions stayed with the 16-bit card. I'm hoping its a faulty card.  pfsense box is now back on the test bench pending arrival of the DLink (hcl compatible) pcmcia card.  I may try to load drivers for the USB adapter just for grins while I wait for the big brown truck to arrive later this week. More later.

You may like to read through this, unsolved, thread and see if the finding there line up with yours: http://forum.pfsense.org/index.php/topic,70773.0.html There was/is a known issue that might be effecting you: https://redmine.pfsense.org/issues/2555 Steve

@opjohnny: @biggsy: @opjohnny: Unless ESXi knows how to send a shutdown command to pfSense … It does.  You just need to select the right shutdown mode: That simply just "cuts power" to the VM before ESXi continues its shutdown process.  It's like holding the power button on a PC or laptop until it turns off. No, it does not just cut the power to the VM.  For me, this is the key benefit of VMware tools.  I used it, quite deliberately, yesterday to shut down a whole host without having to shut down each of the individual VMs beforehand. Of course, if ESXi isn't told to shut down gracefully, as happens in a power-fail with no UPS …

You can restart the monitoring daemon (Status > Services) and it will wipe out the old loss data. We have seen that happen before but have not been able to replicate the circumstances that result in the problem.

@phil.davis: Now I understand your problem - I hadn't looked hard at the "custom provider" bit of your needs. The software that generates the GUI form - services_dyndns_edit.php - is just lacking code to add gateway groups to the dropdown for the "Interface to send update from" field. It really should have the same list as the "Interface to monitor" field. I appreciate the heads up on the pages that you wrote; I'll give those a try shortly. Thanks Phil!

Sounds like you have a flakey line. Try setting up two failover pools and split the clients between them- see if the cable or the DSL people complain more. I've never liked round robin with lines that aren't the same.