PFsense Not Showing Speeds Paid For
-
BTW - Those firebox machines are just insanely expensive!
New they were but now they're cheap. The X-peak-e still command high prices but the Core boxes are <£50 here.
I had assumed the config was being restored or even just the CF card transfered, that's a good point. Is that what happened Ghostshell?
Steve
x1250e using a 2.5 SSD drive with a fresh install, no restore, x5500e 2.5 HDD same config I have been running for a while. I will post specs shortly.
-
I have a hard time believing these x-e boxes wont push that traffic… Of coarse Im running one with a 2.26ghz proc with 533fsb and 2GB RAM. So maybe thats whats helping mine?!?
Im hoping stephenw10 will comment some more as he has more experience pushing these boxes to their limits than probably anyone else here...
ghostshell- you are using the 32bit version correct? Have you tried a 2.2RC snapshot?
I am running 2.1.5 and have not tried 2.2RC
-
Just to be clear we are talking about the X5500e, the peak model, here. It has all 8 ports and the 2GHz Pentium-M as standard. It should have no problems at with 150Mbps. Neither would the X1250e. The fact that both are exhibiting the same throttling indicates some underlying issue.
Did both boxes show the same level of throttling?
Did you try both msk and sk interfaces? Did the speed change?Was your upgrade to 150/100 accompanied by a change of modem/router?
Steve
Per ISP Support my Modem DOCIS 3 will support the new speeds, unless they are lying to me? Since I switched to a biz cust I did get a new modem when I switched and it was swapped out a couple time. Both boxes when directly connected to a port on the FW show 117/98. I had port 8 open and used that to direct connect, I will try when I can one of the 1-4 port using sk and not the msk port which is what I having been using to test.
Could it just be the site I am using to test(speedtest.net)? Anyone got a better option?
-
Yes DOCIS 3will handle your throughput and sounds lik your are doing everything right.
Perhaps its just a matter of units not be being displayed correctly? Or bandwidth not displaying correctly?
If you test with a laptop directly connected to the modem and the same laptop then connected to pfsense lan and the speeds are always different, there is definitely a problem.
-
INFO
1. What kind of internet connection is this? - BROADBAND
2. Model of modem? - MOTO SB6182 ( seperate AP's and of course PFSense FW, modem is only a modem)
3. Does your pfSense box get a public or private space IP address? - I have a static IP assigned to me due to being a biz cust
4. Does the gui show that the port is negotiating at full duplex? - WAN 1000baseT <full-duplex>/ LAN 1000baseT <full- (does="" not="" show="" <full-duplex="">)</full-></full-duplex>
-
Yes DOCIS 3will handle your throughput and sounds lik your are doing everything right.
Perhaps its just a matter of units not be being displayed correctly? Or bandwidth not displaying correctly?
If you test with a laptop directly connected to the modem and the same laptop then connected to pfsense lan and the speeds are always different, there is definitely a problem.
Thats what I thought.
Laptop to modem = 145-148/98
Laptop to Port 8 (MSK) wired = 117/98Up is close enough, down is my concern
-
SPECS: I upgraded mem and CPU in each
CPU Type Intel(R) Pentium(R) M processor 2.13GHz
Version - 2.1.5-RELEASE (i386)
FreeBSD 8.3-RELEASE-p16Memory 2GB
Load average ( at time of check) - { 0.00 0.00 0.00 }
CPU Usage ( at time of check ) - 12.6%
-
What is the model of your modem? Its not a gateway device is it? (Get a modem without a router)
Does your router get a public or private IP? (Gateway device will give you a private address unless you talk the ISP into bridgemode which most gateway modems dont really do.)
Who is your ISP? (cough cough Comcast cough) makes you use a gateway device on commercial accounts only if you need the static IP. If you don't have a static IP then they will let you use your own modem. We wont know unless you tell us.
Sites such as speedtest.net are notorious (at least in the past) for being highly inaccurate.
The best way to test is to download all kinds of stuff across your network (several computers) including the SP3 update for Windows XP (IT version so you get the whole 345+ MB file) and watch the GUI graphs on the dashboard.
Last question- Your computer is the only computer pulling data through the firewall during your speed tests…. right? Antivirus programs are notorious for requesting their updates as soon as they connect to a network.
-
What is the model of your modem? Its not a gateway device is it? (Get a modem without a router)
Does your router get a public or private IP? (Gateway device will give you a private address unless you talk the ISP into bridgemode which most gateway modems dont really do.)
Who is your ISP? (cough cough Comcast cough) makes you use a gateway device on commercial accounts only if you need the static IP. If you don't have a static IP then they will let you use your own modem. We wont know unless you tell us.
Sites such as speedtest.net are notorious (at least in the past) for being highly inaccurate.
The best way to test is to download all kinds of stuff across your network (several computers) including the SP3 update for Windows XP (IT version so you get the whole 345+ MB file) and watch the GUI graphs on the dashboard.
Last question- Your computer is the only computer pulling data through the firewall during your speed tests…. right? Antivirus programs are notorious for requesting their updates as soon as they connect to a network.
See reply #35 above for most of the info you asked for. ISP is COX. Yes, during test all services and devices are shutdown and only the laptop connected to port 8 (MSK) is active.
-
Just to give a heads-up:
I run here similar setup here in the office.
-
Motorola / CV6181E (DOCSIS 3.0)
-
Watchguard x1250e, celeron M cpu (original), MEM upgraded to 2GB, running pfSense 2.1.5 nano on CF.
The modem is directly connected to sk0 (port "0")
No issue hitting my download limit (200Mbps) on a vm when no other traffic is going on, and that is with SNORT running… So I don't think it is to blame on the red box, or at least not in its default config.
Wild guessing here, but did you try swapping back to the default RAM in your box? (it would be my approach to go back to the original HW situation... if that is possible of course)
--edit spelling--
-
-
WAN 1000baseT <full-duplex>/ LAN 1000baseT <full- (does="" not="" show="" <full-duplex="">)</full-></full-duplex>
This could be a clue here. What does ifconfig show? Perhaps you would normally connect the laptop via a switch and are connecting it directly just for this test? That could show up a hardware incompatibility that isn't normally an issue. Are there any errors or collisions showing in Status: Interfaces:
Failed Ethernet speed negotiation is normally toward the top of my list of suspects when throughput is throttled.
To determine whether the throttling is at the WAN or LAN interface I would normally try to run a test from the pfSense box itself. To do that you need to know a test file that can max your WAN bandwidth. Here is a result using a file got from http://www.cloudtestfiles.net/[2.1.5-RELEASE][root@pfsense.fire.box]/root(1): fetch -o /dev/null http://109.123.87.183/speedtest.256mb /dev/null 100% of 256 MB 8444 kBps 00m00sYou will probably need a different address/file.
Steve
-
I'm with stephen on this one.
If you have been using a switch between pfsense and the laptop, try a different switch or no switch.If you have not been testing with switch but have been directly connecting to pfsense lan, try it with a switch.
Also, MAYBE try it with a different cable on the lan. Sounds like you can overcome this.
-
I'm with stephen on this one.
If you have been using a switch between pfsense and the laptop, try a different switch or no switch.If you have not been testing with switch but have been directly connecting to pfsense lan, try it with a switch.
Also, MAYBE try it with a different cable on the lan. Sounds like you can overcome this.
Fingers crossed!
Whats a good site to test with. full-duplex shows up after a sec when going to the main status page, on the terminal all interfaces show full-duplex while the webgui takes a sec to show all.
EDIT: i am trying http://www.cloudtestfiles.net/ now
-
I'd do the same test that's giving you the better speeds with the pfSense out of the picture.
-
I'd do the same test that's giving you the better speeds with the pfSense out of the picture.
cloudtest 256MB file, 15 sec, cant see the speed in firefox dammit!
-
Unplugged all from the network except my laptop directly connected.
1000baseT <full-duplex>The <full- was="" due="" to="" it="" showing:<br="">1000baseT <full-<br>duplex,flowcontrol,rxpause,txpause>
then after about 2 sec it shows
1000baseT <full-duplex>in the web gui.
Current speed test shows improvement: 130.98/111.78
Time to call support I think since if I connect directly to the modem i get the same.</full-duplex></full-<br></full-></full-duplex>
