Throughput from Lan to Wan
-
@johnpoz Before:
TCP Global Parameters ---------------------------------------------- Receive-Side Scaling State : disabled Receive Window Auto-Tuning Level : disabled Add-On Congestion Control Provider : default ECN Capability : disabled RFC 1323 Timestamps : disabled Initial RTO : 1000 Receive Segment Coalescing State : enabled Non Sack Rtt Resiliency : disabled Max SYN Retransmissions : 4 Fast Open : enabled Fast Open Fallback : enabled HyStart : enabled Proportional Rate Reduction : enabled Pacing Profile : offAfter with: netsh int tcp set global autotuninglevel=normal
TCP Global Parameters ---------------------------------------------- Receive-Side Scaling State : disabled Receive Window Auto-Tuning Level : normal Add-On Congestion Control Provider : default ECN Capability : disabled RFC 1323 Timestamps : disabled Initial RTO : 1000 Receive Segment Coalescing State : enabled Non Sack Rtt Resiliency : disabled Max SYN Retransmissions : 4 Fast Open : enabled Fast Open Fallback : enabled HyStart : enabled Proportional Rate Reduction : enabled Pacing Profile : offiperf3.exe -c 192.168.8.248 -V iperf 3.1.3 CYGWIN_NT-10.0 WernerLaptop 2.5.1(0.297/5/3) 2016-04-21 22:14 x86_64 Time: Wed, 22 Feb 2023 13:38:39 GMT Connecting to host 192.168.8.248, port 5201 Cookie: WernerLaptop.1677073117.500543.72f2e TCP MSS: 0 (default) [ 4] local 192.168.8.34 port 2341 connected to 192.168.8.248 port 5201 Starting Test: protocol: TCP, 1 streams, 131072 byte blocks, omitting 0 seconds, 10 second test [ ID] Interval Transfer Bandwidth [ 4] 0.00-1.00 sec 112 MBytes 936 Mbits/sec [ 4] 1.00-2.00 sec 113 MBytes 945 Mbits/sec [ 4] 2.00-3.00 sec 113 MBytes 949 Mbits/sec [ 4] 3.00-4.00 sec 112 MBytes 942 Mbits/sec [ 4] 4.00-5.00 sec 113 MBytes 947 Mbits/sec [ 4] 5.00-6.00 sec 113 MBytes 949 Mbits/sec [ 4] 6.00-7.00 sec 113 MBytes 949 Mbits/sec [ 4] 7.00-8.00 sec 112 MBytes 943 Mbits/sec [ 4] 8.00-9.00 sec 112 MBytes 941 Mbits/sec [ 4] 9.00-10.00 sec 112 MBytes 943 Mbits/sec - - - - - - - - - - - - - - - - - - - - - - - - - Test Complete. Summary Results: [ ID] Interval Transfer Bandwidth [ 4] 0.00-10.00 sec 1.10 GBytes 945 Mbits/sec sender [ 4] 0.00-10.00 sec 1.10 GBytes 945 Mbits/sec receiver CPU Utilization: local/sender 19.2% (2.1%u/17.1%s), remote/receiver 37.3% (1.8%u/35.7%s) iperf Done.iperf3.exe -c 192.168.8.248 -V -R iperf 3.1.3 CYGWIN_NT-10.0 WernerLaptop 2.5.1(0.297/5/3) 2016-04-21 22:14 x86_64 Time: Wed, 22 Feb 2023 13:39:01 GMT Connecting to host 192.168.8.248, port 5201 Reverse mode, remote host 192.168.8.248 is sending Cookie: WernerLaptop.1677073141.672333.7cdce TCP MSS: 0 (default) [ 4] local 192.168.8.34 port 2351 connected to 192.168.8.248 port 5201 Starting Test: protocol: TCP, 1 streams, 131072 byte blocks, omitting 0 seconds, 10 second test [ ID] Interval Transfer Bandwidth [ 4] 0.00-1.00 sec 112 MBytes 942 Mbits/sec [ 4] 1.00-2.00 sec 107 MBytes 898 Mbits/sec [ 4] 2.00-3.00 sec 112 MBytes 939 Mbits/sec [ 4] 3.00-4.00 sec 113 MBytes 945 Mbits/sec [ 4] 4.00-5.00 sec 110 MBytes 919 Mbits/sec [ 4] 5.00-6.00 sec 102 MBytes 855 Mbits/sec [ 4] 6.00-7.00 sec 109 MBytes 917 Mbits/sec [ 4] 7.00-8.00 sec 110 MBytes 923 Mbits/sec [ 4] 8.00-9.00 sec 95.4 MBytes 800 Mbits/sec [ 4] 9.00-10.00 sec 107 MBytes 899 Mbits/sec - - - - - - - - - - - - - - - - - - - - - - - - - Test Complete. Summary Results: [ ID] Interval Transfer Bandwidth Retr [ 4] 0.00-10.00 sec 1.05 GBytes 904 Mbits/sec 0 sender [ 4] 0.00-10.00 sec 1.05 GBytes 904 Mbits/sec receiver CPU Utilization: local/receiver 53.9% (6.7%u/47.1%s), remote/sender 12.8% (3.3%u/9.5%s) iperf Done. -
@orkopaede your scaling is set to disabled still..
"Receive-Side Scaling State : disabled"
-
@johnpoz i see... I changed it also ecn to enabled but still
TCP Global Parameters ---------------------------------------------- Receive-Side Scaling State : enabled Receive Window Auto-Tuning Level : normal Add-On Congestion Control Provider : default ECN Capability : enabled RFC 1323 Timestamps : disabled Initial RTO : 1000 Receive Segment Coalescing State : enabled Non Sack Rtt Resiliency : disabled Max SYN Retransmissions : 4 Fast Open : enabled Fast Open Fallback : enabled HyStart : enabled Proportional Rate Reduction : enabled Pacing Profile : offC:\Users\Werner>Z:\Exchange\Firla\Software\iperf\iperf3.exe -c iperf.par2.as49434.net -p 9232 -V iperf 3.1.3 CYGWIN_NT-10.0 WernerLaptop 2.5.1(0.297/5/3) 2016-04-21 22:14 x86_64 Time: Wed, 22 Feb 2023 13:47:48 GMT Connecting to host iperf.par2.as49434.net, port 9232 Cookie: WernerLaptop.1677073668.275288.50841 TCP MSS: 0 (default) [ 4] local 192.168.8.34 port 2500 connected to 193.177.162.41 port 9232 Starting Test: protocol: TCP, 1 streams, 131072 byte blocks, omitting 0 seconds, 10 second test [ ID] Interval Transfer Bandwidth [ 4] 0.00-1.00 sec 7.00 MBytes 58.5 Mbits/sec [ 4] 1.00-2.01 sec 7.75 MBytes 64.8 Mbits/sec [ 4] 2.01-3.01 sec 7.75 MBytes 64.7 Mbits/sec [ 4] 3.01-4.02 sec 8.00 MBytes 66.9 Mbits/sec [ 4] 4.02-5.00 sec 7.50 MBytes 63.8 Mbits/sec [ 4] 5.00-6.00 sec 7.88 MBytes 66.0 Mbits/sec [ 4] 6.00-7.01 sec 7.88 MBytes 65.6 Mbits/sec [ 4] 7.01-8.00 sec 7.88 MBytes 66.5 Mbits/sec [ 4] 8.00-9.00 sec 7.88 MBytes 66.1 Mbits/sec [ 4] 9.00-10.01 sec 7.75 MBytes 64.8 Mbits/sec - - - - - - - - - - - - - - - - - - - - - - - - - Test Complete. Summary Results: [ ID] Interval Transfer Bandwidth [ 4] 0.00-10.01 sec 77.2 MBytes 64.8 Mbits/sec sender [ 4] 0.00-10.01 sec 77.2 MBytes 64.8 Mbits/sec receiver CPU Utilization: local/sender 1.5% (0.3%u/1.2%s), remote/receiver 4.8% (0.3%u/4.4%s) iperf Done.To be honest, I'm still trying to understand how that NAT ... Thing.. works. what you wrote before... :(
-
@orkopaede it is just an outbound nat.
In my example when pfsense sees traffic to 2.42 from 9.100 it changes the IP from being source of 9.100 to pfsense IP on that interface, in my case 2.253.. Just like what happens when your lan clients goes to something on the internet and pfsense changes the IP to pfsense wan IP (normally some public IP).
notice the states when I do that..
Now when not natting - which pfsense wouldn't do between local networks. See the states
-
@johnpoz okay i think i got it:
Speeds are i would say okay the 8.227 is here probably the bottle neck:
iperf3.exe -c 192.168.8.227 -V iperf 3.1.3 CYGWIN_NT-10.0 WernerLaptop 2.5.1(0.297/5/3) 2016-04-21 22:14 x86_64 Time: Wed, 22 Feb 2023 14:25:03 GMT Connecting to host 192.168.8.227, port 5201 Cookie: WernerLaptop.1677075903.687139.6ae82 TCP MSS: 0 (default) [ 4] local 172.28.0.1 port 1188 connected to 192.168.8.227 port 5201 Starting Test: protocol: TCP, 1 streams, 131072 byte blocks, omitting 0 seconds, 10 second test [ ID] Interval Transfer Bandwidth [ 4] 0.00-1.00 sec 94.8 MBytes 794 Mbits/sec [ 4] 1.00-2.00 sec 96.9 MBytes 813 Mbits/sec [ 4] 2.00-3.00 sec 85.1 MBytes 714 Mbits/sec [ 4] 3.00-4.00 sec 89.8 MBytes 753 Mbits/sec [ 4] 4.00-5.00 sec 89.5 MBytes 751 Mbits/sec [ 4] 5.00-6.00 sec 93.5 MBytes 784 Mbits/sec [ 4] 6.00-7.00 sec 68.8 MBytes 576 Mbits/sec [ 4] 7.00-8.00 sec 72.8 MBytes 609 Mbits/sec [ 4] 8.00-9.00 sec 82.5 MBytes 695 Mbits/sec [ 4] 9.00-10.00 sec 95.4 MBytes 800 Mbits/sec - - - - - - - - - - - - - - - - - - - - - - - - - Test Complete. Summary Results: [ ID] Interval Transfer Bandwidth [ 4] 0.00-10.00 sec 869 MBytes 729 Mbits/sec sender [ 4] 0.00-10.00 sec 869 MBytes 729 Mbits/sec receiver CPU Utilization: local/sender 12.0% (2.0%u/10.0%s), remote/receiver 19.8% (5.2%u/14.6%s) iperf Done.any way it should be enough for the 200Mbit Internet connection
-
@orkopaede I don't see any nat happening in your states.. Notice in my states where it shows 192.168.9.100 changed to the wlan interface IP of 192.168.2.253
If you had an existing state between those 2 IPs, then the nat wouldn't happen. You need to clear any old states between those IPs and that 5201 port so that the new connection would use the nat.
But hard to believe that "nat" alone would be that much of a hit..
edit: just for grins.. Could you disable the Kernel PTI.. That for sure could be a hit on performance.. There is almost zero sort of use case where that would be needed on a firewall.. I find it unlikely it could be such a hit to only see 16mbps vs your 200.. But have a hard time coming up with a use case where you would want/need to enable that on a firewall. And it would be a hit, the level of which is hard to say.. But why have anything enabled that would hinder the performance of your box? Unless you have a very good reason to, and have to take the performance hit.
https://docs.netgate.com/pfsense/en/latest/config/advanced-misc.html#kernel-page-table-isolation-pti
While more secure, this protection can incur a performance penalty. If untrusted users do not have access to run arbitrary code on the firewall, it can be disabled without significant security risk.
-
@johnpoz I will give that a try.
For the NAT problem. I guess the nat didn't work because everything was or is allowed in the firewall rules for this interface. How would the rules have to be set here so that the NAT can still work? Allow everything except LAN interface where the 8.227 is connected?
-
@orkopaede nothing should have to change in the firewall rules, if the traffic was allowed before. But if you had an existing state, you would need to clear out those old states that were not doing nat.. Or just wait til they go away on their own on timeouts.
-
@johnpoz Okay then I'll just reset the stats or wait. I have to go and pick up my child from kindergarten anyway so.
First of all, thank you for your time and help. I'll test further tomorrow. Thank you! -
@johnpoz Finally with the last test
Speeds:
iperf3.exe -c 192.168.8.227 -V iperf 3.1.3 CYGWIN_NT-10.0 WernerLaptop 2.5.1(0.297/5/3) 2016-04-21 22:14 x86_64 Time: Wed, 22 Feb 2023 15:07:53 GMT Connecting to host 192.168.8.227, port 5201 Cookie: WernerLaptop.1677078473.811245.0a2fc TCP MSS: 0 (default) [ 4] local 172.28.0.1 port 5984 connected to 192.168.8.227 port 5201 Starting Test: protocol: TCP, 1 streams, 131072 byte blocks, omitting 0 seconds, 10 second test [ ID] Interval Transfer Bandwidth [ 4] 0.00-1.00 sec 77.4 MBytes 648 Mbits/sec [ 4] 1.00-2.00 sec 81.8 MBytes 686 Mbits/sec [ 4] 2.00-3.00 sec 87.6 MBytes 735 Mbits/sec [ 4] 3.00-4.00 sec 65.5 MBytes 550 Mbits/sec [ 4] 4.00-5.00 sec 96.8 MBytes 812 Mbits/sec [ 4] 5.00-6.00 sec 90.2 MBytes 757 Mbits/sec [ 4] 6.00-7.00 sec 91.9 MBytes 771 Mbits/sec [ 4] 7.00-8.00 sec 75.6 MBytes 635 Mbits/sec [ 4] 8.00-9.00 sec 78.8 MBytes 660 Mbits/sec [ 4] 9.00-10.00 sec 88.4 MBytes 742 Mbits/sec - - - - - - - - - - - - - - - - - - - - - - - - - Test Complete. Summary Results: [ ID] Interval Transfer Bandwidth [ 4] 0.00-10.00 sec 834 MBytes 699 Mbits/sec sender [ 4] 0.00-10.00 sec 834 MBytes 699 Mbits/sec receiver CPU Utilization: local/sender 11.0% (1.4%u/9.6%s), remote/receiver 2.5% (0.6%u/1.9%s)Okay now i have to go.
-
@orkopaede you can always just kill any state in the state table directly. Under diagnostics, state table.
edit:
That seems low for whatever reason.. Notice in mine it was a very small hit to the speed, compared to just lan to lan speed.. And your on an I5.. I would think that is more powerful than my sg4860.. -
-
@orkopaede yeah you need to put the nat on the correct interface for the direction of your traffic flow ;)
-
@johnpoz ok i tested it a bit further and i think it is a windows problem. I repeated the tests on a Linux PC and I always got my 200Mbit without establishing multiple connections (iperf option -P). I don't think I have to bother anyone here with this topic. ;) Thank you once again for the help.
-
@orkopaede said in Throughput from Lan to Wan:
i tested it a bit further and i think it is a windows problem
did you actually enable window scaling - all the posts you show it disabled.edit: never mind looks like you did enable it.
-
@orkopaede I don’t see in the thread that you checked traffic shaping? I’ve seen many threads where an old setting was left enabled.
-
@steveits Hi, traffic shaping is disabled on all Interface. I never touched this part of pfsense.
-
@orkopaede So i guess it is what it is... a Windows problem.
When i find the Problem i will post it here, hopefully with a solution. -
@orkopaede Hi! Run wireshark on the Windows machine and see if anything catches your eye. Also check Windows power-saving for the NIC you are connecting from. For some poorly written drivers, Windows tends to make some bass-ackwards assumptions about what "energy saving" vs "disrupting key functionality" means.
