ePMP4600 Link Capacity - Great! Throughput? Less than great. Help?

Products ePMP
1

Alright so we have a 6GHz 4600 AP on the Bench, with a 4625 Radio linked.
AP configured for TDD, 75/25, 80MHz channel.
SM is running MCS 11, -47dBm.
Link Capacity of 735/266 Mbps, pretty consistent.

After reading some posts about 160MHz channels, we tried TDD and WLR, and the absolute highest we’ve seen is 283/290 on WLR @ 80Mhz. However, with no timing or QoS on WLR, we want to use TDD.

The speed test results leave me wondering am I doing something wrong?

Between Ookla, and IP perf testing, the absolute best we can get was using TDD on 75/25 symmetry, 80MHz channel, was 190/170 Mbps with an Ookla speedtest. IP Perf was usually worse, around 150/120.

So this isn’t normal right? We’ve tried just about everything we can think of and we seem to be unable to figure out where the other 60% of the link capacity has gone?

IP Perf results for illustration:

PS C:\Users\Me\Downloads\iperf-3.19-win64> .\iperf3.exe -c speedtest.*****communications.ca -t 10 -R
Connecting to host speedtest.*****communications.ca, port 5201
Reverse mode, remote host speedtest.*****communications.ca is sending
[  5] local 192.168.1.151 port 63842 connected to xx.xx.xx.160 port 5201
[ ID] Interval           Transfer     Bitrate
[  5]   0.00-1.01   sec  16.9 MBytes   141 Mbits/sec
[  5]   1.01-2.01   sec  16.4 MBytes   137 Mbits/sec
[  5]   2.01-3.01   sec  16.4 MBytes   137 Mbits/sec
[  5]   3.01-4.01   sec  16.4 MBytes   138 Mbits/sec
[  5]   4.01-5.01   sec  15.8 MBytes   132 Mbits/sec
[  5]   5.01-6.01   sec  15.0 MBytes   126 Mbits/sec
[  5]   6.01-7.01   sec  17.9 MBytes   150 Mbits/sec
[  5]   7.01-8.01   sec  18.5 MBytes   155 Mbits/sec
[  5]   8.01-9.00   sec  18.5 MBytes   155 Mbits/sec
[  5]   9.00-10.00  sec  12.9 MBytes   108 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-10.01  sec   167 MBytes   140 Mbits/sec    0            sender
[  5]   0.00-10.00  sec   164 MBytes   138 Mbits/sec                  receiver

iperf Done.

PS C:\Users\Me\Downloads\iperf-3.19-win64> .\iperf3.exe -c speedtest.*****communications.ca -t 10
Connecting to host speedtest.*****communications.ca, port 5201
[  5] local 192.168.1.151 port 63844 connected to xx.xx.xx.160 port 5201
[ ID] Interval           Transfer     Bitrate
[  5]   0.00-1.00   sec  14.1 MBytes   118 Mbits/sec
[  5]   1.00-2.01   sec  13.4 MBytes   112 Mbits/sec
[  5]   2.01-3.01   sec  13.4 MBytes   112 Mbits/sec
[  5]   3.01-4.01   sec  13.4 MBytes   112 Mbits/sec
[  5]   4.01-5.01   sec  13.4 MBytes   112 Mbits/sec
[  5]   5.01-6.01   sec  13.5 MBytes   113 Mbits/sec
[  5]   6.01-7.01   sec  13.5 MBytes   113 Mbits/sec
[  5]   7.01-8.01   sec  13.8 MBytes   115 Mbits/sec
[  5]   8.01-9.00   sec  13.4 MBytes   114 Mbits/sec
[  5]   9.00-10.02  sec  13.6 MBytes   113 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bitrate
[  5]   0.00-10.02  sec   135 MBytes   113 Mbits/sec                  sender
[  5]   0.00-10.02  sec   134 MBytes   112 Mbits/sec                  receiver

iperf Done.

PS C:\Users\Me\Downloads\iperf-3.19-win64> .\iperf3.exe -c speedtest.*****communications.ca -t 10 --bidir
Connecting to host speedtest.*****communications.ca, port 5201
[  5] local 192.168.1.151 port 63846 connected to xx.xx.xx.160 port 5201
[  7] local 192.168.1.151 port 63847 connected to xx.xx.xx.160 port 5201
[ ID][Role] Interval           Transfer     Bitrate
[  5][TX-C]   0.00-1.01   sec  11.1 MBytes  92.5 Mbits/sec
[  7][RX-C]   0.00-1.01   sec  4.12 MBytes  34.3 Mbits/sec
[  5][TX-C]   1.01-2.02   sec  10.5 MBytes  87.5 Mbits/sec
[  7][RX-C]   1.01-2.02   sec  4.62 MBytes  38.5 Mbits/sec
[  5][TX-C]   2.02-3.01   sec  10.2 MBytes  86.2 Mbits/sec
[  7][RX-C]   2.02-3.01   sec  4.25 MBytes  35.7 Mbits/sec
[  5][TX-C]   3.01-4.01   sec  10.9 MBytes  91.1 Mbits/sec
[  7][RX-C]   3.01-4.01   sec  4.25 MBytes  35.6 Mbits/sec
[  5][TX-C]   4.01-5.01   sec  8.75 MBytes  73.4 Mbits/sec
[  7][RX-C]   4.01-5.01   sec  3.38 MBytes  28.3 Mbits/sec
[  5][TX-C]   5.01-6.02   sec  6.75 MBytes  56.6 Mbits/sec
[  7][RX-C]   5.01-6.02   sec  3.00 MBytes  25.1 Mbits/sec
[  5][TX-C]   6.02-7.00   sec  8.25 MBytes  70.2 Mbits/sec
[  7][RX-C]   6.02-7.00   sec  4.25 MBytes  36.2 Mbits/sec
[  5][TX-C]   7.00-8.00   sec  10.0 MBytes  83.6 Mbits/sec
[  7][RX-C]   7.00-8.00   sec  6.00 MBytes  50.2 Mbits/sec
[  5][TX-C]   8.00-9.00   sec  9.75 MBytes  81.8 Mbits/sec
[  7][RX-C]   8.00-9.00   sec  6.62 MBytes  55.6 Mbits/sec
[  5][TX-C]   9.00-10.00  sec  7.88 MBytes  65.9 Mbits/sec
[  7][RX-C]   9.00-10.00  sec  6.75 MBytes  56.5 Mbits/sec
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID][Role] Interval           Transfer     Bitrate         Retr
[  5][TX-C]   0.00-10.00  sec  94.1 MBytes  78.9 Mbits/sec                  sender
[  5][TX-C]   0.00-10.07  sec  93.0 MBytes  77.5 Mbits/sec                  receiver
[  7][RX-C]   0.00-10.00  sec  49.9 MBytes  41.8 Mbits/sec    0            sender
[  7][RX-C]   0.00-10.07  sec  47.2 MBytes  39.4 Mbits/sec                  receiver

iperf Done.

Can anyone shed some light on this? I’m losing hope.

2

Your results do seem strangely low. Let’s start with the basics… are you using 5.10.1 firmware on all radios? Is there any way you can localize the bandwidth tests by using computers directly connected to the radios?

ALSO, possibly try these settings on the AP side under Configuration → Quality of Service:

image
image1139×725 49.1 KB

And these settings under Configuration → Network:

image
image1132×927 70.4 KB

2 Likes
3

Hello @Neil_Capell ,

your test results are really low. If your setup is on the bench, can it be a self-interference? Just because the radios are too close to reach other.

Please, reset statistics, execute WLT 2-3 times and PM me your techsupport files from AP and SM.
I will try to find what can be wrong.

1 Like
4

Having the exact same issue. tried the setting above but no improvement. Have a ticket in. Two radios are 800m apart and all looks great but very poor throughput. Backed the power down, changed freq, dropped the spread down and clean SNR. Any ideas?

5

Hi @techboy,

could you message me your support ticket number?

6

I have exactly similar issue on 4625 PTP and 4600 PTP, great WLT but bad results on real speedtes.net. I have 5.10.1

7

Hi Eric, does this set up apply only for PTP or should I do the same on sectors as well? Should I apply this set up on CPE working as routers?

8

Both PtP/PtMP and AP and CPE, and both bridge and NAT mode (not all settings will be available on all radio mode types)

9

thank you Eric, I see SM isolation you have it disable and LLDP is enable, what are the benefits of doing this?

10

Normally, on a PtMP configuration for an ISP, you’d have SM isolation set to enabled to keep the clients from reaching each other. If you were configuring something like an enterprise, or educational campus network, where there’s a need for clients to talk to each other, then you’d probably want SM isolation set to disabled.

You can read more about LLDP here, but typically, unless you’re using LLDP, you should have it disabled as it introduces more background traffic and could potentially be a security risk. Again, on an ISP network, there’s probably no need for LLDP to be enabled on any equipment.

11

Eric is right about LLDP.
There is a long story behind it. When LLDP is enabled ePMP will pass MNDP, which is the same.
So Mikrotiks can detect loop where it is actually prevented by logical configuration.
It was done so, because some customers really need MNDP pass through for own reasons.
By default 802.1q compliant bridge should not allow it to pass through.

1 Like
12

Ok Thank you. I have my CPE in NAT mode, static IP. Under network, advanced options there is another LLDP option by default is enable. Should I disable as weel on all my CPEs or this only applies for Sector antennas and PTP?

13

Unless you’re specifically using LLDP, you should disable LLDP on ALL devices… this includes both AP’s and SM’s.

1 Like
14

Hi

By default, on your ePMP network, do you enable Traffic Priority and VoIP Priority?

What mtu is used in your epmp 3000 and sm?

15

Hi Junior… yes the QoS question and 1700 mtu in all our networks, in epmp, mikrotik, etc

1 Like
16

Hello

On our Mikrotik routers we use MTU 1492,

and in EPMP we leave the factory default.

17

Eric, what would LLDP be used for on a typical network using 4600/4625 radios? It’s enabled on our system and if we could speed up our system any at all by disabling it, and I don’t need it, that would be great!

John Rayfield, Jr.

18

LLDP is typically not used on an ISP network. It can be disabled.

19

Hello

By default, do you enable priority across your entire network? On both the AP and CPE sides?