throughput reducing each repeater

RouterOS Wireless Networking
1

Hello everyone, I just set up a wireless network based on Mikrotik with 350km long, but I’m experiencing serious performance problems. For example when I run a test band straight from point A to point B the band is expected, run the test from point B to point C all ok, but when I run the test from A to C band drops significantly around the half of what goes in each segment, following drawings and configuration of RBs.

Now appreciate the help.


…Point A…Point B… Point C
…RB 1100 AH…>>>Network cable>>> RB 433AH + R52Hn >>> Wireless Link 34km >>> RB 433AH + R52Hn
…>>>A to B 90mbps>>>…>>>B to C 107mbps>>>
…>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
…A to C 53mbps

and it gets worse every repeater


Config Point B

jan/01/1970 21:53:55 by RouterOS 6.1

software id = 8CBB-BFJK

/interface bridge
add l2mtu=1526 name=PTP protocol-mode=rstp
/interface wireless
set 0 band=5ghz-onlyn channel-width=20/40mhz-ht-above disabled=no frequency=5200 ht-ampdu-priorities=0,1,2,3,4,5,6,7 ht-rxchains=0,1 ht-txchains=0,1 l2mtu=2290 mode=
station-bridge name=RX-MGFG nv2-cell-radius=34 nv2-preshared-key=xxxxxxxxxx nv2-security=enabled ssid=DBT-MGFG wireless-protocol=nv2
/ip neighbor discovery
set RX-MGFG discover=no
/port
set 0 name=serial0
/interface bridge port
add bridge=PTP interface=ether1
add bridge=PTP interface=RX-MGFG
/interface wireless align
set receive-all=yes ssid-all=yes
/ip address
add address=172.16.5.10/24 interface=ether1 network=172.16.5.0
/ip firewall connection tracking
set enabled=no
/ip route
add check-gateway=ping distance=1 gateway=172.16.5.254
add check-gateway=ping distance=2 gateway=172.16.5.1
/ip service
set telnet disabled=yes
set ftp disabled=yes
set www disabled=yes
set api disabled=yes
/system identity
set name=RB_F-GUAIBA
/system ntp client
set enabled=yes mode=unicast primary-ntp=172.16.1.4

Config Point C

jan/02/1970 00:41:53 by RouterOS 6.1

software id = 6IK4-SFQJ

/interface bridge
add l2mtu=2290 name=BR_PTP
/interface wireless
set 0 band=5ghz-onlyn channel-width=20/40mhz-ht-above disabled=no frequency=5260 ht-ampdu-priorities=0,1,2,3,4,5,6,7 ht-rxchains=0,1 ht-txchains=0,1 l2mtu=2290 mode=
station-bridge name=RX-SEMG nv2-cell-radius=74 nv2-preshared-key=xxxxxxxxxxxxxxx nv2-security=enabled ssid=DBT-SEMG tdma-period-size=3 wireless-protocol=nv2
set 1 band=5ghz-onlyn basic-rates-a/g=54Mbps basic-rates-b=11Mbps channel-width=20/40mhz-ht-above disabled=no frequency=5825 ht-ampdu-priorities=0,1,2,3,4,5,6,7
ht-basic-mcs=mcs-10 ht-rxchains=0,1 ht-supported-mcs=mcs-10,mcs-11,mcs-12,mcs-13,mcs-14,mcs-15,mcs-16,mcs-17,mcs-18,mcs-19,mcs-20,mcs-21,mcs-22,mcs-23 ht-txchains=0,1
l2mtu=2290 mode=ap-bridge name=TX-MGFG nv2-cell-radius=34 nv2-preshared-key=xxxxxxxxxxxxxxx nv2-security=enabled rate-set=configured ssid=DBT-MGFG supported-rates-a/g=
54Mbps supported-rates-b=11Mbps wireless-protocol=nv2
/ip neighbor discovery
set RX-SEMG discover=no
set TX-MGFG discover=no
/interface wireless security-profiles
set [ find default=yes ] supplicant-identity=MikroTik
/ip hotspot user profile
set [ find default=yes ] idle-timeout=none keepalive-timeout=2m mac-cookie-timeout=3d
/port
set 0 name=serial0
/interface bridge port
add bridge=BR_PTP interface=ether1
add bridge=BR_PTP interface=RX-SEMG
add bridge=BR_PTP interface=TX-MGFG
/ip address
add address=172.16.5.9/24 interface=BR_PTP network=172.16.5.0
/ip route
add check-gateway=ping distance=1 gateway=172.16.5.254
add check-gateway=ping distance=2 gateway=172.16.5.1
/ip service
set api disabled=yes
/system identity
set name=RB_M-GAVIAO
C from A.png
C from B.png
B from A.png

2

That is normal because “Router B” only has 1 radio… so that radio is 50% of the time talking to “Router A” and then 50% talking to “Router C” (switching back and forth rapidly)… so it cuts your overall bandwidth in half.

If you want to keep your bandwidth at 100% then you need 2 radios in “Router B”. One radio will talk to “Router A” and one radio will talk to “Router C” and they don’t have to switch back and forth.

3

i was just curious to know what is the height of antenna form ground to archive such long distance

4

That is normal because “Router B” only has 1 radio… so that radio is 50% of the time talking to “Router A” and then 50% talking to “Router C” (switching back and forth rapidly)… so it cuts your overall bandwidth in half.

Not in this case. He’s running wired ethernet from device A to device B, so this doesn’t apply.

@araujoprog, I’m not sure which direction these tests are running. Your notes seem to indicate that you are sitting on router A, testing receive speeds to routers B and C. This being the case, should the target IP for “C from A” be 172.16.5.10? Maybe you can elaborate on exactly where you are testing from and to.

Can you post the config extract from point A also please.

Any reason why you’re bridging this link rather than routing?

Do you have any other relevant config? Firewall rules, queues, etc?

Rich

5

hi,araujoprog
i’m just wondering how about the ping delay between the two aps as the distance is 350km.
below 10ms? stable?

6

From reading point A to B is connected by Ethernet and then from B to C is 34Kms - so 350Km is a typo?
Also if test is done from A to C, should it not be testing from before point A and after point C

7

Hello guys,

350km is the entire network are a total of 7 repeaters to cover this distance.

Point A is my edge router to the internet, I’m connected at the end of the network beyond the point C. The entire network is with this problem, now I have hired 20mbps but I can only get 10mbps, and that is the link with the lowest band in the whole network is one that is going to 94km 40Mbps.

8

My problem is that when the traffic goes through a RB it decreases, regardless of whether it is cable or wireless.

9

Are the repeaters bridged or routed?

10

Bridge

11

What is the signal level + CCQ + SNR for each segment , Config Point B freq=5200 - Config Point C freq=5260 ? - what frequency spacing and polarity have you on each segment?

12

I’m also having the same problem, follow the image of my network.

Router A <-----------150 MB ----------------> Router B
Router B <------------80 MB ----------------> Router C
Router C <------------35 MB ----------------> Router D

Router A <----------- 80 MB ----------------> Router C

Problem:

Router A <----------- ~10 MB -------------> Router D
or
Router D <----------- ~10 MB -------------> Router A

All equipment on bridge without firewall or bandwidth control.

All Routerboards are with the latest version of Firmware and RouterOS 5.25, also tested with v6.1.

I need help from someone !!

Thanks
Cyber).jpg

13

I did some more tests, but without success.

Thinking it could be a hardware problem, swapped routers B, C and D for new routers and the problem continues, only’re hard to change the router A is because it is my edge.

14

There are a thousand of possibilities… Its a quite hard to debug wireless problems this way.

First: you have to make a site survey to see how good is your scenario and make the right adjustments.
Second: you need to make sure that all points are pretty tuned!! Network logic, cables, grounding stuff, etc…
Third: you never do TCP tests from these boards as long they need only to forward traffic. Do a TCP test will show you only if the router can or cannot GENERATE traffic and not if it can FOWARD the traffic.
…
…
Its too many things to check in such huge scenario. And maybe you need to double check everything when “you are sure”.

Anyway, good luck!

15

Hello guys, one more information on testing the UDP traffic is normal, only TCP that the problem occurs.

16

Like guilhermeramires says: You should not do the TCP test on the Routerboard. It is way too CPU-hungry for a normal Routerboard (on a fast x86 it works okay).
Connect a normal PC on both ends and run an iperf test. This is much more representative.