pings become high too much

RouterOS Wireless Networking
1

i
have bought 3 Routerboard 112 and I have same problem with all and three.
I have one simple configuration, AP and 15 customers.
I have all IP addresses publics I must render my AP transparent and therefore I realize of the bridges between the net ethernet and the wlan.
All my customers make p2p. to a sure point the pings become high too much compromising the quality of the net.


64 bytes from 192.168.10.107: icmp_seq=15 ttl=64 time=96.9 ms
64 bytes from 192.168.10.107: icmp_seq=16 ttl=64 time=358 ms
64 bytes from 192.168.10.107: icmp_seq=17 ttl=64 time=261 ms
64 bytes from 192.168.10.107: icmp_seq=18 ttl=64 time=2.09 ms
64 bytes from 192.168.10.107: icmp_seq=19 ttl=64 time=58.5 ms
64 bytes from 192.168.10.107: icmp_seq=20 ttl=64 time=72.1 ms
64 bytes from 192.168.10.107: icmp_seq=21 ttl=64 time=85.1 ms
64 bytes from 192.168.10.107: icmp_seq=22 ttl=64 time=2.15 ms
64 bytes from 192.168.10.107: icmp_seq=23 ttl=64 time=14.1 ms
64 bytes from 192.168.10.107: icmp_seq=24 ttl=64 time=123 ms
64 bytes from 192.168.10.107: icmp_seq=25 ttl=64 time=426 ms
64 bytes from 192.168.10.107: icmp_seq=26 ttl=64 time=689 ms

some one can help me?

2

This might help

Do a frequency usage scan on the AP interface, then select a frequency that is less used.

I had a similar problem and this sorted the problem :sunglasses:

3

15-20 users on a RB112? Sounds like CPU utilization is maxed out. Check your system resources. Uninstall any packages you don’t use/need.

4

Turn off NStreme and the problem will go away.

5

Absolutely… a 112 is only really good for a CPE, not an AP

6

I work for an ISP, we also got an RB112 for AP with 10-20 clients. Actually we replaced a Pheenet AP [200mW, prism] with 112, and we got the same problem. 5-10 pings are normal, then periodically comes a peak…
We found that for problematic clients ACK is near 300-400us [set on dynamic] for distance of 1.5-2 km, for others are ok.
We also tried to turn off perodic calibration, but it doesn’t helped. Monitoring the CPU utilization via SNMP, it doesn’t got over 10-20%…
Can anyone help me to solve this??? PLS :astonished: :astonished:

PS: Sorry for my english…

7

the clients with high ack-timeout could increase the latency.
Try to set the ack-timeout to ‘indoors’ and check if you have some improvement.