Ping times?

I have posted some ping results below. The times seem a bit goofy. None of the devices are over a 30 - 40% load. Links are barely used if at all. All wireless links are 5 GHz N-Streme.

— 10.1.5.1 ping statistics —
300 packets transmitted, 300 received, 0% packet loss, time 299248ms
rtt min/avg/max/mdev = 1.265/1.564/2.841/0.308 ms, pipe 2

This is going through a pair of switches to a 532. Not a horrible ping time.

— 10.10.1.1 ping statistics —
300 packets transmitted, 300 received, 0% packet loss, time 299241ms
rtt min/avg/max/mdev = 2.874/19.303/117.095/18.132 ms, pipe 2

This is going through the above router to an Athlon based AP with signal strengths between -75 and -80. Noise floor is -100 to -105. Minimum ping is what I expected. The average and maximum are way above what I expected.

300 packets transmitted, 300 packets received, 0% packet loss
round-trip min/avg/max = 4/23.5/111 ms

This is the same link as above, but from the 532 instead of through it.

300 packets transmitted, 300 packets received, 0% packet loss
round-trip min/avg/max = 2/7.4/45 ms

This is going from a 512 to the Athlon AP. Signal is from -57 to -63 with a noise floor of -105 to -110. The min seems alright, but the average and max seem high.

Test conducted while there were two 1500 byte pings going to or through the router.

532 to Athlon AP

Without pings (repost of previous data)

300 packets transmitted, 300 packets received, 0% packet loss
round-trip min/avg/max = 4/23.5/111 ms

With pings

300 packets transmitted, 300 packets received, 0% packet loss
round-trip min/avg/max = 4/19.0/124 ms


512 to Athlon AP (different sector)

Without pings (repost of previous data)

300 packets transmitted, 300 packets received, 0% packet loss
round-trip min/avg/max = 2/7.4/45 ms


With pings

300 packets transmitted, 300 packets received, 0% packet loss
round-trip min/avg/max = 2/8.9/43 ms


So not much of a difference. Any other ideas?

Those pings look typical. You are sharing 5GHz spectrum with heaps of other devices that interfere. When you have RF interference and loss on these Mikrotik/Atheros units, data retransmits at layer 1, and the loss shows up as jitter.

MT doesn’t report the noise floor higher than -100, meaning a signal to noise of 20 db. I’d expect this to be enough above the noise function to work without the jitter. How can I see if layer 1 retransmissions are what’s causing my jitter?

Is your transmit rate bouncing around at all?

On one link i was just watching, on receive it was 54 most of the time. Ocassionally it would go down to 48. I saw it go to 18 once. In regards to Tx, it was 6 most of the time, with ocassional jumps to 48 and 54.

Another link was mostly 24, but was quite regularly 18 with one drop to 12. The Tx was mostly 6 with regular jumps to 18 and 24. I did see a peak of 36.

I only watched either of these for about 3 or 4 minutes each. Is there a way to log these changes?

We are seeing the same problems with NStreme enabled:

23 packets transmitted, 23 packets received, 0% packet loss
round-trip min/avg/max = 8/18.0/40 ms

signal-strength: -64dBm
tx-signal-strength: -65dBm
noise-floor: -102dBm
signal-to-noise: 38dB
tx-ccq: 96%
rx-ccq: 98%
p-throughput: 37198
wds-link: yes
nstreme: yes
polling: yes
framing-mode: exact-size
framing-limit: 3200
framing-current-size: 2560
routeros-version: “2.9.24”

I set the card’s data rate down to 6 mbit only with 47 db SNR.

100 packets transmitted, 100 packets received, 0% packet loss
round-trip min/avg/max = 2/5.8/38 ms

N-Streme framer-policy is set to none. Why am I getting so much jitter?

               status: connected-to-ess
                 band: 5ghz-10mhz
            frequency: 5810MHz
              tx-rate: 6Mbps
              rx-rate: 6Mbps
                 ssid: "ICS2"
                bssid: 00:15:6D:10:0A:FA
           radio-name: "00156D100AFA"
      signal-strength: -57dBm
   tx-signal-strength: -62dBm
          noise-floor: -106dBm
      signal-to-noise: 49dB
               tx-ccq: 34%
               rx-ccq: 34%
         p-throughput: 5857
             wds-link: no
              nstreme: yes
              polling: yes
         framing-mode: none
     routeros-version: "2.9.24"
              last-ip: 10.10.2.1
  802.1x-port-enabled: yes
  authentication-type: none
           encryption: none
          compression: no
    current-tx-powers: 6Mbps:17,9Mbps:17,12Mbps:17,18Mbps:17,24Mbps:17,36Mbps:14,
                       48Mbps:12,54Mbps:11
  notify-external-fdb: no

Do you have periodic calibration set to enabled? Also, what is your tx/rx ccq values for that link?

Above the CCQ was upper 30s. I believe before I started messing with it today, it was a fair amount higher (2x - 3x). Periodic calibration is set to default.

[admin@Mike's House] > /interface wireless print 
Flags: X - disabled, R - running 
 0  R name="wlan1" mtu=1500 mac-address=00:0B:6B:35:AE:C1 arp=enabled 
      disable-running-check=no interface-type=Atheros AR5213 radio-name="000B6B35AEC1" 
      mode=station ssid="ICS2" area="" frequency-mode=manual-txpower 
      country=no_country_set antenna-gain=0 frequency=5810 band=5ghz-10mhz 
      scan-list=default rate-set=configured supported-rates-b="" 
      supported-rates-a/g=6Mbps basic-rates-b="" basic-rates-a/g=6Mbps 
      max-station-count=2007 ack-timeout=52 tx-power-mode=default 
      noise-floor-threshold=default periodic-calibration=default 
      periodic-calibration-interval=60 burst-time=disabled dfs-mode=none 
      antenna-mode=ant-a wds-mode=disabled wds-default-bridge=none wds-default-cost=100 
      wds-cost-range=50-150 wds-ignore-ssid=no update-stats-interval=disabled 
      default-authentication=yes default-forwarding=yes default-ap-tx-limit=0 
      default-client-tx-limit=0 hide-ssid=no security-profile=default 
      disconnect-timeout=3s on-fail-retry-time=100ms preamble-mode=short compression=yes 
      allow-sharedkey=no

can you temporaly break nstream and test to link p2p and report what you get?

Try to enable periodic-calibration as I think the default == disabled.

As it turns out, both of the AP sectors were set to enabled, but I changed everything else (all CPE) to be enabled. We’ll give it some time to see what that does (like till later tonight). I’ll have to wait to test non-nstreme until I can have someone control both sides of the link to make sure someone doesn’t get stranded.