5.0rc9 "lost connection, medium access timeout"

I’m having a big problem with a 10km radio link, two RB 411AH with r52n and wln200
Every few hours link drops for a while and restart. I don’t know why, i tried to update to 5rc9 but seems worse.
All the signals are good (near -75db), ccq about 90%, modulations stable to 58.5Mbps/58.5Mbps.
Hw-retries set to 15, and disconnect-timeout up to 10s.

In wireless log I can see: “lost connection, medium access timeout”

Here the configurations

name="wlan1" mtu=1500 mac-address=00:80:48:5F:A1:F0 arp=enabled 
      disable-running-check=no interface-type=Atheros 11N 
      radio-name="XXX" mode=bridge ssid="XXX" 
      area="" frequency-mode=regulatory-domain country=italy antenna-gain=0 
      frequency=5500 band=5ghz-a/n channel-width=20mhz scan-list=default 
      wireless-protocol=nv2 rate-set=configured 
      supported-rates-b=1Mbps,2Mbps,5.5Mbps,11Mbps 
      supported-rates-a/g=6Mbps,9Mbps,12Mbps,18Mbps,24Mbps,36Mbps 
      basic-rates-b=1Mbps basic-rates-a/g=6Mbps max-station-count=2007 
      distance=dynamic tx-power-mode=default noise-floor-threshold=default 
      nv2-noise-floor-offset=default periodic-calibration=default 
      periodic-calibration-interval=60 dfs-mode=none wds-mode=disabled 
      wds-default-bridge=none wds-default-cost=100 wds-cost-range=50-150 
      wds-ignore-ssid=no update-stats-interval=disabled bridge-mode=enabled 
      default-authentication=yes default-forwarding=yes default-ap-tx-limit=0 
      default-client-tx-limit=0 proprietary-extensions=post-2.9.25 
      wmm-support=enabled hide-ssid=no security-profile=default 
      disconnect-timeout=10s on-fail-retry-time=100ms preamble-mode=both 
      compression=no allow-sharedkey=no 
      station-bridge-clone-mac=00:00:00:00:00:00 ht-ampdu-priorities=0 
      ht-guard-interval=any ht-extension-channel=disabled 
      ht-supported-mcs=mcs-0,mcs-1,mcs-2,mcs-3,mcs-4,mcs-5,mcs-6,mcs-7,mcs-8,
                 mcs-9,mcs-10,mcs-11 
      ht-basic-mcs=mcs-0,mcs-1,mcs-2,mcs-3,mcs-4,mcs-5,mcs-6,mcs-7 
      ht-txchains=0,1 ht-rxchains=0,1 ht-amsdu-limit=8192 
      ht-amsdu-threshold=8192 tdma-debug=0 tdma-override-rate=disabled 
      tdma-override-size=0 tdma-period-size=2 nv2-queue-count=2 
      nv2-qos=default nv2-cell-radius=30 nv2-security=disabled 
      nv2-preshared-key="" hw-retries=15 frame-lifetime=0 
      adaptive-noise-immunity=ap-and-client-mode 
      hw-fragmentation-threshold=disabled hw-protection-mode=none 
      hw-protection-threshold=0 frequency-offset=0 



name="wlan1" mtu=1500 mac-address=00:0C:42:3A:CC:94 arp=enabled 
      disable-running-check=no interface-type=Atheros 11N 
      radio-name="000C423ACC94" mode=station ssid="XXX" area="" 
      frequency-mode=regulatory-domain country=italy antenna-gain=0 
      frequency=5500 band=5ghz-a/n channel-width=20mhz scan-list=default 
      wireless-protocol=nv2 rate-set=configured 
      supported-rates-b=1Mbps,2Mbps,5.5Mbps,11Mbps 
      supported-rates-a/g=6Mbps,9Mbps,12Mbps,18Mbps,24Mbps,36Mbps,48Mbps,54Mbps 
      basic-rates-b=1Mbps basic-rates-a/g=6Mbps max-station-count=2007 
      distance=dynamic tx-power-mode=default noise-floor-threshold=default 
      nv2-noise-floor-offset=default periodic-calibration=default 
      periodic-calibration-interval=60 dfs-mode=none wds-mode=disabled 
      wds-default-bridge=none wds-default-cost=100 wds-cost-range=50-150 
      wds-ignore-ssid=no update-stats-interval=disabled bridge-mode=enabled 
      default-authentication=yes default-forwarding=yes default-ap-tx-limit=0 
      default-client-tx-limit=0 proprietary-extensions=post-2.9.25 
      wmm-support=disabled hide-ssid=no security-profile=default 
      disconnect-timeout=10s on-fail-retry-time=100ms preamble-mode=both 
      compression=no allow-sharedkey=no 
      station-bridge-clone-mac=00:00:00:00:00:00 ht-ampdu-priorities=0 
      ht-guard-interval=any ht-extension-channel=disabled 
      ht-supported-mcs=mcs-0,mcs-1,mcs-2,mcs-3,mcs-4,mcs-5,mcs-6,mcs-7,mcs-8,mcs-
                 9,mcs-10,mcs-11,mcs-12,mcs-13,mcs-14,mcs-15 
      ht-basic-mcs=mcs-0,mcs-1,mcs-2,mcs-3,mcs-4,mcs-5,mcs-6,mcs-7 
      ht-txchains=0,1 ht-rxchains=0,1 ht-amsdu-limit=8192 
      ht-amsdu-threshold=8192 tdma-debug=0 tdma-override-rate=disabled 
      tdma-override-size=0 tdma-period-size=2 nv2-queue-count=2 nv2-qos=default 
      nv2-cell-radius=30 nv2-security=disabled nv2-preshared-key="" 
      hw-retries=15 frame-lifetime=0 adaptive-noise-immunity=client-mode 
      hw-fragmentation-threshold=disabled hw-protection-mode=none 
      hw-protection-threshold=0 frequency-offset=0

Any ideas? Mikrotik staff?

Thanks.

How long is the time between the medium access timeout disconnect and the reconnect? The timeout message is from the station router log? I would set verbose logging in both routers, wait for another timeout, then check the log.

/system logging
add topics=wireless,debug action=memory

If you have questions, post the log entries from both routers. Mine are both NTP clients and the time is to the second in both routers, so it is easy to match up the event. You don’t need to post all the log entries, just the ones that apply to one disconnect-connect event on both ends. Same event tho!

Ok SurferTim, thanks for your help, the link was up by one hour, logging and client ntp are ok now.
I’m seeing “control frame timeout” from ap and “medium-access timeout” from station.

Here the log, first ap (clock was 1 hour behind), then station.
ap.jpg
station.jpg

Try entering the mac address of the bridge radio in the station connect-list and set “default-authentication=no” in the station.

/interface wireless connect-list
add mac-address=00:80:48:5F:A1:F0 connect=yes interface=wlan1

/interface wireless
set wlan1 default-authentication=no

Odd your log shows just about the same times mine did. That is how I figured it out. Do your between-shows TV commercials start at about ten minutes before the hour? My worst disconnect times were my busiest. Weekend nights at 6:50PM, 7:50PM, 8:50PM, and 9:50PM.

After just those entries, my connection has been up for 4w5d14h39m21s. The exact amount of time since the entries.

Just a word of advice. Insure the mac address is correct before setting the default-authentication. If it isn’t, the connection will go down and stay down until you set the correct mac, or “default-authentication=yes” in the station. You did say 10km?

Thank You!!
I have several links with the same configurations RB800, ROS5.4, Nv2, extension Channel, and AP-bridge-Station. Only one of my links (the longest, and over sea) did loose connections some seconds as described with 2-4 hours cycle. I got different log messages: wlan1 lost connection, medium-access timeout / wlan1 lost connection, synchronization timeout.
Made the changes recommended here and after 9 hours still no lost connections!!

what is the distance of that wireless link?

You are welcome!

Mine has been working great. The uptime on my “bad” connection is 17w2d22h59m55s. This is on a router/radio that could not maintain a connection for even a few hours.

The distance is about 9 kilometers mostly over sea water. Rx level varies between about -50 to -70 dBm coursed by difference in sea levels. Since last ther has been no dropouts at all.
Before changing the log told me that the station was trying to connect to other AP’s present in the area, and I’m wondering if this setup with AP/Bridge-Station is the way to go.
I’ve just checked another mikrotik link, and can see the same problem. This one is about 3 kilometers and sorrounded with other AP’s. I will immidiately make the same settings on this one too.
Maybe it is’nt so much a OS-problem, but neighbour AP’s interference.

@eivind: That is how I came up with this fix. If this is the case, then it would be logical that all users having this problem would have at least two APs showing in this list after enabling verbose logging on the station. My AP for this station was WiFiBridgeMain, but you can see it is also receiving one of my other backhaul’s AP on another frequency (WiFiBridge1).

jan/14 02:46:53 wireless,info 00:0C:42:60:F9:CE@wlan1: lost connection, no beacons received
jan/14 02:46:55 wireless,debug wlan1: must select network
jan/14 02:46:55 wireless,debug 00:0C:42:60:F9:CE: on 5825 > AP: yes SSID WiFiBridgeMain > caps 0x421 rates 0xff00 basic 0x100 MT: yes
jan/14 02:46:55 wireless,debug 00:0C:42:23:32:A0: on 5745 > AP: yes SSID WiFiBridge1 > caps 0x421 rates 0xff00 basic 0x100 MT: yes
jan/14 02:46:55 wireless,info 00:0C:42:60:F9:CE@wlan1 established connection on 5825, SSID WiFiBridgeMain

ADD: I would find it most helpful if anyone on the forum has entries like this with only one AP in the list would post the station log entries from that disconnect. And this does not include disconnects due to “extensive data loss”.

ADD2: Maybe that is why Mikrotik support can not create this problem. It took two conditions for me:

  1. high traffic on primary connection
  2. another AP in range

Some Additions:
I ve just checked two other links, and they too has the same dropouts. The OS on one of them is 4.17. The present log on one of them did not indicate that the station was trying another AP. I could not see more than two drops coursed a lot of other log listings so I don’t know what has happened before.
Now I have only two links left and they are in a relatively “quiet” area with only my own sector AP’s and they have no drops as I can see from present logs.

Wireless debug before changing:
debug.gif

@eivind: If you get another station disconnect with only one AP, check the AP log on that. Is there a disconnect/reconnect on the AP? Maybe the station only disconnects from the primary AP if there is another AP within range.

The primary condition that triggers this on my system is high traffic on this connection. Without high traffic, I do not get this type of disconnect. Period. None. I have backhaul stations on slow network branches that also receive another AP, but traffic is very low on these. No disconnects.

How is your traffic situation when these disconnects happen? My major problem with disconnects was almost always at the commercial breaks between TV shows in the evening. Tourists jump on the net and “there you have it”. Considering the times on that log entry (5:30PM?) would that not be a TV commercial break?

I’ve had two new drops in my modified 9 kilometer link. One at a high traffic point as You said (not possible for me to know if it was at a TV commercial break), and one when most people are in dreamland and almost no traffic. The log says “lost connection, synchronization timeout” and the station got connection after 0,5 seconds, and no other AP’s involved.
The other modified links seems OK.
I think I have to let the case rest for a couple of weeks and just observe what’s happening.
I’m very happy with Mikrotiks price/capasity, and my plan is to install double N-link with full redundancy. The only problem I might face is if the two links looses connection at exactly the same second.

My last solution is to buy an very expensive lisenced link based on a different modulation system. It’s a critical link.

My longest oversea link still gets a lot of drops coursed of synchronization timeout. The solution with changing authentication method works exellent on all other links, but this one is for sure suffering of reflections from calm sea. According to the Rx log it seems for me that the signal is being completely phased out in spesific conditions, and the Dude log don’t register it (only a few milliseconds)
Maybe this problem belongs to another thread?
The Rx levels are jumping 20 dB, and I’m not shure exactly why. The Rx rate is stable on 300 Mbps, so it shouldn’t be coursed by running the tx-mode as default. One big problem might be the 29 dB duopol with one vertical polarization which I believe is far more sensitive for sea reflections than the horizontal. I’ve been working with antennas in my whole lifetime and a often successful solution is to “hide” the sea, but this isn’t easy to do in this installation.
Another problem might be the received level sometimes close to -30 dBm. The spesifications of the R52Hn does not tell anything about maximum input levels, but for me -30 dBm is a very high input level compared to my other links.
I’m a little stuck here since the link is live, and I do not want to disturb it too much. Any ideas??
Station Rx Level.gif

I’m on 5.8 now on both AP and CPE (411Ah + r52Hn on both sides), AP is on a ub 90deg sector and CPE is using 30dB ub dish.

I am getting disconnected very frequently, up to twice an hour.

wlan1: lost connection, synchronization timeout
or
wlan1: lost connection, not responding
or
wlan1: lost connection, medium-access timeout

my signal on these links is -48/-48 on horizontal and -49-49 on vertical. The link is about 2km. Using NV2 with a tx/rx ccq of 90+% and SNR of 60+dB

I don’t know how to log signal strength or CCQ to see if they dip near the disconnects. Any advice?

please write to support and include the support output files from AP and clients right after you see the disconnection.

Just putting some info in here I’m case someone else has this issue .

My problem ended up being pppoe power. I have a 75ft run of ubnt tough cable and the original mikrotik injector. The length of the run was lon enough to cause enough voltage drop to cause power issues. I replaced the injector with on with more amps and it solves the issue. Also, reducing power output on the radio worked as a temporary solution

CASE CLOSED!
My troubleful link is built by relatively big dishes for narrowing the beam, and by that decrease the sea reflections as much as possible. The result was too high input levels.
The most helpful changes was to decrease transmit power on both ends, an maybe some improvements in OS. I still get some random dropouts, and I believe I just discovered why…
BIRDS!!!
These days they are very horny, and recently there was a big flock making babies on the roof in front of the dish. My link went crazy and I got a lot of dropouts. The relatively random dropouts is almoust certainly coursed by birds passing the dish.

Solution: I’ll find a better location…