Been running 5.0RC1 for awhile, upgraded to 5.0rc3 when it came out. Running 5 x 433AH + XR9s with Nstreme. Has been working great, no problems whatsoever. But when I changed to NV2 a LOT of weird stuff started happening to my network, random devices and systems seemed to disconnect from the network entirely, it was kinda like a network loop somewhere was wreaking havoc. I employed RSTP on my network switches and on the Mikrotik WAPs and that seemed to help, but Mikrotik WAPs still seemed to gain and loose connectivity, even though the wireless connections NEVER dropped. When I returned to Nstreme1 everything seemed to return to normal, but that could just be a coincidence, however as I continue hunting through system logs that is appearing less and less likely.

Since then I have upgraded/downgraded to 4.13 and I have enabled the NV2 wireless package, but am still only using Nstreme1, I am a bit hesitant to enable NV2 again until this issue has been nailed down.


Attached I have quick drawing of the WAP topology.


Thanks in advance,

I tried using NV2 again this morning, this time using 4.13 as firmware and got some odd results. Again, I got intermittant connectivity, even to the radio I was directly connected to. Even weirder still, the radio I was directly connected to kept dropping and re-establishing the network connection to my laptop. I do NOT have watchdog enabled and all the settings are the same, only difference is I changed to NV2 instead of Nstreme.

My configuration is pretty simple, I have already posted the map of the basic topology. The "Relay" is set as AP Bridge and I have 4 radios connecting as Station WDS. WDS Mode on all radios are set to Dynamic Mesh and using a Bridge with RSTP enabled. I am employing a Simple Queue on the Radio directly connected to the network switch. The way I'm changing from Nstreme to NV2 is by just changing the Relay Radios "Wireless Protocol" field from "Nstreme" to "NV2". All the station radios have "Any" set in their "Wireless Protocol" field.

Any help would be greatly appreciated.

Thanks,

We've seen a problem on one of our links where in NV2, it would occasionally stall and either stop passing traffic or have very high latency, over the link even though it was still associated. This would occur approximately every 14 minutes (but not exactly the same each time). In 802.11 mode it was fine.

Can you watch your interface traffic rates and/or run a continuous ping across it and see what happens when the problem occurs?

OK, ends up watchdog was rebooting the radio connected to the switch, and that was what was giving the little bit of of intermittant connectivity. With watchdog off, my connection remains up, but I can't pass ANY data to ANY of the radios. Even when i connect directly to them. The only way I can connect is to reboot the radio, it gives me about 30 seconds after it connects to the AP Radio before it stops passing data again.

What ever is happening, it's definitely being caused by something on the radios. I disconnected all the unrelated network items, except for a laptop and I still loose traffic shortly after the radio connects to the AP.

Please Help.

Well, I haven't gotten many replies, so I figured I draw a better picture of the situation. I know the OSI model wasn't strictly adhered to, but this should give the basic idea.

If I use WPA2 with NV2, is that going to cause issues?

>bump<

I'm still having this issue. Any help at all would be greatly appreciated.

please upgrade to RouterOS v5.0rc4 and check if that helps in your situation.
could you tell me if you disable the client wireless interface and enable it back - does it help? Only one station stops passing data or all of them?
Could you try such setup in your lab - can you reproduce it easy?

Hi,
I have same problem, one link pTp, NV2, R52nM, RSTP,RB433AH, Dual polarization antenas, v5.rc4. work a few hours and stop, on Cisco show next message: Nov 23 19:14:06.289: %SW_MATM-4-MACFLAP_NOTIF: Host 000c.421a.501a in vlan 37 is flapping between port Fa0/19 and port Fa0/18


mt1 (eoip)<-------------------->mt2(eoip).......port 18...>(CISCO 3550)<..port 19...........MT Central

With Nstreme and 4.13 version work fine...
Thank you,
去ran

At the moment I'm using the "stable" 4.13 version. I'm a bit reluctant to use the Beta at this point as this is a production network. I will try to schedule some time outside business hours to try this.

I came across exactly the same problem today...

I updated a RB532 with two Atheros 5213 minipci to 4.14, enabled wireless-nv2 and reconfigured the link from nstreme to nv2..

Ran some tests and everything looked very good..
but a couple of hours later i came back and saw that the AP had just rebooted ...
with :

system error critical : router was rebooted without proper shutdown by watchdog timer

and that had happened multiple times for a period of about half an hour !!
看门狗定时器是默认启用我猜?,但t it was never set up on this AP.
moved back to nstreme and its stable since..
the other minipci which runs nstreme was running unchanged.

So its aseriousissue that has to do with NV2 and Mikrotik should look into it!!!

去od news and Bad news.

去od News: The newest release (4.14) did the trick. I'm able to set my radios to NV2 and was able to maintain a connection and pass traffic.

Bad News: NV2 is actually less stable than regular Nstreme and throughput took a significant hit as well.

not for me

i was using 4.14 already