Hello Folks!
I am sorry for a relatively long add here, and not to structured either.. but I am searching for help.
I have seen some discussions around ARM based wifi/nv2 devices and RoS 6.42.X here in forum.
I would be very happy to get listed all the affected device models so we have something to stand on.
We recently had a disaster after upgrading from RoS 6.41.1, 6.41.2, 6.41.3, 6.41.4 (other lower versions also tested) to:
RoS 6.42.1 6.42.2 6.42.3 6.42.4 6.42.5.
Devices are:
RB411 + R52N, RB433 + AR922X, RB333 + 2 * AR922X,
RB SXT G-5HPacD, RB911G-5HPnD, DynaDish and a few other.
So all those seems to be affected, is this true and are there more devices affected ?
We speak wireless communication, the wired routers and CRS, CCR does not have any problems and does not seem to be affected.
WiFi 802.11 issues:
Before the upgrade, all devices were performing excellent and almost no CPU was consumed, a very basic configuration we have been using for years, per:
/interface bridge
add admin-mac=00:0C:42:48:5E:B3 auto-mac=no fast-forward=no mtu=1500 name=\
bridge1
/interface wireless security-profiles
set [ find default=yes ] supplicant-identity=MikroTik
add authentication-types=wpa-psk,wpa2-psk eap-methods="" \
management-protection=allowed mode=dynamic-keys name=fast2 \
supplicant-identity="" wpa-pre-shared-key=XXXXXXXX wpa2-pre-shared-key=\
XXXXXXXX
/interface wireless
set [ find default-name=wlan1 ] adaptive-noise-immunity=ap-and-client-mode \
ampdu-priorities=0,1,2,3 band=2ghz-b/g/n channel-width=20/40mhz-eC \
country=sweden distance=indoors frequency=2457 hw-protection-mode=\
cts-to-self hw-protection-threshold=2347 max-station-count=32 mode=\
ap-bridge preamble-mode=short rx-chains=0,1 security-profile=fast2 ssid=\
fast2 tx-chains=0,1 wireless-protocol=802.11 wmm-support=enabled \
wps-mode=disabled
/interface wireless nstreme
set wlan1 disable-csma=yes
/interface bridge port
add bridge=bridge1 hw=no interface=wlan1
add bridge=bridge1 hw=no interface=ether1
Directly after upgrade all seems ok, but shortly after some wireless 802.11 traffic is passed through the devices they start misbehave.
First all look fine, good throughput over wifi, but after around 15-30 seconds load (here 30-40Mbit/s transferring some files) the speed drops down to first 4-5Mbit/s then down to 30-40Kbit/s and constantly stalling. Shortly after CPU goes sky high and it ends with the devices becomes inaccessible over wire and console. I managed make one supout file that took 2 hours to complete and sent to MT. I also had some autosupout files. A hard powercycle to a device (one device broke beyond repair to.. black screen of death), makes it all happens over again.
Trying to rolling back to stable RoS version become a nightmare to, only way was using netinstall, downgrade ended up in devices did not boot up after the downgrade, complaining about to many nested files/softlinks and some complaining about ssh stuff.
This also happened in a few cases during the upgrade, and devices had to be net installed to RoS6.42.X, but they to missbehaved.
I tried one NV2 p2mp sector to, whole sector become unstable, speeds dropping from 20-40Mbit/s to 20-30Kbit/s down to 0kbit/s constant interruptions, clients were disconnected and connected all the time at random and the NV2 access point went 100%CPU, and become useless.
For the time being all is working optimally, since we rolled back everything except wired routers to RoS6.41.X (.1, .2, .3, .4). Thinking on rolling back to latest bug fix only, but not yet, one should never give up
Is there any magic WiFi and NV2 settings that is needed to get it working ?
I did never manage to keep the devices up so long time that I could evaluate different wifi / nv2 configurations.