Hello All!


I hope, everything goes well everywhere


My someone help pls:

I was so brave, to put our new RB333 to CORE ROUTER for a test period (75m UTP, 24VDC, 1,2A - on top RB measures 23V). What I see, that the wireless link freezes in every 4-5 hours. The link is there, ping no possible. Card dis/ena helps.

60cm Grante parabolic antenna
R52H cards (default TX power)
signal: 58/59
SNR: 58
48Mbit / 48Mbit connected (54Mbit disabled), NStreme enabled
ping is 1 (max. 3ms)

If I disable NStreme, speed drops to max. 2,3Mbit and ping is over 800ms. With NStreme enabled, speed is over 19Mbit.



RB333 on "A" side:
1xCM9 card
1xR52 card


RB333 on "B" side:
1xCM9 card
1xCM10 card
1xSR9 card



If I see the log, on the top it is written: "KERNEL FAILURE - REBOOTING".



Please, any idea?


Many thanks


MZsolt

could you make the support output file after the reboot and send it to support@mikrotik.com
also look if there is an autosupot.rif file generated, if yes, then please send it to us.

sent.....

the "b" side router as well

hmm.....


The link stays still but 0,0 throughput. Ping is not possible.

Now the routers are in "test-only" state, so until the failure is solved, we took out every RB333 from service.

Thanks in advance for ideas


WBR

MZsolt

Hi All!


A possible solution-like setup, we've found out with Uldis:

When You set up watchdog to ping the 2 routers each-other, the router stays stable for at least 12-14 hours (sometimes there is a random reboot).


The only remaining questions is: without NStreme, why the ping in over 1000ms when the max. traffic is under 2Mbit.

Rds


Zsolt

We too may be experiencing this. Pair of RB333, 3x R52H each. Power turned down to 10dBm tx on interfaces in use. NStreme dual between the two. The far end of the link will fail and need a manual power cycle. Bad stuff.

Oh yes, jonbrewer, we are also waiting for the solution.

The only difference is that Our R52H cards came dead after 1-2 days. R52 instead is a good solution. What we do right now is: the watchdog pings the other side on both routers, of no reply, reboot.

We have average 2-4 days uptime maximum.