Here is the first update.

On 28 September 2014 19:29:32 GMT+05:30, Nishant Sharma 
<[email protected]> wrote:

>Made some changes to the radios. HT20, increased beacon intervals,
>lowered the signals...

This has helped stabilize radios a lot and minimise the interference.

>
>Lastly measured the mains electrical voltage that was 25volts above the
>ideal voltage and confirmed the earthing leakage.

Electricals were sorted. But 3 nodes were not on UPS so whenever power tripped, 
network outage happened  until it healed itself by finding new paths to the 
gateway nodes.

It took 5-10 mins sometimes under load.

Most interesting thing that I found was, when we installed the devices, 
warehouse racks were mostly empty and today I found the node serving a 
particular area had surrounding racka and bays filled tobthe brim. This was 
blocking its path to other nodes. Readjusting the antenna solved the problem.

Tomorrow I will relocate the nodes ensuring better visibility between them

Cheers,
Nishant
>
>Suspiscion is on the switches now and the POE of the Internet link
>radio which might be leaking current into the gateway box.
>
>Will keep posted after monitoring for 2-3 days.
>
>Regards,
>Nishant
>
>>>David Lang
>>>
>>>>> I cover a lot of this in the talk I gave at LISA in 2012
>>>>>
>>>https://www.usenix.org/conference/lisa12/building-wireless-network-high-density-users.
>>>>
>>>> Thanks for the link. I am going to study and decide on the
>strategy.
>>>>

-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.
_______________________________________________
openwrt-users mailing list
[email protected]
https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-users

Reply via email to