#3738 BLOC First D: Active Antennas cease to route properly after several weeks of operation

[Zarro Boogs per Child]

#3738: Active Antennas cease to route properly after several weeks of operation
-----------------------+----------------------------------------------------
  Reporter:  wad       |       Owner:  marcelo                            
      Type:  defect    |      Status:  new                                
  Priority:  blocker   |   Milestone:  First Deployment, V1.0             
 Component:  wireless  |     Version:                                     
Resolution:            |    Keywords:  mesh, active antenna, school server
  Verified:  0         |  
-----------------------+----------------------------------------------------
Changes (by jg):

 * cc: javier at cozybit.com, luisca at cozybit.com, mbletsas at laptop.org, GR-
 Wireless-OLPC at marvell.com (added)
  * component:  distro => wireless
  * priority:  normal => blocker
  * milestone:  Untriaged => First Deployment, V1.0
  * owner:  jg => marcelo


Old description:

> On schoolserver.laptop.org, the active antenna stopped providing
> connectivity after a period of continuous operation (over a week).  The
> failure was originally attributed to a poor RF environment, but it was
> cleared up by a power cycle.
>
> The symptoms were that laptops could see the mesh beacon, and could
> associate and discover a mesh portal using DHCP correctly.  After this,
> they could not even ping the mesh portal.   Sometimes moving them closer
> to the server corrected the problem.  When rebooted, even laptops close
> to the mesh portal had problems pinging it.
>
> ifconfig showed that the mesh interface was still present, and iwconfig
> showed that it was at the expected frequency and ESSID.
>
> Is anyone doing longetivity testing of the firmware running on the active
> antennas ?

New description:

 On schoolserver.laptop.org, the active antenna stopped providing
 connectivity after a period of continuous operation (over a week).  The
 failure was originally attributed to a poor RF environment, but it was
 cleared up by a power cycle.

 The symptoms were that laptops could see the mesh beacon, and could
 associate and discover a mesh portal using DHCP correctly.  After this,
 they could not even ping the mesh portal.   Sometimes moving them closer
 to the server corrected the problem.  When rebooted, even laptops close to
 the mesh portal had problems pinging it.

 ifconfig showed that the mesh interface was still present, and iwconfig
 showed that it was at the expected frequency and ESSID.

 Is anyone doing longetivity testing of the firmware running on the active
 antennas ?

--

Comment:

 It should be the same firmware as on the laptops; could you please check
 what firmware you are running?

 I believe Marvell has been running longevity tests, something we're not
 well set up to do in 1cc.

-- 
Ticket URL: <https://dev.laptop.org/ticket/3738#comment:1>
One Laptop Per Child <https://dev.laptop.org>
OLPC bug tracking system