#8666 HIGH Not Tri: Libertas unknown event IDs and resets

[Zarro Boogs per Child]

#8666: Libertas unknown event IDs and resets
------------------------+---------------------------------------------------
   Reporter:  cjb       |       Owner:  jcardona     
       Type:  defect    |      Status:  new          
   Priority:  high      |   Milestone:  Not Triaged  
  Component:  wireless  |     Version:  not specified
 Resolution:            |    Keywords:               
Next_action:  diagnose  |    Verified:  0            
  Blockedby:            |    Blocking:               
------------------------+---------------------------------------------------

Comment(by carrano):

 >
 > Two or three of these guys close to the unit in test will make a hostile
 environment (not a strange one as sometimes we detect at 1cc, though)
 >
 > I'll try to cause the reset with this technique and report.

 I generated a lot of traffic with 3 XOs placed only some centimeters apart
 from the test unit. This is the wispy spectrum read (close to the XO
 position).

 {{{
 Channel Grade (%)       Duty Cycle      Average Floor(dBm)      Average
 Peak (dBm)
 11      9,759895        32,81314        -64,14802       -56,36956
 }}}

 It was not enough to make the XO reset (but enough to cause me a headache
 ;-). He could actually associate to the AP in one of the attempts (the AP
 is in the same channel - 11).

 The failing modes are the same as reported in this ticket's sibling
 (#8667). At least for this test, high contention did not change anything.

-- 
Ticket URL: <http://dev.laptop.org/ticket/8666#comment:13>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system