[Server-devel] Testing 200 XO's in two weeks time for Nepal's pilot

Richard A. Smith richard at laptop.org
Mon Mar 3 20:42:35 EST 2008

Bryan Berry wrote:
> Here is the Rough Test Plan I have in mind
> 1. Boot into firmware and run test-all

Note that the current shipping firmware can sometimes report false 
positives for sticky keys in the keyboard test.  So if you think you 
have sticky keys upgrade the firmware to Q2D13 and also duplicate the 
test under Linux to verify you really have a sticky key.

> I appreciate any other ideas on testing the XO's, particularly testing
> the batteries and the network.

All the battery failures I've seen can be detected with a full 
charge/discharge cycle.  Unplug the laptop from ext power and let it run 
down until you get a red led or it shuts off then plug up to ext power 
and recharge. Re-charge should take at least 1.5 hours.  Then unplug 
again.  Build 656 should run in idle for 3 to 3.5 hours.  If you get a 
green LED in less than 1.5h or the XO shuts off pre-maturely, or if the 
Red light flashes, then RMA that unit.  If you get an early shutoff you 
might repeat the test.  At least one report of a battery getting better 
after a few cycles.  Not sure if this was just user error or really valid.

If the battery-won't-charge problem is there then the charge time drops 
to 10's of minutes so its a quick re-test.

Right now I'm only interested in units that have _flashing_ red led.  So 
if you get a flashing LED contact me please.  I have plenty of data for 
the other problem.

But if you are curious or think you are getting confusing results you 
can run olpc-logbat and watch the numbers or send them to me and we can 
analyze any issues.

Richard Smith  <richard at laptop.org>
One Laptop Per Child

More information about the Server-devel mailing list