gsndelphicoder wrote:

> This was the case. The tablet company will get an earful for this. I
> assume that they configured the tablet this way to increase performance
> and possible save battery life. Thing is though that they did not have
> the power settings configured properly to gracefully hibernate or stand
> by at a pre-set battery level. That's a pretty dangerous config for a
> portable PC dependent on battery power. I'm guessing that even with all
> the inspections e performed that day there was still not enough data to
> warrant the disk cache to commit the data to the HDD?

If I were you I would try to recreate the problem. This is an issue I
have to deal with a lot and I have never seen this. Even with the
Windows disk caching on and yanking the battery from the machine
immediately after commit, I have never seen data lost. That's not to
say it couldn't happen but with the length of time you're talking about
and multiple commits, something seems suspicious here.

It could be specific to the exact tablet or configuration you are using
or it could be a non-obvious problem in your code. I don't rely on just
looking at my code to determine if it's correct. I would get one of
these machines to test and then bash it around. Collect some data then
pull the battery and see if you get any loss and under what
circumstances. Then try the same on another machine (or in a VM). Did
you do this before deploying your application?