Thursday, June 29

testing and testing and ... testing ...

It's 12 AM and it feels like I'm alone on the whole floor (and for those of you who didn't know this, the floor means a few hundred rooms).

What happened?

I stil have the same task from today, yesterday, two three and four weeks ago, but this time, it needs to be ready by tomorrow.

In short (I'm not sure I wrote about this) we have a delivery, with the client expecting to thest their new and shiny data receiving event-processing, world saving, validating, probably-french-fries-baking module, and the module is not validating properly when running under pressure ... err ... something like that.


It's not that I'm getting nowhere ... its that it's taking awfully long to get there ...

In short, the new and shiny data receiving event-processing, world saving, validating, probably-french-fries-baking module is sort of crashing (and by "sort of" I mean only in some conditions difficult to replicate, on a server in a different timezone, and while running a test that is impossible to start from here).


So, I'm going by trial and error:
first, making a change in code on client1 (my local machine);
second, compiling module on server1;
third, moving the module on server2, accross the world;
fourth, connecting on client2 - across the world also, right near server2 (actually, they are probably best buddies or whatever close servers and clients are).
fifth, starting the server daemon with my client module on server2;
sixth, starting the client program on client2, connecting it to server daemon and waiting for them to have a chat;
seventh, waiting for my module to crash; if it does, I have something to worh with; If it doesn't I move the change back and make a smaller one;
eight, go dirrectly to first step; do not pass go; do not collect 200$.

Yup!

Damn slow!

Wish me luck!

No comments: