Artificial Life + OS grid update

The opensim grid was updated to 0.6.9 yesterday or today. I've been running a pool of carnivores for 12 hours and I'm seeing no signs of memory leaks! If things continue to look good it might be time for setting things up to add more species.

Working around memory leaks

A cannibal pool at 11 hours.



I'm an impatient sort, so I've been doing experiments on using listeners and llDie() in an attempt to find out what was causing the leak(s) so I can continue. Moving the llDie() call to a state without an active listener has helped. If you're interested in what I did, click on the pictures below.

Continuing AL, or Little Challenges of OS

I noticed some of the seeds were floating... Did some followup and I think I will wait on this. We're due to have a server software upgrade in the next week or two, and some of the fixes in that may help. (click on the pics if you're interested, there's more info on the Flickr pages.)





The other issue has been with the cannibalistic life. It uses chat to negotiate energy transfers, and that's been causing the region to go down with OOM issues. The software upgrade is supposed to fix several OOM errors and it also offers HTTP server functionality, so hopefully one or the other will work for the negotiations.

Watching the Pools.

It's been interesting to watch the colors develop.



We're talking about making an ecology so I'm modding the scripts so they can eat one another. The first cannibalistic run caused the regions exe to run out of memory, so I'm experimenting with closing the listeners manually instead of letting state change handle it to see if that helps.

I think I'll leave this set overnight...

Grow, my children, grow! (insert mad laughter here)



I wonder what they'll look like after a couple days?

Artificial Life: Bringing a Grid to its Knees

I'm sharing a play-area grid of 4 linked up opensims with a couple friends. Yesterday we decided to explore artificial life. We're starting with a "simple" sessile form and plan to build up as we learn.

Got the basics working. I've also managed to bring the grid to a grinding halt. Now I'm looking for the best way to set things up so we can both keep the experiments contained and clean up the accidents. Ah, hindsight.