AC scripts crash domains


@chris and who ever else might be interested.

As I said in the Friday Alphas meeting today, I have been having crashes when running audio and .hfr recording playback AC scripts on domains.

What I have running are two ambient audio scripts and a bot/recording playback script.

I have tried just the audio scripts and just the .hfr recording playback scripts separately and they still crash. Typically what happens is one AC goes bad and then it is followed by the others.

Avatars can still enter the domain but the sound is not there and the bot/recording is visible but not moving.

Today, after the meeting, I went to the domain and saw that everything had stopped again. But this time there were two recorded “bots” present. Both not moving.

This makes me think that my idea about the problem might be true. When the audio starts to go bad, it sounds like there might be two streams being played back at the same time. And every time the domain crashes I end up with 11 to as many as 19 AC processes running.

I was able to get the log from the domain server here:

But the AC log is 1.3gb and the previous one was over 1gb too. I can send the big file if need be … or crash again and catch it before it gets so big if that would help.

One thing I should probably say is that the audio AC script and the playback script were both modified by me from old examples that were obsoleted by changes in the api. Anyway… this seems similar to the problems I was having with script loading from JS, which I just ended up deleting from the domain.

Any ideas would be appreciated.


I fixed the bad statement a few days ago but these AC scripts still keep crashing. Here is the latest log file from when it crashed this morning.

Since I can not seem to get this to work, I have removed all AC scripts from the domain until things change.


mine run for a while then seem to die of their own accord. I thought maybe they were colliding with each other, but who knows


Thanks, I have logged it to the bug