Mass-related llApplyRotationalImpulse error after sim update?

[Arduenn Schwartzman]

Hi all. As of today I got several complaints about airplane bombs giving error messages. These bombs have worked for over a year. I traced back the issue to what seems to be a mass-related problem. When I use the following script in a physics-enabled plywood box and reduce the size such that the mass is smaller than 0.8, it will give an error after it's clicked. This is the script used:

default{
    touch_end(integer P){
        llSay(0,"Mass = "+(string)llGetObjectMass(llGetKey()));
        llApplyRotationalImpulse(<0,0,5>,TRUE);
    }
}

 No error is given as long as the plywood box has a mass >0.8. Below this mass, it will state:

[13:59] Object: Mass = 0.799572
[13:59] Object: Invalid force in llApplyRotationalImpulse.

Is this because of a sim update (Pathfinding-related)? Is there some deliberate mass-restriction to applyrotationalimpulse? Or is it a bug? If this is an intentional change, a lot of WarBug bomber plane owners will need to update their planes. :-(

[Nalates Urriah]

You are probably going to get more Linden attention by posting this type of question in the Deploys thread.

To get it fixed, write up a bug report and enter it in the JIRA.

[Darien Caldwell]

This was brought up in the SL Beta group awhile back:

 

[2012/07/12 13:48]  Maestro Linden: Yep, that usually triggers when somebody supplies an impulse vector that would result in an unreasonably high angular velocity for the particular object (depends on mass, moment of inertia, etc).

[2012/07/12 14:01]  Mæstro (maestro.linden): Regarding that message triggering unexpectedly... go ahead and file a Jira if you have a repro script and such

[2012/07/12 14:01]  Ariu Kitteh (ariu.arai): /me hasn't found a reproduction yet.

[2012/07/12 14:01]  Mæstro (maestro.linden): it looks like it just triggers if the vector magnitude is too large

 

I don't see a JIRA was ever filed, so if you have a solid reproduction, please file a JIRA.