|
|
|
Hmm that's not good. The LSO VM is designed to ensure that state_entry's always jump to the top of the event queue.
Is this only with scripts compiled under Mono, or does it also affect scripts compiled under the LSL2 engine in 1.24.3? If I'm reading what you wrote correctly, it happens even on scripts that were compiled before 1.24 was on the grid – correct?
Next rolling restart will pick this up
I have found two examples of this error. One in LSL that was compiled months ago, and another on LSL that I was working on at the time of the mono rollout. I can not confirm that this error occures under mono, but the old LSL is suffering from it.
Good work. I see this is being rolled out to the grid already.
Now how about a new function called llOrbitAgent ? no? ok then, how about llOrbitSelf() for when life gets you down? Darling The fix was deployed in 1.24.4, and doesn't seem to have worked.
I'm still leaving this "Fix pending" since we have a new fix for 1.24.5 (this is currently deployed to many regions on aditi, so you can go and verify yourself) Indeed, the timer still misbehaves on 96115.
======= We use this to test: default touch_start(integer total_number) { llSetTimerEvent(0.001); }timer() { llOwnerSay("If you see me when the object rezzes, I am not supposed to be here."); }} -rez Raising the time makes the chance of failure higher. ======= Second Life 1.21.0 (94829) Aug 20 2008 23:45:42 (Second Life Public Nightly) Second Life 1.21.0 (94829) Aug 20 2008 23:45:42 (Second Life Public Nightly)
You are at 255740.2, 255325.0, 20.9 in Sandbox Goguen 1.24 BETA located at sim3008.aditi.lindenlab.com (8.2.33.220:13002) Second Life Beta Server 1.24.5.96378 This server-version has solved the problem |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
SVC-1894so that people can see that it is a different problem.SVC-1894is describing normal behaviour for a timer event, this jira is for abnormal behaviour.