"java.lang.RuntimeException: Waited 5000ms for:" error

General Discussion Other

rox_88

"java.lang.RuntimeException: Waited 5000ms for:" error

in General Discussion • Other • 9 months ago

I'm developing a code for a multi-agent system based on a mesh and sometimes when I start the sketch I get this error:

Exception in thread "Animation Thread" java.lang.RuntimeException: Waited 5000ms for: <4cf221f6, 2c93c4f1>[count 2, qsz 0, owner <AWT-EventQueue-0>] - <Animation Thread>

at jogamp.common.util.locks.RecursiveLockImpl01Unfairish.lock(RecursiveLockImpl01Unfairish.java:197)

at com.jogamp.newt.opengl.GLWindow.display(GLWindow.java:539)

at processing.opengl.PGL.requestDraw(PGL.java:814)

at processing.opengl.PGraphicsOpenGL.requestDraw(PGraphicsOpenGL.java:1566)

at processing.core.PApplet.run(PApplet.java:2020)

at java.lang.Thread.run(Thread.java:662)

After trowing the exception, Processing continues to calculate, till when the setup ends; then it visualize the first frame and freeze. I really cannot figure out what the problem is, especially because it's occurring just sometimes.

I'm using Processing 2.07 and the libraries Toxiclibs, PeasyCam and ControlP5...

Thank you!

Replies(14)

quarks

Re: "java.lang.RuntimeException: Waited 5000ms for:" error

9 months ago

You can't use PeasyCam with 2.0b7 because this version of Processing does not handle mouse wheel events yet and even when it does PeasyCam will have to be updated to use the them correctly.

I suggest that for the moment revert back to 2.0b6

rox_88

Re: "java.lang.RuntimeException: Waited 5000ms for:" error

9 months ago

I know, but recently I've installed this fix:

https://forum.processing.org/topic/peasycam-processing-2-0b7-fix

but I'm still getting the same error...

Do you think the PeasyCam is the reason of the error? I'm not really sure, because I'm using PeasyCam in other sketches and I do not get any problem.

quarks

Re: "java.lang.RuntimeException: Waited 5000ms for:" error

9 months ago

I just tried the PeasyCam fix and it seems to be working fine.

Looking at the error it appears that it is waiting for 5 seconds for something to happen before complaining, but then you say it doen't always crash. There appears to be a problem with 'locks' and the draw event. Is there a lot of calculation in setup? You might try using noLoop/loop to delay drawing until the calculations are done (just a guess) e.g.

void setup() {
size(w,h,OPENGL);
noLoop();
// do all the setup calculations here
loop();
} // end of setup

If that doesn't work you might post the setup() code.

rox_88

Re: "java.lang.RuntimeException: Waited 5000ms for:" error

9 months ago

Tryed the loop, it didn't work...

Yes, actually there is really a lot of calculation in the setup, mainly calculating potentials for each vertex of a mesh of 16000 vertices. Here is the code:

void setup() {
size(1280, 720, OPENGL);
noLoop();
smooth();
frameRate(24);
cam = new PeasyCam(this, 1000);
cam.setMinimumDistance(0.1);
cam.setMaximumDistance(50000);
gfx = new ToxiclibsSupport(this);
timer = new Timer(25);
houses= new ArrayList();
/////////////////////////////////////////////////////////////IMPORT
String[] tempEx = loadStrings("dataIn/roadsPoints.txt");
existingRoads = new ArrayList();
for (int i = 0; i < tempEx.length; i++) {
String[] coord = split(tempEx[i], ",");
Vec3D p = new Vec3D(float(coord[0]), -float(coord[1]), float(coord[2]));
existingRoads.add(p);
}
println("roadsPointsNum: " + existingRoads.size());
mesh =(TriangleMesh)new STLReader().loadBinary(sketchPath("dataIn/130116_terrain_CROP_subdiv1.stl"), STLReader.TRIANGLEMESH);
mesh.flipYAxis();
numV = mesh.getNumVertices();
println("verticesNum: " + numV);
///////////////////////////////////////////////////////////ROAD POTENTIAL CALCULATION
roadPotential = new float[numV];
for (int i=0; i < numV; i ++) {
Vertex v= mesh.getVertexForID(i);
int count =0;
float verDist=0;
for (int j=0; j< existingRoads.size(); j++) {
Vec3D p = (Vec3D) existingRoads.get(j);
float distance= p.distanceTo(v);
if ( distance < roadPotentialDist) {
verDist+= distance;
count++;
}
}
if (verDist>0) {
verDist= verDist/count;
verDist= map( verDist, 0, roadPotentialDist, 1, roadMinTresh);
}
else {
verDist=roadMinTresh;
}
roadPotential[i]= verDist;
if (i%1000 == 0) {
println(i);
}
}
println("verticesPotential... DONE");
for (int i=0; i < numV; i ++) {
Vertex v= mesh.getVertexForID(i);
for (int j=0; j< existingRoads.size(); j++) {
Vec3D p = (Vec3D) existingRoads.get(j);
float distance= p.distanceTo(v);
if ( distance < roadSize) {
roadPotential[i]=0;
}
}
if (i%1000 == 0) {
println(i);
}
}
println("roadsPotentialReset... DONE");
///////////////////////////////////////////////ANGLES CALCULATION
for ( int i=0; i< numV; i++) {
Vertex v= mesh.getVertexForID(i);
Vec3D nxz = new Vec3D( v.normal.x, 0, v.normal.z);
Vec3D xAxis = new Vec3D (1, 0, 0);
float anglex = nxz.angleBetween(xAxis);
anglex = abs(90- degrees(anglex));
if (anglex > maxAngleX) {
maxAngleX = anglex;
}
Vec3D nyz = new Vec3D( 0, v.normal.y, v.normal.z);
Vec3D yAxis = new Vec3D (0, 1, 0);
float angley = nyz.angleBetween(yAxis);
angley = abs(90- degrees(angley));
if (angley > maxAngleY) {
maxAngleY = angley;
}
if (i%1000 == 0) {
println(i);
}
}
println("maxAngleX: " + maxAngleX);
println("maxAngleY: " + maxAngleY);
timer.start();
loop();
}

quarks

Re: "java.lang.RuntimeException: Waited 5000ms for:" error

9 months ago

There is no obvious reason why you should have the problem, I am not sure what to suggest now. Sorry

dreadeye

Re: "java.lang.RuntimeException: Waited 5000ms for:" error

7 months ago

I'm also getting this error. When it happens (not all the time), it's right at the beginning of setup.

Here's the sketch:

http://code.google.com/p/simple-cinema/

marco54321

Re: "java.lang.RuntimeException: Waited 5000ms for:" error

6 months ago

I was also getting this. I believe the problem was as guessed above -- setup() was (sometimes) taking too long. In my case, I was accessing the Capture object without having fully initialized it. This was not crashing, but delaying setup() just long enough to cause the error. Eliminating that removed the problem.

My other plan (which I didn't have to execute) was to put all my setup stuff in a separate function, and call it one time at the start of draw(), so that the slow stuff didn't happen on setup().

GoToLoop

Re: "java.lang.RuntimeException: Waited 5000ms for:" error

6 months ago

You might try using noLoop()/loop() to delay drawing until the calculations are done...

Callback draw() is not run until setup() is fully completed! So they have no effect within setup() itself!

From what I've read about somewhere, not even delay() works within setup()!

Perhaps using some native Java delay type method may work instead.

And, even turning off draw() w/ noLoop(), it always runs at least once after setup() is done!!!

PhiLho

Re: "java.lang.RuntimeException: Waited 5000ms for:" error

6 months ago

" From what I've read about somewhere, not even delay() works within setup()!"
Don't believe everything you read in this forum (everybody, including me, can make mistakes or give obsolete advices).
Little experiment, just reusing a code I gave to somebody:

int cellNb = 10; // Per row and column
int cellSize = 10;
int margin = 5;
void setup() {
size(400, 400);
delay(3000);
for (int i = 0; i < cellNb; i++) {
for (int j = 0; j < cellNb; j++) {
if ((i + j) % 2 == 0)
fill(0);
else
fill(255);
rect(margin + i * cellSize, margin + j * cellSize, cellSize, cellSize);
}
}
}

Commenting out the delay(), the checkerboard appears nearly instantly. Otherwise, it takes 3 more seconds to appear.
Delay is just a call to Java's Thread.sleep(), so it works anywhere.
If I change the number of 7000, it takes more time to display, but I don't have this error. Idem in OpenGL mode (even slower!). With Processing 2.0b8 on Windows 7.

GoToLoop

Re: "java.lang.RuntimeException: Waited 5000ms for:" error

6 months ago

Don't believe everything you read in this forum...

I got it a long time ago from Processing v1.5.1's reference for delay()! It said:

This function has no affect inside setup().

I confess I've never checked whether it was true or not.

[Edit] -> tested in Processing 1.5.1 and it really ignores it inside setup() for this version!

But now, it seems like it got censored again, and axed out from official reference,

even though it actually exists in Processing 2+!

When those developers don' t want something, if they can't get rid of it,

they at least hide its existence or make it difficult to use.

For example, you can't make Processing v2+ to auto import

Processing's non-approved Java libraries which were used before,

and are needed for official Processing books' examples,

even by manually configuring it in preferences.txt. It simply ignores it from now on!

rox_88

Re: "java.lang.RuntimeException: Waited 5000ms for:" error

6 months ago

Thank you, inserting a 1000ms delay between the roadPotentialCalculation() and angleCalculation() functions (the two that were slowing down the execution) worked for me!!!

PhiLho

Re: "java.lang.RuntimeException: Waited 5000ms for:" error

6 months ago

Ah, interesting... Since I never use delay(), I overlooked or forgot this. So the info is in the "obsolete" category.

(I know you don't see 1.5.1 as obsolete...)

Indeed,in 1.5.1, my sketch starts instantly. I wonder if that's because pre-processor just removes it, or what.

dreadeye

Re: "java.lang.RuntimeException: Waited 5000ms for:" error

6 months ago

I took marco54321's advice and created a separate function that is called only once at the beginning of draw(). This seems to solve the problem, but I'm still not clear what was causing the error in the first place.

GoToLoop

Re: "java.lang.RuntimeException: Waited 5000ms for:" error

6 months ago

Here's a hack to get a delay() effect within setup() for Processing v1.5.1.

It's basically a forced Thread.sleep().

But since it can throw an exception, just put that within a try/ catch block!

  try {
  Thread.sleep(3000);    // delay(3000)
  }
  catch (Throwable t) {
  }

Top Reply