Program abort called in file ``../../lib/../src/IB/IBHierarchyIntegrator.cpp

[Ramiro James Rebolledo Cormack]

Hola.

I'm running the examples ex* in IBAMR, and in some of them (IB and IBFE folders) appear the below error. Most of the examples run without any problem.

I'm running the examples on my laptop (Ubuntu 18.04), debug mode, IBAMR version 0,8,0.

Thanks in advance for your help.

--

Ramiro

Error:

+++++++++++++++++++++++++++++++++++++++++++++++++++
At beginning of timestep # 4
Simulation time is 0.00625

At end       of timestep # 4
Simulation time is 0.0078125
+++++++++++++++++++++++++++++++++++++++++++++++++++


+++++++++++++++++++++++++++++++++++++++++++++++++++
At beginning of timestep # 5
Simulation time is 0.0078125
P=00000:Program abort called in file ``../../lib/../src/IB/IBHierarchyIntegrator.cpp'' at line 184
P=00000:ERROR MESSAGE:
P=00000:IBHierarchyIntegrator::preprocessIntegrateHierarchy():  Time step size change encountered.
P=00000:Aborting.
P=00000:
Aborted (core dumped)

--

Ramiro

[Wells, David]

Hi Ramiro,

Unfortunately, the current reality of the situation with examples in IBAMR is that we include the examples provided by authors as-is - a lot of them, for a variety of reasons, have input files that no longer run correctly. Put another way - we know this is a problem but don't have the resources to systematically check all examples.

Do you happen to know which examples crashed? It shouldn't be too hard to fix a few individuals.

Best,

David

---

From: ibamr...@googlegroups.com <ibamr...@googlegroups.com> on behalf of Ramiro James Rebolledo Cormack <ramr...@gmail.com>
Sent: Wednesday, February 10, 2021 11:58 AM
To: ibamr...@googlegroups.com <ibamr...@googlegroups.com>
Subject: [ibamr-users] Program abort called in file ``../../lib/../src/IB/IBHierarchyIntegrator.cpp

 

--
You received this message because you are subscribed to the Google Groups "IBAMR Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ibamr-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ibamr-users/CALsG1Z-KC3oxt1rmd6WNuxziRaG1ng%2Bh%3D4ee-3h_S9FjiGEddA%40mail.gmail.com.

[Haotian Hang]

Oh, it's the same bug which I met. I'm using older version.  (Ubuntu 16.04), debug mode, IBAMR version 0,6,0.

https://groups.google.com/g/ibamr-users/c/LRlipbmO1ZQ

Do you know where exactly the bug comes from? 

[Wells, David]

Hi there,

The message

P=00000:IBHierarchyIntegrator::preprocessIntegrateHierarchy():  Time step size change encountered.

Is not necessarily indicative of a bug in IBAMR - this usually means that the time step chosen in the input file was too large to stably capture the explicit structure dynamics, and as a result the time step size computed elsewhere is going to zero.

Best,

David

---

From: ibamr...@googlegroups.com <ibamr...@googlegroups.com> on behalf of Haotian Hang <haot...@usc.edu>
Sent: Wednesday, February 10, 2021 1:29 PM
To: IBAMR Users <ibamr...@googlegroups.com>
Subject: Re: [ibamr-users] Program abort called in file ``../../lib/../src/IB/IBHierarchyIntegrator.cpp

 

To view this discussion on the web visit https://groups.google.com/d/msgid/ibamr-users/62f54215-a72b-42e1-86a3-8def40118ec7n%40googlegroups.com.

[Ramiro James Rebolledo Cormack]

Thanks for the answer.

For me, just to know that is an error in the input in some specific examples is enough. I'm just running the examples to learn/practice how to use the code.

If you think that could be useful, I can try to run all the examples again and let you know which examples crash.

In particular, I am interested in modeling valves in the aorta (Neumann and Dirichlet boundary conditions), so I was looking in the examples which are more close to this problem.

To view this discussion on the web visit https://groups.google.com/d/msgid/ibamr-users/BN7PR03MB43562AEC1DAB53C9FEFF4A37ED8D9%40BN7PR03MB4356.namprd03.prod.outlook.com.

--

Ramiro

[Haotian Hang]

Hello David,

      Thank you so much for your response. However, in the input file, we do not specify the exact timestep. Instead, we have several parameters regarding this. The grow_dt, cfl_max and dt_max are the parameters I think related to this issue. I guess the actual time step is chosen according to equation (3.5) in professor Boyce Griffith 's thesis in 2005. Is that correct? 

       So is it necessary for us to decrease all these three parameters grow_dt , cfl_max and dt_max when we are introducing finer gird? Or maybe changing one or two parameter is ok? Is there an empirical law to do this? Thank you!

Yours sincerely,

Haotian

[Amneet Bhalla]

Hi Hoatian,

We will add a simple example using IBLevelSetMethod doing a rigid body motion. In that example you can prescribe any rotational/translational velocity that you like. 

Here is the open issue on IBAMR Github

https://github.com/IBAMR/IBAMR/issues/1217

Thanks, 

To view this discussion on the web visit https://groups.google.com/d/msgid/ibamr-users/1fc1ac55-eacd-49f5-9496-214c5552dbdan%40googlegroups.com.

--

--Amneet