| View Issue Details [ Jump to Notes ] | [ Print ] | ||||||||
| ID | Project | Category | View Status | Date Submitted | Last Update | ||||
| 0014291 | ParaView | (No Category) | public | 2013-09-23 17:40 | 2015-12-01 17:56 | ||||
| Reporter | Alan Scott | ||||||||
| Assigned To | Utkarsh Ayachit | ||||||||
| Priority | urgent | Severity | minor | Reproducibility | have not tried | ||||
| Status | closed | Resolution | fixed | ||||||
| Platform | OS | OS Version | |||||||
| Product Version | git-next | ||||||||
| Target Version | 4.4 | Fixed in Version | 4.4 | ||||||
| Summary | 0014291: Extract CTH part crash | ||||||||
| Description | This is the Extract CTH part crash that is being experienced at ARL. ParaView next, with Utkarsh's Extract CTH Part patches. He sent them to me on 9/23/2013. Linux. Remote server (8 processes). * Open Dave's medium CTH dataset. There is some confusion which one is Dave's medium one - it has 32 files in it. All variables on. Apply. * Extract CTH parts. Advanced button. Unselect Remove Ghost Cells. Only turn Material Volume Fraction 1 on. Apply. * Select the reader in the pipeline browser. * Extract CTH parts. Advanced button. Unselect Remove Ghost Cells. Only turn Material Volume Fraction 2 on. Apply. * Select first Extract CTH parts. * Change solid color to red. Change Opacity in Properties to 0.9. * Play. The server will die around the 11th time step. Here is the only output from the server: Warning: In /projects/viz/paraview/src/ParaView/4.1.0-alan-junk/ParaView4/ParaViewCore/ClientServerCore/Core/vtkPVProgressHandler.cxx, line 260 vtkPVProgressHandler (0x16ec9030): vtkSpyPlotReader reported invalid progress (1.15). Value must be between [0, 1]. Clamping to this range. Warning: In /projects/viz/paraview/src/ParaView/4.1.0-alan-junk/ParaView4/ParaViewCore/ClientServerCore/Core/vtkPVProgressHandler.cxx, line 260 vtkPVProgressHandler (0x1fab6030): vtkSpyPlotReader reported invalid progress (1.4). Value must be between [0, 1]. Clamping to this range. Warning: In /projects/viz/paraview/src/ParaView/4.1.0-alan-junk/ParaView4/ParaViewCore/ClientServerCore/Core/vtkPVProgressHandler.cxx, line 260 vtkPVProgressHandler (0x4bdd030): vtkSpyPlotReader reported invalid progress (1.6). Value must be between [0, 1]. Clamping to this range. Generic Warning: In /projects/viz/paraview/src/ParaView/4.1.0-alan-junk/ParaView4/VTK/Parallel/MPI/vtkMPICommunicator.cxx, line 69 MPI had an error ------------------------------------------------ MPI_ERR_TRUNCATE: message truncated ------------------------------------------------ | ||||||||
| Tags | No tags attached. | ||||||||
| Project | Sandia | ||||||||
| Topic Name | |||||||||
| Type | crash | ||||||||
| Attached Files | |||||||||
| Relationships | |
| Relationships |
| Notes | |
|
(0032161) Utkarsh Ayachit (administrator) 2014-01-09 10:43 |
Developer note: -------------- I think the problem is in D3. When it encounters the polydata with ghost-cells not removed resulting in duplicate cells across processes, it seems to be dying. The problem doesn't occur until opacity is changed to 0.9. |
|
(0032162) Utkarsh Ayachit (administrator) 2014-01-09 10:50 edited on: 2014-01-09 10:53 |
Developer note (updated) ------------------------ On further analysis, that's not the case. It has nothing to do with ghost-cells or duplicate cells. The problem only happens when "capping" is ON. The same problem can be reproduced simply by just using 1 ExtractCTH filter and extracting "Volume Fraction 1" alone. The segfault happens only when the following is true: + Opacity < 1 + Cap Surface == 1 + Remove Ghost Cells == 0 If any of the above is not true, it works as expected. |
|
(0035166) Utkarsh Ayachit (administrator) 2015-09-03 16:19 edited on: 2015-09-03 16:26 |
Merge request: https://gitlab.kitware.com/vtk/vtk/merge_requests/625 [^] https://gitlab.kitware.com/paraview/paraview/merge_requests/352 [^] |
|
(0035168) Utkarsh Ayachit (administrator) 2015-09-03 16:23 |
To test the fix, you'll need to disable "Generate Markers" on the CTH reader. This dataset has just 1 marker and that causes the data redistribution code to throw errors/warnings about degenerate volume bounds. |
|
(0035502) Alan Scott (manager) 2015-12-01 17:56 |
Tested master, Linux, 8 remote servers. As stated in the notes, I turned off markers. Markers on gives issues, that is a different bug. |
| Notes |
| Issue History | |||
| Date Modified | Username | Field | Change |
| 2013-09-23 17:40 | Alan Scott | New Issue | |
| 2013-09-24 13:07 | Utkarsh Ayachit | Assigned To | => Utkarsh Ayachit |
| 2013-09-24 13:21 | Utkarsh Ayachit | Target Version | => 4.1 |
| 2014-01-09 10:43 | Utkarsh Ayachit | Note Added: 0032161 | |
| 2014-01-09 10:50 | Utkarsh Ayachit | Note Added: 0032162 | |
| 2014-01-09 10:53 | Utkarsh Ayachit | Note Edited: 0032162 | |
| 2014-02-07 21:47 | Utkarsh Ayachit | Target Version | 4.1 => 4.2 |
| 2014-07-31 13:08 | Utkarsh Ayachit | Status | backlog => todo |
| 2014-11-14 22:52 | Utkarsh Ayachit | Target Version | 4.2 => 4.4 |
| 2015-07-08 16:30 | Utkarsh Ayachit | Status | todo => backlog |
| 2015-07-08 16:36 | Utkarsh Ayachit | Status | backlog => todo |
| 2015-09-03 16:19 | Utkarsh Ayachit | Note Added: 0035166 | |
| 2015-09-03 16:23 | Utkarsh Ayachit | Note Added: 0035168 | |
| 2015-09-03 16:23 | Utkarsh Ayachit | Status | todo => gatekeeper review |
| 2015-09-03 16:23 | Utkarsh Ayachit | Fixed in Version | => git-master |
| 2015-09-03 16:23 | Utkarsh Ayachit | Resolution | open => fixed |
| 2015-09-03 16:26 | Utkarsh Ayachit | Note Edited: 0035166 | |
| 2015-09-04 11:22 | Utkarsh Ayachit | Status | gatekeeper review => customer review |
| 2015-09-06 12:17 | Utkarsh Ayachit | Fixed in Version | git-master => 4.4 |
| 2015-12-01 17:56 | Alan Scott | Note Added: 0035502 | |
| 2015-12-01 17:56 | Alan Scott | Status | customer review => closed |
| Issue History |
| Copyright © 2000 - 2018 MantisBT Team |