CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > OpenFOAM Running, Solving & CFD

Mesquite - Adaptive mesh refinement / coarsening?

Register Blogs Members List Search Today's Posts Mark Forums Read

Like Tree16Likes

Reply
 
LinkBack Thread Tools Display Modes
Old   January 14, 2014, 12:17
Default
  #81
Member
 
joegi
Join Date: Nov 2009
Location: genoa
Posts: 68
Rep Power: 8
joegi.geo is on a distinguished road
Hi,

Well is a way way way much slower, probably 5 times. In any case, all my previous cases are not working fine. At the moment I am playing with the length scales to see if I find the right values.


Regards,

Joel
joegi.geo is offline   Reply With Quote

Old   January 14, 2014, 13:23
Default
  #82
New Member
 
olivier Skurtys
Join Date: Aug 2012
Location: Santiago, Chile
Posts: 23
Rep Power: 5
olivier_au_chili is on a distinguished road
Dear Sandeep:

I have run different cases on 4 processors. I have the same problem.

First: dynamicTopoFvMesh run well
but after various time steps > 200, I have this error

Code:
 ~~~ Mesh Quality Statistics ~~~ 
 Min: 0.551996
 Max: 0.999803
 Mean: 0.900042
 Cells: 386866
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 

 Topo modifier time: 1.29494 s
 Bisections :: Total: 0, Surface: 0
 Collapses  :: Total: 0, Surface: 0
 Swaps      :: Total: 3, Surface: 0
 Mapping time: 1.1e-05 s
[manjaro:04482] *** Process received signal ***
[manjaro:04482] Signal: Floating point exception (8)
[manjaro:04482] Signal code:  (-6)
[manjaro:04482] Failing at address: 0x3e800001182
[manjaro:04482] [ 0] /usr/lib/libc.so.6(+0x35390) [0x7faf21a21390]
[manjaro:04482] [ 1] /usr/lib/libc.so.6(gsignal+0x39) [0x7faf21a21319]
[manjaro:04482] [ 2] /usr/lib/libc.so.6(+0x35390) [0x7faf21a21390]
[manjaro:04482] [ 3] /home/olivier/OpenFOAM/OpenFOAM-1.6-ext/lib/linux64Gcc46DPOpt/libfiniteVolume.so(_ZNK4Foam19coupledFvPatchFieldINS_6TensorIdEEE5writeERNS_7OstreamE+0x122) [0x7faf2494d962]
[manjaro:04482] [ 4] /home/olivier/OpenFOAM/OpenFOAM-1.6-ext/lib/linux64Gcc46DPOpt/libfiniteVolume.so(_ZN4FoamlsINS_6TensorIdEENS_12fvPatchFieldENS_7volMeshEEERNS_7OstreamES6_RKNS_14GeometricFieldIT_T0_T1_EE+0x15a) [0x7faf249d763a]
[manjaro:04482] [ 5] /home/olivier/OpenFOAM/olivier-1.6-ext/lib/linux64Gcc46DPOpt/libdynamicTopoFvMesh.so(_ZNK4Foam11coupledInfoINS_17dynamicTopoFvMeshEE4sendINS_14GeometricFieldINS_6TensorIdEENS_12fvPatchFieldENS_7volMeshEEEEEvRKNS_4ListINS_4wordEEERKSB_RKNSA_IiEERNS_8OSstreamE+0x88) [0x7faf1411f078]
[manjaro:04482] [ 6] /home/olivier/OpenFOAM/olivier-1.6-ext/lib/linux64Gcc46DPOpt/libdynamicTopoFvMesh.so(_ZN4Foam17dynamicTopoFvMesh18initFieldTransfersERNS_4ListINS_4wordEEERNS1_IS3_EERNS1_INS1_IcEEEES9_+0x8c8) [0x7faf140ba9b8]
[manjaro:04482] [ 7] /home/olivier/OpenFOAM/olivier-1.6-ext/lib/linux64Gcc46DPOpt/libdynamicTopoFvMesh.so(_ZN4Foam17dynamicTopoFvMesh9resetMeshEv+0x468) [0x7faf13ff82a8]
[manjaro:04482] [ 8] /home/olivier/OpenFOAM/olivier-1.6-ext/lib/linux64Gcc46DPOpt/libdynamicTopoFvMesh.so(_ZN4Foam17dynamicTopoFvMesh6updateEv+0x1ae) [0x7faf13ffaf5e]
[manjaro:04482] [ 9] sixDOFFoam() [0x419632]
[manjaro:04482] [10] /usr/lib/libc.so.6(__libc_start_main+0xf5) [0x7faf21a0db05]
[manjaro:04482] [11] sixDOFFoam() [0x41d239]
[manjaro:04482] *** End of error message ***
--------------------------------------------------------------------------
mpirun noticed that process rank 1 with PID 4482 on node manjaro exited on signal 8 (Floating point exception).
I will attempt to run with debug mode, to obtain more details.

Regards

Olivier
olivier_au_chili is offline   Reply With Quote

Old   January 16, 2014, 05:30
Default
  #83
New Member
 
olivier Skurtys
Join Date: Aug 2012
Location: Santiago, Chile
Posts: 23
Rep Power: 5
olivier_au_chili is on a distinguished road
Dear Sandeep

I tested your latest version (01/15) in debug mode: fatal error: cells already calculated. When I run on one processor I have not this error.

Code:
[0]  Mapping Faces: 267
[0]  Mapping Cells: 228
[0]  Load starts: 4(0 66 133 200)
[0]  Load sizes: 4(66 67 67 67)
[0]  Load starts: 4(0 57 114 171)
[0]  Load sizes: 4{57}
[1]  Mapping Faces: 55
[1]  Mapping Cells: 30
[1]  Load starts: 4(0 13 27 41)
[1]  Load sizes: 4(13 14 14 14)
[1]  Load starts: 4(0 7 15 23)
[1]  Load sizes: 4(7 8 8 7)
[manjaro:04287] *** Process received signal ***
[manjaro:04287] Signal: Segmentation fault (11)
[manjaro:04287] Signal code:  (-6)
[manjaro:04287] Failing at address: 0x3e8000010bf
[2]  Mapping Faces: 16
[2]  Mapping Cells: 8
[2]  Load starts: 4(0 4 8 12)
[2]  Load sizes: 4{4}
[2]  Load starts: 4(0 2 4 6)
[2]  Load sizes: 4{2}
[3]  Mapping Faces: 87
[3]  Mapping Cells: 66
[3]  Load starts: 4(0 21 43 65)
[3]  Load sizes: 4(21 22 22 22)
[3]  Load starts: 4(0 16 33 50)
[3]  Load sizes: 4(16 17 17 16)
*** Error in `sixDOFFoam': double free or corruption (fasttop): 0x0000000001daab60 ***
[3] 
[3] 
[3] --> FOAM FATAL ERROR: 
[3] cells already calculated
[3] 
[3]     From function primitiveMesh::calcCells() const
[manjaro:04290] *** Process received signal ***
[3]     in file meshes/primitiveMesh/primitiveMeshCells.C at line 114.
[3] 
FOAM parallel run aborting
[3] 
[manjaro:04290] Signal: Segmentation fault (11)
[manjaro:04290] Signal code:  (-6)
[manjaro:04290] Failing at address: 0x3e8000010c2
[manjaro:04289] *** Process received signal ***
[manjaro:04289] Signal: Segmentation fault (11)
[manjaro:04289] Signal code:  (-6)
[manjaro:04289] Failing at address: 0x3e8000010c1
[manjaro:04289] [ 0] /usr/lib/libc.so.6(+0x35390) [0x7f792029a390]
[manjaro:04289] [ 1] /usr/lib/libc.so.6(gsignal+0x39) [0x7f792029a319]
[manjaro:04289] [ 2] /usr/lib/libc.so.6(+0x35390) [0x7f792029a390]
[manjaro:04289] [ 3] /home/olivier/OpenFOAM/OpenFOAM-1.6-ext/lib/linux64Gcc46DPOpt/libOpenFOAM.so(_ZN4Foam16invertManyToManyINS_4ListIiEES2_EEviRKNS_5UListIT_EERNS1_IT0_EE+0x66) [0x7f79213d6736]
[manjaro:04289] [ 4] /home/olivier/OpenFOAM/OpenFOAM-1.6-ext/lib/linux64Gcc46DPOpt/libOpenFOAM.so(_ZNK4Foam13primitiveMesh10cellPointsEv+0x7f) [0x7f79213da99f]
[manjaro:04289] [ 5] /home/olivier/OpenFOAM/olivier-1.6-ext/lib/linux64Gcc46DPOpt/libdynamicTopoFvMesh.so(_ZNK4Foam16cellSetAlgorithm8containsEi+0x25) [0x7f79129aece5]
[manjaro:04289] [ 6] /home/olivier/OpenFOAM/olivier-1.6-ext/lib/linux64Gcc46DPOpt/libdynamicTopoFvMesh.so(_ZN4Foam17dynamicTopoFvMesh21computeCoupledWeightsEiiRNS_4ListIiEERNS_5FieldIdEERNS4_INS_6VectorIdEEEEb+0x269) [0x7f7912917689]
[manjaro:04289] [ 7] /home/olivier/OpenFOAM/olivier-1.6-ext/lib/linux64Gcc46DPOpt/libdynamicTopoFvMesh.so(_ZN4Foam17dynamicTopoFvMesh14computeMappingEdbbiiii+0x405) [0x7f79128b30d5]
[manjaro:04289] [ 8] /home/olivier/OpenFOAM/olivier-1.6-ext/lib/linux64Gcc46DPOpt/libdynamicTopoFvMesh.so(_ZN4Foam17dynamicTopoFvMesh20computeMappingThreadEPv+0x126) [0x7f79128b65b6]
[manjaro:04289] [ 9] /home/olivier/OpenFOAM/OpenFOAM-1.6-ext/lib/linux64Gcc46DPOpt/libOpenFOAM.so(_ZN4Foam13multiThreader10poolThreadEPv+0x6e) [0x7f79215548fe]
[manjaro:04289] [10] /usr/lib/libpthread.so.0(+0x80a2) [0x7f791fbd10a2]
[manjaro:04289] [11] /usr/lib/libc.so.6(clone+0x6d) [0x7f792034a3dd]
[manjaro:04289] *** End of error message ***
Regards,

Olivier
olivier_au_chili is offline   Reply With Quote

Old   January 19, 2014, 18:31
Default
  #84
Senior Member
 
Sandeep Menon
Join Date: Mar 2009
Location: Amherst, MA
Posts: 387
Rep Power: 16
deepsterblue will become famous soon enough
Olivier,

I've posted a fix for the floating-point exception issue. I'm not sure what's going on with the cellCells error - does it occur only in Debug?

[EDIT:]
Also - are you running this multi-threaded (i.e., threads > 1)?

Thanks
__________________
Sandeep Menon
University of Massachusetts Amherst
https://github.com/smenon

Last edited by deepsterblue; January 19, 2014 at 18:36. Reason: Add info about threads
deepsterblue is offline   Reply With Quote

Old   January 20, 2014, 12:53
Default
  #85
New Member
 
olivier Skurtys
Join Date: Aug 2012
Location: Santiago, Chile
Posts: 23
Rep Power: 5
olivier_au_chili is on a distinguished road
Sandeep,

I tested your latest version. The error is still there.

does it occur only in Debug?
==> no

are you running this multi-threaded (i.e., threads > 1)?
==> yes

Regards
olivier_au_chili is offline   Reply With Quote

Old   January 20, 2014, 12:59
Default
  #86
Senior Member
 
Sandeep Menon
Join Date: Mar 2009
Location: Amherst, MA
Posts: 387
Rep Power: 16
deepsterblue will become famous soon enough
Okay.. I'll look into the multi-threaded issue. Can you try with single-threaded for now, and see if the rest works?

Thanks
__________________
Sandeep Menon
University of Massachusetts Amherst
https://github.com/smenon
deepsterblue is offline   Reply With Quote

Old   January 20, 2014, 14:16
Default
  #87
Senior Member
 
Sandeep Menon
Join Date: Mar 2009
Location: Amherst, MA
Posts: 387
Rep Power: 16
deepsterblue will become famous soon enough
I've fixed the multi-threading issue as well. Hopefully, it should work now.
hua1015 likes this.
__________________
Sandeep Menon
University of Massachusetts Amherst
https://github.com/smenon
deepsterblue is offline   Reply With Quote

Old   January 20, 2014, 22:04
Default
  #88
New Member
 
olivier Skurtys
Join Date: Aug 2012
Location: Santiago, Chile
Posts: 23
Rep Power: 5
olivier_au_chili is on a distinguished road
Sandeep:

Yes it works very fine, congratulation. However, now after various time steps, I have a problem at the boundaries, sorry

Quote:
~~~ Mesh Quality Statistics ~~~
Min: 0.351905
Max: 1
Mean: 0.832525
Cells: 159046
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

[1]
[2]
[2]
[2] --> FOAM FATAL ERROR:
[2] face 30 area does not match neighbour by 0.0101978% -- possible face ordering problem.
patch: procBoundary2to1 my area:4.1039e-09 neighbour area: 4.10432e-09 matching tolerance: 0.0001
Mesh face: 76429 vertices: 3((0.0111005 0.0171076 0.0104443) (0.0110425 0.0170291 0.0105041) (0.0111254 0.0170277 0.0104461))
Rerun with processor debug flag set for more information.[1]
[1] --> FOAM FATAL ERROR:
[1] face 30 area does not match neighbour by 0.0101978% -- possible face ordering problem.
patch: procBoundary1to2 my area:4.10432e-09 neighbour area: 4.1039e-09 matching tolerance: 0.0001
Mesh face: 79637 vertices: 3((0.0111005 0.0171076 0.0104443) (0.0111254 0.0170277 0.0104461) (0.0110425 0.0170291 0.0105041))
Rerun with processor debug flag set for more information.
[1]
[1] From function processorPolyPatch::calcGeometry()
[1] in file meshes/polyMesh/polyPatches/constraint/processor/processorPolyPatch.C at line 216.
[1]
FOAM parallel run exiting
[1]

[2]
[2] From function processorPolyPatch::calcGeometry()
[2] in file meshes/polyMesh/polyPatches/constraint/processor/processorPolyPatch.C at line 216.
[2]
FOAM parallel run exiting
[2]
--------------------------------------------------------------------------
MPI_ABORT was invoked on rank 2 in communicator MPI_COMM_WORLD
with errorcode 1.

NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes.
You may or may not see output from other processes, depending on
exactly when Open MPI kills them.
--------------------------------------------------------------------------
--------------------------------------------------------------------------
mpirun has exited due to process rank 1 with PID 5264 on
node manjaro exiting without calling "finalize". This may
have caused other processes in the application to be
terminated by signals sent by mpirun (as reported here).
--------------------------------------------------------------------------
[manjaro:05262] 1 more process has sent help message help-mpi-api.txt / mpi-abort
[manjaro:05262] Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / error messages
Tomorrow, I will run in debug mode

Regards,

Olivier
olivier_au_chili is offline   Reply With Quote

Old   January 21, 2014, 10:06
Default
  #89
New Member
 
olivier Skurtys
Join Date: Aug 2012
Location: Santiago, Chile
Posts: 23
Rep Power: 5
olivier_au_chili is on a distinguished road
Sandeep:

I carried out a run in debub mode:

Quote:
~~~ Mesh Quality Statistics ~~~
Min: 0.351905
Max: 1
Mean: 0.832525
Cells: 159046
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


From function void fvMesh::movePoints(const mapPolyMesh& meshMap)
in file fvMesh/fvMesh.C at line 594
Grabbing old cell volumes.
tmp<scalarField> polyMesh::movePoints(const pointField&) : Moving points for time 0.00042 index 42
Moving object leastSquaresVectors
Moving object mesquiteMotionSolver
Done moving
[1] [2]
[1]
[1] --> FOAM FATAL ERROR:
[1]
[2]
[2] --> FOAM FATAL ERROR:
[2] face 30 area does not match neighbour by 0.0101978% -- possible face ordering problem.
patch: procBoundary2to1 my area:4.1039e-09 neighbour area: 4.10432e-09 matching tolerance: 0.0001
Mesh face: 76429 vertices: 3((0.0111005 0.0171076 0.0104443) (0.0110425 0.0170291 0.0105041) (0.0111254 0.0170277 0.0104461))
Rerun with processor debug flag set for more information.
[2]
[2] From function processorPolyPatch::calcGeometry()
[2] in file meshes/polyMesh/polyPatches/constraint/processor/processorPolyPatch.Cface 30 area does not match neighbour by 0.0101978% -- possible face ordering problem.
patch: procBoundary1to2 my area:4.10432e-09 neighbour area: 4.1039e-09 matching tolerance: 0.0001
Mesh face: 79637 vertices: 3((0.0111005 0.0171076 0.0104443) (0.0111254 0.0170277 0.0104461) (0.0110425 0.0170291 0.0105041))
Rerun with processor debug flag set for more information.
[1]
[1] From function processorPolyPatch::calcGeometry()
[1] in file meshes/polyMesh/polyPatches/constraint/processor/processorPolyPatch.C at line at line 216.
[2]
FOAM parallel run exiting
[2]
216.
[1]
FOAM parallel run exiting
Regards,

Olivier
olivier_au_chili is offline   Reply With Quote

Old   January 21, 2014, 10:56
Default
  #90
Senior Member
 
Sandeep Menon
Join Date: Mar 2009
Location: Amherst, MA
Posts: 387
Rep Power: 16
deepsterblue will become famous soon enough
Could you give me a test case that I can work with?
__________________
Sandeep Menon
University of Massachusetts Amherst
https://github.com/smenon
deepsterblue is offline   Reply With Quote

Old   April 28, 2014, 19:07
Default
  #91
New Member
 
olivier Skurtys
Join Date: Aug 2012
Location: Santiago, Chile
Posts: 23
Rep Power: 5
olivier_au_chili is on a distinguished road
Hi Sandeep,

A good day to you! I have modified the solver (it can work now in parallel and take in account viscous forces). Now, I am working with foam3.0 and I use parMetis method.

The solver works fine during various time steps, about 30000, up to an error:
<
Code:
ExecutionTime = 179843 s  ClockTime = 183110 s

Courant Number mean: 0.000429068 max: 0.007494 velocity magnitude: 0.346252
Time = 0.28947


 ~~~ Mesh Quality Statistics ~~~ 
 Min: 0.074796
 Max: 0.999982
 Mean: 0.849104
 Cells: 132815
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 

 Topo modifier time: 0.687138 s
 Bisections :: Total: 0, Surface: 0
 Collapses  :: Total: 0, Surface: 0
 Swaps      :: Total: 3, Surface: 2
 Mapping time: 7e-06 s
 Reordering time: 0.03978 s
[1] 
[1] 
[1] --> FOAM FATAL ERROR: 
[1]  Mapping for inserted boundary face is incorrect. Found an empty masterObjects list.
 Face: 46348
 Patch: particle1
[1] 
[1]     From function void topoPatchMapper::calcInsertedFaceAddressing() const
[1]     in file fieldMapping/topoPatchMapper.C at line 137.
[1] 
FOAM parallel run aborting
[1]
code>

When I run another test case, I have the same error but on another patch.

Have you an idea?

Regards,

Olivier
olivier_au_chili is offline   Reply With Quote

Old   April 29, 2014, 16:31
Default
  #92
Senior Member
 
Sandeep Menon
Join Date: Mar 2009
Location: Amherst, MA
Posts: 387
Rep Power: 16
deepsterblue will become famous soon enough
Olivier,

If you are able to reproduce this consistently, then it would be helpful if you could write out the mesh at the time-step just prior to the crash, and then restart the simulation with a higher debug level (valid up to level 5) to produce more output. This should write out a few VTK files (related to the swaps) that you can visualize, and would also help me figure out what might be wrong.

Glad to know that you're finding this useful.
__________________
Sandeep Menon
University of Massachusetts Amherst
https://github.com/smenon
deepsterblue is offline   Reply With Quote

Old   May 4, 2014, 19:32
Default
  #93
New Member
 
Join Date: Mar 2014
Posts: 9
Rep Power: 4
Mibte is on a distinguished road
Hi Sandeep,

I have the same error than the one at post #91 from this page when running the GDI engine case. Bruno (wyldckat) have been helping me to resolve the issues with relation to the case, from the problem with the load balancing process and the loading of the parMetis libraries to the one with debbuging the dynamic mesh. Here are the results:

problems with turbDyMEngineFoam using dynamicTopoFvMesh and load balancing

I have sent to you through a private message the link to the debug package "turbDyMEngineFoam4subdomains.debug_at_358.3_3 58.4 .tar.gz" as well as the original case.

Thank you very much, Alvaro.
Mibte is offline   Reply With Quote

Old   September 19, 2014, 04:42
Default Tutorial for openFoam 2.3.0
  #94
Senior Member
 
Paritosh Vasava
Join Date: Oct 2012
Location: Lappeenranta, Finland
Posts: 591
Rep Power: 13
vasava will become famous soon enough
I am using openFoam 2.3.0 and I managed to compile both the mesquite and the dynamicTopoFvMesh. I could run the text example circCylinder3d by Mooney (7th openFoam workshop) after initial hitch.

However I do not understand the other tutorials.
  1. ballTranslation: The controlDict file does not have solver name. When I try to run the case with the 'Allrun' file that came with the case, it creates an empty log file and stops.
  2. needleValveClosure: The controlDict file has line 'application icoFoamAutoMotion' which is suppose is the solver name. When I I try to run the case with the 'Allrun' file that came with the case, it creates a log file with following line.
    Code:
    /opt/openfoam230/bin/tools/RunFunctions: line 52: icoFoamAutoMotion: command not found
  3. projectile: The controlDict file does not have solver name and there is no 'Allrun' file with the case.

Here is link to Kyle Mooney's files from the workshop.

Of course the tutorials were for old openFoam versions but can someone can please suggest modifications for openFoam2.3.0.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

In addition I have a case where a cube with in another cube is rotating with some speed. I want the mesh to update while the cube is rotating. Of course with current setup it does not. It would be nice if someone can help me fix this case. Here is link to the case.

http://www.mediafire.com/download/mb...bject_mod3.zip
vasava is offline   Reply With Quote

Old   January 27, 2016, 10:40
Question Points between two slip boundaries do not move correctly with mesquiteMotionSolver
  #95
New Member
 
Lu ZHOU
Join Date: Jul 2014
Location: Lyon, France
Posts: 11
Rep Power: 3
lzhou is on a distinguished road
Hello ,

I would like to use Mesquite Motion Solver coupled with dynamicTopoFvMesh to simulate a case with a deforming boundary (which is deformed based on calculated data) .

I created a simple case based on the circCylinder3dHex tutorial but I got a problem. The points on the line between two slip boundaries do not move correctly as showed in pictures. I uploaded my case and please help me find out where did I make a mistake...(I am using Foam-extend 3.1)

By the way, I would like to use Mesquite Motion solver with interDyMFoam. However, the alpha1 value is not correct when the mesh is deformed. Could someone give me some suggestions about how to solve this problem ?

Thanks in advance !

Best regards !

Lu ZHOU
Attached Images
File Type: jpg origin.jpg (142.7 KB, 9 views)
File Type: jpg problem.jpg (152.8 KB, 9 views)
Attached Files
File Type: zip case_mesquite2.zip (15.2 KB, 1 views)

Last edited by lzhou; January 28, 2016 at 10:15.
lzhou is offline   Reply With Quote

Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
No layers in a small gap bobburnquist OpenFOAM Native Meshers: snappyHexMesh and Others 6 August 26, 2015 09:38
snappyHexMesh refinement regions ignored guitarbren OpenFOAM Meshing & Mesh Conversion 2 April 9, 2013 03:59
non-smooth mesh Svensson OpenFOAM Native Meshers: snappyHexMesh and Others 11 January 18, 2012 10:13
fluent add additional zones for the mesh file SSL FLUENT 2 January 26, 2008 12:55
basic of mesh refinement arya CFX 4 June 19, 2007 12:21


All times are GMT -4. The time now is 03:26.