Error snappyhexmesh - Multiple outside loops
following is the error I am getting in smoothing process of snappyhexmesh.
I have meshed the same geometry with fine grids and it worked. Now with coarse meshing levels I am getting this error .................................................. .................................................. ...... Multiple outside loops:0() From function combineFaces::getOutsideFace(const indirectPrimitivePatch&) in file polyTopoChange/polyTopoChange/combineFaces.C at line 423. FOAM aborting #0 Foam::error::printStack(Foam::Ostream&) in "/opt/openfoam230/platforms/linux64GccDPOpt/lib/libOpenFOAM.so" #1 Foam::error::abort() in "/opt/openfoam230/platforms/linux64GccDPOpt/lib/libOpenFOAM.so" #2 Foam::combineFaces::getOutsideFace(Foam::Primitive Patch<Foam::face, Foam::IndirectList, Foam::Field<Foam::Vector<double> > const&, Foam::Vector<double> > const&) in "/opt/openfoam230/platforms/linux64GccDPOpt/lib/libdynamicMesh.so" #3 Foam::combineFaces::validFace(double, Foam::PrimitivePatch<Foam::face, Foam::IndirectList, Foam::Field<Foam::Vector<double> > const&, Foam::Vector<double> > const&) in "/opt/openfoam230/platforms/linux64GccDPOpt/lib/libdynamicMesh.so" #4 Foam::combineFaces::getMergeSets(double, double, Foam::HashSet<int, Foam::Hash<int> > const&) const in "/opt/openfoam230/platforms/linux64GccDPOpt/lib/libdynamicMesh.so" #5 Foam::meshRefinement::mergePatchFacesUndo(double, double, Foam::List<int> const&, Foam::dictionary const&, Foam::List<int> const&) in "/opt/openfoam230/platforms/linux64GccDPOpt/lib/libautoMesh.so" #6 Foam::autoSnapDriver::doSnap(Foam::dictionary const&, Foam::dictionary const&, double, double, Foam::snapParameters const&) in "/opt/openfoam230/platforms/linux64GccDPOpt/lib/libautoMesh.so" #7 in "/opt/openfoam230/platforms/linux64GccDPOpt/bin/snappyHexMesh" #8 __libc_start_main in "/lib/x86_64-linux-gnu/libc.so.6" #9 in "/opt/openfoam230/platforms/linux64GccDPOpt/bin/snappyHexMesh" .................................................. .................................................. .................... What is the reason? |
1 Attachment(s)
DEar Foamers
I solved the problem. yeah! I further increased the scale for edge feature refinement from 3 to 4 . and also increased refinement level for surface corners. It worked superb. Hence the problem but be related to edges of the geometry. Still I would like to know exactly when that problem arises. If anyone could help. I am attaching snapshots of mesh. |
I have the same problem happening and I would say this is related to the fact that I just installed ParaView 4.0. I was running the same exact case on 3.14 and I had no problem.
It seems that is like you said, I increased my refinement level from 4 to 5 and the problem was gone. Yet, I don't quite understand the reason behind. Cheers |
Greetings to all!
@avinashjagdale: If you can provide a test case that reproduces this problem, I can have a look into it. Without a test case, I can only guess that it's a bug and possibly one that has already been fixed in OpenFOAM 2.3.x. @Lucas: As I stated in the other thread (http://www.cfd-online.com/Forums/ope...tml#post488338 post #8), without knowing which steps you've taken to change between versions of ParaView, I'm not able to deduce why this happened in your case, because OpenFOAM should not be affected by ParaView. Well, there is a possibility: you might have installed both OpenFOAM 2.2 and 2.3 from Deb packages and now have a mixed shell environment; this can lead to libraries being used between OpenFOAM versions and I find it strange that it hasn't crashed something more along the way... Best regards, Bruno |
Hey Bruno,
Thanks for the response. I have installed two different versions of OpenFOAM because my ParaView 3.12 has crashed as I explained in the post you have just referenced. It is interesting that by installing ParaView 4.0 it overwrites 3.12. Cheers |
Hello,
I am having the same error as reported in this thread: Merging all faces of a cell --------------------------- - which are on the same patch - which make an angle < 180 degrees (cos:-1) - as long as the resulting face doesn't become concave by more than 90 degrees (0=straight, 180=fully concave) [56] [56] [56] --> FOAM FATAL ERROR: [56] Multiple outside loops:0() [56] [56] From function combineFaces::getOutsideFace(const indirectPrimitivePatch&) [56] in file polyTopoChange/polyTopoChange/combineFaces.C at line 423. [56] This happens at the beginning of the Layer phase (running SHM 2.3.x updated about 2 weeks ago). I have checked the snapped mesh and it looks fine and passes all checkmesh criteria. Furthermore I have meshed very similar geometries without any issues.. Any hints about what could be the problem?(sorry I can post the geoemtry becuase it is confedential). Thanks Matteo |
Greetings Matteo,
Quote:
Code:
checkMesh -allTopology -allGeometry Quote:
But did you try the solution mentioned in the posts above? Namely to increase the refinement level, previous to the layer adding step? In addition, this reminds me of this bug report: http://www.openfoam.org/mantisbt/view.php?id=1376 By the way, proper visual mesh diagnosis is explained here: http://openfoamwiki.net/index.php/FA...is_in_ParaView Best regards, Bruno |
SnappyHexmesh using all my RAM and not finalising
1 Attachment(s)
i have started building my case and trying to run my mesh. however it doesn’t work completely.
i started of modifying the snappyhexmesh contained in the snappy multi region heater case but it ended up taking allot of time and using up my most of my RAM, and when ever it ended, the meshing doesnt finalise and when i view it on paraview, it all seems like lego assembly and the surfaces not properly refined. Also some portion of the geometry,s seems erased. After multiple trials, i decided to use a copy of the snappyhezmeshdict made by douglas, the mesh of the geometries got better with less portions erased but still looks like lego assembly. i have attached my new snappyhexmesh file (snappyhexmeshdict ) - source http://www.calumdouglas.ch/openfoam-...snappyhexmesh/ and the old modified (snappyhexmeshdict-) - source multi region heater case, to help you to have a clearer picture of what am trying to do. Note: the filenames are modified with ( - ) symbol at the end of the file name. ( - ) meaning modified snappyhexmesh from multi region heater case. thanks Also, when ever i run my mesh using this procedure: Procedure for Meshing in Parallel with snappyHexMesh (SHMesh) (replace the number “8” with how ever many cores you have & make sure your decomposeParDict matches) ---------------------------------------------------------------------- 1 Rename 0 folder 0.org This prevents SHMesh interfering with it 2 <blockMesh> Creates background mesh for SHMesh 3 <surfaceFeatureExtract> So the mesher knows where to snap to 4 <decomposePar> Divides mesh into one section per CPU core 5 <mpirun -np 8 snappyHexMesh -overwrite -parallel> Runs mesher in parallel 6 <reconstructParMesh -constant> Puts the mesh back together again 7 delete all processor folders Clear old mesh data 8 delete folder 0 This was a dummy folder for SHMesh 9 rename folder 0.org to 0 Reactivate the folder for the solver to use ----------------------i get the following results: also this is what i get at the end of my meshing: and for the past 2 days, i have not been able to reolve the problem, i ve tried changing the quality parameters, enable and disable layers but nothing seem to work. i would appreciate ur guidance on this issue. thanks Code:
Morph iteration 19 |
How much memory does your machine have? Have you tried making the mesh coarser and seeing if that completes?
|
Hello Bruno, I am having the same issue when i select face ype as boundary or baffle in SHMDict file. i can provide my test case if anyone can help highlight the reason the mesh is failing to snap. i have looked at mesh in paraview, and it seems like thin structure with thickness of 0.02mm have poorly refined cellzones, but the face zone seems well refined. i have my edge and surface refinement level set to 7 and (5-5) respective.
i would really appreciate some guidance. thanks |
hello i have the same issue and i don't know why it keeps failing to snap. please help, have a look at my log bellow, i am happy to provide a test case if someone wants to have a look
Code:
Morph iteration 9 |
Hi Nasir,
I can try and take a look at your case, although it would help a lot if you have a much smaller test case, i.e. something that doesn't need 5 million cells :( I say this because by machine at home only has 6 GB of RAM and won't be able to mesh this. Best regards, Bruno |
I could mesh this if wyldckat can use his superior foam-knowledge to help solve the problem!
|
Hi me3840,
Are you willing to help in literally hunt down the bug? We can try doing an assisted debugging session, where:
I say this because this looks like a pretty crazy bug that I would like to hunt down, because the error message claims: Code:
Multiple outside loops:0() The annoying part is that I've got the very vague feeling I've also tripped over this bug sometime in the past, but I can't remember when or why or how I solved it :( I can only guess that it was in fact the problem that avinashjagdale mentioned in the first post. Best regards, Bruno |
Wyldckat,
Sounds like fun, let's give it a go. Are all the files in post 8, and are you just using the source from earlier in the thread? |
Hello Bruno,
Thanks for your prompt reply, i have been away and couldn't reply. i could create a simple case but i reckon, it will be more efficient to use team viewer and you can remotely use my pc and have discussions quite easily. let me know if it is ok with you and i we can arrange anytime thats most convenient for you and i will email you my username and password + (Skype) or something else. so far i couldn't resolve the issue but i was able to avoid but not revolve it by removing face type boundary from snappyhexmeshdict. i am carefully using the word avoid because, even though my mesh gets to finalise, for some reason the internal mesh has some parts missing. when i set Code:
allowFreeStandingZoneFaces false; Code:
allowFreeStandingZoneFaces true; i am happy to help in resolving the bug, but i am a newbie and ve no experience writing c++ codes. so i am sorry. kind regards nas |
Greetings to all!
I'll try to answer to your both: Quote:
Then ask Nasir via private message for the case. Please let me know when you have everything ready, but in the meantime I'll prepare a branch of OpenFOAM 2.4.x in my repository for debugging the issue. Quote:
Either way, the most efficient way to diagnose the issue is to have a small test case. In addition, I have several OpenFOAM versions installed in my machine, which easily allows me to test with various versions to assess if the bug didn't exist in older or newer versions or even in variants of OpenFOAM, which is why I also prefer a smaller test case. Quote:
And Nasir, if you prefer to build and test this yourself in your machine, instead of asking me3840 to assist in debugging the problem, you can try as well by using the instructions I'll provide in a few minutes. Best regards, Bruno |
Greetings to all once again!
It took me a bit longer than I had planned, as I had to take care of a few other things. The instructions for getting the code I uploaded a few minutes ago is as follows and should only be used with OpenFOAM 2.4.x: Code:
foam Code:
mpirun -np 8 snappyHexMesh -overwrite -parallel > log.snappyHexMesh 2>&1 Code:
gzip < log.snappyHexMesh > log.snappyHexMesh.gz Best regards, Bruno |
Hello,
I will try to follow, the procedure by Bruno, however, i reckon, the main problem is that SnappyHexMesh is not capable of handling multi volume mesh for surfaces with thickness bellow 0.025mm. as you can see the thickness of s1...s15 is 0.025, but some part of the internal cell zone seems erased..would appreciate your contributions. for some reason, i can't attach files in private message. https://www.dropbox.com/s/86zdqugxi6...htMRF.zip?dl=0 kind regards Nas |
installing openfoam 2.4.x
Hello Bruno, now trying to install the 2.4.x however i keep getting stuck in
step 6 of the installation, when ever i run the command, i get: Code:
parallels@ubuntu:~/OpenFOAM$ source $HOME/OpenFOAM/OpenFOAM-2.4.x/etc/bashrc WM_NCOMPPROCS=16 and afterwards, i get the following Code:
parallels@ubuntu:~$ of24x Also wondering if i should try the version 3, which i just realised, has been released. thanks |
All times are GMT -4. The time now is 08:11. |