View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0002728 | OpenFOAM | Bug | public | 2017-10-18 16:57 | 2017-10-20 15:21 |
| Reporter | Kerghan | Assigned To | will | ||
| Priority | normal | Severity | minor | Reproducibility | always |
| Status | resolved | Resolution | fixed | ||
| Platform | Unix | OS | GNU/Linux Ubuntu 16.04 | OS Version | (please specify) |
| Summary | 0002728: nearWallFields fails to produce field on some meshes | ||||
| Description | It looks like there is a bug in point generation for probing locations near walls to estimate near-wall fields in nearWallFields function object. More precisely, it's probably somewhere in point tracking. | ||||
| Steps To Reproduce | 1. Prepare $FOAM_TUTORIALS/incompressible/simpleFoam/airFoil2D 2. In $YOUR_CASE/system/controlDict add functions { nearField { type nearWallFields; libs ("libfieldFunctionObjects.so"); writeControl timeStep; writeInterval 1; fields ((U UNear1em6)); patches (walls); distance 1e-6; } } 3. Run simpleFoam | ||||
| Additional Information | In case of OpenFOAM-5.x one should get warning messages: ..........snip.......... Starting time loop nearWallFields nearField: Sampling 1 fields --> FOAM Warning : From function void Foam::particle::locate(const vector&, const vector*, Foam::label, bool, Foam::string) in file particle/particle.C at line 507 "Particle initialised with a location outside of the mesh." --> FOAM Warning : From function void Foam::particle::locate(const vector&, const vector*, Foam::label, bool, Foam::string) in file particle/particle.C at line 507 "Particle initialised with a location outside of the mesh." ..........snip.......... --> FOAM Warning : From function void Foam::particle::locate(const vector&, const vector*, Foam::label, bool, Foam::string) in file particle/particle.C at line 507 "Particle initialised with a location outside of the mesh." --> FOAM Warning : From function bool Foam::functionObjectList::read() in file db/functionObjects/functionObjectList/functionObjectList.C at line 675 Caught FatalError --> FOAM FATAL ERROR: Cell not found for particle position (-17.2841 -0.406463 0.025). From function void Foam::particle::locate(const vector&, const vector*, Foam::label, bool, Foam::string) in file particle/particle.C at line 455. Time = 1 smoothSolver: Solving for Ux, Initial residual = 1, Final residual = 0.0611362, No Iterations 2 smoothSolver: Solving for Uy, Initial residual = 1, Final residual = 0.0585177, No Iterations 2 GAMG: Solving for p, Initial residual = 1, Final residual = 0.0940365, No Iterations 7 time step continuity errors : sum local = 0.000301165, global = 6.11097e-18, cumulative = 6.11097e-18 smoothSolver: Solving for nuTilda, Initial residual = 1, Final residual = 0.0452916, No Iterations 2 ExecutionTime = 0.18 s ClockTime = 0 s ..........snip.......... Not sure if solver should keep running after fatal error, but it does. Eventually case doesn't have requested near-wall field, everything else is ok. In case of OpenFOAM-4.x solver crashes right away: ..........snip.......... Starting time loop nearWallFields nearField: Sampling 1 fields #0 Foam::error::printStack(Foam::Ostream&) at ??:? #1 Foam::sigSegv::sigHandler(int) at ??:? #2 ? in "/lib/x86_64-linux-gnu/libc.so.6" #3 double Foam::particle::trackToFace<Foam::findCellParticle::trackingData>(Foam::Vector<double> const&, Foam::findCellParticle::trackingData&) at ??:? #4 Foam::findCellParticle::move(Foam::findCellParticle::trackingData&, double) at ??:? #5 void Foam::Cloud<Foam::findCellParticle>::move<Foam::findCellParticle::trackingData>(Foam::findCellParticle::trackingData&, double) at ??:? #6 Foam::functionObjects::nearWallFields::calcAddressing() at ??:? #7 Foam::functionObjects::nearWallFields::read(Foam::dictionary const&) at ??:? #8 Foam::functionObjects::nearWallFields::nearWallFields(Foam::word const&, Foam::Time const&, Foam::dictionary const&) at ??:? #9 Foam::functionObject::adddictionaryConstructorToTable<Foam::functionObjects::nearWallFields>::New(Foam::word const&, Foam::Time const&, Foam::dictionary const&) at ??:? #10 Foam::functionObject::New(Foam::word const&, Foam::Time const&, Foam::dictionary const&) at ??:? #11 Foam::functionObjects::timeControl::timeControl(Foam::word const&, Foam::Time const&, Foam::dictionary const&) at ??:? #12 Foam::functionObjectList::read() at ??:? #13 Foam::Time::loop() at ??:? #14 Foam::simpleControl::loop() at ??:? #15 ? at ??:? #16 __libc_start_main in "/lib/x86_64-linux-gnu/libc.so.6" #17 ? at ??:? Segmentation fault (core dumped) | ||||
| Tags | No tags attached. | ||||
|
|
The function object does all sorts of wacky fixes to account for particles that are slightly outside or just along face-diagonals. These are what is failing. None of that is needed any more. The version 5+ tracking just handles it. There are still some NaN-s in the output, which is a little worrying. I haven't investigated why. This crash is fixed, though. Resolved in dev by commit aa24a97c, and in 5.x by 26b8a502. |
|
|
FYI, the NaN issue has now been dealt with, too, in dev, by commit 6d06f737. There were some other issues in the way that the tracking was being done which prevented some tracks from registering. That has been dealt with now. |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2017-10-18 16:57 | Kerghan | New Issue | |
| 2017-10-20 11:52 | will | Assigned To | => will |
| 2017-10-20 11:52 | will | Status | new => resolved |
| 2017-10-20 11:52 | will | Resolution | open => fixed |
| 2017-10-20 11:52 | will | Fixed in Version | => 5.x |
| 2017-10-20 11:52 | will | Note Added: 0008905 | |
| 2017-10-20 15:21 | will | Note Added: 0008911 |
