[pvpnrao]

I am trying to run a mesh independence study on channelflow395 tutorial.
The tutorial in its pristine state has a 60000 cells, I am mapping the initial field from 60,000 to 1080,000 cells. But when I run it in parallel I am getting the following error log. Did anyone else encoutered it?

Please help! The case is running file in serial mode

[1] [0] #0 Foam::error:rintStack(Foam::Ostream&)#0 Foam::error:rintStack(Foam::Ostream&)[2] #0 Foam::error:rintStack(Foam::Ostream&)[3] #0 Foam::error:rintStack(Foam::Ostream&)[6] #0 Foam::error:rintStack(Foam::Ostream&)[7] #0 Foam::error:rintStack(Foam::Ostream&) in in in in in in """""/"//w/w/wwoooorrrwkkkr///o00kr0/33k30/63686080830/06//p8ppv0vvpp/pnnpnrrarvoap/oaOopn//eOrOpnapeFoneOFnOAF/OMAA/OMMpO/p/eOnOepnFpeOeFnnAFFMOO-AAOM2A./2M.MOxp--/22/p..le2.nawFxoO/tprflAkoa/Mr0-3mt2s6f./8o2r0lm.isx//n/puplxl6vap4tinfInrocuacxrD6moP4sOI//plcOiptec/nnlDFuOPiAOxMb6//pO4tplIeicnbcF/OODlAPpMiObep-/ntlF2i/OblA.iOMb.2/p.slexon/i"pFblOOaAptM
feo.nrsFomO"sA/[7] Ml.#
isnou"x64Ic
1cD POpt/li[3] b#/l[2] i#b1O pen1FO AM.so"Foam::sigSegv::sigHandler(int)
Foam::sigSegv::sigHandler(int)[6] #Foam::sigSegv::sigHandler(int)1 Foam::sigSegv::sigHandler(int)680/pvpnrao/OpenFOAM/OpenFOAM-2.2.x/platforms/linux64IccDPOpt/lib/libOpenFOAM.so"
[0] #1 Foam::sigSegv::sigHandler(int)2.x/platforms/linux64IccDPOpt/lib/libOpenFOAM.so"
[1] #1 Foam::sigSegv::sigHandler(int) in in in """///w in wwooorrrkkk//"/0 in 0/303w6638o60 in "88/r/00/p/"kv/p/pwvwpoonvrprpkn/an0ro3ar/6oaO/8oOp0p//eOeppnvenpnFFnFOOrAOAMAMa//MoO/O/ppOOeppeeennnF0FOFnO3AAOMM6-A/82OM0.-p/22e.p.n2vFx.pO/nAxpM/r-lpaa2lo.ta/2ftO.ofpxroe/mnprFslOa/AtlMmf/isoOn/rpulmexisnn6/Fu4lOIxiAcn6Mc4-uDI2xcP.6cO4D2pIP.tcOx/cp/lDpitlP/baO/tllfpiiotbr/b/mlOlsipi/bebl/nOilFpneiOnbAuOFMxpO.6eAs4nMoIF."cOscAoD
M"P.Ospot
"/l[0]
i#b/li[7] b#O2p[1] e#n2FO 2AM .so"

[3] #2
rk/03680/pvpnrao/OpenFFOAM/OOpAeMn-F2O.A2M.-x2/.p2l.axt/fpolramtsf/olrimnsu/xl6i4nIucxc6D4PIOcpctD/PlOipbt//lliibbO/pleinbFOOpAeMn.FsOoA"M.so"

[6] #2[2] #2

[3] [0] [6] [2] [1] [7] at at at at sigaction.c:0sigaction.c:0sigaction.c:0 at sigaction.c:0 at
sigaction.c:0sigaction.c:0

and so on...

[pvpnrao]

I was able to fix the issue for time being. From various treads on this forum I understood that my case is getting non-physical values somewhere, i.e. somewhere after decomposition one of the chunks is having division by zero.
As my case was running fine in serial, I thought the key is in the way the work is being divided to the processors.

The tutorial channel395 in its pristine form decomposes the domain using 'scotch'.

So, all I did is...I switch the decomposition from scotch to simple, and hardwired the decomposition by specifying divisions in x, y and z directions in the domain (this needs to be done in "simplecoeff" section of the system/decomposeparDict file).

The case runs fine now.

Hope this helps someone.