Re: [gmx-users] Signal: Floating point exception (8) Signal code: Floating point divide-by-zero (3)

2019-12-12 Thread Dave M
Hi Paul, Thanks. My replies below: On Thu, Dec 12, 2019 at 2:32 AM Paul bauer wrote: > Hello, > > I'll have a look later today. > Can you also give as information about your build configuration? > If I am not wrong you mean this: -DGMX_BUILD_OWN_FFTW=ON -DREGRESSIONTEST_DOWNLOAD=on

Re: [gmx-users] Signal: Floating point exception (8) Signal code: Floating point divide-by-zero (3)

2019-12-12 Thread Paul bauer
Hello, I'll have a look later today. Can you also give as information about your build configuration? If you could upload a log file from mdrun to Redmine it should be enough. Thanks Paul On 12/12/2019 11:28, Dave M wrote: Hi Mark, Thanks just opened new issue. My all simulations are

Re: [gmx-users] Signal: Floating point exception (8) Signal code: Floating point divide-by-zero (3)

2019-12-12 Thread Dave M
Hi Mark, Thanks just opened new issue. My all simulations are performed using gromacs2019.4 and to be consistent to report the results I would like to keep same version. But this specific case is creating trouble. What is best I can do? if I use a different gromacs version for this specific case

Re: [gmx-users] Signal: Floating point exception (8) Signal code: Floating point divide-by-zero (3)

2019-12-12 Thread Mark Abraham
Hi, That sounds very much like a bug, but it's hard to say where it comes from. Can you please open an issue at https://redmine.gromacs.org/ and attach your .tpr files plus a log file from a failing run and the above stack trace? Mark On Thu, 12 Dec 2019 at 08:37, Dave M wrote: > Hi All, > >

[gmx-users] Signal: Floating point exception (8) Signal code: Floating point divide-by-zero (3)

2019-12-11 Thread Dave M
Hi All, I am getting errors which I don't understand. I ran a simulation (gromacs 2019.4) which completed fine. Then, with exactly the same system and mdp files, I ran three repeats with different velocity seeds (so initial configuration is same) but each simulation ended incomplete at different