[monkey1]

Hi Everyone.
Since we installed Ansys Release 16 on our HPC, I'm facing a peculiar problem. Whilst with the parallely installed V15 it was and still is possible to continue a simulation from a previous run (.res or .bak file), with V16 it doesn't work anymore. In fact the solver starts, does all the Stuff he usually has to do, finds the initial values file, reads it, interpolates it on the "new" case" and all that but then it stalls somewhere before calculating the first time step. In fact it stalls after "Mesh Statistics" and "User defined Monitor Information" at the point "Reference Pressure Information". There neither comes a crash down, nor an error message or anything...it just looks like frozen but the cpu's are still busy / under load. And yes I tried to wait a longer time period (I think >12 Hrs should be long enough), yes it occured with different simulation files, dealing with different geometries and setups. And I'm simulating the release of a gas through a source point in a closed geometry/ a room.

In Fact as I just noticed I didn't formulate any question yet
Is there anyone who had the same phenomenon?
If yes, is there a solution?

Thx, Have anice day and nice WE!

[Lance]

Not quite the same problem, but we've also had problems with R16 while R15 works fine. I dont know all the details, but we cant get the MPI to work on linux HPC. Ansys are looking into it right now.

[monkey1]

Hmm thats interesting.
We are running our HPC also under LInux, and everything works fine (continue from file excluded of course ).
But I know that our Administrators had some problems because we are using script files to setup the runs. And between v15 and v16 there have been some relevant changes in the syntax of the command line options, that lead to the problem, taht our v15 script files did not work on v16 anymore. Unfortunately I don't know in Detail what problem they had.

[Mads]

Sorry to revive this thread, but I've just encountered the exact same problem using Ansys Release 17. Since the problem have now been solved, I feel obligated to let people know.

I was doing a transient simulation of a bioreactor with water in the bottom and air in the top. I ran the simulation on a linux HPC, starting the simulation using a shell script containing the command-line options. When I tried to start a simulation using a .def file and continue from a .res file, the solver halted at the point where it says " Buoyancy Reference Information". I tried waiting for several hours, but nothing seemed to happen. The same thing happened for all .def files I tried to start by continuing from a .res file.

It turned out that switching to Ansys 18.2 solves the problem, so it seems to have been fixed as of v18.2, and possibly earlier.