open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

analyzing telemac result files with telemac2d v7p1r0 in BK i have the problem, that Blue Kenue shows a weird timeline output after the first timestep, but only if I run projects in parallel. I checked it even with v7p0r1. Here there is no problem just with v7p1r0.
Are there any ideas to solve this? I attached the testfile with the weird timeline output.

Thank you
Engelbert

Hi,

I've stepped through the Blue Kenue code while reading your slf file. It seems that the start date is read as: "StartDate -1348282224/32767/-251727307 -249753600:32623:-254986853"

This seems incorrect. Maybe some kind of problem on the file creation side?

Kindly,
Alan

Hi,

In this file IPARAM(10) is set to 1
This indicates that the next record should be the start date
defined by 6 four byte integers (year,month,day,hour,minute,second)

The contents of these integers are as noted by Alan above.
There is definitely something wrong with how this file was written.

Cheers... Martin

Hello,

thank you Alan and Martin for your answers. Since I used the telemac2d example "cavity", and I did not change anything but reducing the timesteps and setting the number of processors to 4 I ask myself if it is just a problem I have or if someone else has the same problem with computing parallel. If its just on my setup I guess there went something wrong with compiling telemac?

Thank you
Engelbert

Hi

This is a known bug of the last version in parallel computation.
This has been solved on the trunk of development and will be available in the next release.
Sorry for that!

Regards

Hi,

I am running v7p1r1 and get the same error. When you say it will be available in the next release do you mean v7p2?

I understand it is not advisable to compile and run the trunk revision of code. Is there any way I can overcome the problem in the meantime?

Many thanks
Jon