Error: input file needs to contain GRiB1 formattedmessages

[nanditha]

I find the following error message while executing the model.
*ERROR: input file needs to contain GRiB1 formattedmessages

I found it is due to gridcheck.f90, line number 383. I am using ECMWF data downloaded via copernicus using API request. So I have a sinfle grib file that contains u and v component of wind, and vertical velocity at 3 hour duration for a month.Please tell me whether it is due to any compatibility issue of grib file with FLEXPART 9.0.
Kndly guide me in fixing the issue.
Im attaching my AVAILABLE and pathname files.

[pesei]

As you can see in FpRoadmap, grib2 is supported from v9.2beta on.

[nanditha]

Thanks for the response. But if I use flex_extract version 7 with flexpart 9.0.3, will the issue be solved?

[pesei]

Replying to nanditha:

Thanks for the response. But if I use flex_extract version 7 with flexpart 9.0.3, will the issue be solved?

No, please look at the Roadmap (FpRoadmap)!
You could use either 9.2beta or the latest v10.

[mradenz]

How can the source for version 9.2beta be downloaded?
The ​Roadmap only refers to a changeset.

Just using the source of the recent most trunk ​https://www.flexpart.eu/browser/trunk/src?

In term of using v10 i would like to emphasize issue #254.

Thanks for your help.

[pesei]

Replying to mradenz:

How can the source for version 9.2beta be downloaded?
The ​Roadmap only refers to a changeset.

If you follow the link to the changeset, you will see a download option on that page.

[mradenz]

The download at ​changeset31 only refers to netcdf_output_mod.f90
However, that file does not fix the grib issue.

When using 9.2beta from the downloads page ​https://www.flexpart.eu/downloads/47, the original
problem #255 with sorry: t not in [k] still persists. Both for the FNL 1 degree data as well as the 0.25 GFS data.

Sorry, but I am becoming increasingly clueless how to fix this problem.
Anyhow, thanks for your help.