Increasing maxNoGauges from 200_i4 to 2000_i4 causes segmentation fault (core dumped)

[cuneyd]

Hello,

I am getting segmentation fault (core dumped) error when I try to run mHM with 1856 gauges.

I changed 200 to 20000 in mo_mrm_constants.F90 in mRM folder and recompiled mHM but failed to run even for test domain.

E:\mhm-v5.13.2\src\mRM\mo_mrm_constants.F90

#ifdef CYGWIN
  integer(i4), public, parameter :: maxNoGauges = 50_i4 ! maximal number of gauges allowed
#else
  integer(i4), public, parameter :: maxNoGauges = 2000_i4 ! maximal number of gauges allowed

Is there other places to update to be able to include more gauges or we can use max 200 gauges in a setup?

Cheers,
Cüneyd

Screenshot Ubuntu LTS 22.04:

[MuellerSeb]

Hey Cüneyd,

this will change in mHMv6 (we already did tests with 10.000 stations).

What you can try in the meantime is to reduce maxNoDomains to compensate.
My guess is that all the arrays that have maxNoGauges as a dimension are taking too much memory in your case, since all these arrays are 2D with dimension(maxNoDomains, maxNoGauges).

So maybe reduce the number of maxNoDomains to your actual number and see if that helps in the meantime.

Cheers,
Sebastian

[cuneyd]

Hey Sebastian, Thank you very much for your detailed reply and tips. Have a fruitful 2026. Kind Regards,Cüneyd On Tuesday, March 24, 2026 at 05:22:58 PM GMT+3, Sebastian Müller ***@***.***> wrote: Hey Cüneyd, this will change in mHMv6 (we already did tests with 10.000 stations). What you can try in the meantime is to reduce maxNoDomains to compensate. My guess is that all the arrays that have maxNoGauges as a dimension are taking too much memory in your case, since all these arrays are 2D with dimension(maxNoDomains, maxNoGauges). So maybe reduce the number of maxNoDomains to your actual number and see if that helps in the meantime. Cheers, Sebastian — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: ***@***.***>