Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Segmentation fault when running image driver with LAKES = TRUE (cells with and without lakes) #928

Open
ednisle opened this issue Oct 17, 2022 · 1 comment

Comments

@ednisle
Copy link

ednisle commented Oct 17, 2022

Hello, I am trying to run VIC with the image driver for the submeridional lowlands (45000 km2, Argentina) with cells of ~5 km *5 km. The region has numerous lakes and wetlands, whose representation implies that there are cells with and without lake/wetland.
I got the following error when trying to run the model on a certain year:

Invalid MIT-MAGIC-COOKIE-1 keyVIC Driver : Image
VIC Version : 5.0.1 February 1, 2017
VIC Git Tag :
Compiled : by root on edna-MS-7A15 (LINUX) Oct 11 2022 21:36:34
Compiler : gcc
version : 9.4.0

Variable Infiltration Capacity (VIC) macroscale hydrologic
model version 5.0.1, Copyright (C) 2016 Computational
Hydrology Group, Dept. of Civil and Environmental Engineering,
University of Washington. VIC comes with ABSOLUTELY NO
WARRANTY. This is free software, you may redistribute it
under certain conditions; see LICENSE.txt for details.

Report Bugs and Issues to : https://github.com/UW-Hydro/VIC/issues
VIC Users Email Listserve : [email protected]

[INFO] ../shared_all/src/vic_log.c:139: Logging to stderr
[DEBUG] ../shared_image/src/get_global_domain.c:98: 5881 active grid cells found in run_cell in the parameter file.
[edna-MS-7A15:385101] *** Process received signal ***
[edna-MS-7A15:385101] Signal: Segmentation fault (11)
[edna-MS-7A15:385101] Signal code: Address not mapped (1)
[edna-MS-7A15:385101] Failing at address: 0x7ffcf0fe6000
[edna-MS-7A15:385101] [ 0] /lib/x86_64-linux-gnu/libpthread.so.0(+0x143c0)[0x7f89cd7983c0]
[edna-MS-7A15:385101] [ 1] ./vic_image.exe(+0x1ff9d)[0x5646cfdb1f9d]
[edna-MS-7A15:385101] [ 2] ./vic_image.exe(+0x47dd1)[0x5646cfdd9dd1]
[edna-MS-7A15:385101] [ 3] ./vic_image.exe(+0x1b952)[0x5646cfdad952]
[edna-MS-7A15:385101] [ 4] ./vic_image.exe(+0xb2aef)[0x5646cfe44aef]
[edna-MS-7A15:385101] [ 5] ./vic_image.exe(+0x4d8ab)[0x5646cfddf8ab]
[edna-MS-7A15:385101] [ 6] ./vic_image.exe(+0x4dbca)[0x5646cfddfbca]
[edna-MS-7A15:385101] [ 7] /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf3)[0x7f89cd5b60b3]
[edna-MS-7A15:385101] [ 8] ./vic_image.exe(+0x3e2e)[0x5646cfd95e2e]
[edna-MS-7A15:385101] *** End of error message ***
Violación de segmento

This error occurs only when LAKES = TRUE, otherwise the model runs successfully for all times.
To find out if the problem was in having cells with and without wetland, I tried to run the model with a domain of 4 cells with lake/wetland and the model ran but until the penultimate time:

[ERROR] ../shared_image/src/get_nc_field.c:48: errno: NetCDF: Index exceeds dimension bound: Error getting values for CHANNEL_IN in /home/edna/Escritorio/datos_BBSS/forcing_file_2003.nc

It's very weird because I have a single netcdf forcing and the variables (PREC, AIR_TEMP..., CHANNEL_IN) share the same time dimension (the model runs successfully when LAKES=FALSE).

@KMarkert
Copy link

KMarkert commented Apr 6, 2023

Try using gcc version 8.* That seemed to have fixed the issue for me. I suspect newer versions of gcc have stricter memory safety features that throw seg faults with the lake model.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants