cheyenne error loading module 'tcl2lua'

[thoar]

I have downloaded tag release-cesm2.2.0 and am trying to build a 3 member multi-instance (multi-driver) case and am running into the following:


ERROR: module command /glade/u/apps/ch/opt/lmod/7.5.3/lmod/lmod/libexec/lmod python load ncarenv/1.3 cmake intel/19.0.5 esmf_libs mkl failed with message:
/glade/u/apps/ch/opt/lmod/8.1.7/bin/lua: error loading module 'tcl2lua' from file '/glade/u/apps/ch/opt/lmod/7.5.3/lmod/lmod/libexec/../lib/tcl2lua.so':
libimf.so: cannot open shared object file: No such file or directory
stack traceback:
[C]: ?
[C]: in function 'require'
...u/apps/ch/opt/lmod/7.5.3/lmod/lmod/libexec/utils.lua:828: in function 'build_runTCLprog'
...u/apps/ch/opt/lmod/7.5.3/lmod/lmod/libexec/utils.lua:833: in main chunk
[C]: in function 'require'
...pps/ch/opt/lmod/7.5.3/lmod/lmod/libexec/colorize.lua:43: in main chunk
[C]: in function 'require'
...lade/u/apps/ch/opt/lmod/7.5.3/lmod/lmod/libexec/lmod:110: in main chunk
[C]: ?

ERROR: Case could not be set up.

My loaded modules are:

Currently Loaded Modules:
1) ncarenv/1.3 2) intel/19.0.5 3) ncarcompilers/0.5.0 4) mpt/2.22 5) netcdf/4.7.4 6) mkl/2019.0.5

Has anyone run into this? More specifically ... how do I fix it?

 

[fischer]

Is this a repeatable error? I was able to build a multi-instance and multi-driver test using release-cesm2.2.0 on cheyenne.

Chris

 

[thoar]

VERY repeatable ... what modules do you have loaded ... and what modules is the test using?
Perhaps also ... what shell are you running?

 

[fischer]

The cesm scripts purge the user modules, and loads it's own modules that specified in env_mach_specific.xml.
You can see the commands issued in .env_mach_specific.sh for bash or .env_mach_specific.csh for csh/tcsh.
I'm using tcsh, but the shell shouldn't make a difference.

Can you point me to your case on cheyenne, then I can try to repeat your failure.

Thanks
Chris

 

[thoar]

Grrrrr .... system configuration change, apparently. In short, something between my .login and my .tcshrc (both of which have worked fine since the dawn of cheyenne) now conflicts. I am tracking it down, but I seem to be on my way. Essentially, I had to remove my .tcshrc and just use the .login. As I say, I am exploring why my previous environment that worked for years is no longer working.

Sorry to bother anyone.

 

[mlague]

Not a bother, I'm now having this issue also! I'll try disposing of by .tcshrc (sad) and see if that fixes my problems... :(

 

[Ziqi Yin]

Hi Chris, I am creating a new B1850G case under the 1pctCO2 scenario using CESM2.1.3 and I also ran into a similar 'error loading module 'tcl2lua'':

cesm model version found: cesm2.1.3-rc.01
Batch_system_type is pbs
job is case.run USER_REQUESTED_WALLTIME None USER_REQUESTED_QUEUE None WALLTIME_FORMAT %H:%M:%S
job is case.st_archive USER_REQUESTED_WALLTIME None USER_REQUESTED_QUEUE None WALLTIME_FORMAT %H:%M:%S
Creating Case directory /glade/scratch/ziqiyin/release-cesm2.1.3.cism_B1850G1PCT_f09_g17_2160pes_230302_test0
This component includes user_mods /glade/u/home/ziqiyin/src/POPARCTIC/release-cesm2.1.3.cism/components/cice/cime_config/usermods_dirs/cmip6
Adding user mods directory /glade/u/home/ziqiyin/src/POPARCTIC/release-cesm2.1.3.cism/components/cice/cime_config/usermods_dirs/cmip6
ERROR: No variable ROOTPE_IAC found in case
ERROR: module command /glade/u/apps/ch/opt/lmod/7.5.3/lmod/lmod/libexec/lmod python purge failed with message:
/glade/u/apps/ch/opt/lmod/8.1.7/bin/lua: error loading module 'tcl2lua' from file '/glade/u/apps/ch/opt/lmod/7.5.3/lmod/lmod/libexec/../lib/tcl2lua.so':
libtcl8.6.so: cannot open shared object file: No such file or directory
stack traceback:
[C]: ?
[C]: in function 'require'
...u/apps/ch/opt/lmod/7.5.3/lmod/lmod/libexec/utils.lua:828: in function 'build_runTCLprog'
...u/apps/ch/opt/lmod/7.5.3/lmod/lmod/libexec/utils.lua:833: in main chunk
[C]: in function 'require'
...pps/ch/opt/lmod/7.5.3/lmod/lmod/libexec/colorize.lua:43: in main chunk
[C]: in function 'require'
...lade/u/apps/ch/opt/lmod/7.5.3/lmod/lmod/libexec/lmod:110: in main chunk
[C]: ?
cp: target ‘/glade/scratch/ziqiyin/release-cesm2.1.3.cism_B1850G1PCT_f09_g17_2160pes_230302_test0/run/’ is not a directory

My source code is at /glade/u/home/ziqiyin/src/POPARCTIC/release-cesm2.1.3.cism. I work with Adam Herrington, and he can't reproduce this error. So, I am wondering if this is a problem of my own environment. May I ask how I could fix this? Thanks very much!

Best,
Ziqi

 

[aherring]

wondering if this has anything to do with trying (and failing) to compile cesm on casper? my understanding is that's a no no ...

 

[Ziqi Yin]

Just tested that's true for my case... I got the error loading module 'tcl2lua' since I logged in through Casper and compiled CESM. The error was gone when I logged in through Cheyenne. I forgot about this. Thanks Adam!