[Met_help] [rt.rap.ucar.edu #99067] History for METplus error from Cheyenne update?

Julie Prestopnik via RT met_help at ucar.edu
Tue Mar 16 16:45:33 MDT 2021


----------------------------------------------------------------
  Initial Request
----------------------------------------------------------------

Hello,
I am now trying to run the feature-relative use case again (without user
diagnostics;
https://dtcenter.org/metplus-practical-session-guide-version-3-0/session-5-trkintfeature-relative/use-case-feature-relative).
I successfully ran this program ~2-3 weeks ago, but now get errors.

I am working under /glade/u/home/nleonardo/METplus_Tutorial/ATTEMPT001/ and
entered:

master_metplus.py -c ${METPLUS_TUTORIAL_DIR}/tutorial.conf -c
${METPLUS_TUTORIAL_DIR}/ATTEMPT001/TCStat_SeriesAnalysis_fcstGFS_anlGFS_TCrelative.conf
-c
dir.OUTPUT_BASE=${METPLUS_TUTORIAL_DIR}/ATTEMPT001/output/feature_relative_by_init;


The error does not make sense in that I indeed have the existing path
"MET_INSTALL_DIR = /glade/p/ral/jntp/MET/MET_releases/9.1" correctly set in
the top tutorial.conf file (and then even set it as an environmental
variable in my .bashrc):

03/11 23:09:13.302 metplus (config_launcher.py:564) ERROR: Path must exist:
/glade/p/ral/jntp/MET/MET_releases/9.1

INPUT_BASE = /glade/u/home/nleonardo/ISAIAS2020_data
03/11 23:09:13Z run-METplus-metplus: ERROR:  Path must exist:
/glade/p/ral/jntp/MET/MET_releases/9.1

INPUT_BASE = /glade/u/home/nleonardo/ISAIAS2020_data
03/11 23:09:13.302 metplus (met_util.py:128) ERROR: MET_INSTALL_DIR must be
set correctly to run METplus
03/11 23:09:13Z run-METplus-metplus: ERROR:  MET_INSTALL_DIR must be set
correctly to run METplus


I get the same thing even if I run with v9.1.1, though I notice "The
following have been reloaded with a version change:
  1) met/9.1.1 => met/9.1" prints whenever I log in.  Attached is the log
from the previous run that worked.

Lindsay is currently helping me troubleshoot this, but we also wanted to
check if the error has to do with the recent update to Cheyenne.  Thank you.
-Nick


----------------------------------------------------------------
  Complete Ticket History
----------------------------------------------------------------

Subject: METplus error from Cheyenne update?
From: Julie Prestopnik
Time: Tue Mar 16 12:06:45 2021

Hi Nick.

I see you are having trouble with METplus on Cheyenne.  Can you please
tell
me what you are loading for running METplus?

I see on our Existing Builds page
<https://dtcenter.org/community-code/metplus/metplus-3-1-existing-
builds>:
module use /glade/p/ral/jntp/MET/METplus/modulefiles
module load metplus/3.1
ncar_pylib

Is that what you are loading?

Can you please run "which master_metplus.py" to verify the location of
master_metplus.py and let me know that result as well?

I will start looking into this as well.

Julie

On Tue, Mar 16, 2021 at 11:13 AM Nicholas Leonardo via RT
<met_help at ucar.edu>
wrote:

>
> Tue Mar 16 11:13:01 2021: Request 99067 was acted upon.
> Transaction: Ticket created by nleonardo87 at gmail.com
>        Queue: met_help
>      Subject: METplus error from Cheyenne update?
>        Owner: Nobody
>   Requestors: nleonardo87 at gmail.com
>       Status: new
>  Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=99067 >
>
>
> Hello,
> I am now trying to run the feature-relative use case again (without
user
> diagnostics;
>
> https://dtcenter.org/metplus-practical-session-guide-version-3-
0/session-5-trkintfeature-relative/use-case-feature-relative
> ).
> I successfully ran this program ~2-3 weeks ago, but now get errors.
>
> I am working under
/glade/u/home/nleonardo/METplus_Tutorial/ATTEMPT001/ and
> entered:
>
> master_metplus.py -c ${METPLUS_TUTORIAL_DIR}/tutorial.conf -c
>
>
${METPLUS_TUTORIAL_DIR}/ATTEMPT001/TCStat_SeriesAnalysis_fcstGFS_anlGFS_TCrelative.conf
> -c
>
>
dir.OUTPUT_BASE=${METPLUS_TUTORIAL_DIR}/ATTEMPT001/output/feature_relative_by_init;
>
>
> The error does not make sense in that I indeed have the existing
path
> "MET_INSTALL_DIR = /glade/p/ral/jntp/MET/MET_releases/9.1" correctly
set in
> the top tutorial.conf file (and then even set it as an environmental
> variable in my .bashrc):
>
> 03/11 23:09:13.302 metplus (config_launcher.py:564) ERROR: Path must
exist:
> /glade/p/ral/jntp/MET/MET_releases/9.1
>
> INPUT_BASE = /glade/u/home/nleonardo/ISAIAS2020_data
> 03/11 23:09:13Z run-METplus-metplus: ERROR:  Path must exist:
> /glade/p/ral/jntp/MET/MET_releases/9.1
>
> INPUT_BASE = /glade/u/home/nleonardo/ISAIAS2020_data
> 03/11 23:09:13.302 metplus (met_util.py:128) ERROR: MET_INSTALL_DIR
must be
> set correctly to run METplus
> 03/11 23:09:13Z run-METplus-metplus: ERROR:  MET_INSTALL_DIR must be
set
> correctly to run METplus
>
>
> I get the same thing even if I run with v9.1.1, though I notice "The
> following have been reloaded with a version change:
>   1) met/9.1.1 => met/9.1" prints whenever I log in.  Attached is
the log
> from the previous run that worked.
>
> Lindsay is currently helping me troubleshoot this, but we also
wanted to
> check if the error has to do with the recent update to Cheyenne.
Thank
> you.
> -Nick
>
>

--
Julie Prestopnik (she/her)
Software Engineer
National Center for Atmospheric Research
Research Applications Laboratory
Email: jpresto at ucar.edu

My working day may not be your working day.  Please do not feel
obliged to
reply to this email outside of your normal working hours.

------------------------------------------------
Subject: METplus error from Cheyenne update?
From: Nicholas Leonardo
Time: Tue Mar 16 12:22:50 2021

Hello Julie,
Yes, I was loading the existing build for NCAR/Cheyenne, though when I
first went over it I swore it said METv9.1.1 (which definitely worked
for
me before in conjunction with METplus3.1).  Perhaps I'll quickly test
loading 9.1.2 instead.  Regardless, the following are all modules I
loaded
via my .bashrc:

module use /glade/p/ral/jntp/MET/MET_releases/modulefiles
module use /glade/p/ral/jntp/MET/METplus/modulefiles
module load met/9.1.1
module load metplus/3.1
ncar_pylib


I am using /glade/p/ral/jntp/MET/METplus/METplus-
3.1/ush/master_metplus.py.

Thank you.

-Nick




On Tue, Mar 16, 2021 at 2:06 PM Julie Prestopnik via RT
<met_help at ucar.edu>
wrote:

> Hi Nick.
>
> I see you are having trouble with METplus on Cheyenne.  Can you
please tell
> me what you are loading for running METplus?
>
> I see on our Existing Builds page
> <https://dtcenter.org/community-code/metplus/metplus-3-1-existing-
builds>:
> module use /glade/p/ral/jntp/MET/METplus/modulefiles
> module load metplus/3.1
> ncar_pylib
>
> Is that what you are loading?
>
> Can you please run "which master_metplus.py" to verify the location
of
> master_metplus.py and let me know that result as well?
>
> I will start looking into this as well.
>
> Julie
>
> On Tue, Mar 16, 2021 at 11:13 AM Nicholas Leonardo via RT <
> met_help at ucar.edu>
> wrote:
>
> >
> > Tue Mar 16 11:13:01 2021: Request 99067 was acted upon.
> > Transaction: Ticket created by nleonardo87 at gmail.com
> >        Queue: met_help
> >      Subject: METplus error from Cheyenne update?
> >        Owner: Nobody
> >   Requestors: nleonardo87 at gmail.com
> >       Status: new
> >  Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=99067 >
> >
> >
> > Hello,
> > I am now trying to run the feature-relative use case again
(without user
> > diagnostics;
> >
> >
> https://dtcenter.org/metplus-practical-session-guide-version-3-
0/session-5-trkintfeature-relative/use-case-feature-relative
> > ).
> > I successfully ran this program ~2-3 weeks ago, but now get
errors.
> >
> > I am working under
/glade/u/home/nleonardo/METplus_Tutorial/ATTEMPT001/
> and
> > entered:
> >
> > master_metplus.py -c ${METPLUS_TUTORIAL_DIR}/tutorial.conf -c
> >
> >
>
${METPLUS_TUTORIAL_DIR}/ATTEMPT001/TCStat_SeriesAnalysis_fcstGFS_anlGFS_TCrelative.conf
> > -c
> >
> >
>
dir.OUTPUT_BASE=${METPLUS_TUTORIAL_DIR}/ATTEMPT001/output/feature_relative_by_init;
> >
> >
> > The error does not make sense in that I indeed have the existing
path
> > "MET_INSTALL_DIR = /glade/p/ral/jntp/MET/MET_releases/9.1"
correctly set
> in
> > the top tutorial.conf file (and then even set it as an
environmental
> > variable in my .bashrc):
> >
> > 03/11 23:09:13.302 metplus (config_launcher.py:564) ERROR: Path
must
> exist:
> > /glade/p/ral/jntp/MET/MET_releases/9.1
> >
> > INPUT_BASE = /glade/u/home/nleonardo/ISAIAS2020_data
> > 03/11 23:09:13Z run-METplus-metplus: ERROR:  Path must exist:
> > /glade/p/ral/jntp/MET/MET_releases/9.1
> >
> > INPUT_BASE = /glade/u/home/nleonardo/ISAIAS2020_data
> > 03/11 23:09:13.302 metplus (met_util.py:128) ERROR:
MET_INSTALL_DIR must
> be
> > set correctly to run METplus
> > 03/11 23:09:13Z run-METplus-metplus: ERROR:  MET_INSTALL_DIR must
be set
> > correctly to run METplus
> >
> >
> > I get the same thing even if I run with v9.1.1, though I notice
"The
> > following have been reloaded with a version change:
> >   1) met/9.1.1 => met/9.1" prints whenever I log in.  Attached is
the log
> > from the previous run that worked.
> >
> > Lindsay is currently helping me troubleshoot this, but we also
wanted to
> > check if the error has to do with the recent update to Cheyenne.
Thank
> > you.
> > -Nick
> >
> >
>
> --
> Julie Prestopnik (she/her)
> Software Engineer
> National Center for Atmospheric Research
> Research Applications Laboratory
> Email: jpresto at ucar.edu
>
> My working day may not be your working day.  Please do not feel
obliged to
> reply to this email outside of your normal working hours.
>
>

------------------------------------------------
Subject: METplus error from Cheyenne update?
From: Julie Prestopnik
Time: Tue Mar 16 12:38:15 2021

Hi Nick.

Thanks!  I copied your files over and was able to reproduce the
problem.  I
believe I have found the problem and that the given error was
misleading.

Looking in your metplus_final.conf file at:
/glade/u/home/nleonardo/METplus_Tutorial/ATTEMPT001/output/feature_relative_by_init/metplus_final.conf


for MET_INSTALL DIR, I saw strange formatting:

> [dir]
> OUTPUT_BASE =
>
/glade/u/home/nleonardo/METplus_Tutorial/ATTEMPT001/output/feature_relative_by_init
> STAGING_DIR = {OUTPUT_BASE}/stage
> MET_INSTALL_DIR = /glade/p/ral/jntp/MET/MET_releases/9.1.1
>




                  INPUT_BASE = /glade/u/home/nleonardo/ISAIAS2020_data
> MET_BIN_DIR = {MET_INSTALL_DIR}/bin


Note the extra lines and spaces above and before "INPUT_BASE".

Looking in your tutorial.conf file located
at /glade/u/home/nleonardo/METplus_Tutorial/tutorial.conf, I see that
there
is a space before "INPUT_BASE":

> # Location of input data required to run use cases
> # INPUT_BASE = {ENV[METPLUS_DATA]}
>  INPUT_BASE = /glade/u/home/nleonardo/ISAIAS2020_data
> #INPUT_BASE = /glade/p/ral/jntp/MET/METplus/METplus-4.0_sample_data


In my copy of your files, I removed the space before "INPUT_BASE", ran
the
case, and started getting output.  Please try removing the space and
seeing
if it works for you as well.

It looks like we need to improve our error checking for cases like
this.
Thank you for bringing this to our attention.  I will create a GitHub
Issue
to enhance our error checking.

Julie

On Tue, Mar 16, 2021 at 12:22 PM Nicholas Leonardo via RT
<met_help at ucar.edu>
wrote:

>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=99067 >
>
> Hello Julie,
> Yes, I was loading the existing build for NCAR/Cheyenne, though when
I
> first went over it I swore it said METv9.1.1 (which definitely
worked for
> me before in conjunction with METplus3.1).  Perhaps I'll quickly
test
> loading 9.1.2 instead.  Regardless, the following are all modules I
loaded
> via my .bashrc:
>
> module use /glade/p/ral/jntp/MET/MET_releases/modulefiles
> module use /glade/p/ral/jntp/MET/METplus/modulefiles
> module load met/9.1.1
> module load metplus/3.1
> ncar_pylib
>
>
> I am using /glade/p/ral/jntp/MET/METplus/METplus-
3.1/ush/master_metplus.py.
>
> Thank you.
>
> -Nick
>
>
>
>
> On Tue, Mar 16, 2021 at 2:06 PM Julie Prestopnik via RT
<met_help at ucar.edu
> >
> wrote:
>
> > Hi Nick.
> >
> > I see you are having trouble with METplus on Cheyenne.  Can you
please
> tell
> > me what you are loading for running METplus?
> >
> > I see on our Existing Builds page
> > <https://dtcenter.org/community-code/metplus/metplus-3-1-existing-
builds
> >:
> > module use /glade/p/ral/jntp/MET/METplus/modulefiles
> > module load metplus/3.1
> > ncar_pylib
> >
> > Is that what you are loading?
> >
> > Can you please run "which master_metplus.py" to verify the
location of
> > master_metplus.py and let me know that result as well?
> >
> > I will start looking into this as well.
> >
> > Julie
> >
> > On Tue, Mar 16, 2021 at 11:13 AM Nicholas Leonardo via RT <
> > met_help at ucar.edu>
> > wrote:
> >
> > >
> > > Tue Mar 16 11:13:01 2021: Request 99067 was acted upon.
> > > Transaction: Ticket created by nleonardo87 at gmail.com
> > >        Queue: met_help
> > >      Subject: METplus error from Cheyenne update?
> > >        Owner: Nobody
> > >   Requestors: nleonardo87 at gmail.com
> > >       Status: new
> > >  Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=99067
> >
> > >
> > >
> > > Hello,
> > > I am now trying to run the feature-relative use case again
(without
> user
> > > diagnostics;
> > >
> > >
> >
> https://dtcenter.org/metplus-practical-session-guide-version-3-
0/session-5-trkintfeature-relative/use-case-feature-relative
> > > ).
> > > I successfully ran this program ~2-3 weeks ago, but now get
errors.
> > >
> > > I am working under
/glade/u/home/nleonardo/METplus_Tutorial/ATTEMPT001/
> > and
> > > entered:
> > >
> > > master_metplus.py -c ${METPLUS_TUTORIAL_DIR}/tutorial.conf -c
> > >
> > >
> >
>
${METPLUS_TUTORIAL_DIR}/ATTEMPT001/TCStat_SeriesAnalysis_fcstGFS_anlGFS_TCrelative.conf
> > > -c
> > >
> > >
> >
>
dir.OUTPUT_BASE=${METPLUS_TUTORIAL_DIR}/ATTEMPT001/output/feature_relative_by_init;
> > >
> > >
> > > The error does not make sense in that I indeed have the existing
path
> > > "MET_INSTALL_DIR = /glade/p/ral/jntp/MET/MET_releases/9.1"
correctly
> set
> > in
> > > the top tutorial.conf file (and then even set it as an
environmental
> > > variable in my .bashrc):
> > >
> > > 03/11 23:09:13.302 metplus (config_launcher.py:564) ERROR: Path
must
> > exist:
> > > /glade/p/ral/jntp/MET/MET_releases/9.1
> > >
> > > INPUT_BASE = /glade/u/home/nleonardo/ISAIAS2020_data
> > > 03/11 23:09:13Z run-METplus-metplus: ERROR:  Path must exist:
> > > /glade/p/ral/jntp/MET/MET_releases/9.1
> > >
> > > INPUT_BASE = /glade/u/home/nleonardo/ISAIAS2020_data
> > > 03/11 23:09:13.302 metplus (met_util.py:128) ERROR:
MET_INSTALL_DIR
> must
> > be
> > > set correctly to run METplus
> > > 03/11 23:09:13Z run-METplus-metplus: ERROR:  MET_INSTALL_DIR
must be
> set
> > > correctly to run METplus
> > >
> > >
> > > I get the same thing even if I run with v9.1.1, though I notice
"The
> > > following have been reloaded with a version change:
> > >   1) met/9.1.1 => met/9.1" prints whenever I log in.  Attached
is the
> log
> > > from the previous run that worked.
> > >
> > > Lindsay is currently helping me troubleshoot this, but we also
wanted
> to
> > > check if the error has to do with the recent update to Cheyenne.
Thank
> > > you.
> > > -Nick
> > >
> > >
> >
> > --
> > Julie Prestopnik (she/her)
> > Software Engineer
> > National Center for Atmospheric Research
> > Research Applications Laboratory
> > Email: jpresto at ucar.edu
> >
> > My working day may not be your working day.  Please do not feel
obliged
> to
> > reply to this email outside of your normal working hours.
> >
> >
>
>

--
Julie Prestopnik (she/her)
Software Engineer
National Center for Atmospheric Research
Research Applications Laboratory
Email: jpresto at ucar.edu

My working day may not be your working day.  Please do not feel
obliged to
reply to this email outside of your normal working hours.

------------------------------------------------
Subject: METplus error from Cheyenne update?
From: Nicholas Leonardo
Time: Tue Mar 16 13:46:52 2021

Excellent!  That appears to have done the trick.

With this error in mind, I'll retry the versions of MET/python used
for the
PV/IVT diagnostics, assuming my previous failed attempt was from the
same
problem.  That way, we have one working version for all we need.

Afterwards, I can try to insert my own diagnostics, such as
calculating the
900-700-hPa potential stability.  I gather this can be accomplished by
modifying the config file to accommodate python scripts formatted
similar
to those developed by Lindsay and Taylor.

Thank you all very much.

-Nick

On Tue, Mar 16, 2021 at 2:38 PM Julie Prestopnik via RT
<met_help at ucar.edu>
wrote:

> Hi Nick.
>
> Thanks!  I copied your files over and was able to reproduce the
problem.  I
> believe I have found the problem and that the given error was
misleading.
>
> Looking in your metplus_final.conf file at:
>
>
/glade/u/home/nleonardo/METplus_Tutorial/ATTEMPT001/output/feature_relative_by_init/metplus_final.conf
>
>
> for MET_INSTALL DIR, I saw strange formatting:
>
> > [dir]
> > OUTPUT_BASE =
> >
>
/glade/u/home/nleonardo/METplus_Tutorial/ATTEMPT001/output/feature_relative_by_init
> > STAGING_DIR = {OUTPUT_BASE}/stage
> > MET_INSTALL_DIR = /glade/p/ral/jntp/MET/MET_releases/9.1.1
> >
>
>
>
>
>                   INPUT_BASE =
/glade/u/home/nleonardo/ISAIAS2020_data
> > MET_BIN_DIR = {MET_INSTALL_DIR}/bin
>
>
> Note the extra lines and spaces above and before "INPUT_BASE".
>
> Looking in your tutorial.conf file located
> at /glade/u/home/nleonardo/METplus_Tutorial/tutorial.conf, I see
that there
> is a space before "INPUT_BASE":
>
> > # Location of input data required to run use cases
> > # INPUT_BASE = {ENV[METPLUS_DATA]}
> >  INPUT_BASE = /glade/u/home/nleonardo/ISAIAS2020_data
> > #INPUT_BASE = /glade/p/ral/jntp/MET/METplus/METplus-
4.0_sample_data
>
>
> In my copy of your files, I removed the space before "INPUT_BASE",
ran the
> case, and started getting output.  Please try removing the space and
seeing
> if it works for you as well.
>
> It looks like we need to improve our error checking for cases like
this.
> Thank you for bringing this to our attention.  I will create a
GitHub Issue
> to enhance our error checking.
>
> Julie
>
> On Tue, Mar 16, 2021 at 12:22 PM Nicholas Leonardo via RT <
> met_help at ucar.edu>
> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=99067 >
> >
> > Hello Julie,
> > Yes, I was loading the existing build for NCAR/Cheyenne, though
when I
> > first went over it I swore it said METv9.1.1 (which definitely
worked for
> > me before in conjunction with METplus3.1).  Perhaps I'll quickly
test
> > loading 9.1.2 instead.  Regardless, the following are all modules
I
> loaded
> > via my .bashrc:
> >
> > module use /glade/p/ral/jntp/MET/MET_releases/modulefiles
> > module use /glade/p/ral/jntp/MET/METplus/modulefiles
> > module load met/9.1.1
> > module load metplus/3.1
> > ncar_pylib
> >
> >
> > I am using
> /glade/p/ral/jntp/MET/METplus/METplus-3.1/ush/master_metplus.py.
> >
> > Thank you.
> >
> > -Nick
> >
> >
> >
> >
> > On Tue, Mar 16, 2021 at 2:06 PM Julie Prestopnik via RT <
> met_help at ucar.edu
> > >
> > wrote:
> >
> > > Hi Nick.
> > >
> > > I see you are having trouble with METplus on Cheyenne.  Can you
please
> > tell
> > > me what you are loading for running METplus?
> > >
> > > I see on our Existing Builds page
> > > <
> https://dtcenter.org/community-code/metplus/metplus-3-1-existing-
builds
> > >:
> > > module use /glade/p/ral/jntp/MET/METplus/modulefiles
> > > module load metplus/3.1
> > > ncar_pylib
> > >
> > > Is that what you are loading?
> > >
> > > Can you please run "which master_metplus.py" to verify the
location of
> > > master_metplus.py and let me know that result as well?
> > >
> > > I will start looking into this as well.
> > >
> > > Julie
> > >
> > > On Tue, Mar 16, 2021 at 11:13 AM Nicholas Leonardo via RT <
> > > met_help at ucar.edu>
> > > wrote:
> > >
> > > >
> > > > Tue Mar 16 11:13:01 2021: Request 99067 was acted upon.
> > > > Transaction: Ticket created by nleonardo87 at gmail.com
> > > >        Queue: met_help
> > > >      Subject: METplus error from Cheyenne update?
> > > >        Owner: Nobody
> > > >   Requestors: nleonardo87 at gmail.com
> > > >       Status: new
> > > >  Ticket <URL:
> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=99067
> > >
> > > >
> > > >
> > > > Hello,
> > > > I am now trying to run the feature-relative use case again
(without
> > user
> > > > diagnostics;
> > > >
> > > >
> > >
> >
> https://dtcenter.org/metplus-practical-session-guide-version-3-
0/session-5-trkintfeature-relative/use-case-feature-relative
> > > > ).
> > > > I successfully ran this program ~2-3 weeks ago, but now get
errors.
> > > >
> > > > I am working under
> /glade/u/home/nleonardo/METplus_Tutorial/ATTEMPT001/
> > > and
> > > > entered:
> > > >
> > > > master_metplus.py -c ${METPLUS_TUTORIAL_DIR}/tutorial.conf -c
> > > >
> > > >
> > >
> >
>
${METPLUS_TUTORIAL_DIR}/ATTEMPT001/TCStat_SeriesAnalysis_fcstGFS_anlGFS_TCrelative.conf
> > > > -c
> > > >
> > > >
> > >
> >
>
dir.OUTPUT_BASE=${METPLUS_TUTORIAL_DIR}/ATTEMPT001/output/feature_relative_by_init;
> > > >
> > > >
> > > > The error does not make sense in that I indeed have the
existing path
> > > > "MET_INSTALL_DIR = /glade/p/ral/jntp/MET/MET_releases/9.1"
correctly
> > set
> > > in
> > > > the top tutorial.conf file (and then even set it as an
environmental
> > > > variable in my .bashrc):
> > > >
> > > > 03/11 23:09:13.302 metplus (config_launcher.py:564) ERROR:
Path must
> > > exist:
> > > > /glade/p/ral/jntp/MET/MET_releases/9.1
> > > >
> > > > INPUT_BASE = /glade/u/home/nleonardo/ISAIAS2020_data
> > > > 03/11 23:09:13Z run-METplus-metplus: ERROR:  Path must exist:
> > > > /glade/p/ral/jntp/MET/MET_releases/9.1
> > > >
> > > > INPUT_BASE = /glade/u/home/nleonardo/ISAIAS2020_data
> > > > 03/11 23:09:13.302 metplus (met_util.py:128) ERROR:
MET_INSTALL_DIR
> > must
> > > be
> > > > set correctly to run METplus
> > > > 03/11 23:09:13Z run-METplus-metplus: ERROR:  MET_INSTALL_DIR
must be
> > set
> > > > correctly to run METplus
> > > >
> > > >
> > > > I get the same thing even if I run with v9.1.1, though I
notice "The
> > > > following have been reloaded with a version change:
> > > >   1) met/9.1.1 => met/9.1" prints whenever I log in.  Attached
is the
> > log
> > > > from the previous run that worked.
> > > >
> > > > Lindsay is currently helping me troubleshoot this, but we also
wanted
> > to
> > > > check if the error has to do with the recent update to
Cheyenne.
> Thank
> > > > you.
> > > > -Nick
> > > >
> > > >
> > >
> > > --
> > > Julie Prestopnik (she/her)
> > > Software Engineer
> > > National Center for Atmospheric Research
> > > Research Applications Laboratory
> > > Email: jpresto at ucar.edu
> > >
> > > My working day may not be your working day.  Please do not feel
obliged
> > to
> > > reply to this email outside of your normal working hours.
> > >
> > >
> >
> >
>
> --
> Julie Prestopnik (she/her)
> Software Engineer
> National Center for Atmospheric Research
> Research Applications Laboratory
> Email: jpresto at ucar.edu
>
> My working day may not be your working day.  Please do not feel
obliged to
> reply to this email outside of your normal working hours.
>
>

------------------------------------------------
Subject: METplus error from Cheyenne update?
From: Julie Prestopnik
Time: Tue Mar 16 16:45:16 2021

Great!  I'm so glad it worked for you.

George, a METplus developer, modified the code to check directory
paths.
In our next release the error you would receive would look something
like
this:

ERROR: run_metplus  failed: Invalid value for [config] MET_INSTALL_DIR
> (/d1/projects/MET/MET_releases/met-10.0.0-beta4
> INPUT_BASE = /d1/projects/METplus/METplus_Data). Hint: Check that
next
> variable in the config file does not start with a space


which would hopefully be more helpful.

I'll go ahead and resolve this ticket.  Please feel free to start a
new
email to met_help at ucar.edu with any other questions or issues you
encounter.

Julie

On Tue, Mar 16, 2021 at 1:46 PM Nicholas Leonardo via RT
<met_help at ucar.edu>
wrote:

>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=99067 >
>
> Excellent!  That appears to have done the trick.
>
> With this error in mind, I'll retry the versions of MET/python used
for the
> PV/IVT diagnostics, assuming my previous failed attempt was from the
same
> problem.  That way, we have one working version for all we need.
>
> Afterwards, I can try to insert my own diagnostics, such as
calculating the
> 900-700-hPa potential stability.  I gather this can be accomplished
by
> modifying the config file to accommodate python scripts formatted
similar
> to those developed by Lindsay and Taylor.
>
> Thank you all very much.
>
> -Nick
>
> On Tue, Mar 16, 2021 at 2:38 PM Julie Prestopnik via RT
<met_help at ucar.edu
> >
> wrote:
>
> > Hi Nick.
> >
> > Thanks!  I copied your files over and was able to reproduce the
> problem.  I
> > believe I have found the problem and that the given error was
misleading.
> >
> > Looking in your metplus_final.conf file at:
> >
> >
>
/glade/u/home/nleonardo/METplus_Tutorial/ATTEMPT001/output/feature_relative_by_init/metplus_final.conf
> >
> >
> > for MET_INSTALL DIR, I saw strange formatting:
> >
> > > [dir]
> > > OUTPUT_BASE =
> > >
> >
>
/glade/u/home/nleonardo/METplus_Tutorial/ATTEMPT001/output/feature_relative_by_init
> > > STAGING_DIR = {OUTPUT_BASE}/stage
> > > MET_INSTALL_DIR = /glade/p/ral/jntp/MET/MET_releases/9.1.1
> > >
> >
> >
> >
> >
> >                   INPUT_BASE =
/glade/u/home/nleonardo/ISAIAS2020_data
> > > MET_BIN_DIR = {MET_INSTALL_DIR}/bin
> >
> >
> > Note the extra lines and spaces above and before "INPUT_BASE".
> >
> > Looking in your tutorial.conf file located
> > at /glade/u/home/nleonardo/METplus_Tutorial/tutorial.conf, I see
that
> there
> > is a space before "INPUT_BASE":
> >
> > > # Location of input data required to run use cases
> > > # INPUT_BASE = {ENV[METPLUS_DATA]}
> > >  INPUT_BASE = /glade/u/home/nleonardo/ISAIAS2020_data
> > > #INPUT_BASE = /glade/p/ral/jntp/MET/METplus/METplus-
4.0_sample_data
> >
> >
> > In my copy of your files, I removed the space before "INPUT_BASE",
ran
> the
> > case, and started getting output.  Please try removing the space
and
> seeing
> > if it works for you as well.
> >
> > It looks like we need to improve our error checking for cases like
this.
> > Thank you for bringing this to our attention.  I will create a
GitHub
> Issue
> > to enhance our error checking.
> >
> > Julie
> >
> > On Tue, Mar 16, 2021 at 12:22 PM Nicholas Leonardo via RT <
> > met_help at ucar.edu>
> > wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=99067 >
> > >
> > > Hello Julie,
> > > Yes, I was loading the existing build for NCAR/Cheyenne, though
when I
> > > first went over it I swore it said METv9.1.1 (which definitely
worked
> for
> > > me before in conjunction with METplus3.1).  Perhaps I'll quickly
test
> > > loading 9.1.2 instead.  Regardless, the following are all
modules I
> > loaded
> > > via my .bashrc:
> > >
> > > module use /glade/p/ral/jntp/MET/MET_releases/modulefiles
> > > module use /glade/p/ral/jntp/MET/METplus/modulefiles
> > > module load met/9.1.1
> > > module load metplus/3.1
> > > ncar_pylib
> > >
> > >
> > > I am using
> > /glade/p/ral/jntp/MET/METplus/METplus-3.1/ush/master_metplus.py.
> > >
> > > Thank you.
> > >
> > > -Nick
> > >
> > >
> > >
> > >
> > > On Tue, Mar 16, 2021 at 2:06 PM Julie Prestopnik via RT <
> > met_help at ucar.edu
> > > >
> > > wrote:
> > >
> > > > Hi Nick.
> > > >
> > > > I see you are having trouble with METplus on Cheyenne.  Can
you
> please
> > > tell
> > > > me what you are loading for running METplus?
> > > >
> > > > I see on our Existing Builds page
> > > > <
> > https://dtcenter.org/community-code/metplus/metplus-3-1-existing-
builds
> > > >:
> > > > module use /glade/p/ral/jntp/MET/METplus/modulefiles
> > > > module load metplus/3.1
> > > > ncar_pylib
> > > >
> > > > Is that what you are loading?
> > > >
> > > > Can you please run "which master_metplus.py" to verify the
location
> of
> > > > master_metplus.py and let me know that result as well?
> > > >
> > > > I will start looking into this as well.
> > > >
> > > > Julie
> > > >
> > > > On Tue, Mar 16, 2021 at 11:13 AM Nicholas Leonardo via RT <
> > > > met_help at ucar.edu>
> > > > wrote:
> > > >
> > > > >
> > > > > Tue Mar 16 11:13:01 2021: Request 99067 was acted upon.
> > > > > Transaction: Ticket created by nleonardo87 at gmail.com
> > > > >        Queue: met_help
> > > > >      Subject: METplus error from Cheyenne update?
> > > > >        Owner: Nobody
> > > > >   Requestors: nleonardo87 at gmail.com
> > > > >       Status: new
> > > > >  Ticket <URL:
> > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=99067
> > > >
> > > > >
> > > > >
> > > > > Hello,
> > > > > I am now trying to run the feature-relative use case again
(without
> > > user
> > > > > diagnostics;
> > > > >
> > > > >
> > > >
> > >
> >
> https://dtcenter.org/metplus-practical-session-guide-version-3-
0/session-5-trkintfeature-relative/use-case-feature-relative
> > > > > ).
> > > > > I successfully ran this program ~2-3 weeks ago, but now get
errors.
> > > > >
> > > > > I am working under
> > /glade/u/home/nleonardo/METplus_Tutorial/ATTEMPT001/
> > > > and
> > > > > entered:
> > > > >
> > > > > master_metplus.py -c ${METPLUS_TUTORIAL_DIR}/tutorial.conf
-c
> > > > >
> > > > >
> > > >
> > >
> >
>
${METPLUS_TUTORIAL_DIR}/ATTEMPT001/TCStat_SeriesAnalysis_fcstGFS_anlGFS_TCrelative.conf
> > > > > -c
> > > > >
> > > > >
> > > >
> > >
> >
>
dir.OUTPUT_BASE=${METPLUS_TUTORIAL_DIR}/ATTEMPT001/output/feature_relative_by_init;
> > > > >
> > > > >
> > > > > The error does not make sense in that I indeed have the
existing
> path
> > > > > "MET_INSTALL_DIR = /glade/p/ral/jntp/MET/MET_releases/9.1"
> correctly
> > > set
> > > > in
> > > > > the top tutorial.conf file (and then even set it as an
> environmental
> > > > > variable in my .bashrc):
> > > > >
> > > > > 03/11 23:09:13.302 metplus (config_launcher.py:564) ERROR:
Path
> must
> > > > exist:
> > > > > /glade/p/ral/jntp/MET/MET_releases/9.1
> > > > >
> > > > > INPUT_BASE = /glade/u/home/nleonardo/ISAIAS2020_data
> > > > > 03/11 23:09:13Z run-METplus-metplus: ERROR:  Path must
exist:
> > > > > /glade/p/ral/jntp/MET/MET_releases/9.1
> > > > >
> > > > > INPUT_BASE = /glade/u/home/nleonardo/ISAIAS2020_data
> > > > > 03/11 23:09:13.302 metplus (met_util.py:128) ERROR:
MET_INSTALL_DIR
> > > must
> > > > be
> > > > > set correctly to run METplus
> > > > > 03/11 23:09:13Z run-METplus-metplus: ERROR:  MET_INSTALL_DIR
must
> be
> > > set
> > > > > correctly to run METplus
> > > > >
> > > > >
> > > > > I get the same thing even if I run with v9.1.1, though I
notice
> "The
> > > > > following have been reloaded with a version change:
> > > > >   1) met/9.1.1 => met/9.1" prints whenever I log in.
Attached is
> the
> > > log
> > > > > from the previous run that worked.
> > > > >
> > > > > Lindsay is currently helping me troubleshoot this, but we
also
> wanted
> > > to
> > > > > check if the error has to do with the recent update to
Cheyenne.
> > Thank
> > > > > you.
> > > > > -Nick
> > > > >
> > > > >
> > > >
> > > > --
> > > > Julie Prestopnik (she/her)
> > > > Software Engineer
> > > > National Center for Atmospheric Research
> > > > Research Applications Laboratory
> > > > Email: jpresto at ucar.edu
> > > >
> > > > My working day may not be your working day.  Please do not
feel
> obliged
> > > to
> > > > reply to this email outside of your normal working hours.
> > > >
> > > >
> > >
> > >
> >
> > --
> > Julie Prestopnik (she/her)
> > Software Engineer
> > National Center for Atmospheric Research
> > Research Applications Laboratory
> > Email: jpresto at ucar.edu
> >
> > My working day may not be your working day.  Please do not feel
obliged
> to
> > reply to this email outside of your normal working hours.
> >
> >
>
>

--
Julie Prestopnik (she/her)
Software Engineer
National Center for Atmospheric Research
Research Applications Laboratory
Email: jpresto at ucar.edu

My working day may not be your working day.  Please do not feel
obliged to
reply to this email outside of your normal working hours.

------------------------------------------------


More information about the Met_help mailing list