Email Archive, November 2001.

One more thing I just noticed, The EMS_FILE directory is referenced in uirundes.sas but it should reference ems_run. All the run_desc sas data sets are wrong.
I have attached a newer yet version of solaris.bat that should fix the problem. I just added a line to the batch file that would set EMS_FILE to equal EMS_RUN.

Mark.

At 10:00 AM 12/19/01 -0500, Eric E. Zalewsky NYSDEC wrote:

Hi Mark,

I think we got the problem straightened out here with the untaring and
uncompressing. We ended up trying an older version of "tar" and it worked
no problem. (Winston is the man!)
Anyway, I just wanted to make sure I have everything in the
$EMS_RAW directory. Here is what was extracted from the 3 ems_raw?.Z
files...

biome3/common with 30 files
canada/common with 21 files
met/common with 3 files
net96_v4 with 7 files (just point and area ff's aren't there any mobile ff's???)

Is this everything?
Thanks,

Eric

________________________________________________________________________________________________

Right, I forgot about that.....

Mark

At 09:07 AM 12/20/01 -0500, Eric E. Zalewsky NYSDEC wrote:

Thanks Mark. You still need to change your "if" statement
when you create $EMS right? It also checks for the presence of
fipsstcy in $EMS_CAT as a check for the creation of $EMS...

Eric

______________________________________________________________________________________________

Eric,

yes $EMS_GRD is supposed to be empty until you run subgrid.sas then it will start to fill up with goodies.

Thanks for the info on ls I'll remove it. I would suggest that you move the "./" part of your path from the front to the back. For other reasons besides this one.

Mark .

At 01:27 PM 12/20/01 -0500, Eric E. Zalewsky NYSDEC wrote:

$EMS_GRD is supposed to be empty?

There's a file in $EMS_BIN called ls, I thought I was losing my mind...
whenever I'd type ls in bin, nothing would happen, but when I listed bin
from another directory, I got the listing.... :)

__________________________________________________________________________________________

Easting and Northing when the value of "projectn" is LAMBERT or UTM and LAT/LON when the value of projectn is GEOGRAPHIC. The units are whatever is listed in "projunit" it is either DD for decimal degrees or METERS for lambert or UTM.

The example run_desc files sent with the model should should have everything set correctly.

Mark.

At 10:21 AM 12/21/01 -0500, Eric E. Zalewsky NYSDEC wrote:

so are the values of utmxorig and utmyorig (at the sw corner) in the run_desc file
eastings and northings in (meters or km) or lat lon in (degrees)?

eric

___________________________________________________________________________________________

____________________________________________________________________________________________

The following installation errors occured and this is a chain of emails that followed between Mark Janssen and a State.

First for version 1.1 of EMS2001 make sure you are using solaris.bat from this web page NOT THE ONE ON THE CD

For Error:
############ Finished Cimport of SAS datasets
ERROR: EMS_CAT not created: File /storm1/ems/ems2001/project/unified/lookup/fipsstcy.ssd01 does not exist
ERROR: EMS not created: File /storm1/ems/ems2001/project/unified/lookup/fipsstcy.ssd01 does not exist

Check the Log file in $EMS_BIN/unix_cimport.log

Check for "ERROR" It might happen that the export file is bad or your version of SAS did not like it. Let me know what it says

For this one:

ERROR: EMS_RUN not created for 910718 , File does not exist: /storm1/ems/ems2001/project/unified/gridspec/unified12/scenario/910718/sas/run_desc.ssd01
ERROR: EMS_RUN not created for 960719 , File does not exist: /storm1/ems/ems2001/project/unified/gridspec/unified12/scenario/960719/sas/run_desc.ssd01

For this one Check to see that the directory exists:

/storm1/ems/ems2001/project/unified/gridspec/unified12/scenario/910718/sas/

Next look to see that in the directory listed above there is a file named

run_desc.in

If that file exists, Then check the log file $EMS_BIN/run_desc.log for ERRORS. If they exist try and figure out why or email me the entire log file. If anything does not exist anywhere along the way, We should try and figure out why it does not. Let me know what you find.

Resolution

I successfully re-installed EMS2001, under SAS version 8 and Solaris 8, with the
following changes to the "solaris.bat" installation script:

1) SAS version 8 under Solaris uses the "sas7bdat" extension:

setenv SAS_EXTENSION sas7bdat

2)The line

setenv EMS_FILE $EMS_RUN

was added to the script immediatedly before the invocation of "uirundes.sas".

Thanks for your help.