[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[McIDAS #HPW-941884]: What am I doing wrong?

Hi Gilbert,

re:
> Well, as it turns out, I deleted your last response to me, which
> isn't on your support server yet...weather3's hard
> drives are failing, our network outside of NIU Weather in my departmental
> office blew up, so I had to help with that...and the phone to the hard
> drive vendor for warranty replacement is temporarily down. Can you play a
> country song for me here? ;-)

What's up with you and hard disks?  We rarely see a disk go out...

re:
> OK, but here's the deal: RTIMAGES, which is what stopped on October 1, I
> think, should point to weather2. I just looked on weather.admin at the
> config, and it shows it is pointing there, so I have no idea why it is
> pointing to weather for an outdated dataset.

OK, I am confused.  I logged onto weather.admin as 'mcidas' and did a
'crontab -l' to see what is run out of cron.  Here is what I see:

# DO NOT EDIT THIS FILE - edit the master and reinstall.
# (/tmp/crontab.5141 installed on Thu Aug 18 10:43:43 2005)
# (Cron version -- $Id: crontab.c,v 2.13 1994/01/17 03:20:37 vixie Exp $)
00,12,27,42,57 * * * * /home/mcidas/kickoff /home/mcidas/workdata CFDGONLY.BAT
#00,6,12,18,24,30,36,42,48,54 * * * * /home/mcidas/kickoff 
/home/mcidas/workdata CFDGONLY.BAT
00,3,5,8,10,13,15,18,20,23,25,27,30,32,35,38,40,43,45,47,50,53,55 * * * * 
/home/mcidas/kickoff WEATHERWATCH.BAT
1,2,4,6,7,9,11,12,14,16,19,21,22,24,26,28,29,31,33,34,36,37,39,41,42 * * * * 
/home/mcidas/kickoff WEATHERWATCH.BAT
44,46,48,49,51,52,54,56,57,58,59 * * * * /home/mcidas/kickoff WEATHERWATCH.BAT

It looks like you have 3 entries that each run WEATHERWATCH.BAT from your 
/home/mcidas/kickoff
script.  The first uncommented entry, however, is apparently trying to run 
/home/mcidas/workdata??
Your kickoff script expects the 1st passed parameter to be the name of a McIDAS 
BATCH file:

#!/bin/sh -f

# Set environment variables using Bourne shell syntax

MCHOME=/home/mcidas
MCDATA=$MCHOME/workdata
MCPATH=${MCDATA}:$MCHOME/data:$MCHOME/help

MCLOG=$MCDATA/crontab.log
MCTABLE_READ="${MCHOME}/data/ADDESITE.TXT"

PATH=$MCHOME/bin:$PATH
LD_LIBRARY_PATH=/usr/local/lib:/lib:$MCHOME/lib:/usr/lib

# export the environment variables so that they can be used 

export MCHOME MCDATA MCPATH MCTABLE_READ PATH LD_LIBRARY_PATH

# set logging of stdin and stdout to $MCLOG

exec 2>$MCLOG 1>&2
#exec 2>$MCLOG 2>&1
#exec 2>/dev/null 1>&2

# cd to $MCDATA and run the script whose name was passed in on the command line

cd $MCDATA
mcenv << EOF
  batch.k $1
  exit
EOF

# Done!
exit 0

The first entry should, therefore, always fail.

Also, I looked at WEATHERWATCH.BAT and see that it basically runs WWDISP which
uses by default data from the RTWXTEXT dataset.  Your setup is pointing to
weather2.admin.niu.edu for RTWXTEXT:

<as 'mcidas' on weather.admin>
cd $MCDATA
dataloc.k LIST RTWXTEXT

Group Name                    Server IP Address
--------------------         ----------------------------------------
RTWXTEXT                     WEATHER2.ADMIN.NIU.EDU

<LOCAL-DATA> indicates that data will be accessed from the local data directory.
DATALOC -- done

I see that the GIF file to be created by WEATHERWATCH.BAT does, in fact, get
created (I changed the logging entry in /home/mcidas/kickoff to write to the
log file /home/mcidas/workdata/crontab.log; it was writing to /dev/null).

So, given your crontab entries, the content of /home/mcidas/kickoff, and the
content of /home/mcidas/workdata/WEATHERWATCH.BAT, I am not sure what is
failing for you.

Question:

- which system (weather, weather2 or weather3) is running a cron-initiated
  script that is failing?

re:
> That should point to
> weather2.  I save no GRID data, so that should point to you by default.

OK.

re:
> All text weather info should point to the local machines; I do save things
> like watches, warnings, METARS.

Again, the setup in the 'mcidas' account on weather.admin is to point at
weather2.admin for the weather text dataset, RTWXTEXT.

re:
> All images though, as appropriate, should
> point to weather2 where the datasets aren't sent to the LDM (for example,
> SSEC's 30 minute image update files).

Which machine is running a cron job that is supposed to access image data?

Which BATCH file is supposed to process image data?
I think it must be CFDGONLY.BAT (/home/mcidas/workdata/CFDGONLY.BAT).

A quick look at the copy of CFDGONLY.BAT on weather and weather2 shows
that it will try to access images from the GINIEAST dataset, and
presumably, it will try to access this dataset on weather2.admin.

When I try to list images from the GINIEAST dataset, I get nothing:

<as 'mcidas' on weather2>

dataloc.k LIST GINEAST

weather2-niu Mci-163$ dataloc.k LIST GINIEAST

Group Name                    Server IP Address
--------------------         ----------------------------------------
GINIEAST                     WEATHER2.ADMIN.NIU.EDU

<LOCAL-DATA> indicates that data will be accessed from the local data directory.
DATALOC -- done
cd $MCDATA
weather2-niu Mci-164$ imglist.k GINIEAST/GE1KVIS
Image file directory listing for:GINIEAST/GE1KVIS
imglist.k: done

However, when I try to access the data locally, it works:

dataloc.k ADD GINIEAST LOCAL-DATA

Group Name                    Server IP Address
--------------------         ----------------------------------------
GINIEAST                     <LOCAL-DATA>

<LOCAL-DATA> indicates that data will be accessed from the local data directory.
DATALOC -- done
weather2-niu Mci-167$ imglist.k GINIEAST/GE1KVIS
Image file directory listing for:GINIEAST/GE1KVIS
 Pos Satellite/         Date       Time      Center   Band(s)
     sensor                                 Lat  Lon
 --- -------------  ------------  --------  ---- ---- ------------
  71  G-13 IMG      11 NOV 10315  22:45:19    41   88 1
imglist.k: done


So, the BIG question is why remote access to the dataset is failing while
local access succeeds while at the same time local and remote access to
RTWXTEXT and RTIMAGES works!?

I have been looking into this strangeness for the past two hours now, and
I still don't have an answer for you.  I will keep looking (and scratching
my head).

Until I can figure out what is going on, you may want to point at a different
ADDE server for the GINIEAST data;  I suggest adde.ucar.edu.  Here is
what you should do:

<as 'mcidas' on the machine(s) running the script(s) that display GINIEAST data>
cd $MCDATA
dataloc.k ADD GINIEAST adde.ucar.edu



Cheers,

Tom
--
****************************************************************************
Unidata User Support                                    UCAR Unidata Program
(303) 497-8642                                                 P.O. Box 3000
address@hidden                                   Boulder, CO 80307
----------------------------------------------------------------------------
Unidata HomePage                       http://www.unidata.ucar.edu
****************************************************************************


Ticket Details
===================
Ticket ID: HPW-941884
Department: Support McIDAS
Priority: Normal
Status: Closed