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

[McIDAS #WUE-272422]: GOES 12 or 13

Hi Gustavo,

re:
> Thank you a lot! Very useful your help and work!!!.

No worries.

re:
> Specially about GEMPAK's software and mail
> it was producing by scripts where I had no idea.

Yes, things run out of cron will generate email to the user running
the cron jobs unless the messages are sent to the "bit bucket" or
(in Linux) if the MAILTO="" line is added.  It is always best to
write log output to a specific log file which is, in turn, "rotated"
periodically.  I added the "rotation" of the GEMPAK log files to
'ldm's crontab on unidata.at.fcen.uba.ar right after Michael installed
GEMPAK v6.2.0.  Here is the entry that I am talking about:

#
# Rotate GEMPAK logs
#
00 21 * * * util/dcrotatelog.csh >/dev/null 2>&1

In this action, any/all informational comments are explicitly sent to
the "bit bucket" (/dev/null).  This specific redirection of log
comments is no longer needed since the MAILTO="" line says to _not_
send email.  I left in this syntax so you could see how the redirection
of log messages is done on systems where the MAILTO="" construct is
not available.

re:
> I'll add a few comments and questions
> 
> (I) GOES, a doubt with sat-solo.sh
> then all images (from Argentina and its limits)
> come from GOES -East (GOES 13)?

Yes.  The GOES-West satellite sectors in the UNIWISC (aka MCIDAS)
datastream do not cover any portion of Argentina; only the GOES-East
ones provide coverage.

re:
> Sorry ... It is not clear form me: then GOES 12
> is no any more available? even if using Mcidas??

The GOES-East satellite sectors (images) in the UNIWISC datastream
are composites -- they contain information from both GOES-13 and
GOES-12.  Since GOES-13 is the operational GOES-East satellite, I
chose to use as much of its coverage in the composites as possible.
The information from GOES-12 is added to the primary image in areas
were GOES-13 does not scan when the time difference in GOES-13 and
GOES-12 scans is less than or equal to 15 minutes.  Because GOES-13
provides the majority of the information in the composites, the images
are labeled as being from GOES-13

NB: You can always tell which portion of the GOES-East sectors are
from GOES-12 since the edge of that portion of the image will have
a jagged, sawtooth border.

re:
> (II) Disk
> all right!!, I started with backup in order to free disk space.

Very good.  You still had disk space, but I foresaw that it would
disappear as new images get saved and new GIFs get created.

re:
> (III) Cron - mail
> I think they are all right, yet I'll check all of them.
> I'll check rmtcuba's mail.

rmtcuba's mail file is very large.  I am willing to bet that nobody
is reading the mail messages, so they are probably not that important.
Clearing out the mail file (if the messages are not important) is
very easily done by either the user or 'root' using:

cat /dev/null > /var/spool/mail/rmtcuba

re:
> Additional tips
> ==========
> 
> (IV) By the way, local time
> 
> [ldm@unidata ~]$ date
> Fri Dec  3 11:00:28 ARST 2010
> [ldm@unidata ~]$ date -u
> Fri Dec  3 13:00:31 UTC 2010
> 
> I think it's wrong; there is an hour difference with our real time.
> 
> This must be UTC's time less 3 hours..

Hmm... Is there no daylight savings time in Argentina?

> In this example, it should be:
> 10:00 ARST
> 
> Can we change it? Or it can influence or damage
> part of the work you've done??
> 
> By the time being here there is no difference
> between winter and summer's time.
> 
> To have real local time is important
> since there are scripts using it
> as they display such a time.

Just so you know, I did not setup the time zone on unidata.at.fcen.uba.ar.
If what is there is wrong (the UTC time _is_ correct; I just checked), then
someone there needs to address the issue.  Please be careful in doing this,
however... the important thing for the LDM is that the indicated UTC time
is correct.

re:
> (V) In order to plot Argentina's states
> (or province) using gempak could you please
> tell me where I can find more information?

GEMPAK should have a map file that contains states for Argentina.  I will
check later today to see if this is true, and if it isn't how they can
be added.

re:
> Here, for instance with  Golden Software Surfer,
> we have an ASCII's file with the point that shows Argentina's
> provinces... Do you know a way to implement
> this inside Unidata's software?

Again, I think that GEMPAK has a map file that contains the states of
Argentina, but I am not 100% positive.  I will check later today and
get back to you.

re:
> Well, again thank you.

No worries.

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: WUE-272422
Department: Support McIDAS
Priority: Normal
Status: Closed