20030602: McIDAS installation at UFRJ (cont.)

>From:  David Garrana Coelho <address@hidden>
>Organization:  UFRJ
>Keywords:  200305280404.h4S44lLd024244 McIDAS ADDE IDD-Brazil

Hi David,

re: recommended configuration for a new IDD/ADDE machine at UFRJ

>       We will be working with $1000 tops, so I dont believe we can get
>dual processor machine. And we don't have 250Gb disks available on local
>market, i was thinking about 2 of 80Gb working as LVM (Logical Volume
>Manager) making a single one of 160Gb.

OK.  If you find that your site becomes very popular for IDD and ADDE
traffic as IDD-Brazil grows, you may want to reevaluate your hardware.

>About OS, never had experience with
>FreeBSD, do you believe its easy to migrate (management wise) from RedHat
>to it?

FreeBSD _is_ different in a number of ways from Linux, so it will take
some amount of effort to adapt to it.  If you have system
administration support, and if they are familiar with BSD Unix, then
you might consider adopting FreeBSD straight off.  If you are like many
universities and you are your system administration support (;-), and
if you are familar/comfortable with Linux, then you will probably want
to stay with it.

re: We would appreciate it if you reference
your participation in the Unidata community and our funding agency, the
US National Science Foundation.  We don't expect this on the images
themselves (it is nice, however), but we really do like to see it
somewhere on our sites' web pages.

Just so you know, I want to reassure you that this is no different than
what we ask of our US university sites.


>       Deal!

OK, good.

re: grabbing the McIDAS docs

>       Tried both ftp and sftp, connection closed after passing password

This is very odd.  I grabbed the McIDAS distribution from the same
machine while logged onto brisa as 'mcidas'.  I had no problems with
the FTP at all.

I just logged onto brisa as 'mcidas' and grabbed the PostScript (tm)
version of the Users Guide:

<login as 'mcidas'>
mkdir docs
mkdir docs/users_guide docs/xcd_admin
mkdir docs/users_guide/ps docs/users_guide/pdf docs/xcd_admin/ps 
docs/xcd_admin/pdf

cd docs/users_guide/ps
ftp ftp.unidata.ucar.edu
  <user> umcidas
  <pass> XXXXXX
  cd unix/2002/docs/users_guide/ps
  prompt
  get *.Z
  lcd ../../xcd_admin/ps
  get *.Z
  quit

You now have the PostScript versions of the McIDAS Users Guide and XCD
Administrators Guide on brisa.

>       On a side note, I created (not intentionally, of course) a bug
>which is preventing me to restart mcidas. I tried to put SYNOP number
>instead of METAR name in station choosing under meteogram menu, it started
>a infinite loop of error messages, crashed, and now it prints a number of
>error messages when I type "mcidas" on prompt. If it helps, i tried to put
>83746 instead of SBGL on it. Hope its not very bad....=/

This is an odd one.  While logged onto brisa as 'mcidas', I took a look
at the contents of the McIDAS String Table.  The String Table is a file
that stores the values of environment variables used in McIDAS.  These
variables are known in McIDAS as 'strings', hence the name String
Table.

The MCGUI defines a number of default values for variables that it will
use in things like image loads, etc.  Those values get written out to
the String Table (instantiated in the file STRTABLE) for use from then
on.

To make the story short, the String Table contents on brisa in the
'mcidas' account are all messed up!  And, they are messed up in a way I
have never seen before.  The only way out of this that I can come up
with at the moment is to delete the String Table, and let it get
created the next time you start a McIDAS session:

<still as 'mcidas'>
cd ~/workdata
rm -f STRTABLE

>       Best Regards,

Have you been looking at the imagery available through the MCGUI?  I
was hoping that you would be looking at the sector being created over
your modeling domain to see if it meets your needs/expectations.

Also, it is interesting to look at the LDM real time statistics pages
for brisa to see how much real time image data is being received there:

http://www.unidata.ucar.edu/staff/chiz/rtstats/siteindex.shtml?brisa.meteoro.ufrj.br

Click on 'volume' for the UNIWISC feed.  The numbers represent the
volume of the imagery data _after_ it has been PNG compressed for IDD
transmission.  Since we get about a 50% reduction in size of VIS images
and up to 80% size reduction in some IR images (e.g. Water Vapor),
these number are small by a factor of 2 to 3.  I am feeling like the
generation of the images and transfer down to brisa are working well,
but I would like to get your perspective on this before feeling too
comfortable.

Tom
--
+-----------------------------------------------------------------------------+
* Tom Yoksas                                             UCAR Unidata Program *
* (303) 497-8642 (last resort)                                  P.O. Box 3000 *
* address@hidden                                   Boulder, CO 80307 *
* Unidata WWW Service                             http://www.unidata.ucar.edu/*
+-----------------------------------------------------------------------------+