[McIDAS #JTG-722928]: installation error

Hi Justin,

I drove the three emails you sent to me personally into our inquiry tracking
system...

re: installation of ADDE remote server

> If you have the opportunity, would you also be able to help me setup the
> adde server?
> 
> My next question is do I really need to setup this server? Is this
> basically to allow others access to my data?

If you will want multiple McIDAS-X users on your Linux machine or
IDV users on a variety of machines to use data that you have (and
can be served by ADDE), it is simplest to best/easiest to setup
the remote ADDE server.  If you do not have local data that you
want to serve, or if you will only be accessing the data from a
single account, then you do not need the remote ADDE server setup.
This is why I asked you if you have local data that you want to
serve.

re:
> Do I need to setup a scheduler to update the gui for new images or
> does it automatically poll for data from other adde servers?

McIDAS does not have auto update capability built in like GEMPAK,
the IDV or AWIPS-II.  If you want to update displays periodically,
then you will need to run routines at intervals to reload existing
displays with new data.  This can be done from the McIDAS scheduler
or with a little more work with cron.

By the way, the way that McIDAS has (or had) been used operationally
at NWS sites (e.g., Aviation Weather Center) was in the way that you
are asking about: a _large_ McIDAS-X session (meaning hundreds of
display frames) would be kept running, and the sets of frames would
be dedicated to display of particular things (e.g., loops of VIS
imagery, meteorograms, what have you).  Users of this system would
take advantage of being able to map keyboard function key sequences
(e.g., F1, CTRL-F5, ALT-Fn) to jumping to a particular display
sequence (e.g., the IR loop of GOES-East IR images in frames 123-200,
etc.) or to reload a particular kind of display.

re:
> If you feel I would still need the mcadde server, just let me know and
> I can give you access to my server again.

Whether or not you can benefit from running the ADDE remote server depends
entirely on what you want to accomplish.  Again, the simplest thing to
decide on is how many different users would you like to benefit from
access to datasets held on the machine on which McIDAS is installed.
If the answer is a single user, then the answer is simple: no.  If
you were trying to support a classroom of students, the answer is also
simple: yes.

re: ADDE datasets
> I was trying to retrieve data from satepsanone.nesdis.noaa.gov
> <http://satepsanone.nesdis.noaa.gov>.
> 
> I used the following command to add the data location.
> 
> dataloc.k ADD MTSAT SATEPSANONE.NESDIS.NOAA.GOV
> <http://SATEPSANONE.NESDIS.NOAA.GOV>
> 
> When I typed in "dsinfo.k IMAGE MTSAT" it says no datasets were found,
> but I see them available on the website. Am I entering something incorrect?

ADDE dataset names are composed of two pieces: group and descriptor.
An example that you will find on all ADDE server installations that
we maintain is RTIMAGES/GE-IR:

RTIMAGES -> group name
GE-IR    -> descriptor name

One users DATALOC to tell his/her McIDAS session where to find all
datasets that share the group name:

DATALOC ADD RTIMAGES atm.ucar.edu

In the case of ATEPSANONE.NESDIS.NOAA.GOV, the group name of the
publicly available dataset is PUB:

DATALOC ADD PUB SATEPSANONE.NESDIS.NOAA.GOV

The datasets hosted in the PUB group can then be seen using DSINFO:

DSINFO ALL PUB

You will see that PUB contains datasets of type IMAGE and POINT.
In McIDAS, there are four classes of data:

IMAGE
GRID
POINT
TEXT

re:
> Is wxdata.db.erau.edu still available for adde use?
> I tried to access it and wasn't able to get through.

Hmm... I have not received any notification that wxdata.db.erau.edu
has been decommissioned.  I will need to check with the folks there
to see what may be going on.

Here is something you should know about ADDE datasets:

DSSERVE is used to create a dataset definition.  The definition specifies
the group and descriptor names, the type of data, where on disk to find
the data, the extent of the dataset, and optionally a description of
the dataset.  It does not, however, guarantee that the dataset has
any content.  As one becomes familiar with McIDAS, it will become
clear that lots of ADDE administrators define ADDE datasets, and some
of those datasets are empty.  Here is an example from the PUB group
at NESDIS:

IMGLIST PUB/LI-EC
Image file directory listing for:PUB/LI-EC
IMGLIST: No images satisfy the selection criteria

IMGLIST: done
IMGLIST failed, RC=2

The PUB/LI-EC dataset is defined, but it has no content.  This kind of
situation is both good and bad:  good since it allows the content of
datasets to change dynamically; bad since the end-user may be confused
by the result of there not being any data.
 

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: JTG-722928
Department: Support McIDAS
Priority: Normal
Status: Closed