20010215: 7.704 upgrade and NEXRAD ADDE configuration

>From: Gilbert Sebenste <address@hidden>
>Organization: NIU
>Keywords: 200102151852.f1FIq6L29876 McIDAS-X 7.704 NEXRAD NOWrad

Gilbert.

>I fired up McIDAS today and noticed that it was version 7.704. But the
>latest I had on there was 7.702? I noticed a few logins from you over the
>past week...did you put 7.704 on there?

Yes.

>Or is this Nyquil haze I am in making me loopy?

Here is the story.  I have found a problem on RedHat 6.2-7.0 Linux
systems where the Linux system is serving as the remote host for
sounding data:  it does not work reliably.  The problem did not/does
not exist on RedHat 5.2 (verified).  I had a hunch that the problem may
be related to Linux sites running a SMP kernel,and I remembered that
you were running a 6.2 SMP kernel, so I wanted to quickly jump onto
weather and conduct a few tests locally.

Your machine exhibits the same problems (bummer).  I have been beating
my head on this problem for about a week now, and it is driving me
crazy.  All of the tests that I have run seem to indicate that the problem
is in the OS, not in McIDAS code.  If this is true, it will be _very_
hard to find and fix.

While I was on weather, I took a quick look at how you had reorganized
your NEXRAD Level III product filing.  It is much better than before!
I then looked to see if you had setup ADDE access to those data, and I
saw that you had not gotten around to it yet.  Since this kind of stuff
only takes me 5 minutes or so, I went ahead and setup access to the
datasets RTNEXRAD and RTNOWRAD.  It was then that I noticed that you
were at addendum 7.702.  In either 7.703 or 7.704 I released a fix for
situations were there could be an arm of the directory NEXRAD hierarchy
in which there would be a directory with no data files.  This situation
would cause:

IMGLIST RTNEXRAD/xxx ID=LIST     (fill in xxx with N0R, NCR, etc.)

to fail.  So, I upgraded weather to the latest released addendum, 7.704.

Now, you can serve NEXRAD and NOWrad (tm) data off of your own
machine.  To see what I did, please take a look at the files
NIUNEXR.CFG and NIUNOWR.CFG ~mcidas/workdata.  This demonstrates how to
setup a configuration file for those data.  Also, do a

DSSERVE LIST RTNEXRAD

and 

DSSERVE LIST RTNOWRAD

To see how the datasets were defined.

I tried accessing your data from my machine at Unidata and verified
that things work well (albeit a little slow given the 'rad' processing
that seems to consume a lot of CPU).

I meant to blast you a note about the changes I made on Wednesday, but
I ran out of time and took yesterday off to go skiing.

Hope you don't mind the mods...

Tom