[Datastream #XOM-655184]: Radar question

Hi Kevin,

I just started looking into the following:

re:
> Here is another thing we found today.
> See attached file ind_20100811_1542.n0q

> This file appeared on both of my test machines.
> When we do an IMGLIST on this dataset:
> IMGLIST WSR/BREF1.ALL ID=IND, this 1542 file is skipped over.

In reviewing the code in nexradir.cp, I am reminded of my thinking
that bad data files are treated more-or-less as if they do not
exist.

re:
> When I try to do IMGDISP
> IMGDISP WSR/BREF1 ID=IND ALL=1 20 STA=IND
> then it fails to plot anything after this 1542 file.

Hmm...  This is very strange given that the code in nexraget is
the same as in nexradir: if there is a failure in extracting
image directory information, the next image in the list of images
should be processed.  I can not see how/why other images in the
dataset temporarily _after_ the bad would not be processed.

Any chance you have a set of images in which one or more are bad
and the others are good so I can use them as a test dataset?

By the way, one strange thing that I see in the image you provided
is its date/time:

<output from trce file>
 ...
in read_proddesc
GetNexrTime:: product vs_date/vs_time:  14833/56538
GetNexrTime:: date: 2010223 time: 154218 tsec: 1281541338
...

This says that the image was from February 23, not from August 11.

re:
> DEV=CCC shows "Read of server has failed. Wanted 256 Did -1"
> a TRACE=1 yields on the test machine:
> 
> ReadNexrInfo:: file size, name: 12452,
> /data/xcd/wsr/IND/N0Q/ind_20100811_1542.n0q
> GetNexrDirs:: Error unbzip2ing buffer, rc = -7
> GetNexrDirs:: error opening NEXRAD file:
> /data/xcd/wsr/IND/N0Q/ind_20100811_1542.n0q
> nexraget:: error reading area directory:
> /data/xcd/wsr/IND/N0Q/ind_20100811_1542.n0q  ERR:
> nexraget:: binding off transaction

I do not have a dataset available that has a bad image in a sequence of
good ones.  If you can provide this, it will help me figure out what is going
on.

re:
> I'm assuming the attached file is bad for some reason, but shouldn't the
> IMGDISP command be able to run to completion.

As I read the code, it should run to completion.

re:
> As it stands now, there
> really isn't any indication that anything has gone wrong unless you do
> the DEV=CCC.

Something is very weird here.  Hopefully, you can give me a set of files
that I can use to replicate the problem and fix the cause.

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: XOM-655184
Department: Support McIDAS
Priority: Emergency
Status: Closed