[LDM #MNK-989154]: comparing, contrasting two ldm servers

Hi SJ,

Quick update on some of our activities yesterday afternoon on measwx...

Three of us (Michael, Steve and I) were looking at the segmentation
log files on measwx and using the debugger to examine the core files.
We discovered at least one problem that we fixed:

dcnldn   - a character buffer into which an 'sprintf' was writing
           information was too small to accommodate the fully
           qualified name of the '.gem' files on measwx.

           After increasing the buffer size; rebuilding the dcnldn
           6.6.0 binary from source; and "installing" it (i.e.,
           copying the newly built binary to the ~ldm/decoders
           directory), it stopped dumping core

dcncprof - we (Michael actually) have not yet figure out what is
           going on with this decoder

Another BIG thing we discovered yesterday is that the LDM on measwx
was configured to run redundant invocations of GEMPAK decoders.
The effect of this was that two invocations of the same decoder
would try to write to the same output '.gem' file.  This was
causing most of the segmentation violations log messages you
were seeing (we think).

As we left things yesterday evening, the pattern-action file actions
for dcncprof were commented out, and the LDM was left running.

More as we figure out what is going on with dcncprof...

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: MNK-989154
Department: Support LDM
Priority: Normal
Status: Closed