20030821: [awipsldm] LDM Crashing... (cont.)

>From: "Brad Schick" <address@hidden>
>Organization: National Weather Service - Goodland Kansas
>Keywords: 200308202105.h7KL51Ld020001 LDM

Brad,

>Our problem has been solved.  Seems that due to bandwidth restraints,
>some products that we were requesting was causing our connection to
>time-out due to the amount of time it would take to receive the data,
>thus eventually causing LDM to crash.

From our point of view this is still a big problem that needs to be
investigated.  The LDM is designed to work in adverse environments the
least of which is a slow connection.  At worst the receiving LDM
process should eventually give up and send a reclass to the upstream
feed site.

>After removing the product from the ldmd.conf file, its been running fine.

It would really help us to know what version of the LDM you are using,
what the contents of your ldmd.conf file are, what operating system you
are running on, and what compilers you used to build the LDM if you
built from source.  Since we have not had any feedback from our
university IDD users that indicates a problem like the one you are
seeing, I am particularly interested in finding out exactly what is
causing your problem.

If you are using an older version of the LDM (e.g., LDM-5.x) I would
strongly encourage you to upgrade to the latest release, LDM-6.0.14, as
it is much more efficient in moving data and more robust in the face of
network dropouts.  Even though you did not initiate the inquiry about
the problems you are having with your LDM installation, it is in our
best interest to get to the bottom of your problem especially since the
NWS is adopting use of the LDM for movement of NEXRAD Level II data.

>A little FYI for other Weather Service that may experience the same
>problem.  Any of the following in you ldmd.conf file may cause this
>problem.
>
>WMO, PPS, NEC, ANY

I am not sure what NEC is, but the other feed types are standard ones
used by the entirety of the Unidata IDD community to move hundreds of
gigabytes of data in near real time.

>Thanks for your assistance,

Please help us by working with us to get to the real cause of your
problem!

Thanks in advance for your help...

Tom