[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[IDD #GOK-818269]: Help -- NEXRAD3 feed no longer working, along with some GOES imagery from UNIWISC



Hi Marcus,

re:
> Yes, there is definitely packet shaping going on, just as before;
> we are just trying to find out how bad it is and
> what limits we are working with.

OK.

re:
> We want to make sure that the problems are solely due to the bottleneck
> before taking mitigating action on this side.

It looks like you have already done the easiest thing to mitigate somewhat
the effect of packet shaping: splitting your feed requests into separate
ldmd.conf REQUESTs.  The next step (other than getting your network folks
to ease up on the packet shaping) would be to determine what subset of each
feed you want/need.

re:
> It looks like we are getting HDS data, but it is not getting processed;
> so, could it be that we are always getting incomplete
> files, 

The way the LDM works is that individual products are sent and they are either
received or not.  There should never be an instance of a part of a product being
received.

re:
> or is it that we have something incorrect in the pqact...?

This will take digging on our side.  The better question is what do your log
files say?  For instance, are you seeing any indication of problems in
the LDM log file (~ldm/etc/ldmd.conf), and/or are you seeing any indications
of problems in the log file(s) for the GEMPAK decoders you are running
(~ldm/data/gempak/logs/*.log)?

re:
> I also noticed that the rtstats output pages think we are using ldm 6.8.1, 
> but the main list of hosts
> show us using 6.9.8 ... probably doesn't affect anything, but an fyi.

Thanks for the FYI.  Your surmise is correct, however: you would see no 
difference
running LDM 6.9.8 than with 6.8.1.

re:
> We are hopeful that we can get a consistent 3Mbit/s or more soon...which
> should be a small improvement.

3 Mbps should be enough to get the data you are currently requesting.

Ideas:

- think about what NEXRAD3 data you do _not_ want/need

- think about what model output you can do without

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: GOK-818269
Department: Support IDD
Priority: Normal
Status: Closed