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

[LDM #YFA-661250]: LDM product time question

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: [LDM #YFA-661250]: LDM product time question
  • Date: Mon, 06 Jul 2009 09:09:48 -0600
Art,

> If LDM B feeds from LDM A, and LDM C feeds from LDM A and B (as primary
> and secondary), should I see products on LDM C with identical product
> times to those coming in on LDM B?

If there's only one source for the data-products (i.e., one ingester setting 
the product creation-time) and the XDR layer and C runtime functions (e.g., 
"strftime()", "sprintf()") are working correctly, then the string 
representation of the product creation-time should be the same on all systems.

> I've done a test case where I see
> some products on C that have slightly different times than those on B...
> is this an allowable occurrence or a problem?  E.g. (from pqutil -v -w):
> 
> LDM B:
> Jul 03 00:00:02 pqutil INFO:     2645 20090703000000.639     HDS 596019
> SDUS84 KEPZ 022356 /pDPAHDX
> 
> LDM C:
> Jul 03 00:00:00 pqutil INFO:     2645 20090703000000.885     HDS 596019
> SDUS84 KEPZ 022356 /pDPAHDX

This is a puzzle.

The HDS feed comes from NOAAPORT and there are several ingesters for that 
data-stream.  It could be that the, otherwise, identical data-products 
originated at different sites.

> I'm investigating this because I'm trying to use this kind of
> configuration for redundancy but my downstream LDM's (which would be LDM D
> feeding off LDM C) is losing random data (but only sometimes) and I'm
> wondering if this time variance could cause a problem.

The time variance shouldn't cause a loss of data because the MD5 checksum 
signatures of the data products should be identical, and it's the signatures 
that the LDM system uses to identify data-products.

> Thanks...
> 
> Art
> 
> Arthur A. Person
> Research Assistant, System Administrator
> Penn State Department of Meteorology
> email:  address@hidden, phone:  814-863-1563


Regards,
Steve Emmerson

Ticket Details
===================
Ticket ID: YFA-661250
Department: Support LDM
Priority: Normal
Status: Closed