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

[LDM #KMT-181014]: ldmd.conf request entry clarification



> Full Name: Mike
> Email Address: address@hidden
> Organization: SJSU
> Package Version: LDM6.6.5
> Operating System: LINUX
> Hardware:
> Description of problem: Hi,
> I'm have the following entry in ldmd.conf, which requests gfs model data 
> every 12 hours:
> request        CONDUIT  
> "prod/gfs.*pgrb[^2].*(F000|F012|F024|F036|F048|F060|F072|F084|F096)" 
> 130.65.97.148
> 
> This works fine, except that I also get a trickle of F003, F006, F009 data as 
> well. This tells me that occasionally somewhere in the headers for these odd 
> hours the patterns F000 or F012, etc, show up and thus are matched.
> 
> My entry in pqact.conf is general, so it will decode or file anything.
> 
> 1) Does this make sense?
> 2) Any idea how I might clarify my request so I do not get any of the odd 
> hour data?
> 
> thanks,
> 
> -Mike
> 
> 
> 
Hi Mike, 

Chiz ~was our CONDUIT expert, but I will take a crack at what i see ~may be an 
issue.

Please see:

http://www.unidata.ucar.edu/data/conduit/ldm_idd/gfs_files.html

Note that for some grids the forecast hour goes from a 3 digit filed to a 2 
digit field. I also noticed that the files you get:

F003, F006, F009

would match the same as 12 hour forecast times:

F036||F060|F096...if they were using a 2 digit field F06 vs F006, and F06 would 
match F06 and F060.

One way to get around this would be a more strict reg exp to not allow F060 to 
match F06.

Again, this is a stab, but I think it may be the crux of the issue :)

Hope this helps,

Jeff
Jeff Weber
Unidata User Support
http://www.unidata.ucar.edu

Ticket Details
===================
Ticket ID: KMT-181014
Department: Support IDD
Priority: Normal
Status: Open