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

20001002: CONDUIT ETA 212 Data


Daryl,

The .status.* file for each product giving the injected size is the
last product sent for each forecast time. If your file has reached this
size, then you know it is complete. However, you cannot guarantee
that this .status part will arrive after every other product fore
the forecast time. The best solution would be a combination of checking that
the .status file has arrived and your local output file is no longer 
growing. If the data reception are problematic so that
you are not receiving complete files (such as the 9MB files shown below),
then you may not get the .status file either, so a check on a .timestamp
file (using find -newer for example) would let you determine when
you have received all the data you are going to.


Steve Chiswell
Unidata User Support


On Mon, 2 Oct 2000, Daryl Herzmann wrote:

> Hello,
>       I have been working on a project to use the ETA 212 data from our
> LDM CONDUIT feed as initialization data for the Workstation ETA.  I have
> run into a few problems/questions.  Here is my LDM pqact entry in which I
> save F00-F39 data from our CONDUIT feed.
>  
> NMC2    eta\.T(..)Z\.AWIP3D([0-3][0-9])\.tm00
>         FILE    -close  data/eta212/eta.T\1Z.AWIP3D\2.tm00
>  
> Question -> How can I tell when the files are completed?  
> 
> Since I want to feed the initialization data to the model runs as soon as
> possible, I want to know when the file is complete.  For example, this
> mornings files look like.
>  
>  9953099 Oct  2 05:01 eta.T06Z.AWIP3D30.tm00
> 14148121 Oct  2 05:01 eta.T06Z.AWIP3D15.tm00
> 14148121 Oct  2 05:01 eta.T06Z.AWIP3D12.tm00
> 14154077 Oct  2 05:01 eta.T06Z.AWIP3D06.tm00
>  9988893 Oct  2 05:01 eta.T06Z.AWIP3D36.tm00
>  9195475 Oct  2 05:00 eta.T06Z.AWIP3D39.tm00
> 13569393 Oct  2 04:55 eta.T06Z.AWIP3D00.tm00
> 14088449 Oct  2 04:55 eta.T06Z.AWIP3D21.tm00
> 14022813 Oct  2 04:54 eta.T06Z.AWIP3D24.tm00
>  9791997 Oct  2 04:52 eta.T06Z.AWIP3D27.tm00
> 14124259 Oct  2 04:49 eta.T06Z.AWIP3D18.tm00
>  9982927 Oct  2 04:45 eta.T06Z.AWIP3D33.tm00
> 13998949 Oct  2 04:43 eta.T06Z.AWIP3D03.tm00
> 14171985 Oct  2 04:42 eta.T06Z.AWIP3D09.tm00
> 
> but the 0z data looks like.
> 
> 14306879 Oct  1 23:11 eta.T00Z.AWIP3D21.tm00
> 14136829 Oct  1 23:09 eta.T00Z.AWIP3D33.tm00
> 14253179 Oct  1 23:07 eta.T00Z.AWIP3D36.tm00
> 14205447 Oct  1 23:07 eta.T00Z.AWIP3D30.tm00
> 14262141 Oct  1 23:07 eta.T00Z.AWIP3D06.tm00
> 14264927 Oct  1 23:07 eta.T00Z.AWIP3D03.tm00
> 14247225 Oct  1 23:04 eta.T00Z.AWIP3D12.tm00
> 13644441 Oct  1 23:04 eta.T00Z.AWIP3D00.tm00
> 14250011 Oct  1 23:04 eta.T00Z.AWIP3D39.tm00
> 14172651 Oct  1 23:04 eta.T00Z.AWIP3D09.tm00
> 14277049 Oct  1 23:04 eta.T00Z.AWIP3D24.tm00
> 14247027 Oct  1 23:04 eta.T00Z.AWIP3D15.tm00
> 14196323 Oct  1 23:04 eta.T00Z.AWIP3D27.tm00
> 14294941 Oct  1 22:54 eta.T00Z.AWIP3D18.tm00
>  
>  
>       Currently my scripts look at the F39 file and check to see if it
> is over 12 MB in size, if so, then I ship what ever (F00-F36) data I have
> over to the model runs.  The 0z run was fine using this method, but the 6z
> never ran since the F39 file was only ~9 MB in size.
> 
>       What is puzzling is when I look on the NCEP ftp server at these
> same files, they are all about 13 MB in size. 
>       Anybody have any ideas?  Am I missing something?
>  
>  
> Thanks much,
>       Daryl
> 
>