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

[GEMPAK #ZWT-869984]: SFCFIL errors



> Thanks for the speedy reply--I didn't even expect to get a support ticket
> for this, as I thought I sent this over gembud.

Daniel,

Your question was bounced to support since it included a very large attachment.
Given that emails have to be replicated by the number of recipients on the list,
our systems do not allow large attachments to be sent via our lists.
In the future, you may just want to post your large files on the web and 
reference
the email list recipients to that location. I suspect that many sites that 
receive these
emails may also limit the size of messages they allow to be sent to them.

> 
> I didn't even catch those errors in the input file, we're filtering out
> redundant obs and there was a bug in writing out the lats and lons that I
> didn't catch. Thanks.
> 
> I'm still getting errors when I try running sfcfil with the settings you
> specified. They're the same errors I was getting before.

See if you already have a test.gem file in that directory and remove it. If you
still have trouble, send me the output of your sfcfil command history.
> 
> [FL -3]  Cannot create file test.gem.
> [DM -1]  File test.gem cannot be created.
> [SF -1]  File test.gem could not be created.
> 
> I fixed the input file, as well as the bufr.pack file I was using to fix
> the mistakes I had, but still getting the same errors. Permissions should
> not be an issue, because the entire directory has read/write access as any
> user.
> 
> Now, I'll be using sfcfil on lots of files... do I need the number of
> stations in TIMSTN to be exact? Not all the files have 8672 observatins in
> them. Can I just put a large arbitrary number into that parameter that
> would  never be exceeded, or does it need to be exact?

No. The maximum number of entries in a ship file is 1/29999 and you don't
have to be exact. Note that you could also use a standard surface file if you 
created a 
station table separate from your SFEDIT input file, which may make sense if you 
have the same stations
at many times (and you have less than 29,700 stations and 300 times). The 
default number of
headers in a surface file in GEMPAK is set at compile time.

Steve Chiswell
Unidata User Support

> 
> Thanks,
> 
> Dan
> 
> 
> >
> > Daniel,
> >
> > In sfcfil, use:
> >  SFOUTF   = test.gem
> >  SFPRMF   = bufr.pack
> >  STNFIL   =
> >  SHIPFL   = y
> >  TIMSTN   = 1/8762
> >  SFFSRC   = UNKN
> >  GEMPAK-SFCFIL>
> >
> > Your sffsrc must be one of the recognized types, not the name of a data
> > file as you had previously.
> >
> > For your data set, the packing file can be:
> > SELV         0.      10000.      1.
> > TMPC       -50.         50.     .1
> > DWPC       -50.         50.     .1
> > SPED         0.        60.      1.
> > DRCT         0.        360.     10.
> > PRES       600.       1080.     .1
> >
> > Don;t store the SLAT and SLON as a parameter here.
> >
> > In your input SFEDIT file, you have a few small corrections:
> > The PARM line had comma instead of a ";". Use:
> >  PARM = SLAT;SLON;SELV;TMPC;DWPC;SPED;DRCT;PRES
> >
> >
> > Change your times from 4 digit years to 2 digit years (I did a quick
> > global replace of 20060713 with 060713).
> >
> > In your SLAT and SLON data currently, you have -104.59 as SLON for every
> > station, but
> > have what looks like SLON specified for the SLAT column, so you just need
> > to fix that.
> >
> > I ran SFEDIT as:
> >  SFEFIL   = 2006071313_rtmagemob.dat
> >  SFFILE   = test.gem
> >  GEMPAK-SFEDIT>r
> >
> >  SFEDIT PARAMETERS:
> >
> >  Edit file:           2006071313_rtmagemob.dat
> >  Output surface file: test.gem
> >
> >  Parameters to be added to file:
> >
> >  SLAT SLON SELV TMPC DWPC SPED DRCT PRES
> > Enter <cr> to accept parameters or type EXIT:
> >  Parameters requested: SFEFIL,SFFILE.
> >  GEMPAK-SFEDIT>
> >
> > Steve Chiswell
> > Unidata User Support
> >
> >
> >
> >> Hello,
> >>
> >> I'm trying to put together a file that will be used to plot station
> >> observations in GEMPAK. The station observations are from the mesonet,
> >> so
> >> a lot of the stations are not built into the GEMPAK station location
> >> tables--I know I must create a ship file for this, but I've been getting
> >> loads of errors when I try running SFCFIL. My settings for SFCFIL are:
> >>
> >> SFOUTF    Output surface file               test.gem
> >> SFPRMF    Surface parameter packing file    ./bufr.pack
> >> STNFIL    Station information file
> >> SHIPFL    Ship data file flag               YES
> >> TIMSTN    Times/additional stations         1/8762
> >> SFFSRC    Surface file source               2006071313_rtmagemob.dat
> >> Parameters requested: SFOUTF,SFPRMF,STNFIL,SHIPFL,TIMSTN,SFFSRC.
> >>
> >> And the error codes from the run:
> >>
> >> [FL -3]  Cannot create file test.gem.
> >> [DM -1]  File test.gem cannot be created.
> >> [SF -1]  File test.gem could not be created.
> >>
> >> Now, I'm I correct in thinking that TIMSTN is the number of times and
> >> stations in the file? In my case, I believe that's 1/8762, because I
> >> have
> >> one reporting time (20060713/1300) and 8762 observations (1 ob per
> >> station). I've tried all sorts of combinations for TIMSTN... and I'm
> >> just
> >> not getting this to work. I've attached the bufr.pack and my surface
> >> source file to this email. I'd appreciate any help you all can provide.
> >>
> >> Thanks,
> >>
> >> Dan Tyndall
> >>
> >>
> >
> >
> > Ticket Details
> > ===================
> > Ticket ID: ZWT-869984
> > Department: Support GEMPAK
> > Priority: Normal
> > Status: Closed
> >
> >
> 
> 
> 


Ticket Details
===================
Ticket ID: ZWT-869984
Department: Support GEMPAK
Priority: Normal
Status: Closed