[daip] FILLM message

[Eric Greisen]

cchandle at nrao.edu wrote:
> Hi Eric,
> 
> I'm trying to load a VLA dataset /home/pasilla/cjc/linux/TP026_1 and
> I'm expecting there to be data up to 17:11 UT (according to the scan
> listing the archive and the observe file), but FILLM stops at around
> 13:12 UT with a message 
> 
> FILLM1  FLDKIO: LBYTES=   10240 BUT IERR = 0
> 
> What does this mean?  Is there a problem with the archive file?  How
> can I find out if the other 4 hours of data are even in the archive
> file?

This error arises either when the IO gives a record that is not 2048*n 
bytes (not your case) or that the pointers within the data say that the 
record should be longer than the record read.  Since the disk file is 
just a byte stream, the only way that the read can return less than the 
requested number of bytes is at the end of data.  This means that the 
distributed file has been truncated.  Note - if you used ftp - did you 
remember to say binary when copying these data?

Cheers,

Eric Greisen