[daip] New Ticket - [!MNC-216660]: Tables corrupted, probably by REBYTE

[Craig Walker]

New Ticket: Tables corrupted, probably by REBYTE

I am working with files converted from PPC MAC to Intel MAC with REBYTE.  They were subject to the real/imaginary swap so I am trying to run SWAPR.  I have found that the BP tables in the REBYTE output are corrupted to the point were some of the programs that read them die.  SWAPR is one of those routines.  UVCOP is another.  See messages below:

SWAPR1: Task SWAPR  (release of 31DEC11) begins
AIPS 1: Resumes
>SWAPR1: Doing no flagging this time
SWAPR1: Create BW088A      .MULTI .   1 (UV)  on disk  5  cno    1
SWAPR1: Copied AT file from vol/cno/vers  5  195   1 to  5    1   1
SWAPR1: Copied CT file from vol/cno/vers  5  195   1 to  5    1   1
SWAPR1: Copied OB file from vol/cno/vers  5  195   1 to  5    1   1
SWAPR1: Updating tables for IF/FREQID/channel selection
SWAPR1: Copied AN file from vol/cno/vers  5  195   1 to  5    1   1
SWAPR1: TABIO: BAD LRNO =       181 LIMIT       180
SWAPR1: TABBP: TABIO ERROR  2
SWAPR1: BPSEL: ERROR   2 READ INPUT TABLE
SWAPR1: Appears to have ended successfully
SWAPR1: localhos 31DEC11 TST: Cpu=     36.0  Real=     63  IO=      1352

When I look at the BP table with PRTUV, I only see the first row, and the data are constant which is not right.  Looking at the binary files crudely (with less) indicated that maybe only the first row got copied.  BP tables on the same files transferred through FITS are ok.

Following this up, I also found that there was a change in the GC tables.  Values that appear in PRTAB as INDE in the tables transferred through FITS have an actual number in the tables transferred through REBYTE.  In this case, that might really make a difference as it may be helping distinguish between a gain vs elevation table and polynomial coefficients.

I have looked for other tables with problems but have not found any yet, but the search has not been exhaustive.  So far I am concluding that I need to transfer my MULTI and TASAV files by FITS.  I may keep the REBYTE/SWAPR files for the single source files - their tables seem to be ok as does the data.

I realize that, with Eric out of town, this may not be looked at for a while and my memory may fade.  It my memory needs jogging to find examples, the offending files are with the data for BW088A on lacie3 (an external disk).



Ticket Details
===================
Ticket ID: MNC-216660
Department: AIPS Data Processing
Priority: Default
Status: Open
Link:  https://help.nrao.edu/staff/index.php?_m=tickets&_a=viewticket&ticketid=705