[daip] question on running VBGLU

[Eric Greisen]

Shen Z.-Q. writes:
 > I'm trying to run VBGLU to combine two data which are only different in frequency
 > due to two different correlator passes, but the output message said that these CL
 > tables were not matched. I'm wondering whether or not I can ignore this by simply
 > running INDXR to generate a CL table; usually I found it is not necessary to re-run 
 > INDXR after VBGLU.
 >  
 > Zhiqiang
 >  
 > VBGLU1: File BS158B      .MATCH .   1 on disk   2 covers
 > VBGLU1: time range   0/  5 49 29 -   0/ 12 50 54 and has   399355 vis
 > VBGLU1: File BS158B_PASS3.MSORT .   1 on disk   2 covers
 > VBGLU1: time range   0/  5 49 29 -   0/ 12 50 54 and has   399498 vis
 > VBGLU1: Wrote     1897 visibilities with data merged in from 1 files
 > VBGLU1: Wrote   398478 visibilities with data merged in from 2 files

    This says that the 2 files did not exactly match in the times of
samples and the like.  INDXR will be required to get a new NX table in
any case.

 > VBGLU1: GLUTAB TABLE: AT MISSING REQUIRED COLUMNS, FILE 1
 > VBGLU1: IM table copied:   1811 rows containing data from 2 input files
 > VBGLU1: CL table copied:   7815 rows containing data from 1 input files
 > VBGLU1: CL table copied:  19430 rows containing data from 2 input files
 > VBGLU1:     THIS SUGGESTS THAT THESE CL TABLES WERE NOT WELL MATCHED

    This is a significant mismatch of the times in the CL table.

 > VBGLU1: MC table copied:   4146 rows containing data from 2 input files
 > VBGLU1: CTINI: 'TIME INT' COLUMN NOT FOUND
 > VBGLU1: CTINI: 'TIME INT' COLUMN NOT FOUND
 > VBGLU1: Copied CT file from vol/cno/vers  2    9   1 to  2   10   1
 > VBGLU1: TABINI: KEY COUNTS     32     51 OR COLS     12     11 DON'T MATCH
 > VBGLU1: CTINI: ERROR   1 FROM TABINI OPCODE = WRIT
 > VBGLU1: CTINI: ERROR   1 INITIALIZING CALC TABLE FOR WRIT
 > VBGLU1: ZFIO: LUN = 28 NOT OPEN IN FTAB
 > VBGLU1: TABIO: I/O ERROR    1 ON READ PHYS REC         7
 > VBGLU1: TABKEY: ERROR   3 READING TABLE FILE KEYWORDS
 > VBGLU1: TABKEY: ERROR OCCURRED DURING TABLE KEYWORD ACCESS, OPCODE=WRIT
 > VBGLU1: UPDKEY: ERROR   3 UPDATING TABLE KEYWORDS
 > VBGLU1: GLUTAB: ERROR   3 FROM UPDKEY OPCODE = UPDT

     The CT file underwent a format change - it would appear that
these data were read in by FITLD some time ago before the format
change.

 > VBGLU1: Run INDXR to create a new index (NX) table for the output file
 > VBGLU1: Appears to have ended successfully 

You can run INDXR to create a new CL table which should have values in
it based on the IM and MC tables.  It would be wise to check the CL
tables of your inputs to VBMRG with the CL table output by INDXR to
make sure that they are similar in their atmospheric and other group
delays and clock offsets.


Eric Greisen