[daip] CLCAL failure

[Eric Greisen]

On 07/28/2016 03:03 AM, James M. Anderson wrote:
> Hello again daip,
>
> This time I am having a problem with CLCAL.  I have a number of SN
> tables, from 3 to 18, each of which only has valid calibration
> information for 1 of my 16 IFs.  I want to merge them together (to get
> the calibration for all IFs together) and apply them to get a new CL
> table.  All of the calibration information is listed in entries with
> subarray=1, and I want to apply that to all of the subarrays in my
> dataset.  So I am running CLCAL with
>
> opcode="CALP"
> opcode = "CALP"
> interpol="SELF"
> subarray = -32000
> snver = 3
> inver = 18
> gainver = 8
> gainuse = 9
>
> And then CLCAL consistently dies.
>
> CLCAL2: Task CLCAL  (release of 31DEC16) begins
> CLCAL2: Using interpolation mode SELF
> CLCAL2: Copied CL file from vol/cno/vers  1    2   8 to  1    2   9
> CLCAL2: CALSEL: Making room for new CL data by
> CLCAL2: Deleting CL table no.    9
> CLCAL2: Processing SN tables   3 through  18
> CLCAL2: Adjusting solutions to a common reference antenna =  3
> CLCAL2: CALREF: CANNOT RELATE ANTENNA  6 TO  3 IN SUBARRAY  1
> CLCAL2: CALREF: CANNOT RELATE ANTENNA  6 TO  3 IN SUBARRAY  1
> CLCAL2: Some solutions were not adjusted to the common
> CLCAL2: reference antenna. This may be because one or more
> CLCAL2: antennas were not used. Check your data before
> CLCAL2: proceeding.
> CLCAL2: SNMRG: Merging SN table
> CLCAL2: SNMRG: Write   19377 merged records from   69604 input records
> CLCAL2: SN2CL: Applying SN tables to CL table   8, writing CL table  9
> CLCAL2: TABIO: BAD LRNO= -2147483648 LIMIT           1
> CLCAL2: SN2CL: TABIO ERROR  2 APPLYING SN TO CL TABLE
> CLCAL2: Purports to die of UNNATURAL causes
> CLCAL2: kg122 31DEC16 TST: Cpu=      9.3  Real=     10  IO=      6900

In tomorrow's MNJ please find:

14871.  July 29, 2016             CLCAL                Eric
         Changed:
         SNAPP    To return a logical if any phases, delays, rates are
                  valid and non-zero
         SN2CL    Poor logic allowed the SN table record number to be
                  counted beyond 2 billion and so to warp to negative.
                  At that point it thought it could read more SN table
                  data and died badly.
         CLCAL    Fixed an array declaration to use MAXANT rather than
                  a hard-coded 90.  Changed call to SNAPP and set it up
                  to call CALREF only when phases actually need to be
                  re-referenced.  This will save alarming messages when
                  processing APCAL SN tables.
         Moved nowhere.

Eric Greisen