[daip] (no subject)

[Eric Greisen]

Katrina Exter writes:
 > >      The 2nd and 3rd above are an integer multiple of 2880 bytes as
 > > they must be, the first is not.  Did you use "binary" when doing the
 > > ftp?  That is absolutely required.
 > yup, all 3 datafiles were ftp'd at the same time with "bin" set.  only the 
 > C-band data will not fitld properly into aips, it's also the only one that 
 > is a different size on my disc compared to on the ftp disc.
 > 
 > actually I have now ftp'd the datafile over 3 times and it is now 
 > 33288 -rw-r--r--    1 kme      staff    34042178 Jul 17 19:51 AE146.020502.C
 > only on the 4th ftp (going for the 'resume' option) does ftp report that 
 > the file is all there, ie 
 >         Local:      34042178 bytes, dated Wed Jul 17 19:51:06 BST 2002.
 >         Remote:     34042178 bytes, dated Wed Jul 17 19:51:06 BST 2002.
 > 

    I just went downstairs and looked at the size of the file on disk
in our ftp site.  It is 33943680, an integer multiple of 2880 and
98498 bytes fewer than the number above.  The only way I know to get a
bigger size is to have the mode be ascii rather than binary so that
LFs are added to all apparent CRs.

 > and when I try to fitld this file I get still errors
 > FITLD1: TAPIO: RECORD LENGTH    578 INCONSISTENT WITH BLOCK SIZE  2880

as you should with a wrong file size.  I am still suprised that a
record gap seems to occur inside a byte stream, but maybe one of the
CR-LFs or a form feed or something gets taken as a gap...

 > FITLD1: WARNING:    505 unknown records skipped while reading tape

since the error occurs in the middle of the file.

 > >go prtuv
 > ERROR  3 ON UNIT TTDEV01         , GETTING HOST/SERVICE BY NAME
 >  ZMSGOP: ZTTOPN RETURNS ERROR =   1 FOR LUN =   6
 > PRTUV1: Finding the scaling parameters to set formats
 > PRTUV1: Full dynamic range of weights cannot be printed
 >  dakika    PRTUV(31DEC01)     50     22-JUL-2002  13:26:01    Page    1
 > TEST        .C BAND.   5  Vol= 1  User=   50  Channels=   1 to    1
 > Source= MULTI      RA   =  00 00  0.00   DEC  =  00 00  0.0 IF =  1
 > Freq=  4.885099862 GHz   Ncor=  4   No. vis=    254497   Sort order= TB
 > Weights have been multiplied by  0.0E+00
 > 
 > MULTI     4.885099862 TB      1 RR          1 LL          1 RL          1 
 > LR
 >  Vis #     IAT      Ant    Amp Phas Wt   Amp Phas Wt   Amp Phas Wt   Amp 
 > Phas Wt
 > 
 >      1  0/12:33:00 22-23    0.0 -79  0    0.0   0  0    0.0   0  0    0.0  
 > 41  0
 >      2  0/12:33:00 13-23    0.0 130  0    0.0   0  0    0.0   0  0    0.0 
 > 149  0
 >      3  0/12:33:00  2-23    0.0   0  0    0.0   0  0    0.0   0  0    0.0 
 > -94  0

    These are clearly all wrong...

 > my K-band file
 > 
 > >go prtuv
 > 
 > PRTUV1: Finding the scaling parameters to set formats
 >  dakika    PRTUV(31DEC01)     50     22-JUL-2002  13:11:22    Page    1
 > TEST        .K BAND.   1  Vol= 1  User=   50  Channels=   1 to    1
 > Source= 20178+38   RA   =  00 00  0.00   DEC  =  00 00  0.0 IF =  1
 > Freq= 22.485099364 GHz   Ncor=  4   No. vis=    245325   Sort order= TB
 > Weights have been multiplied by  1.0E-03
 > 
 > 2015+371 22.485099364 TB      1 RR          1 LL          1 RL          1 
 > LR
 >  Vis #     IAT      Ant    Amp Phas Wt   Amp Phas Wt   Amp Phas Wt   Amp 
 > Phas Wt
 > 
 >      1  0/12:21:40 20-23  0.009  -4 26  0.006  -8 26  0.006 162 26  
 > 0.004-123 26
 >      2  0/12:21:40 17-23  0.019 162 18  0.019 162 18  0.010  43 18  0.004 
 > -16 18
 >      3  0/12:21:40  2-23  0.026-159 22  0.000   0  0  0.005-118 22  0.000   
 > 0  0
 >      4  0/12:21:40 23-24  0.031 -36 23  0.000   0  0  0.000   0  0  0.004 
 > 170 23

This looks fine.  Try the ftp again.  Say binary not bin, do a dir
first to see sizes and try the copy.  It is striking that you had to
try several times to get it to think it had worked...

ERic