IBM Mainframe Forum

Hi,

I am copying a FB file to a VB file. In the first method, I used “alike” command in ISPF. The record length of the file is 1000 and the block size is 27900. I ftp’ed the file to Unix server and following is the ftp log.

ftp> get 'tvdt.n2.vrol.rua5bc02.vb2' testrfile2
200 OK, Ready
150-Dataset opened; data connection starting.
Data transfer Type is ASCII. Structure is File. Mode is Stream.
Dataset name: TVDT.N2.VROL.RUA5BC02.VB2 Dataset attributes: Dsorg=PS
150 Recfm=VB Lrecl=1000 Blksize=27900 Volser=TSS286 Unit=3390
226-Transfer complete. 10621 bytes sent in 0.05 seconds (212420 bytes/s)
Dataset name: TVDT.N2.VROL.RUA5BC02.VB2 User=Z014721
Data bytes read: 10229.
226 Disk tracks read: 1.
local: testrfile2 remote: 'tvdt.n2.vrol.rua5bc02.vb2'
10621 bytes received in 0.27 seconds (38.73 Kbytes/s)

We can see that only 10621 bytes transferred.

In the other method, I used sort jcl to create a VB dataset from FB dataset. Following is the jcl.

//STEP01 EXEC SORTD5
//SYSPRINT DD SYSOUT=*
//SYSOUT DD SYSOUT=*
//SORTIN DD DSN=TVDT.N2.VROL.RUA5BC02,DISP=SHR
//SORTOUT DD DSN=TVDT.N2.VROL.RUA5BC02.VBUNIX,
// DISP=(NEW,CATLG,DELETE),UNIT=SYSDA,
// SPACE=(CYL,(10,5),RLSE),
// DCB=(RECFM=VB,BLKSIZE=27900,LRECL=1000)
//SYSIN DD *
SORT FIELDS=COPY
OUTFIL FTOV
/*

Again I ftp’ed the output file of the above step to unix server and following is the ftp log.

ftp> get 'tvdt.n2.vrol.rua5bc02.vbunix' testrfile
200 OK, Ready
150-Dataset opened; data connection starting.
Data transfer Type is ASCII. Structure is File. Mode is Stream.
Dataset name: TVDT.N2.VROL.RUA5BC02.VBUNIX Dataset attributes:
150 Dsorg=PS Recfm=VB Lrecl=1000 Blksize=27900 Volser=TSS410 Unit=3390
226-Transfer complete. 195608 bytes sent in 0.13 seconds (1504676 bytes/s)
Dataset name: TVDT.N2.VROL.RUA5BC02.VBUNIX User=Z014721
Data bytes read: 195216.
226 Disk tracks read: 4.
local: testrfile remote: 'tvdt.n2.vrol.rua5bc02.vbunix'
195608 bytes received in 0.35 seconds (545.80 Kbytes/s)

But here in this case, 195608 bytes have been transferred. When I open the file in Unix server, there are spaces between each record.
Can you please let me know what could have caused this difference?

Thanks

Hello,

I suppose one major difference might be that you not have used the same file for both tests?

there are spaces between each record.

Does this mean there is a blank record between each data record? If yes, this is often caused by the transfer terminating a line due to length and the first character of the "next record" is a new-line.

I figured out the issue with this. Actually i logged onto to mainframe from unix server manually and tried doing a "get" from there. Today i did it in the other way and there are no spaces between records.

Thanks.

Thanks for letting us know.

d