I am not sure that the following is the reason, but it is likely. When you do a PUT from a z/OS system (client) to another system (z/OS or not), the z/OS client ftp does a SENDSITE command which relays the DCB information of the file being sent to the receiving z/OS system. Other ftp servers tend to flag this and return a command not understood type error.
As to why IBM has those as the ftp defaults? I don't have any real notion. My uneducated guess is that since the default transfer method is ASCII (text), and when a file comes from a non-z/OS system, all text records are general variable in length that IBM decided that VB made sense. I have no idea why they thought that 256 was a good number, unless it is because that it the maximum number of bytes which can be copied with a single MVC instruction. 6233 is a real stumper. I would have used 0 so that SDB would find the best fit for the receiving device. On Mon, Oct 14, 2013 at 10:09 AM, Mike Wojtukiewicz <[email protected]>wrote: > I've been using MVS FTP for the longest time. I've encountered this > problem but it never presented much of a problem until I had a problem with > someone. Our company accepts SMF data that's been compressed via TRSMAIN > (LRECL=1024,RECFM=FB). I wrote SMS routines such that if a user signs in > using a certain RACF userid it gets assigned a DATACLAS such that it > assigns the correct DCB. I also have the ftp parm file (/etc/ftp.parm) set > up to default to LRECL=80 RECFM=FB BLKSIZE=27920 seeing as how MOST ftps > I''ve seen is moving XMIT files between MVS and other platforms. My > question is this. Why when I use a DOS Window and client to sign on does it > assign the correct DATACLAS but use the default LRECL=80 but when I go from > MVS to MVS it works correctly. As an aside..I've NEVER understood why IBM > sends LRECL 256 RECFM VB BLKSIZE 6233 as the defaults. > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN > -- This is clearly another case of too many mad scientists, and not enough hunchbacks. Maranatha! <>< John McKown ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
