I found the most simple one we've got. Much easier for you to unravel. Some notes
- PIPEs gives a INMR123 REXX stage to block pipe records into NETDATA format . We modified this a little, and named that NETDATA+ REXX. - The sample is SMFMVS EXEC. It reads a VM:OPER logfile to extract all LOGON/LOGOFF records and send those to MVS in an SMF-like format. The reason: initially we were asked to send RACF"s SMF log to MVS where they would extract LOGON/LOGOFF events to see who was logged on at a system in which time period. As RACF didn't collect LOGOFFs in its SMF file, we scanned the VM operator log and mimicked a bit an SMF file. - At line 201 we store some PIPE stages in a REXX variable: in testmode, the stages are simple: create a CMS file, otherwise the stages create MVS JCL and call NETDATA+ to format the SMF-like records we created from the operator log. - At line 266 we read the operator log, and at line 288 the pipe stages created in the previous step are inserted in the pipeline. - At line 370 the data punched by the pipe are sent to MVS via RSCS. 2011/3/8 Shumate, Scott <scshum...@bbandt.com> > That would be great. Can I take a look at it? > > > Thanks > Scott > > > ------------------------------ > *From:* The IBM z/VM Operating System [mailto:IBMVM@LISTSERV.UARK.EDU] *On > Behalf Of *Kris Buelens > *Sent:* Tuesday, March 08, 2011 3:47 PM > *To:* IBMVM@LISTSERV.UARK.EDU > *Subject:* Re: Sending files to JES > > On our VM systems, we had some processes that sent a CMS file embedded in > an MVS job. The MVS job started TSO in batch to receive the file. At last > that single job could carry multiple CMS files. As a result, the REXX code > became less than easy to read for beginners. I can dig that up, on > condition that the receiver is prepared to unravel that a bit (I'm convinced > I already sent that once to someone). > > 2011/3/8 David Boyes <dbo...@sinenomine.net> > >> On 3/8/11 2:21 PM, "Ward, Mike S" <mw...@ssfcu.org> wrote: >> >> >Why not FTP it? >> >> Simple enough: Automating NJE transfer is trivial -- NJE is >> fire-and-forget in that the system daemons handle routing, retries, and >> guaranteed delivery (short of someone clearing spool for some stupid >> reason). If you've got the NJE connection, it's really, really easy to do >> this. NJE also doesn't require an active userid on the remote system to >> transfer files, and it doesn't tie up your userid while the file transfer >> is going on. >> >> Automating FTP transfer is a PITA, especially if you care if it actually >> arrives and/or if you hit a problem mid-transfer. Out of space errors (or >> any other trivial problem) are hard to recover from in a FTP script >> > > > > -- > Kris Buelens, > IBM Belgium, VM customer support > -- Kris Buelens, IBM Belgium, VM customer support
