Re: SYSPROC Concatenation?
On 2016-03-20, at 12:19, Jesse 1 Robinson wrote: > AFAIK the allocation done by EX(EC) lasts only for the duration of the script > execution. Of course if the data set has been independently allocated to > SYSPROC or SYSEXEC, then that allocation will persist. > Don't trust IBM on such matters. Long ago, a colleague did a TSO TRANSMIT of a data set otherwise previously allocated. TRANSMIT dynalloced it; transmitted it; and did FREE DSN() when finished, thereby usurping his prior allocation. He went to SR; got WAD. I don't know that it was ever fixed. -- gil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SYSPROC Concatenation?
AFAIK the allocation done by EX(EC) lasts only for the duration of the script execution. Of course if the data set has been independently allocated to SYSPROC or SYSEXEC, then that allocation will persist. . . . J.O.Skip Robinson Southern California Edison Company Electric Dragon Team Paddler SHARE MVS Program Co-Manager 323-715-0595 Mobile 626-302-7535 Office robin...@sce.com -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Paul Gilmartin Sent: Saturday, March 19, 2016 11:31 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: (External):Re: SYSPROC Concatenation? On Sat, 19 Mar 2016 15:28:30 +, J R wrote: >> o Isn't the argument to EX a DDNAME, not a DSNAME? > >No! > I stand corrected; I had it exactly backward (I intended the reverse.) But the prior ply used "ex sysproc(logon)", and SYSPROC is conventionally a DDNAME. And by Jesse's clarification, that means: EXEC 'tso-pref.SYSPROC.CLIST(LOGON)' >> o What causes that PDS to be allocated? > >The EX command. > Am I correct that allocation is temporary, and FREEd when EXEC completes? -- gil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SYSPROC Concatenation?
On Mar 19, 2016, at 9:25 AM, Paul Gilmartin wrote: On Fri, 18 Mar 2016 23:28:46 -0500, Ed Gould wrote: The way I learned to do it was that logon procs did a universal ex sysproc(logon) and if you had a pds that had the member logon in it o Isn't the argument to EX a DDNAME, not a DSNAME? o What causes that PDS to be allocated? That was the OP's question. THE LOGON PROC *OR* execution of an exec on the execution (exec pgm=ikjeft01 or the session manager name) statement with a standard logon proc. Or is the assumption that there's an optional 'prefix.SYSPROC(LOGON)' and that failure of the EXEC command is tolerable. it would exec that clist and in that clist you could allocate the libraries to your liking. With the understanding that if didn't have IBM libraries first don't come trouble shooting to us. On Fri, 18 Mar 2016 23:30:27 -0500, Ed Gould wrote: Don't blame MVS. *YOU* can set it up to your own local standards. Tunnel vision. That's fine if the "*YOU*" involved is the arbiter of those "local standards"; not so much for a lowly end user applications programmer with no authority to modify a logon procedure. It is semi local standard *IF* the local choose to use it. It is documented in several TSO manuals. But on experiment I believe the COMMAND field meets the OP's needs: o It's controlled by the end user. Both.. if the installation wants it that way. o It persists from session to session. It does if you use the logon proc to exec a clist. -- gil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SYSPROC Concatenation?
On Mar 19, 2016, at 1:30 PM, Paul Gilmartin wrote: On Sat, 19 Mar 2016 15:28:30 +, J R wrote: o Isn't the argument to EX a DDNAME, not a DSNAME? No! I stand corrected; I had it exactly backward (I intended the reverse.) But the prior ply used "ex sysproc(logon)", and SYSPROC is conventionally a DDNAME. And by Jesse's clarification, that means: EXEC 'tso-pref.SYSPROC.CLIST(LOGON)' o What causes that PDS to be allocated? The EX command. Am I correct that allocation is temporary, and FREEd when EXEC completes? No. -- gil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SYSPROC Concatenation?
On Sat, 19 Mar 2016 15:28:30 +, J R wrote: >> o Isn't the argument to EX a DDNAME, not a DSNAME? > >No! > I stand corrected; I had it exactly backward (I intended the reverse.) But the prior ply used "ex sysproc(logon)", and SYSPROC is conventionally a DDNAME. And by Jesse's clarification, that means: EXEC 'tso-pref.SYSPROC.CLIST(LOGON)' >> o What causes that PDS to be allocated? > >The EX command. > Am I correct that allocation is temporary, and FREEd when EXEC completes? -- gil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SYSPROC Concatenation?
The EX(EC) command allocates the library according to several rules. -- EX MYLIB(SCRIPT) attempts to run member SCRIPT in library 'tso-pref.MYLIB.CLIST' -- EX MYLIB(SCRIPT) EXEC attempts to run member SCRIPT in library 'tso-pref.MYLIB.EXEC' -- EX 'hlq.SOMELIB.PROC(SCRIPT)' attempts to run member SCRIPT in the library exactly as named, no assumptions. -- Parameters passed to the exec must be enclosed in single quotes. This means that any single quotes passed in the parms must be doubled: -- EX MYLIB(SCRIPT) 'DS(''hlq.a.b.c(member)'' ' These are two single quotes, not a double quote. This can be a major typing challenge. Best executed from another script. . . . J.O.Skip Robinson Southern California Edison Company Electric Dragon Team Paddler SHARE MVS Program Co-Manager 323-715-0595 Mobile 626-302-7535 Office robin...@sce.com -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of J R Sent: Saturday, March 19, 2016 8:29 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: (External):Re: SYSPROC Concatenation? > o Isn't the argument to EX a DDNAME, not a DSNAME? No! > o What causes that PDS to be allocated? The EX command. From: IBM Mainframe Discussion List on behalf of Paul Gilmartin <000433f07816-dmarc-requ...@listserv.ua.edu> Sent: Saturday, March 19, 2016 10:25 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: SYSPROC Concatenation? On Fri, 18 Mar 2016 23:28:46 -0500, Ed Gould wrote: >The way I learned to do it was that logon procs did a universal ex >sysproc(logon) and if you had a pds that had the member logon in it o Isn't the argument to EX a DDNAME, not a DSNAME? o What causes that PDS to be allocated? That was the OP's question. Or is the assumption that there's an optional 'prefix.SYSPROC(LOGON)' and that failure of the EXEC command is tolerable. >it would exec that clist and in that clist you could allocate the >libraries to your liking. With the understanding that if didn't have >IBM libraries first don't come trouble shooting to us. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SYSPROC Concatenation?
> o Isn't the argument to EX a DDNAME, not a DSNAME? No! > o What causes that PDS to be allocated? The EX command. From: IBM Mainframe Discussion List on behalf of Paul Gilmartin <000433f07816-dmarc-requ...@listserv.ua.edu> Sent: Saturday, March 19, 2016 10:25 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: SYSPROC Concatenation? On Fri, 18 Mar 2016 23:28:46 -0500, Ed Gould wrote: >The way I learned to do it was that logon procs did a universal ex >sysproc(logon) and if you had a pds that had the member logon in it o Isn't the argument to EX a DDNAME, not a DSNAME? o What causes that PDS to be allocated? That was the OP's question. Or is the assumption that there's an optional 'prefix.SYSPROC(LOGON)' and that failure of the EXEC command is tolerable. >it would exec that clist and in that clist you could allocate the >libraries to your liking. With the understanding that if didn't have >IBM libraries first don't come trouble shooting to us. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SYSPROC Concatenation?
On Fri, 18 Mar 2016 23:28:46 -0500, Ed Gould wrote: >The way I learned to do it was that logon procs did a universal ex >sysproc(logon) and if you had a pds that had the member logon in it o Isn't the argument to EX a DDNAME, not a DSNAME? o What causes that PDS to be allocated? That was the OP's question. Or is the assumption that there's an optional 'prefix.SYSPROC(LOGON)' and that failure of the EXEC command is tolerable. >it would exec that clist and in that clist you could allocate the >libraries to your liking. With the understanding that if didn't have >IBM libraries first don't come trouble shooting to us. On Fri, 18 Mar 2016 23:30:27 -0500, Ed Gould wrote: > >Don't blame MVS. *YOU* can set it up to your own local standards. > Tunnel vision. That's fine if the "*YOU*" involved is the arbiter of those "local standards"; not so much for a lowly end user applications programmer with no authority to modify a logon procedure. But on experiment I believe the COMMAND field meets the OP's needs: o It's controlled by the end user. o It persists from session to session. -- gil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SYSPROC Concatenation?
On Wed, 16 Mar 2016 10:58:28 -0500, John McKown wrote: >> >> Q). How did I get my CLIST Concatenated permanently? I would like to do >> it again. > >Actually, nobody here can tell you -- exactly. ... There is not "IBM standard" way to do this. > Pretty much confirmed by: Writing a logon procedure z/OS TSO/E Customization SA32-0976-00 Grrr... I hate MVS! It seems to have a philosophy of denying flexibility to end users and elevating it to administrators. The default PARM for IKJEFT01 ought to be something such as "EXEC 'prefix.CLIST(PROFILE)'" The chapter cited above mentions the COMMAND field of the LOGON panel. I believe this is persistent and could be a script to perform the concatenation you want. Don't forget SYSEXEC. --gil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SYSPROC Concatenation?
IBM does this exact same thing in an OOTB ServerPac install. It allocates libraries in the logon proc then calls a clist to reallocate them. For no apparently good reason. I've done that when I wanted more control, for example, who gets what libraries, maybe certain users get dev libraries allocated first, the possibilities are endless. Br, Lindy -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Paul Gilmartin Sent: keskiviikkona 16. maaliskuuta 2016 17.22 To: IBM-MAIN@LISTSERV.UA.EDU Grrr... I hate MVS! It seems to have a philosophy of denying flexibility to end users and elevating it to administrators. The default PARM for IKJEFT01 ought to be something such as "EXEC 'prefix.CLIST(PROFILE)'" The chapter cited above mentions the COMMAND field of the LOGON panel. I believe this is persistent and could be a script to perform the concatenation you want. Don't forget SYSEXEC. --gil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SYSPROC Concatenation?
On Wed, Mar 16, 2016 at 10:51 AM, Hansen, Dave L - Eagan, MN < 00d83826d62b-dmarc-requ...@listserv.ua.edu> wrote: > Hi, > > I went to https://listserv.ua.edu/cgi-bin/wa and did a search on > IBM-MAIN and IBM-MAIN-ARCHIVES (1986-2004). I did a search for SYSPROC and > got no match. > > On one system I have ISRDDN showing SYSPROC with my CLIST concatenated. > I can use ALTLIB (or SPROC) to add it in to the other system. But I have > to manually add these and repeat it. > > My logon Procedure is xxxACCNT. I looked in SYS2.LOG and > SYS2.CLIST(SPFALLOC) for my userid to find userid.CLIST but am not finding > it. > >I looked through the userid.lpar.ISPF .PROFILE datasets. ISRPLIST > shows the last 30 referenced data sets. > > Q). How did I get my CLIST Concatenated permanently? I would like to do > it again. > Actually, nobody here can tell you -- exactly. We do the same. In our logon PROCEDURE, we have a //TSOISPF EXEC PGM=IKJEFT01,PARM='LOGON'. In the SYSPROC DD concatenation, we have a member called LOGON which is what that executes. It is a REXX exec. One of the things that it does is re-allocate the SYSPROC DD to include, if it exists, the equivalent of DSN=&SYSUID.TSO.CLIST . Something like that is most likely what is happening at your shop. But you'll need to track down the CLIST/REXX programs invoked by the "xxxACCNT" logon procedure to determine what you need to do. There is not "IBM standard" way to do this. > > >Thank you, Dave > -- A fail-safe circuit will destroy others. -- Klipstein Maranatha! <>< John McKown -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SYSPROC Concatenation?
On Mar 16, 2016, at 11:22 AM, Paul Gilmartin wrote: On Wed, 16 Mar 2016 10:58:28 -0500, John McKown wrote: Q). How did I get my CLIST Concatenated permanently? I would like to do it again. Actually, nobody here can tell you -- exactly. ... There is not "IBM standard" way to do this. Pretty much confirmed by: Writing a logon procedure z/OS TSO/E Customization SA32-0976-00 Grrr... I hate MVS! It seems to have a philosophy of denying flexibility to end users and elevating it to administrators. The default PARM for IKJEFT01 ought to be something such as "EXEC 'prefix.CLIST(PROFILE)'" The chapter cited above mentions the COMMAND field of the LOGON panel. I believe this is persistent and could be a script to perform the concatenation you want. Gil: Don't blame MVS. *YOU* can set it up to your own local standards. Ed -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SYSPROC Concatenation?
The way I learned to do it was that logon procs did a universal ex sysproc(logon) and if you had a pds that had the member logon in it it would exec that clist and in that clist you could allocate the libraries to your liking. With the understanding that if didn't have IBM libraries first don't come trouble shooting to us. Ed On Mar 16, 2016, at 10:58 AM, John McKown wrote: On Wed, Mar 16, 2016 at 10:51 AM, Hansen, Dave L - Eagan, MN < 00d83826d62b-dmarc-requ...@listserv.ua.edu> wrote: Hi, I went to https://listserv.ua.edu/cgi-bin/wa and did a search on IBM-MAIN and IBM-MAIN-ARCHIVES (1986-2004). I did a search for SYSPROC and got no match. On one system I have ISRDDN showing SYSPROC with my CLIST concatenated. I can use ALTLIB (or SPROC) to add it in to the other system. But I have to manually add these and repeat it. My logon Procedure is xxxACCNT. I looked in SYS2.LOG and SYS2.CLIST(SPFALLOC) for my userid to find userid.CLIST but am not finding it. I looked through the userid.lpar.ISPF .PROFILE datasets. ISRPLIST shows the last 30 referenced data sets. Q). How did I get my CLIST Concatenated permanently? I would like to do it again. Actually, nobody here can tell you -- exactly. We do the same. In our logon PROCEDURE, we have a //TSOISPF EXEC PGM=IKJEFT01,PARM='LOGON'. In the SYSPROC DD concatenation, we have a member called LOGON which is what that executes. It is a REXX exec. One of the things that it does is re- allocate the SYSPROC DD to include, if it exists, the equivalent of DSN=&SYSUID.TSO.CLIST . Something like that is most likely what is happening at your shop. But you'll need to track down the CLIST/REXX programs invoked by the "xxxACCNT" logon procedure to determine what you need to do. There is not "IBM standard" way to do this. Thank you, Dave -- A fail-safe circuit will destroy others. -- Klipstein Maranatha! <>< John McKown -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN