I'm assuming that OP has no other functioning OS, no ability to edit or zap the failing system externally. I've never actually done this and don't have any example, but you might be able to use the IPL LOADPARM to force operator prompting at NIP. That lives in HMC on the IPL screen. If you could you respond 'properly' to the prompt, you might be able to nullify the poison exit.
See https://www.ibm.com/support/knowledgecenter/en/SSB27H_6.2.0/fa2sc_r_ipload.html . . J.O.Skip Robinson Southern California Edison Company Electric Dragon Team Paddler SHARE MVS Program Co-Manager 323-715-0595 Mobile 626-543-6132 Office ⇐=== NEW robin...@sce.com -----Original Message----- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Carmen Vitullo Sent: Wednesday, November 21, 2018 9:01 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: (External):Re: Region size for OMVS tasks if your using the PROG00 member to add the exit, deleting the member does nothing, you will probably need to issue a SETPROG EXIT DELETE,EXITNAME=xxxxxx too late now but remove this EXIT ADD if it exists maybe this will help Carmen Vitullo ----- Original Message ----- From: "Sean Gleann" <sean.gle...@gmail.com> To: IBM-MAIN@LISTSERV.UA.EDU Sent: Wednesday, November 21, 2018 10:56:00 AM Subject: Re: Region size for OMVS tasks Thanks, Allan, but I can't get any tasks of any nature started - no VTAM, no TSO, so no copy is possible. (I'll squirrel that 'SYS1.AOSB3' info away, though, for possible future reference - hopefully never) Sean On Wed, 21 Nov 2018 at 16:38, Allan Staller <allan.stal...@hcl.com> wrote: > If you have customized IEFUSI, reassemble you customized exit. > IF not, copy SYS1.AOSB3 to SYS1.LLPALIB and re-ipl. > > -----Original Message----- > From: IBM Mainframe Discussion List <IBM-MAIN@LISTSERV.UA.EDU> On > Behalf Of Sean Gleann > Sent: Wednesday, November 21, 2018 10:20 AM > To: IBM-MAIN@LISTSERV.UA.EDU > Subject: Re: Region size for OMVS tasks > > Trying to follow-up on the stuff that Peter has suggested, I'm still > chasing that elusive '54M' - Who sets it? Where is it set? How can I > override it? > So, I created an IEFUSI that, whatever size is requested, whether it > be below, above or MEMLIMIT, you get '0M'. > Which, I agree, is probably a bit extreme, so I armed myself with an > IDCAMS job in 'HOLD' on the input queue to do a DELETE > xxx.LPALIB(IEFUSI), that I could release if things went pear-shaped. > System had a certain amount of … umm … 'difficulty' in restarting, but > JES itself started, so $AJnnnn worked OK. > But all the inits go to a 922 abend when they try to start. so the > delete job never got a chance. > Now I've got a system that won't IPL at all beyond a certain point. > Have tried alternate members from SYS1.IPLPARM, but none of then > specify sufficient paging space. > > Up the creek sans paddle, right now > > Sean > > On Tue, 20 Nov 2018 at 20:01, Peter Hunkeler <p...@gmx.ch> wrote: > > > > Where did you get all this from? Is it documented somewhere, or is > > > it > > something you've 'soaked up through your fingertips' over the years? > > > > > > So I went back to the FMs. For any question of kind "How does that > > UNIX service behave regarding MVS resources, attributes, etc.", I > > always read what the "z/OS UNIX System Services Programming - > > Assembler Callable Services Reference" tells me. If something is to > > be said, there is a section called "Usage Notes". Here are some > > interesting > snippet: > > > > > > For fork() it says: > > > > - The child process inherits the MEMLIMIT of the parent. > > - The child address space inherits the following address space > > attributes of the parent address space: region size and time limit. > > > > > > For spawn() it says: > > > > - The region size of the parent is inherited by the child, unless > > the INHESETREGIONSZ flag in the inheritance structure is set on to > > indicate that the value specified in INHEREGIONSZ is to be used to > > determine the child's region size. For more information, see What > > are soft limits? in z/OS UNIX System Services Planning and > > Inheriting soft limits in z/OS UNIX System Services Planning. > > > > - The MEMLIMIT of the parent is inherited by the child, unless the > > INHESETMEMLIMIT flag in the inheritance structure is set on to > > indicate that the value specified in INHEMEMLIMIT is to be used to > > determine the child's MEMLIMIT. > > > > > > Another useful source of information is "z/OS UNIX System Services > > Planning". This FM says about MAXASSIZE: > > > > MAXASSIZE is the maximum region size (in bytes) for an address space > > that was created by rlogind, telnetd, and other daemons. The > > MAXASSIZE value limits the combined size of above and below the 16 M line > > storage. > > If MAXASSIZE is greater than LDALIMIT (the <16M limit), then the > > LDAELIM (the >16M limit) is set to MAXASSIZE - LDALIM. > > > > You can set a system-wide limit in BPXPRMxx and then set higher > > limits for individual processes. Use the RACF ADDUSER or ALTUSER > > command to specify the ASSIZEMAX limit on a per-process basis as follows: > > ALTUSER userid OMVS(ASSIZEMAX(nnnn) > > > > > > This manual also has a couple of pages discussing all about limits. > > Of special interest to this discussion it the section called "How > > are limits handled after an identity change?", and the following sections. > > Its too much to copy&paste here. I suggets you have a read if your > > interested. (I've got the impression this part has not always be > > there in that detail. Great it is now) > > > > > > -- > > > > Peter Hunkeler ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
