Re: IEFU83, IEFU84 and IEFU85
Charles If you are installing IEFU83 via CSVDYNEX then the following points may be of interest : (o) "IEFU83" must be specified in the EXITS() keyword for each SMF subsystem where the record can be written (o) Note that this is specified as "IEFU83" in SMFPRMxx and *not* the name of your exit module - all you are doing here is enabling the exit point, the name of your exit module is set via the CSVDYNEX service later on. (o) You *must* ensure that IEFU83 is enabled in each SMF subsystem that the record can be written from. There is a "gotcha" here when a site has defined either or both of the "STC" and "JES2" subsystems as you need to ensure that SYSSTC.IEFU83 and SYSJES2.IEFU83 have been defined along with SYS.IEFU83. There might be other SMF subsystems in play as well and you might have to cater for those as well. (o) Normally I would not expect to have to alter the PROGxx or EXITxx PARMLIB members when installing/defining IEFU83 via CSVDYNEX. (o) I would advise sticking in some diagnostic/accounting capabilities in the IEFU83 exit while you have the hood up. (o) When debugging this sort of thing for a customer site, it is always worth getting a copy of their SMFPRMxx and the output from a "D SMF,O" command. Rob Scott Lead Developer Rocket Software 275 Grove Street * Newton, MA 02466-2272 * USA Tel: +1.617.614.2305 Email: rsc...@rs.com Web: www.rocketsoftware.com -Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of Charles Mills Sent: 18 July 2010 21:44 To: IBM-MAIN@bama.ua.edu Subject: Re: IEFU83, IEFU84 and IEFU85 > I.o.w.: check that site's SMFPRMxx I hope someone is reading this on a Sunday ... Please have patience with me. I'm a developer not a sysprog and I know many things but I am not (yet!) an SMF exit expert. I didn't write the code in question. I'm trying to figure out the right question to ask a customer. I've spent a lot of time reading SMFPRMxx, PROGxx, and EXITxx documentation. If an IEFU83 exit is to be added via CSVDYNEX REQUEST=ADD, what has to exist or not exist in SMFPRMxx, PROGxx, and EXITxx in order for the exit to be invoked? (Please for now assume that the CSVDYNEX and the exit itself are valid, and that SMF is configured to write Type 80 records.) I'm trying to understand the statement "EXITS specifies which SMF exits are to be invoked. A maximum of 15 exits is allowed; if an exit is not specified, it is not invoked. If this parameter is not specified, SMF behaves as if this parameter is specified with all 15 exits listed here and all SMF system exits are invoked." How do you specify that an exit "is to be invoked" if the name is not known at IPL time? If a customer were not currently using U83 but had an SMFPRMxx member that specified other exits and not U83, what change would the customer have to make in order for the CSVDYNEX ADD to be effective and have the added exit module invoked? Charles bm-main.html -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Detecting when a READ/GET crosses a concatenation boundary
At 6:35 PM -0400 on 07/18/10, Thompson, Steve wrote about Re: Detecting when a READ/GET crosses a concatenation bound: Take a look at "Using Non-VSAM User-Written Exit Routines" with the specifics of "Open/Close/EOV Standard User Label Exit". One potential GOTCHA - The EOV exit will be driven not only when you reach the end of a file and switch to the next file in the concatenation but ALSO during a volume switch of a multi-volume sequential dataset. Thus you need to check for this to avoid false positives if all you care about is switching to a new dataset as opposed to also being signaled about a new volume of the current dataset. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: IEFU29 - Intermi ttently do es not get control
Peter Relson wrote: ...You might even try using a name other than IEFU29 (there is no need to use the name IEFU29 for your exit routine) with appropriate EXIT ADD statements in PROGxx. My own view is marginally different; I would strongly urge you to use different and distinct, more perspicuous names for each of these exits. John Gilmore Ashland, MA 01721-1817 USA _ The New Busy is not the old busy. Search, chat and e-mail from your inbox. http://www.windowslive.com/campaign/thenewbusy?ocid=PID28326::T:WLMTAGL:ON:WL:en-US:WM_HMP:042010_3 -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: IEFU29 - Intermittently does not get control
It seems unlikely that the system is failing to call the exit routines in the circumstances in which it is documented to do so. Have you tried putting a SLIP trap wherever the call is initiated to confirm that your expectation of calling the exit matches the reality? Are you certain that your IEFU29 is defined properly? There is not one single IEFU29 exit. There is, for example, SYS.IEFU29 and SYSSTC.IEFU29 etc. Doing various DISPLAY PROG,EXIT's can make sure that you have all of the IEFU29's defined that you want and that your exit routine is associated properly with each. You might even try using a name other than IEFU29 (there is no need to use the name IEFU29 for your exit routine) with appropriate EXIT ADD statements in PROGxx. Peter Relson z/OS Core Technology Design -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Detecting when a READ/GET crosses a concatenation boundary
-Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of Kenneth J. Kripke Sent: Sunday, July 18, 2010 5:09 PM To: IBM-MAIN@bama.ua.edu Subject: Detecting when a READ/GET crosses a concatenation boundary Hello; I wish to detect when a BSAM read or QSAM GET crosses a concatenation boundary. I am guessing I need to code an OPEN and EOV DCB exit to get control. The input data sets can reside on DASD or TAPE, but, the environment does not mix unlike devices...all one way or the other. Purpose: Be able to the extract the dsn actively being processed. Environment has specifc dsn naming conventions that tie back to a cycle number and the generating jobstream. Any assistance/tips would be greatly appreciated. "DFSMS Using Data Sets" gives some tips in this area. Look at the area covering "Concatenating Data Sets Sequentially". The OPEN exit is driven for each data set in the concat as the Access Method switches to it. So from this, you should be able to get the DSN. And the Label exits get driven as well. Take a look at "Using Non-VSAM User-Written Exit Routines" with the specifics of "Open/Close/EOV Standard User Label Exit". So you are on the right track. Beware of mixed mode (24/31 bit addressing), it can get you really lost when trying to diagnose a problem. Also, look out for LBI -- you have said something about tapes. They can be in LARGE BLOCK (>32K) and certain things may be different for the exits (I can't remember exactly how that works -- I get to diagnose problems with our code from time to time). Regards, Steve Thompson -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Detecting when a READ/GET crosses a concatenation boundary
Hello; I wish to detect when a BSAM read or QSAM GET crosses a concatenation boundary. I am guessing I need to code an OPEN and EOV DCB exit to get control. The input data sets can reside on DASD or TAPE, but, the environment does not mix unlike devices...all one way or the other. Purpose: Be able to the extract the dsn actively being processed. Environment has specifc dsn naming conventions that tie back to a cycle number and the generating jobstream. Any assistance/tips would be greatly appreciated. Sincerely Yours; Ken Kripke kkri...@mindspring.com -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: IEFU83, IEFU84 and IEFU85
> I.o.w.: check that site's SMFPRMxx I hope someone is reading this on a Sunday ... Please have patience with me. I'm a developer not a sysprog and I know many things but I am not (yet!) an SMF exit expert. I didn't write the code in question. I'm trying to figure out the right question to ask a customer. I've spent a lot of time reading SMFPRMxx, PROGxx, and EXITxx documentation. If an IEFU83 exit is to be added via CSVDYNEX REQUEST=ADD, what has to exist or not exist in SMFPRMxx, PROGxx, and EXITxx in order for the exit to be invoked? (Please for now assume that the CSVDYNEX and the exit itself are valid, and that SMF is configured to write Type 80 records.) I'm trying to understand the statement "EXITS specifies which SMF exits are to be invoked. A maximum of 15 exits is allowed; if an exit is not specified, it is not invoked. If this parameter is not specified, SMF behaves as if this parameter is specified with all 15 exits listed here and all SMF system exits are invoked." How do you specify that an exit "is to be invoked" if the name is not known at IPL time? If a customer were not currently using U83 but had an SMFPRMxx member that specified other exits and not U83, what change would the customer have to make in order for the CSVDYNEX ADD to be effective and have the added exit module invoked? Charles bm-main.html -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: IND$FILE question
Ed, There were separate products released for MVS, VM, and VSE. All were released in 1983. On MVS, the product number was 5665-311, and the FMID was HFX1112. John P. Baker -Original Message- From: IBM Mainframe Discussion List [mailto:ibm-m...@bama.ua.edu] On Behalf Of Ed Gould Sent: Sunday, July 18, 2010 1:58 AM To: IBM-MAIN@bama.ua.edu Subject: IND$FILE question I vaguely remember (A LONG TIME AGO mind you) that there was a separately installable FMID for IND$FILE (bonus points for the year it came out) Does anyone remember this? As a side question there was a separate help member in sys1.help (vague recollection here) Does(did?) IBM still ship the Help member for it? I seemed to remember that it was dropped along the way and it was never put back. Is this true? I just vaguely remember creating a usermod to add it back. Ed -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: IND$FILE question
Ed Gould wrote: >I vaguely remember (A LONG TIME AGO mind you) that there was a separately installable FMID for IND$FILE (bonus points for the year it came out) My last few remaining brain cells can't help you with that... ;-D >Does anyone remember this? As a side question there was a separate help member in sys1.help (vague recollection here) Does(did?) IBM still ship the Help member for it? I seemed to remember that it was dropped along the way and it was never put back. Is this true? My searches for you, yielded NOTHING... SYS1.MACLIB, MODLIB, SYS1.HELP, the usual concatenation lot Nada zilch, nothing... Oh yes. IND$FILE (AM and RM still at 24!) on z/OS v1.10 is found in SYS1.CMDLIB with the usual disclaimer text: INDFT0 IBM File Transfer Program: Version 1.1.1 INDFT0 PTF: UR43604 Date: 05/09/95 Size: 50C8 Bytes Somewhat old ... My emulator supported it, but says a little about it in the help screens. Check your emulator software if they support it or document something about IND$FILE. >I just vaguely remember creating a usermod to add it back. ??? Usermod??? First time I read about this (or probably forgot about it) Groete / Greetings Elardus Engelbrecht -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html