Re: SMF exit IEFU086 work area size
@Peter, it is not so documented, but is it safe to assume that one has exclusive use of a given work area for the duration of the exit routine's execution? In other words, that the work area might be reused later, but is not shared among several tasks or processors? The documentation does not say that, but a "global" or "static" work area would be fairly useless, so it seems like a reasonable assumption. Charles -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Peter Relson Sent: Friday, October 11, 2019 5:26 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: SMF exit IEFU086 work area size When is this area obtained and freed? I.e. is this area unique for each invocation of IEFU86, or unique for each individual record, or "global". I'm wondering if I can put something in it on one invocation of IEFU86 and access it in a different invocation. Standard response, which applies here too: if it doesn't say that you are allowed to do so then do not assume that you can. The area is not unique to your exit routine. This should not be a surprise given that there can be many exit routines associated with a given exit, and SMF is not in charge of tracking what those exit routines are. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SMF exit IEFU086 work area size
On Fri, 11 Oct 2019 07:30:42 -0500, John McKown wrote: >OK, I'll remember that. "That which is not specifically allowed is >forbidden." More like "That which is not documented as working in a particular way cannot be assumed to work in that way, even if testing seems to show that it works that way." One can never realistically test all possible permutations. One possible permutation that can never be tested is the change that is made in the future. -- -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SMF exit IEFU086 work area size
On Fri, Oct 11, 2019 at 7:26 AM Peter Relson wrote: > > When is this area obtained and freed? I.e. is this area unique for each > invocation of IEFU86, or unique for each individual record, or "global". > I'm wondering if I can put something in it on one invocation of IEFU86 and > access it in a different invocation. > > > Standard response, which applies here too: if it doesn't say that you are > allowed to do so then do not assume that you can. > OK, I'll remember that. "That which is not specifically allowed is forbidden." > The area is not unique to your exit routine. This should not be a surprise > given that there can be many exit routines associated with a given exit, > and SMF is not in charge of tracking what those exit routines are. > > Peter Relson > z/OS Core Technology Design > > -- I find television very educational. The minute somebody turns it on, I go into the library and read a good book -- Groucho Marx 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: SMF exit IEFU086 work area size
When is this area obtained and freed? I.e. is this area unique for each invocation of IEFU86, or unique for each individual record, or "global". I'm wondering if I can put something in it on one invocation of IEFU86 and access it in a different invocation. Standard response, which applies here too: if it doesn't say that you are allowed to do so then do not assume that you can. The area is not unique to your exit routine. This should not be a surprise given that there can be many exit routines associated with a given exit, and SMF is not in charge of tracking what those exit routines are. Peter Relson z/OS Core Technology Design -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SMF exit IEFU086 work area size
I had the same question on the CNZ_WTOMDBEXIT work area. There they say you cannot store information from one invocation to another, but not to leave anything sensitive behind. IIRC. I assumed (yes, I know) that it was effectively "private" to the invoked exit instance (either serialized or a unique member of a set) and so far the code has not crashed or shown evidence of a collision. Charles -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of John McKown Sent: Thursday, October 10, 2019 5:16 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: SMF exit IEFU086 work area size On Thu, Oct 10, 2019 at 7:04 AM Peter Relson wrote: > > I can't find any documentation in KC on how to customize the size of the > workarea in SMPRMxx or what the minimum size of the area is. > > > The size is not customizable. 1024 is the size. The commentary on > SMXP_WORKAREA_LENGTH ought to say so (but doesn't, currently -- it will in > the future). > If the size were to change in the future (there are no plans to do so), > then 1024 would be the minimum. > > Peter Relson > z/OS Core Technology Design > When is this area obtained and freed? I.e. is this area unique for each invocation of IEFU86, or unique for each individual record, or "global". I'm wondering if I can put something in it on one invocation of IEFU86 and access it in a different invocation. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SMF exit IEFU086 work area size
On Thu, Oct 10, 2019 at 7:04 AM Peter Relson wrote: > > I can't find any documentation in KC on how to customize the size of the > workarea in SMPRMxx or what the minimum size of the area is. > > > The size is not customizable. 1024 is the size. The commentary on > SMXP_WORKAREA_LENGTH ought to say so (but doesn't, currently -- it will in > the future). > If the size were to change in the future (there are no plans to do so), > then 1024 would be the minimum. > > Peter Relson > z/OS Core Technology Design > When is this area obtained and freed? I.e. is this area unique for each invocation of IEFU86, or unique for each individual record, or "global". I'm wondering if I can put something in it on one invocation of IEFU86 and access it in a different invocation. -- I find television very educational. The minute somebody turns it on, I go into the library and read a good book -- Groucho Marx 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: SMF exit IEFU086 work area size
I can't find any documentation in KC on how to customize the size of the workarea in SMPRMxx or what the minimum size of the area is. The size is not customizable. 1024 is the size. The commentary on SMXP_WORKAREA_LENGTH ought to say so (but doesn't, currently -- it will in the future). If the size were to change in the future (there are no plans to do so), then 1024 would be the minimum. Peter Relson z/OS Core Technology Design -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SMF exit IEFU086 work area size
Absolutely. Today is the first day of my new career as an IEFU8x consultant. Charles -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of David Crayford Sent: Tuesday, October 8, 2019 9:27 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: SMF exit IEFU086 work area size On 2019-10-09 9:55 AM, Charles Mills wrote: > No, I retired from the SMF exit business on March 31. Of course, you have a wealth of expertise in this area from your Correlog days. I might ping you off list if I have any questions if that's alright? ;) > If I were writing the product today there is no question I would use it: only > one exit point to maintain, handles the new > 127 SMF records whenever they > come, and the work area is a plus. I used a cell pool for work areas in my > exit so it was pretty efficient, but "free" is even better. > > BTW, IEFU86, right? No 0? My bad :) -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SMF exit IEFU086 work area size
On 2019-10-09 9:55 AM, Charles Mills wrote: No, I retired from the SMF exit business on March 31. Of course, you have a wealth of expertise in this area from your Correlog days. I might ping you off list if I have any questions if that's alright? ;) If I were writing the product today there is no question I would use it: only one exit point to maintain, handles the new > 127 SMF records whenever they come, and the work area is a plus. I used a cell pool for work areas in my exit so it was pretty efficient, but "free" is even better. BTW, IEFU86, right? No 0? My bad :) Charles -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of David Crayford Sent: Tuesday, October 8, 2019 6:27 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: SMF exit IEFU086 work area size Hi Charles, Yes, it is! SMXP_WORKAREA_LENGTH_V1 EQU 1024 This IEFU086 exit is great! Have you used it? David On 2019-10-08 10:27 PM, Charles Mills wrote: It's not defined in IFAEXITP? Charles -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of David Crayford Sent: Tuesday, October 8, 2019 5:52 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: SMF exit IEFU086 work area size z/OS 2.3 introduced the new IEFU086 SMF exit which is passed a work area which is great. I can't find any documentation in KC on how to customize the size of the workarea in SMPRMxx or what the minimum size of the area is. Can anybody shed any light on this? -- 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 -- 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 -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SMF exit IEFU086 work area size
Should have written > 255, of course. Charles -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Charles Mills Sent: Tuesday, October 8, 2019 6:56 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: SMF exit IEFU086 work area size No, I retired from the SMF exit business on March 31. If I were writing the product today there is no question I would use it: only one exit point to maintain, handles the new > 127 SMF records whenever they come, and the work area is a plus. I used a cell pool for work areas in my exit so it was pretty efficient, but "free" is even better. BTW, IEFU86, right? No 0? Charles -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of David Crayford Sent: Tuesday, October 8, 2019 6:27 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: SMF exit IEFU086 work area size Hi Charles, Yes, it is! SMXP_WORKAREA_LENGTH_V1 EQU 1024 This IEFU086 exit is great! Have you used it? David -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SMF exit IEFU086 work area size
No, I retired from the SMF exit business on March 31. If I were writing the product today there is no question I would use it: only one exit point to maintain, handles the new > 127 SMF records whenever they come, and the work area is a plus. I used a cell pool for work areas in my exit so it was pretty efficient, but "free" is even better. BTW, IEFU86, right? No 0? Charles -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of David Crayford Sent: Tuesday, October 8, 2019 6:27 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: SMF exit IEFU086 work area size Hi Charles, Yes, it is! SMXP_WORKAREA_LENGTH_V1 EQU 1024 This IEFU086 exit is great! Have you used it? David On 2019-10-08 10:27 PM, Charles Mills wrote: > It's not defined in IFAEXITP? > > Charles > > > -Original Message- > From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On > Behalf Of David Crayford > Sent: Tuesday, October 8, 2019 5:52 AM > To: IBM-MAIN@LISTSERV.UA.EDU > Subject: SMF exit IEFU086 work area size > > z/OS 2.3 introduced the new IEFU086 SMF exit which is passed a work area > which is great. > > I can't find any documentation in KC on how to customize the size of the > workarea in SMPRMxx or what the minimum size of the area is. > > Can anybody shed any light on this? > > -- > 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 -- 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: SMF exit IEFU086 work area size
Hi Charles, Yes, it is! SMXP_WORKAREA_LENGTH_V1 EQU 1024 This IEFU086 exit is great! Have you used it? David On 2019-10-08 10:27 PM, Charles Mills wrote: It's not defined in IFAEXITP? Charles -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of David Crayford Sent: Tuesday, October 8, 2019 5:52 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: SMF exit IEFU086 work area size z/OS 2.3 introduced the new IEFU086 SMF exit which is passed a work area which is great. I can't find any documentation in KC on how to customize the size of the workarea in SMPRMxx or what the minimum size of the area is. Can anybody shed any light on this? -- 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 -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SMF exit IEFU086 work area size
It's not defined in IFAEXITP? Charles -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of David Crayford Sent: Tuesday, October 8, 2019 5:52 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: SMF exit IEFU086 work area size z/OS 2.3 introduced the new IEFU086 SMF exit which is passed a work area which is great. I can't find any documentation in KC on how to customize the size of the workarea in SMPRMxx or what the minimum size of the area is. Can anybody shed any light on this? -- 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: SMF exit IEFU086 work area size
Thank you Larre! This looks definitive :) On 2019-10-08 9:42 PM, Larre Shiller wrote: David - Based on some ESP notes that I have from our z/OS 2.3 install, it looks like the work area size is 1024 bytes. The parameter list is mapped by MACLIB(IFAEXITP). I see no evidence to indicate that the length can be set/modified by the installation. Larre US Social Security Administration "The opinions expressed in this e-mail are mine personally and do not necessarily reflect the opinion of the US Social Security Administration and/or the US Government." -- 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: SMF exit IEFU086 work area size
David - Based on some ESP notes that I have from our z/OS 2.3 install, it looks like the work area size is 1024 bytes. The parameter list is mapped by MACLIB(IFAEXITP). I see no evidence to indicate that the length can be set/modified by the installation. Larre US Social Security Administration "The opinions expressed in this e-mail are mine personally and do not necessarily reflect the opinion of the US Social Security Administration and/or the US Government." -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SMF exit IEFU086 work area size
David, I guess you will use a mapping for your work area in your program. So you can always have code which compares the length needed with the length available. I just don't know what you do when your needs are greater than they supply! Issue a message? Yes, but how many messages will you get? (Ans: Millions!) Give up without a message? Not very useful at all. Assuming this is a dynamic exit, you could issue a message and then deactivate the exit. None of these is very satisfactory. I suspect the length is always the same, but IBM reserve the right to increase it. Come on IBM. Give us the answer! Lennie Dymoke-Bradshaw | Security Lead | RSM Partners Ltd Web: www.rsmpartners.com ‘Dance like no one is watching. Encrypt like everyone is.’ -Original Message- From: IBM Mainframe Discussion List On Behalf Of David Crayford Sent: 08 October 2019 14:22 To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: [IBM-MAIN] SMF exit IEFU086 work area size Thank you Lennie, You make some very good observations. I think the salient point of the work area is so the exit routine doesn't need to allocate any storage for obvious reasons (performance). I'm leery of using an exit with a work area without knowing it's length! On 2019-10-08 9:13 PM, Lennie Dymoke-Bradshaw wrote: > When I look in the System Exits manual I see a structure passed as a > parameter list which has an offset to the work area, and a length specified > for it. Assuming the length is specified in bytes it must fit in a halfword. > So max would be 65536 bytes. > > I agree it would be sensible for IBM to give an indication of how much there > is. Without that one could be reduced to getting ones own work area . > > Can you dump out the parameter list to see what is passed? Can we have any > indication from IBM of a minimum size? > > Lennie Dymoke-Bradshaw | Security Lead | RSM Partners Ltd -- 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: SMF exit IEFU086 work area size
Thank you Lennie, You make some very good observations. I think the salient point of the work area is so the exit routine doesn't need to allocate any storage for obvious reasons (performance). I'm leery of using an exit with a work area without knowing it's length! On 2019-10-08 9:13 PM, Lennie Dymoke-Bradshaw wrote: When I look in the System Exits manual I see a structure passed as a parameter list which has an offset to the work area, and a length specified for it. Assuming the length is specified in bytes it must fit in a halfword. So max would be 65536 bytes. I agree it would be sensible for IBM to give an indication of how much there is. Without that one could be reduced to getting ones own work area . Can you dump out the parameter list to see what is passed? Can we have any indication from IBM of a minimum size? Lennie Dymoke-Bradshaw | Security Lead | RSM Partners Ltd -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SMF exit IEFU086 work area size
When I look in the System Exits manual I see a structure passed as a parameter list which has an offset to the work area, and a length specified for it. Assuming the length is specified in bytes it must fit in a halfword. So max would be 65536 bytes. I agree it would be sensible for IBM to give an indication of how much there is. Without that one could be reduced to getting ones own work area . Can you dump out the parameter list to see what is passed? Can we have any indication from IBM of a minimum size? Lennie Dymoke-Bradshaw | Security Lead | RSM Partners Ltd Web: www.rsmpartners.com ‘Dance like no one is watching. Encrypt like everyone is.’ -Original Message- From: IBM Mainframe Discussion List On Behalf Of David Crayford Sent: 08 October 2019 13:52 To: IBM-MAIN@LISTSERV.UA.EDU Subject: [IBM-MAIN] SMF exit IEFU086 work area size z/OS 2.3 introduced the new IEFU086 SMF exit which is passed a work area which is great. I can't find any documentation in KC on how to customize the size of the workarea in SMPRMxx or what the minimum size of the area is. Can anybody shed any light on this? -- 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
SMF exit IEFU086 work area size
z/OS 2.3 introduced the new IEFU086 SMF exit which is passed a work area which is great. I can't find any documentation in KC on how to customize the size of the workarea in SMPRMxx or what the minimum size of the area is. Can anybody shed any light on this? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
