Re: z/OS 2.1 to 2.4 [EXTERNAL]
ASM is Adabas Parallel Services, I'm not sure why they didn't call it APS, but possibly they already had one by that name. ASM provides Compression, decompression, format buffer translation, sorting, retrieving, searching and updating operations all occur in parallel. So it gives better distribution of your workload across the available processors. You might already be running it and just don't know it, or just don't have the extra address space running that Natural will need. Brian -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Case TS002648607 (PMR 76523,082,000) - Compiler abend
> Why? Because it treats trailing commas as continuation chararacters. > I find Rexx pleasantly consistent there: I find REXX unpleasantly consistent there; I find, e.g., PL/I, pleasantly consistent in the use of semicolons. > Compare POSIX shell conventions. Why? The Bourne shell and its children are poster children for bad language design. I'd rather compare REXX continuation to, e.g., Algol 60, C, Pascal, PL/I, Ruby. Personally, I believe that nobody understands a language unless he can identify the flaws in it. -- Shmuel (Seymour J.) Metz http://mason.gmu.edu/~smetz3 From: IBM Mainframe Discussion List on behalf of Paul Gilmartin <000433f07816-dmarc-requ...@listserv.ua.edu> Sent: Friday, August 30, 2019 6:03 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: Case TS002648607 (PMR 76523,082,000) - Compiler abend On Fri, 30 Aug 2019 17:33:56 +, Seymour J Metz wrote: >Be careful what you ask for - you might get it. It's one of the things that I >don't like about REXX. > >ObHamlet "And make us rather bear those semicolons we have, then fly to >continuation conventions that we know not of" > ( C 'then' 'than') Why? I find Rexx pleasantly consistent there: An instruction is terminated by: o A newline not preceded by a comma o Or a semicolon Spaces are irrelevant Newline and semicolon are highly interchangeable. Compare POSIX shell conventions. Are they even documented? For example: 542 $ for I in 1 2 > do > echo $I > done OK, but: 544 $ for I in 1 2; do; echo $I; done -sh: syntax error near unexpected token `;' -- 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: SYNCSORT and STEPLIB/JOBLIB/LINKLIB issue
Adding that library second in the STEPLIB using MODS with no DD name allows the alternate version in the first STEPLIB library allows the alternate version to be loaded. A good demonstration of the danger of assumptions. Apparently the module was not fetched from the LNKLST, but was fetched from the STEPLIB. I didn't notice anyone asking if the module in question was the copy that fetching from STEPLIB would have chosen. Anyway, I'm glad you figured it out. The other "normal" case if the observation were correct, is that the LOADer is authorized and the tasklib/steplib/joblib concatenation is not authorized, in which case the taslkib/steplib/joblib would not have been eligible. 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: z/OS 2.1 to 2.4 [EXTERNAL]
Thank you to Brian and others. I am OOO next week. We are actually pretty current on EntireX. I thought I knew all SAG products, but ASM isn't one I remember. > -Original Message- > From: IBM Mainframe Discussion List On > Behalf Of Brian Westerman > Sent: Friday, August 30, 2019 10:44 PM > To: IBM-MAIN@LISTSERV.UA.EDU > Subject: Re: z/OS 2.1 to 2.4 [EXTERNAL] > > Yes, you need to have current maintenance on your old 2.1 system, but your > ability to fall back based on what you have stated in your response makes it > so that your fallback (if necessary) will not be an issue for you. Depending > on > your product mix, I think that SAG might have a simple work around for you > for allowusekeycsa(yes), and so long as you are using the Version 8 SVC > (which still supports version 7 Adabas products like cluster services and > parallel services) you will be okay for all but a small number of products. > There are some products that you will have to update but they are likely > minor in your case (older EntireX Brokers), at least they should be because > very little depends on the actual release of the broker. There are some > special PPT entries you have to perform as well that depend on your > products, but there is an SAG page that tell you all of them. > > The only problem you will hit is with Natural, which needs to be at 4.2.4 to > not need the allowuserkeycsa(yes), this will require SAG's ASM to do this, > but I think it's free. > > I have performed several upgrades to old levels of Adabas (back to version > 7.1) for just the Natural component, and they are truly dead simple, and if > done correctly won't require you to recatalog everything in the natural > libraries either (at least not manually depending on how far back you are). > Plus, again with proper planning, you have a fallback that you can implement > within a few minutes. Remember if you upgrade Natural, you still have to > install ASM. The suggested solution SAG is to install ASM Authorized Services > Manager with all versions of Natural that use the older buffer pools and want > to convert to allowuserkeycsa(no). > > Brian > > -- > 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
