Re: How many cost a cpu second?
Or, to expand on R.S.'s remarks, if your installation software charges are based on peak 4-hour MSU* average during the month and this application not only must run during the peak time but is sufficient to raise that peak, then your software license charges will go up for that month and there is a real dollar cost involved. On the other hand, if your installation is on peak 4-hour MSU billing and the application is easily scheduled for slack periods and does not contribute to the monthly peak, the CPU time used is essentially free. If you are not on peak 4-hour MSU billing and the application is scheduled to use otherwise idle CPU and I/O resources, then the physical costs are $0; but if it must be run during peak loads and contributes to forcing an earlier upgrade of hardware to handle that load, then hardware upgrade costs could be significant and should be at least partially attributed to that application. Another cost not mentioned is end-user waiting time. If cutting the resources required by the application significantly improves the turn around time to the end user, those improvements could be justified by improvements to end-user productivity, which could have other desirable side effects. JC Ewing *See other discussions on hardware-dependent relationship between CPU seconds and Millions of Service Units (MSU). On 06/12/2012 04:29 PM, Charles Mills wrote: I think you are not getting an answer -- actually as I recall you got several answers to this effect -- because the question is effectively "how long is a piece of string?" I use several LPARs for development. For various business reasons a CPU second costs me and my employer nothing. OTOH if you had a contract at a service bureau where you paid $5000 a month up to 20,000 CPU seconds and $20,000 thereafter, then you just saved $15,000 per month. If I improve my car's mileage from 20 to 30 miles per gallon how much money do I save? Well, it depends on how many miles I drive and how much I pay for gas. Charles -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Salva Carrasco Sent: Tuesday, June 12, 2012 2:13 PM To: IBM-MAIN@bama.ua.edu Subject: Re: How many cost a cpu second? Apart from the syntax error "How many" vs. "How much", My question is not having much success. In other terms: If I have a STC consuming 25,000 cpus / secs per month, and I spend two days optimizing it to reduce to 15,000. Am I gaining money or loosing time ? ... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: SMS Dataset Alloc Issue
And since no one else seems to have mentioned it yet, this whole thread is an excellent argument for NOT using a "SYS1." high level for naming installation datasets you want system-specific, like the this SCDS dataset -- instead using some unique high-level qualifier associated with the master catalog of each system, say MCTA and MCTB. Then you can simply create and maintain your installation System-B datasets from System-A by temporarily connecting the System-B MCAT as a "User Catalog" to System-A and defining "MCTB" (in System-A MCAT) as an ALIAS pointing to the System-B MCAT. With those conventions and working with datasets with an MCTB high-level from System-A, everything gets correctly pointed to the System-B MCAT without having to play any bizarre games, and utilities like ISPF 3.4 will behave more predictably as well. When System-B is up and running on its Master Catalog, there will be no ALIAS for MCTB defined in the System-B MCAT, so a search for MCTB.anything from System-B would default to the System-B Master Catalog. J C Ewing On 05/26/2012 12:12 PM, Lizette Koehler wrote: Hello, Thanks for help. But I still have once confusion is, When I am trying to delete the Data component of these datset, for which cluster part is not visible , system is not allowing me to delete. I get error " DATASET not Found". I think, I am getting this error because that particular datset is not part of current master catalog. But if Still I want to delete the data component I created, Can you please help to perform this task. Regards Saurabh Some questions 1) Is the catalog where this SYS1.SMS.SCDS is cataloged a USER CAT on the system you are trying to do the display? 2) Is the system where the SYS1.SMS.SCDS is cataloged - up or able to be IPL'D? If you have system up with MCATA and MCATB is a usercat in MCATA then you may be able to delete the dataset with the CATALOG parm in IDCAMS DEL. If you can bring up the system where MCATB is the Master Cat then you can do the changes from the running system IIRC SETSMS command can point SMS to different SCDS or ACDS datasets. So if you have a running system that you can logon, then you can fix from that system. Or could you create a new SYS1.SMS2.SCDS dataset to the correct MCAT and then update the SYS1.PARMLIB member to point to the new SCDS dataset and IPL? You may have several options to correct this issue. Lizette ... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: IEW4005I FETCH FOR MODULE Failed
On 05/12/2012 01:27 AM, Lim Ming Liang wrote: Pardon my ignorance, where can I get this Mods Tape You referring to ? Regards Lim ML On 12-May-12 11:57 AM, Skip Robinson wrote: Mods Tape ... See PDS 8.6 at http://www.cbttape.org/ These "mods" haven't actually been distributed as a mainframe tape for years, but it's still called the "CBT Tape" because for years prior to being converted to on-line and CDROM formats, versions 1 through 321 of this collection of code were managed and distributed on mainframe tape by Arnold Casinghino, who worked at Connecticut Bank and Trust Company. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: ### of GDG Entries
But,... While mod'ing onto an existing file can work, for a production process this can create a nightmare for restart after a process failure if you don't plan for process restart up front -- highly recommended to either first make a backup for recovery before each append process or copy the old file to a new and append the additional records to the new file as part of that copy process. My personal preference for a report-retention requirement like this is to come up with some technique that allows the use of variable dataset names rather than trying to work around the limitations of GDGs. Or, if all the reports have the same DS attributes and are relatively small, it may be advantageous to use a PDS or PDS/E with each report as a separate member, using dynamic allocation of a variable member name (possibly variable DS name also) from within the report writing program. With variable names, you will most likely also need your own techniques for purging reports that are no longer needed. It is not necessary to roll your own SVC99 interface to do dynamic allocation: just use facilities accessible from REXX or other scripting languages, or use a free subroutine like CBT's DYNALLOC from compiled languages. Joel C Ewing On 05/10/2012 07:41 PM, Cris Hernandez #9 wrote: does it have to be a gdg? what about mod'ing onto the file then emptying it out when it's processed? back it up with a gdg with that batch job at regular intervals suggestion. From: "Donnelly, John" To: IBM-MAIN@bama.ua.edu Sent: Thursday, May 10, 2012 5:36 PM Subject: Re: ### of GDG Entries Thankyou all John Donnelly Texas Instruments SVA 2900 Semiconductor Drive Santa Clara, CA 95051 408-721-5640 408-470-8364 Cell john.p.donne...@ti.com -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Linda Sent: Thursday, May 10, 2012 2:28 PM To: IBM-MAIN@bama.ua.edu Subject: Re: ### of GDG Entries Hi John, I have had I similar GDG 'thing'. It would be helpful to know more details... In my case, we were receiving a widely varying number of reports that we were to backup and print for our customer. We scheduled the print job to run every 8 hours. The operations staff fit the actual printing in to meet customer needs. Would that work for you? Another option might include merging the generations to another dataset. HTH, Linda Sent from my iPhone On May 10, 2012, at 1:12 PM, "Donnelly, John" wrote: We have a business application that creates literally 100s of GDGs a day; please don't ask. Is there any way to create or pretend to create a GDG base greater than 255... John Donnelly Texas Instruments SVA 2900 Semiconductor Drive Santa Clara, CA 95051 408-721-5640 408-470-8364 Cell john.p.donne...@ti.com ... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: RES: XPAF replacement
On 05/09/2012 01:40 PM, Paul Gilmartin wrote: On Wed, 9 May 2012 13:12:52 -0500, Walt Farrell wrote: On Wed, 9 May 2012 12:37:18 -0500, Joel C. Ewing wrote: Another one of those message formats that appears to Thunderbird as "garbage". I can read it fine on the ibm-main online archive, where is shows Ituriel's readable message followed by the ibm-main trailer lines, but it must be arriving in a format that ibm-main doesn't fully understand, as its re-broadcast seems to have Ituriel's message encoded as base64, followed by the ibm-main trailer lines, also encoded as a separate base64 block but with no message heading structure appropriate to the sending of two base64 blocks. I gather some Email clients may tolerate this, but Thunderbird does not and just displays the base64 encoded data. With Firefox 12.0 I see it garbled on the LISTSERV at: http://bama.ua.edu/cgi-bin/wa?A2=ind1205&L=ibm-main&P=320205 In my experience, Joel, it's often a question of -your- (that is, each recipient's) personal settings at the list server. If you query your settings, look at the header options you have. If you have anything other than FULLHDR you may find some messages that don't have enough header info to allow your client (Thunderbird) to process the message properly. -- gil I know about the FULLHDR option and did have problems with that many years back. Did a list QUERY just be be sure I didn't somehow lose that option, and confirmed I still have FULLHDR on. This appears to be something else. On my received Email, all the usual initial message headers appear to be there (which isn't the case without FULLHDR), but it's as if the message is being sent as a multi-part MIME message but with all the internal MIME headers around the parts missing. I vaguely remember reading at one time that the list server didn't support MIME and multi-part formats. If that is true and this is what is being sent to the list in the problem cases, that might be the cause and perhaps an issue with the senders Email options. Ituriel's latest 9 May @1234 and @1248 postings do show up as garbled for me on the LISTSERV archive also. I must have been looking at the 8 May posting on the archive instead, which is OK. It looks like the only posts with the problem are replies to previous postings and that his initial post that started the thread came through clean. Perhaps that is significant. The only other thing I see unusual is that Ituriel's company adds a VERY lengthy appended legal disclaimer, which contains a several html tags in a context where html is not appropriate. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: RES: XPAF replacement
Another one of those message formats that appears to Thunderbird as "garbage". I can read it fine on the ibm-main online archive, where is shows Ituriel's readable message followed by the ibm-main trailer lines, but it must be arriving in a format that ibm-main doesn't fully understand, as its re-broadcast seems to have Ituriel's message encoded as base64, followed by the ibm-main trailer lines, also encoded as a separate base64 block but with no message heading structure appropriate to the sending of two base64 blocks. I gather some Email clients may tolerate this, but Thunderbird does not and just displays the base64 encoded data. I think we've been down this path before. The original message format must be partly responsible, but this also looks like a bug in the ibm-main list server logic: it should never think it reasonable to append its trailer in a way that sends out two base64 blocks back-to back, rather than, say, trying to merge the the data into a single base64 block, or resend the whole thing un-decoded with 8-bit MIME Email conventions. JC Ewing On 05/09/2012 10:34 AM, ITURIEL DO NASCIMENTO NETO wrote: ‰íz{S©ì}êÄ�ÊŠxjǺà*... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Progress Toward z/OS Personal Use License
On 04/25/2012 09:38 AM, Farley, Peter x23353 wrote: Pardon me if I misinterpreted, but your very short responses, each followed by a period, said "each of these is an issue for me". Perhaps I need more coffee before I write such a question... :) Peter -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Paul Gilmartin Sent: Wednesday, April 25, 2012 10:30 AM To: IBM-MAIN@bama.ua.edu Subject: Re: Progress Toward z/OS Personal Use License On Wed, 25 Apr 2012 10:19:19 -0400, Farley, Peter x23353 wrote: I'll grant you the dongle issue (but it's probably unavoidable) and possibly the APAR submission issue (which can be anything from a non-issue to a business killer), but why is Linux/Intel hosting a problem to you rather than a solution? Just curious. Who used the word "problem" or "issue"? -- ... A dongle definitely could be an issue for some. Might be less of an issue on Linux, but my experiences on Windoze has been less than ideal and makes me regard any application that requires a dongle as more of a gamble. While the dongle may be regarded as "nice license insurance" from the software vendors standpoint, it is essentially just another point of failure for the user and lowers the value of the product. My wife has some very expensive Embroidery software that requires a dongle. The license does entitle her to run the software on multiple platforms, both her laptop and desktop, since the dongle prevents concurrent use. After a year or so the dongle case became too loose to remove the dongle from the USB port - the only way now is grasp and pull the dongle base with a pair of needle-nose pliers, which works, but is certainly not the advertised convenience. The only "support" provided by the application vendor to remedy this situation is to re-purchase the software at full price to get a new dongle. Other than using standard Windows GUI interfaces, this software does nothing that special at the Operating System level, except for the dongle support that requires a hardware driver written by yet a different vendor. Logic would suggest that this application should be able to migrate from Win XP to Win 7 without a problem, provided one can find support for the dongle on Win 7. My initial attempts to migrate have so far failed because the dongle vendor's current drivers for Win 7 are not compatible with the older version dongle that came with the application. I haven't given up, but unless I can locate a compatible driver that is also compatible with Win 7 this expensive application is toast on Win 7. A nice result for the application vendor if I'm forced to do an otherwise unnecessary upgrade at great cost, but from the user's standpoint this is a very poor outcome, apparently forced by the decision to require a dongle. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Progress Toward z/OS Personal Use License
On 04/25/2012 01:53 AM, Paul Gilmartin wrote: On Tue, 24 Apr 2012 23:36:19 -0700, Edward Jaffe wrote: The "march" toward a personal use z/OS license takes another step forward ... http://www.ibm.com/common/ssi/rep_ca/5/897/ENUS212-145/ENUS212-145.PDF "... Additionally, the [Rational Developer and Test Environment for System z] product can now be purchased as a stand-alone entry point into the Rational development solutions for System z. This lowers the cost of initial purchase, opening the environment for use by developers, testers, and operations personnel, and provides an easier path to adoption for traditional mainframe developers looking to modernize their development and test processes and infrastructure." Dongle. Intel/Linux hosted. No APAR support. -- gil ... Dongle, don't like but could live with. Intel/Linux hosting, not unreasonable. The big unanswered question not mentioned anywhere in the pdf document is the cost. The big problem with something like zPDF was that it still had a minimum $20K - $30K per year cost. A quick read of this latest offering suggests it still has an annual license charge per user, but if there was any clue on price range I missed it. Perhaps there are other on-line resources that clarify. Having a personal z/OS to occasionally play with would be so cool. But, the intended target here still appears to be businesses, which no doubt means it's priced accordingly and out of range for casual personal use! Speaking for those of us not in the top 1%, even $5K per year would be way more than I currently budget for all my home personal computing, so I doubt this new offering yet approaches what I could justify as an entertainment expense. I baulked at MicroSoft's concept that in response to MS's agenda, and not mine, I should be willing to shell out $100's per home platform every several years for the privilege of having to re-learn all the familiar user interfaces, force-upgrade other application software, and still expend significant resources on "protection software" -- which is why my primary home systems have been SE Linux (Fedora) for several years. A cost of $1000's per year for "cool" wouldn't fly for me. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: ZOS 1.13 SMPTABL Mystery
On 04/23/2012 04:35 PM, Shmuel Metz (Seymour J.) wrote: In<4f90d295.6000...@acm.org>, on 04/19/2012 at 10:05 PM, "Joel C. Ewing" said: If you have multiple people who might have to back each other up and be able to take over and eventually complete a maintenance project started by another using SMP/E ISPF dialogs, then they had better share the same SMPTABL dataset. Otherwise, the only way to continue their maintenance would be to manually start a new "project", determine what SYSMODS should be selected, determine the last step done by the prior person, and spin through the already-done dialog steps without actually submitting jobs in order to get to the right starting step. The flip side of that is that if you have multiple people working concurrently the variables tracking the work of one are highly inappropriate for the work of another. If I have to back up somebody who has started to install service, the CSI will tell me what I need to know. Chances are that I will receive updated HOLDDATA before doing anything else, so I may not care what he had previously selected. SMP/E dialogs do not work that way. Users do not share the same variables directly, they share the same list of "named" maintenance projects within each distinct target zone name, and variable values are associated with a specific maintenance project, not with a user (internally, SMP/E dialogs generate unique project ISPF table names in SMPTABL as needed for storing the variables). If you want a new independent project, you "start" a new project and give it a name to distinguish it and it has its own set of variables. If you want to resume/continue a previously started project (yours or someone else's), you instead select an existing project from the list and the dialog automatically picks up all the saved variables appropriate to the options and state of that project. When projects are "completed" either by advancing to the final step or by an explicit "CANCEL", the project variables and corresponding ISPF table members are discarded. It works quite well for tracking the state of multiple projects to multiple zones from one maintainer or multiple maintainers, and I have also used it to pick up and resume maintenance started by another when the other went on vacation and priority of the project was "raised" in their absence. It is especially useful when there is a relatively long lag between APPLY and ACCEPT and you want to be sure to ACCEPT the same SYSMOD set as in APPLY, because a maintenance configuration has proven to be a stable one and you want it as a potential RESTORE point. There is nothing to prevent SMP/E ISPF dialogs with a shared SMPTABL DS from correctly tracking the distinct variables of two different SysProgs that are concurrently working in ISPF on two independent projects. This is true even if both projects are within the same TZONE/DZONE; although, I wouldn't recommend that last level of concurrency as a standard practice as it complicates the issue of when affected libraries should be backed up, and potentially their batch jobs would waste initiator time with enqueue delays waiting for exclusive access to the same CSI datasets. Yes, you can deduce most of the state information from the CSI, possibly with help from the SMP/E log datasets; but it takes much more work and adds unnecessary opportunity for human error. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: ZOS 1.13 SMPTABL Mystery
On 04/19/2012 08:16 AM, Shmuel Metz (Seymour J.) wrote: In<4f8eb53d.9000...@us.ibm.com>, on 04/18/2012 at 08:36 AM, Kurt Quackenbush said: http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/gimusr51/3.6.3?SHELF=gim2bk90&DT=20110811181158 The only explanation given for "installation-wide" is "SMP/E uses this table data set to save process status information for the SYSMOD management dialogs." What problems occur[1] if you don't use a shared SMPTABL? I've run into problems when it's shared. [1] Assuming that the unshared SMPTABL is in the ISPTLIB concatenation. If you have multiple people who might have to back each other up and be able to take over and eventually complete a maintenance project started by another using SMP/E ISPF dialogs, then they had better share the same SMPTABL dataset. Otherwise, the only way to continue their maintenance would be to manually start a new "project", determine what SYSMODS should be selected, determine the last step done by the prior person, and spin through the already-done dialog steps without actually submitting jobs in order to get to the right starting step. Also, if a different SMPTABL is used, if the original person were to later use the SMP/E dialogs, it would still look like he had an incomplete project and he might attempt to complete it and attempt to run maintenance jobs that are no longer needed or appropriate. If you have multiple people applying maintenance to the same global/target/dlib zones, sharing SMPTABL may be advisable so you can be fully aware of other activity that might affect or have some impact on the same zones and libraries you are changing. If you have other people doing maintenance with the SMP/E dialogs to other global zone, if there is any chance there are prereq/coreq requirements between those zones and zones of interest to you, a shared SMPTABL is occasionally useful to make it easier to check if there is maintenance in progress to those zones. We had a small enough number of System Programmers that used the SMP/E dialogs that we just found it simpler to share the same SMPTABL among all SysProgs (and there is no reason for anyone other than a SysProg to have SMPTABL allocated). It wasn't that difficult to manually coordinate on rare occasions when library compression or expansion was required. YMMV. Occasionally one may have to clean up abandoned maintenance projects and obsolete SMPTABL members associated with abandoned target zones, but that can happen whether the SMPTABL dataset is shared or not. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: JCL help (yes, with COND)
On 04/18/2012 12:43 PM, Frank Swarbrick wrote: I have the following job (cut down to include only the relevant parts): //VAPPROC4 JOB //BACKUP1 EXEC PROC=SORTBKUP,COND=(4,LT) //EP4INEXEC PROC=EP4IN,COND=(4,LT) //E4INPRT EXEC PROC=E4INPRT // The EP4IN PROC is a vendor supplied proc as follows: //EP4IN PROC //STEP01 EXEC PGM=IEFBR14,COND=(20,LT) //STEP01A EXEC PGM=IEFBR14,COND=(20,LT) //STEP02 EXEC PGM=EPZOS,COND=(20,LT) //STEP03 EXEC PGM=IDCAMS,COND=(7,LT) //STEP04 EXEC PGM=EPTBLUPD // IF ( STEP04.RC = 3 ) THEN //STEP05 EXEC PGM=IEBGENER,COND=((7,LT),(4,LT,STEP03),(4,LT,STEP04)) // ENDIF // IF ( STEP04.RC = 0 | STEP04.RC = 4 ) THEN //STEPLAST EXEC EPLINK99 // ENDIF //EP4IN PEND //STEP06 EXEC PGM=IEBGENER,COND=(24,LT) //STEP07 EXEC PGM=EPX99,COND=(20,LT) //EP4IN PROC I've already solved my issue (which I will detail below), but I'm still unclear as to why what I had was wrong, and why my fix actually fixes it. The issue is that EP4IN.STEP06 and EP4IN.STEP07 were not being executed: IEF202I VAPPROC4 STEP06 EP4IN - STEP WAS NOT RUN BECAUSE OF CONDITION CODES IEF272I VAPPROC4 STEP06 EP4IN - STEP WAS NOT EXECUTED. IEF202I VAPPROC4 STEP07 EP4IN - STEP WAS NOT RUN BECAUSE OF CONDITION CODES IEF272I VAPPROC4 STEP07 EP4IN - STEP WAS NOT EXECUTED. STEPNAME PROCSTEPRC BACKUP1 BACKUP 00 EP4INSTEP01 00 EP4INSTEP01A 00 EP4INSTEP02 00 EP4INSTEP03 05 EP4INSTEP04 FLUSH EP4INSTEP05 FLUSH STEPLAST STEP16 FLUSH STEPLAST STEP17 FLUSH EP4INSTEP06 FLUSH EP4INSTEP07 FLUSH (It's expected that EP4IN.STEP04, EP4IN.STEP05, STEPLAST.STEP16 and STEPLAST.STEP17 do not execute under these conditions, becasue EP4IN.STEP03 resulted in RC=05; they execute only when RC=03 for this step.) Anyway, the solution is to remove the COND parameter from the "EXEC PROC=EP4IN". My new result follows: BACKUP1 BACKUP 00 EP4INSTEP01 00 EP4INSTEP01A 00 EP4INSTEP02 00 EP4INSTEP03 05 EP4INSTEP04 FLUSH EP4INSTEP05 FLUSH STEPLAST STEP16 FLUSH STEPLAST STEP17 FLUSH EP4INSTEP06 00 EP4INSTEP07 00 E4INPRT REPORT 00 What really was the issue and why did my solution resolve it? My reason for including this parameter is so that EP4IN should be bypassed if BACKUP1 fails. Once again I ponder the sanity of the inventor of COND. Thanks, Frank The EP4IN PROC seems to have an embedded PEND and then a spurious PROC statement at the end (where the PEND should be ?). I suspect the problem is that combining COND on an EXEC PROC with embedded use of COND inside a PROC does not interact well and seldom does what one might expect/want. The COND on the "EXEC PROC" statement does not determine whether that EXEC is performed -- it instead overrides the COND parameter on every EXEC within the PROC and determines when steps within the PROC will be executed, completely ignoring any original carefully thought out conditional logic using COND on EXECs within the PROC definition itself. As a side issue, mixing IF/THEN/ELSE conditional execution with use of COND is so highly confusing, we always recommended not introducing the new forms without converting old EXEC COND usage to IF/THEN/ELSE statements at the same time (except possibly for COND on the JOB statement). Also one can use IF/THEN/ELSE in the main JCL stream without the bizarre override issues you get with COND on the "EXEC PROC". -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Execute certain steps based on input parm
On 04/18/2012 10:51 AM, Paul Gilmartin wrote: On Wed, 18 Apr 2012 09:47:05 -0500, Joel C. Ewing wrote: On 04/18/2012 08:46 AM, Paul Gilmartin wrote: On Wed, 18 Apr 2012 09:30:59 -0400, Veilleux, Jon L wrote: According to the JCL manual that won't work: The following keywords are the only keywords supported by IBM and recommended for use in relational-expressions. Any other keywords, even if accepted by the system, are not intended or supported keywords. The syntax as "formally" described in the JCL Reference (z/OS 1.12) is demonstrably incomplete just based on the supplied examples. Something like // IF (&SYM = value) THEN is clearly valid by the manual's description if&SYM has a numeric value and "value" is also some numeric constant ... No. In: http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/iea2b6b0/17.1.4.2 17.1.4.2 Comparison Operators Use comparison operators in a relational-expression to compare a keyword with a numeric value. ... There is no mention of using a comparison operator to compare two numeric values. The "keywords" are listed in: http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/iea2b6b0/17.1.4.5 Note that "TRUE" and "FALSE" are not themselves keywords nor constant values; they appear only (and superfluously) in complex tokens such as "ABEND=TRUE" (which means exactly the same as "ABEND". Quine would shudder). I think I'll submit another RCF. But, in their own examples IBM uses as valid constant values things like "Unnn", "Snnn", "TRUE", "FALSE", none of which are described as These are not used as constant values, but as lexical parts of "keywords". keywords and which clearly are not numeric in the normal sense of the word! Obviously some special alphanumeric constants are acceptable, which begs the question why other arbitrary alphanumeric constants that can't be confused with keywords should not be explicitly allowed as operands as well. (Maybe they work even though undocumented, but usage in that case would be risky!) If the relational expressions directly supported by JCL are found too restrictive, one could always write a fairly trivial utility ... JCL tailoring with ISPF skeletons? -- gil ISPF skeletons would only be an alternative if dynamically submitting another batch job is acceptable. The referenced 17.1.4.2 section is very definitely in conflict with the actual IF JCL implementation. I successfully used for years such forms as IF (&OPT = 1) THEN and IF (&OPT = 0) THEN where &OPT was set on a prior SET statement to have a value of "0" or "1", to allow various job steps to be executed conditionally based on a set of "option" statements at the start of the job stream. Much easier than hunting for the job steps, and the expected results of "1 = 1" and "0 = 0" being taken as TRUE and "1 = 0" and "0 = 1" being taken as FALSE were as intuitively expected. So does IBM change the docs to agree with the implementation (the cheaper solution), or do they change the code to not conflict with the docs (which may break existing usage)? In any event, the existing JCL documentation should not be incomplete, confusing, and inconsistent with the implementation. There is a lot to be said for more formal syntax and semantic descriptions. I confess to having less familiarity with ABENDCC keyword forms, never having had a need to use one. So, "U001" in "ABENDCC=U001" is a "lexical part" of a keyword? Good Lord, who invented this syntax? So in the context of an "IF" statement relational expression, "=" is both a relational operator and also a lexical part of such key words as the above. A rational person familiar with practically any other programming language (and with other JCL relational expressions) would intuitively expect given "ABENDCC=U001" as valid, for "ABENDCC = U001", "ABENDCC EQ U001", and "ABENDCC NE U001" to all be legal syntax, which they apparently are not. It is a confusing and unnatural syntactic act to permit the same special symbol to serve as both an operator and as part of a keyword! -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Execute certain steps based on input parm
On 04/18/2012 08:46 AM, Paul Gilmartin wrote: On Wed, 18 Apr 2012 09:30:59 -0400, Veilleux, Jon L wrote: According to the JCL manual that won't work: The following keywords are the only keywords supported by IBM and recommended for use in relational-expressions. Any other keywords, even if accepted by the system, are not intended or supported keywords. Also you need to change DNS to DSN ... You will have to get more creative. Perhaps pass a parm that causes an abend in a first step? Or contrive a JCL symbol that evaluates to one of the supported forms. -- gil The syntax as "formally" described in the JCL Reference (z/OS 1.12) is demonstrably incomplete just based on the supplied examples. Something like // IF (&SYM = value) THEN is clearly valid by the manual's description if &SYM has a numeric value and "value" is also some numeric constant or if "&SYM" resolves to a legal keyword value for which "value" is compatible, as legal operands are described as keywords or numeric values, and "&SYM" in this context would not be a keyword but simply whatever value replaces it. But, in their own examples IBM uses as valid constant values things like "Unnn", "Snnn", "TRUE", "FALSE", none of which are described as keywords and which clearly are not numeric in the normal sense of the word! Obviously some special alphanumeric constants are acceptable, which begs the question why other arbitrary alphanumeric constants that can't be confused with keywords should not be explicitly allowed as operands as well. (Maybe they work even though undocumented, but usage in that case would be risky!) If the relational expressions directly supported by JCL are found too restrictive, one could always write a fairly trivial utility (perhaps CBT site already has one) that would accept as a PARM value or an input record a more general relational expression (which could include parameter references) with syntax that fully supports character string comparison and produces a 0|1 (false|true) step condition code that could be tested by subsequent conditional JCL statements. Such a utility could even be generalized to allow conditional setting of arbitrary step completion codes or even conditionally ABENDing with some user ABEND code. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: GO TO "cobol"
And of course there were languages of the time like FORTRAN, which encouraged unconstrained use of "GO TO" and the creation of "spaghetti code". FORTRAN at that point had very limited support for structured programming: one very restrictive loop construct and a conditional branch which was essentially a three-way "GO TO". There were even cases where the conventions being followed might require a "GO TO" to reach corrective additions which would then use a "GO TO" to resume the original statement stream, either to preserve statement number sequencing conventions, or to avoid resequencing the entire program deck when all cards had sequence numbers. I think more people interpreted Dijkstra's remarks as a complaint against the "unstructured" use of "GO TO" and the lack of language support for structured programming constructs that forced frequent "GO TO" use, rather than an arbitrary total ban. The complaint was primarily one about use of the "GO TO" degrading comprehension of the program structure, not about program performance. Ultimately, all code is reduced to the Assembler or machine language level, where the machine-level equivalent of the "GO TO" is essential and unavoidable. ACM SIGPLAN (Special Interest Group on Programming Languages) Notices tended to publish humor, especially near April 1. One proposal many years ago to totally eliminate the FORTRAN "GO TO" was to replace it with a "COME FROM" statement. This totally eliminated the confusion over whether any FORTRAN statement with a statement number might be the target of some unseen remote branch -- and replaced it with the even more confusing concept that every statement with a statement number might actually be a disguised branch to some remote "COME FROM" statement. JC Ewing On 04/16/2012 07:42 AM, McKown, John wrote: What??? "monopolizes the CPU"??? GO TO was made a pariah by an article by Edgar Dijkstra. http://en.wikipedia.org/wiki/Considered_harmful And, of course, management went stupid (again) and came up with "you cannot use the GOTO in any code at all". Which actually makes some COBOL more complicated due to the requirement of nesting IF statements within IF statements. And before the END-IF, that could be very complicated. I've see old code like: IF ... THEN ... IF ... THEN ... ELSE NEXT SENTENCE ... IF ... THEN ... IF ... THEN ... ELSE NEXT SENTENCE ELSE ... . Each internal IF had to have a corresponding ELSE with only NEXT SENTENCE in it. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Query about overcoming -- debunking, countering, and burying --mainframe myths
On 04/11/2012 11:12 AM, McKown, John wrote: Not legally. But I've heard rumors that it does. -- John McKown Systems Engineer IV IT Administrative Services Group HealthMarkets(r) 9151 Boulevard 26 * N. Richland Hills * TX 76010 (817) 255-3225 phone * john.mck...@healthmarkets.com * www.HealthMarkets.com Confidentiality Notice: This e-mail message may contain confidential or proprietary information. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. HealthMarkets(r) is the brand name for products underwritten and issued by the insurance subsidiaries of HealthMarkets, Inc. -The Chesapeake Life Insurance Company(r), Mid-West National Life Insurance Company of TennesseeSM and The MEGA Life and Health Insurance Company.SM -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Paul Gilmartin Sent: Wednesday, April 11, 2012 8:50 AM To: IBM-MAIN@bama.ua.edu Subject: Re: Query about overcoming -- debunking, countering, and burying --mainframe myths On Wed, 11 Apr 2012 07:04:45 -0500, Tom Marchant wrote: A renowned industry expert ... [said] the 4341 might not be compatible with the 148 Not what you are looking for, but in the late 1970's, when I was an Amdahl SE someone said that they understood that the Amdahl was IBM-compatible, but would it run Cobol? I understand that Hercules is zSeries compatible, but will it run z/OS? -- gil ... I realize it would be ridiculous from a performance and cost standpoint to do so, but as a test of Hercules functionality I have always wondered if it would be possible to run z/OS legally under Hercules if you did it under a Linux system running in an LPAR on GP processors on a z box already licensed for running z/OS natively? -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Batch utility to show DCB info for files
On 04/11/2012 11:36 AM, Bill Ashton wrote: Hello! Is there a standard IBM batch utility that can show the DCB and Space atributes for a file? I tried LISTCAT, but it didn't give me this data. I would like to generate a report for a whole list of files, so as we shift these to another location, we can have the metadata, too. Coding a Rexx or other program is not an option...the requirement is to use standard, already existing utilities. Thanks! Billy See IBM IEHLIST utility, LISTVTOC command. Will have to separately list VTOC for each DASD volume containing files of interest. Output tends to be verbose, but at least it's a standard utility. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: VSAM help wanted for random reads
On 04/04/2012 10:37 AM, Chip Grantham wrote: We have an application like this, that is multiple record types in the same KSDS. We found that if we had a FD for the type '4' records and a FD for the type '5' records (that is two DDs pointing to the same file), that each kept a separate sequence set in storage and it ran faster. You might try it. Chip Grantham | Ameritas | Sr. IT Consultant | cgrant...@ameritas.com 5900 O Street, Lincoln NE 68510 | p: 402-467-7382 | c: 402-429-3579 | f: 402-325-4030 ... Unless you have something at your installation that automatically tunes VSAM buffer allocation, some kind of manual tuning in the JCL is almost always recommended, as the default VSAM buffer allocations tend to be terrible for performance. Just specifying an BUFFNI INDEX buffer count large enough to accommodate all index levels, plus additional buffers if the access pattern has multiple localities of reference, can do wonders for random access performance, even without going to BLSR. The default used to "guarantee" that random access to any VSAM file with data in more than one CA (and hence at least two levels of index) would require re-reading CI's for all the various index levels for each data record access. Just providing a few additional index buffers in such cases might be enough to cut the physical I/O's by a significant factor. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Malicious Software Protection
w. Regards, Tom Conley On 3/27/2012 1:06 PM, David Cole wrote: At 3/27/2012 11:19 AM, Pinnacle wrote: There is a mainframe product that protects against malicious software. It's called SAF, and it interfaces with ESM's like RACF, or ACF2, or TopSecret. "SAF" is not a product. It stands for "System Access Facility" and it is nothing more than an interface within z/OS into which a security system (such as ACF2, TopSecret and any ryo security system) can plug into to receive and respond to security calls. It really has nothing to do with anti-virus protection. SAF is not a product, you're right. Please forgive my use of the term "product", I should have said "feature". I do take issue with your last sentence. SAF and an ESM have everything to do with anti-virus protection, provided they are configured to correctly protect APF-authorized resources. It [z/OS] is the only operating system out there with built-in anti-virus protection. On top of that, the hardware itself actively protects against damage through storage keys, protected memory, etc. You have to explain to the auditors that anti-virus software is not needed on z/OS, because it's intrinsic to the operating system and the hardware. I think you seriously misunderstand what a virus is... Yes, z/OS has exceptional security (and integrity and reliability) features for protecting against non-authorized programs. But I must emphasize... -->NON<--authorized programs! When it comes to AUTHORIZED programs, z/OS's integrity (which is what you are talking about with "storage keys" and such) is very good, but of course not bulletproof. Worse though, when it comes to SECURITY, there are some real problems! Because with the proper knowledge, it is TRIVIALLY EASY FOR AN AUTHORIZED PROGRAM TO SUBVERT SECURITY COMPLETELY! This is what mainframers constantly forget regarding security. For authorized programs there is no security. All that is necessary for a malicious program to do is to Trojan-horse its way (with the AC(1) attribute) into an authorized library, and you're done for! I've never forgotten this. That's why my APF-authorized libraries are severely limited in scope, and audited for any and all updates. As far as I know there is no serious anti-virus program for mainframes. I believe strongly that there needs to be one, but I don't know of one. And at this stage of the mainframe culture, I would be seriously suspicious of the efficacy of any program that claimed to be anti-virus. I don't think that a serious mainframe anti-virus program can exist unless and until IBM itself makes a commitment to support an effort to make the mainframe anti-virus proof. ... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: A z/OS Redbook Corrected - just about!
On 03/26/2012 04:20 PM, John Gilmore wrote: Yes, your pronunciation appears to be an acceptable one---'correct' is too normative a word for most linguists---for an educated-circa-2012-somewhere-in-Texas speaker. Still, I don't suppose that I will be expected to forego my usual plea for the use of the IPA instead of such expedients in these situations; and I will not. John Gilmore, Ashland, MA 01721 - USA No doubt a subtle form of humor for this group to use another overloaded acronym, "IPA". I made the mistake of getting IPA by accident once when intending beer/ale. Have never forgotten what India Pale Ale is like (yuk) and won't make that mistake again; but I'm sure in sufficient quantities it would put an end to any argument over "acceptable" or "correct" pronunciation - hence, I submit, this could be a plausible interpretation of IPA in this context -- though knowing John's posts, I'm sure the interpretation of one of the later Google links was intended. :) -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Leaving IBM
Best wishes. Your contributions over the years have been appreciated! JC Ewing On 03/26/2012 08:52 AM, Walt Farrell wrote: I mentioned this over on RACF-L the other day, so for some of you this will be old news. I've been an IBMer for 28 years and have had a lot of fun with RACF and MVS, and I've had a lot of fun interacting with you folks on RACF-L and IBM-MAIN. But the time has come for me to retire and have fun with other things. I've enjoyed the discussions here, and working with many of you to plan enhancements or resolve problems. I'm sure I'll still read both lists for awhile, and probably even participate from a personal email address. But after Wednesday morning I will no longer be an active IBM employee and I'll speak about z/OS and RACF even less officially than than I do now. It's been a great 28 years, but my family and other activities are calling to me more and more strongly, and it's time to spend more time with them. Best wishes to you all, Walt ... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Pre-Friday fun: Halon dumps and POK Resets
On 03/23/2012 04:47 PM, Shmuel Metz (Seymour J.) wrote: In , on 03/22/2012 at 01:33 PM, zMan said: Who else has stories to share? EDS, at a government facility. Halon dumps, everybody ordered out. One operator decides to be a hero and to shut down the equipment in an orderly fashion, which he did. It turns out that "non-toxic" is a relative term; he did require medical care. I don't know whether he got a commendation or a reprimand. There are a couple of other list members who were there at the time; [perhaps they recall the details. Halon had a minimum concentration of 5% to be effective and a maximum concentration of 7%, over which it starts to have toxic effects on the nervous system. When ozone-depletion concerns caused discontinuation of use of Halon, one of the replacement suppression agents was FM200, or heptafluoropropane. FM200 has a slightly broader band of useful-safe concentration, 6.25% to 9%. But, above 9% it is described as causing "cardiac senstivity", which doesn't sound like a good thing for long-term exposure either. With both of these suppression systems, if through design error the dispersal system is over-sized for the area, or if the agent isn't distributed uniformly, personnel remaining in the area could be at risk from overexposure. The recommendations for maximum exposure to both of these agents is based on the assumption you are not in a room with an active fire. These compounds break down while doing their suppression job in the presence of fire and other compounds could be released that are much more toxic -- not to mention that the typical by products of the type of electrical fire one might expect in a computer room would by themselves be toxic in a closed area. That's why one should always assume the worse in the event of a discharge and exit. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: host codepge 0037 and the obscure "not sign"
On 03/13/2012 03:11 AM, Hunkeler Peter (KIUP 4) wrote: Try quote site sbdataconn(IBM-037,ISO8859-1) before your get or put command. This tells the ftp server what translation to apply. If it help, find out what is setup at your location. Have a look at the FTP data configuration file (http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/f1a1b4a0/18. 3?SHELF=f1a1bkc0&DT=20100607113411) -- Peter Hunkeler And if appropriate, you can make the sbdataconn value shown above the FTP default for your installation in TCPPARMS. When I last checked, the default FTP translation table not only handled some characters "incorrectly" for us, but was not even a one-to-one mapping for 8-bit characters, so that FTP transmission of 8-bit text to z/OS and then back to the original non-EBCDIC system could result in changed characters from the original. In some cases you might want to specify the code set IBM-1047 rather than IBM-037. There are unfortunately [at least] two conflicting EBCDIC-based code sets in common use on z/OS: IBM-037 for applications from the 3270 tradition, IBM-1047 for applications from the C Language or UNIX tradition, mainly differing in code points for brackets, caret, and not sign. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Program FLIH backdoor - This is a criminal breach of security!
CA SYSMODs at the time. While only intended to affect SMP/E maintenance for CA products, the exit was typically globally accessible, and there was no easy way to verify that it did not have the potential to compromise SMP/E behavior for all other z/OS product maintenance. Needless to say that approach was not well received once it became known (and was discontinued). -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Writing article on telework/telecommuting
A session manager on the mainframe (CL/Supersessions or equivalent functionality) is your friend! This protects you not only from remote failures but local glitches as well. The current screen from multiple logged-in applications is maintained by the session manager and unaffected by drop outs occurring between the end user and the session manager. Applications like TSO don't see a session failure that only affects the link between the user and the session manager (unless it lasts so long that application sessions time out from inactivity). Joel C Ewing On 03/04/2012 12:25 PM, David Betten wrote: One thing I'll add to that is that if your internet service periodially drops, it's a real pain if you're connected to a host 3270 session. For example, my wife primarally does email and web browsing while working from home. So if our internet signal drops for a few minutes and then comes back, she's not likely to even notice. However, if I'm scrolling through code or a hex dump and the sevice drops for just a few seconds, it's a major headache getting loging back on and hoping my sesson reconnects to where I was. Our latest VPN client seems to offer a bit better recovery from that by maintaining the session but a few years ago it was a major headache for me. Have a nice day, Dave Betten DFSORT Development, Performance Lead IBM Corporation email: bet...@us.ibm.com DFSORT/MVSontheweb at http://www.ibm.com/storage/dfsort/ IBM Mainframe Discussion List wrote on 03/04/2012 08:49:57 AM: From: Martin Packer To: IBM-MAIN@bama.ua.edu, Date: 03/04/2012 10:01 AM Subject: Re: Writing article on telework/telecommuting Sent by: IBM Mainframe Discussion List One experience from teleworking which should appeal to mainframers: By and large 3270 is the least demanding data stream - so TSO / ISPF goes fast even on "broadband" as crummy as mine. (It's all the other junk that runs really slowly when the wet string dries out.) Now I may be in a minority but I bet this counts for lots of people. Anyhow, having telecommuted for more than 10 years I'm looking forward to this article: "You are not alone" is a useful thing to hear. :-) Cheers, Martin Martin Packer, Mainframe Performance Consultant, zChampion Worldwide Banking Center of Excellence, IBM +44-7802-245-584 email: martin_pac...@uk.ibm.com Twitter / Facebook IDs: MartinPacker Blog: https://www.ibm.com/developerworks/mydeveloperworks/blogs/MartinPacker From: Gabe Goldberg To: IBM-MAIN@bama.ua.edu, Date: 03/03/2012 21:43 Subject: Writing article on telework/telecommuting Sent by: IBM Mainframe Discussion List I'm writing article for Destination z<http://destinationz.org/ on telework/telecommuting. I think this partitions in two dimensions -- technology vs. mindset and worker vs. employer. There's abundant information -- and blather -- about this subject. But Destination z is mainframe focused so I'm especially interested in relevant System z tips for all four quadrants: technology/mindset/worker/employer. Again, this is a "tips" article so won't include positive/negative anecdotes. But they're still welcome -- they can suggest tips, they're interesting, and I might write a longer piece on this sometime. As usual, extra credit for sending to me directly (in addition to list, if you're so inclined) so I needn't pluck from digests. Thanks, as always.. -- Gabriel Goldberg, Computers and Publishing, Inc. g...@gabegold.com 3401 Silver Maple Place, Falls Church, VA 22042 (703) 204-0433 LinkedIn: http://www.linkedin.com/in/gabegoldTwitter: GabeG0 ... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: JES2 STCCLASS and TSUCLASS - documented
On 03/04/2012 11:29 AM, Bob Rutledge wrote: Bob Rutledge wrote: Andrew Metcalfe wrote: Folks Anyone know in which manual can I find the STCCLASS and TSUCLASS initialisation parameters documented please? Not in 1.12 JES2 Init & Tuning. I suspect that they are synonymous with JOBCLASS(STC) and JOBCLASS(TSU). I have inherited support for a JES2 system that has STCCLASS and TSUCLASS specified in the init deck, so just looking for definitive description. STCCLASS AUTH=ALL, BLP=YES, COMMAND=EXECUTE, MSGCLASS=W, MSGLEVEL=(1,1), REGION=512K, TIME=(60,0) TSUCLASS AUTH=ALL, BLP=YES, COMMAND=EXECUTE, MSGCLASS=W, MSGLEVEL=(1,1), REGION=512K, TIME=(60,0) JES2 Init & Tuning Guide/Rererence, SA22-7532/3 Also commands $DSTCCLASS and $DTSUCLASS don't appear to be documented. JES2 Commands, SA22-7526 Sorry. The definition isn't expressly stated. It's hinted at in the explanation for $HASP837. Bob ... The syntax for JES2 initialization parameters and commands have changed a number of times over the years, and some of those changes very definitely involved unifying extremely similar and related parameters and commands into a common syntax with an additional parameter to distinguish. I strongly suspect the STCCLASS and TSUCLASS to JOBCLASS() are an example of this, and while the old forms may still be supported for compatibility, they would not be documented if their use is deprecated, and the old forms may cease to be supported at some arbitrary point in the future. I suspect someone there ignored some JES2 migration instructions in the past. I believe the JOBCLASS parameters and corresponding commands others have mentioned are the current incarnation and the versions that should be in use rather than the obsolete forms. If the sub-parameter migration is not obvious, you may have to find a manual old enough to have documentation on the old forms (or perhaps search for STCCLASS/TSUCLASS migration information on-line). -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: IKT100I USERID CANCELED immediately after TN3270 connection fail
This sounds like a user training problem. Users should be taught in the scenario you describe to use EDIT on the new member name, within EDIT use COPY to pull in the old member contents, then do editing. The design intent of VIEW is to view without altering the original contents, and default attributes such as RECOVERY should be expected to be set OFF accordingly for VIEW. They should not be using VIEW when intending extensive modifications, but should instead be using EDIT, and have only themselves (and their trainer) to blame when they lose data using VIEW in an inappropriate context. It shouldn't take a whole lot of thought to realize that the commands are named differently for a reason, and that the names actually do convey the intended use! J. C. Ewing On 03/03/2012 10:27 AM, chen lucky wrote: YES, this is one option. But some user have the habit that do a lot of edition in view mode, then create a new member on behalf of it. So if the user can reconnect to the remain session after connection failure, this should be the best choice. thanks. 2012/3/3 retired-mainfra...@q.com If the data to be saved is from ISPF Edit, have you considered setting the Recover option for your users? - Original Message - From: "chen lucky" To: IBM-MAIN@bama.ua.edu Sent: Saturday, March 3, 2012 2:08:17 AM Subject: IKT100I USERID CANCELED immediately after TN3270 connection fail Hi List, Thanks for your help. Recently I encounter a problem that IKT100I USERID CANCELED immediately after TN3270 connection fail, and It is unacceptable in our shop, because users will lose their work that do not save in time. ... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Why _TZ put times 7 minutes off?
On 03/02/2012 09:41 PM, George Kozakos wrote: On 02/03/2012 08:36 PM, Joel C. Ewing wrote: In absence of sysplex timer or the like, the processor TOD clock is set only at POR and is set based on the HMC clock, which may in turn sync once a day with the SE clock. The SE is sync'd to the CEC TOD and the HMC is sync'd to the SE. The SE is sync'd to the CEC TOD every 24 hours, 4 hours or 1 hour depending on MCL level. The current best practice is to keep the CEC TOD accurate via an external time source which in turn keeps the SE and HMC accurate. This requires STP or sysplex timers. George Kozakos z/OS Software Service, Level 2 Supervisor The CEC sync on more recent machines had slipped my memory. On those systems, to see a significant jump in TOD time at POR, one would have to manually set the SE time, do a poor job of it, and do it close enough to POR so SE time doesn't get reset by the CEC TOD, yet far enough away (several minutes?) to be sure it propagated to HMC. Particularly now that STP is just a matter of code rather than hardware, it makes less and less sense (from the customer's viewpoint of course) for this to be a chargeable feature, which was still the case when I last checked. As long as it is a chargeable feature it is hard to cost-justify unless you are running a multi-system sysplex environment that requires it or you have some unusual requirement that really demands your TOD clock be that accurate. That it is the "best practice" and the only sure way for keeping the TOD clock accurate makes it nice to have for a number of reasons; but if management asks if it is a "must have" additional expense or a feature you can live without, in many cases the latter response must be given. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Why _TZ put times 7 minutes off?
On 03/02/2012 06:44 PM, Charles Mills wrote: And the answer from those who know is "It happened during the POR last Thursday and we're talking with IBM to figure out why a POR would do that to us." Thanks all for your patience with YATQ (yet another time question). Charles In absence of sysplex timer or the like, the processor TOD clock is set only at POR and is set based on the HMC clock, which may in turn sync once a day with the SE clock. If you don't have any procedure to sync the HMC/SE clocks to UTC or verify they are reasonably accurate when you know a POR is imminent, odds are they have drifted from reality and a POR will propagate that error to the processor TOD clock, which in turn propagates to all LPAR TOD clocks as they are activated. It would seem your operators must be setting local time explicitly at IPL (rather than using a fixed offset from TOD), or the error would have been more obvious after the IPL.My recollection is it has worked this way ever since IBM mainframes have had HMCs for control. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Originality
On 03/01/2012 10:42 AM, Clark Morris wrote: On 1 Mar 2012 08:21:46 -0800, in bit.listserv.ibm-main you wrote: 17 USC 106: "Subject to sections 107 through 122, the owner of copyright under this title has the exclusive rights to do and to authorize any of the following: (1)to reproduce the copyrighted work in copies or phonorecords; (2)to prepare derivative works based upon the copyrighted work;" quoted from http://www.law.cornell.edu/uscode/text/17/106 Is anyone here using an IBM mainframe for personal work? The rest of us work(ed) for companies which probably have legal departements. That is where this discussion belongs. Or, since mainframes are probably used in some hospitals and health insurance companies, we can also discuss the virtues and limitations of self-performed amputations. Although I am semi-retired (will take contracts) and probably long past any statute of limitations, I have used as is or modified SAMPLIB members, placing them in production and in addition I may well have put said modifications on the MICHMODS, JES3 or CBT tapes. For those of you still active, not using SAMPLIB members could be a drastic change in the way things are done. I believe the intent of those members is that they be used as templates for organizations to customize the system and that sharing can be a part of that customization. Since I am fairly certain that there are installations within IBM and other vendors that have copies of various MODS tapes and/or their successors, this probably is a non-issue. However, I'm just a bumped up DOS360 COBOL payroll programmer, not a lawyer. Clark Morris Considering what a small percentage of the totality that is z/OS is represented by SAMPLIB examples and default PARMLIB members, I would be astonished if even an anal-retentive lawyer would consider quoting with attribution in another "published" work the unmodified or modified versions of these members as anything other than "fair use" of z/OS (but then there are lawyers that surpass anal-retentive). Considering that IBM advises customers to customize these for their own use, and that it is in IBM's advantage for customers to have access to as much assistance and insight from other installations as possible in that process, it would make no sense for IBM to contest such sharing. If it were brought to IBM's attention that someone were attempting to distribute bad mods or other customization examples with the deliberate intent of tricking IBM customers into compromising their z/OS integrity, I'm sure IBM would take exception at that; but there surely are other statutes more potent than copyright infringement that could be brought to bear at that point. Joel C Ewing - Original Message - From: "Shmuel Metz (Seymour J.)" To: IBM-MAIN@bama.ua.edu Sent: Thursday, March 1, 2012 9:52:04 AM Subject: Re: Originality (was: Duplicating SYSOUT ...) In<040601ccf67d$d7ea4060$87bec120$@mcn.org>, on 02/28/2012 at 05:02 PM, Charles Mills said: Creating derivative works is a right reserved to the copyright holder. In what country? Do you have a citation for the section of the US Code that prohibits creating, as opposed to distributing or performing, a copyrighted work? ... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Stupid JCL trick?
On 02/21/2012 10:16 AM, McKown, John wrote: Do you ever want to "retire" some step(s) from a job? But you don't really want to remove the step(s) "just in case"? I don't remember this being mentioned before, so I thought I would. It will work for any step, other than the first one in the job. Find a step, any step, before the step(s) you want to bypass. Wrap the step(s) you want to bypass with: // IF (stepname.RUN=TRUE AND stepname.RUN=FALSE) THEN steps to be bypassed // ENDIF Replace "stepname" with the name of the step you selected which exists before the bypassed step(s). This works if the previous step ran or didn't run, regardless of the step's return code if it did run. Sorry if this was obvious. Maybe my brain has retired already. John McKown Systems Engineer IV IT Administrative Services Group HealthMarkets(r) 9151 Boulevard 26 * N. Richland Hills * TX 76010 (817) 255-3225 phone * john.mck...@healthmarkets.com * www.HealthMarkets.com ... But be aware there can still be some interactions with the DSNs referenced in the DDs for the skipped step. Whether a DSN enqueue is done, the type of enqueue, and how long the enqueue is held may still be influenced by the reference to the DSN in the skipped step. Also, job restart managers (ZEBB, CA-11,...) that are configured for auto delete of existing datasets at start of job if the first reference to that DSN in the job is with DISP=(NEW...) may be fooled if that first reference is in a job step that will be skipped. If you really want guaranteed zero effects from the unwanted step without complete deletion of step JCL, changing the statements to comments is the only sure way. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: O/T but curious (Re: Archaic allocation in JCL (Was: Physical record size query) )
On 02/19/2012 11:40 AM, Shmuel Metz (Seymour J.) wrote: In , on 02/18/2012 at 07:06 PM, Mike Schwab said: Neither Windows or Linux have a Catalog concept to find a dataset on What do you think a directory is? Under Windows, a directory is closer functionally to the MVS/DOS concept of a VTOC, as each volume has its own directory and you have to somehow know which volume to consult -- although admittedly in a windows system the number of volumes is typically very low. In Linux, if all volumes are mounted, the directory plays a similar functional role to that of the MVS catalog(s) and VTOCs combined. But in either case they are obviously structurally different: finding an file entry in Windows or Linux requires a progressive search through multiple directory levels rather than just a single lookup of the full path name as with a data set name in an MVS catalog. And in both Windows and Linux, in many cases the user thinks of a file by its file name and not its full path, and the onus in on the user to remember under what directory the file was placed. That issue does not arise in MVS because dataset names are always referenced by the full name -- roughly the equivalent to always requiring the full path name in Win/Linux -- and that makes direct lookup in a "catalog" possible. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Batch process VS Started task
The described record arrival rate averages at just under 58 records per second. Somehow I don't think XBM was designed with that high of transaction arrival rate in mind, processing each record as a separate transaction. I would think that a high-speed transaction processing environment (CICS, IMS) would do better than XBM, although those would probably still involve more overhead than a specialized transaction-processing STC optimized just for your own peculiar transactions (but I agree a separate specialized STC would be more of a maintenance headache). There is also an intermediate approach where you collect records by some means and then periodically fire off a process or transaction to process records collected since the last time the process was run. Another consideration: If the current batch processing is done at a time of lower system system load and involves significant work per record, moving this processing into peak processing times could have significant impact on your peak MSU requirement and a significant impact on your costs. I suspect your record arrival rate is not a constant throughout the day but also has its peaks. If those peak record arrival rates actually correspond to current periods of peak system load, then the impact on the peak system MSU requirements of moving this load would be even greater than one would expect from the average record rate alone. In other words, if there is a perceived business need to process the records in a more timely manner, be sure those footing the bill are aware it may not be a "free lunch". JC Ewing On 02/19/2012 04:04 PM, Ed Gould wrote: Magen: Not sure if this is what you are looking for but There is a facility called "execution batch monitoring" (at least in JES2) You "feed" it input and it creates output (to your specifications). It was originally designed for batch compiles but it could be adapted to something like you want (?). Ed On Feb 19, 2012, at 3:25 PM, Magen Margalit wrote: Hi list. We have a daily betch job that is processing as input records which has been collected all day. volume of records is about 5 millions for 24 hours. In order to make systems more "online" we are looking for a way to run the process for each record all day long instead of a daily run, and doing so with minimum as possible application changes. One idea that came up is to convert the process to a "self developed" STC which will be triggered by a record on an MQ queue and will run as STC all the batch process programs To me it seems like a bad idea because having a "self developed" STC in production create a maintenance gap (and where there is one STC a second one will soon to follow...)... Are there other advantages / dis-advantages regarding a "self developed" STC ? Are there any "self developed" STC's at your shop? Any other ideas on how to approach this issue? Thanks in advanced. Magen -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Archaic allocation in JCL (Was: Physical record size query)
On 02/13/2012 04:26 PM, Paul Gilmartin wrote: On Mon, 13 Feb 2012 16:11:53 -0600, Joel C. Ewing wrote: (I'd prefer, for legibility, SPACE=(1,54.000.000.000) with European thousands separators.) Not sure about your reference to half-word chunks. Although there are 16Mi limits on max numerical value for primary-qty parameter, "SPACE=(1,54000),AVGREC=M" is perfectly legitimate for allocating 54,000 MiB (only about 5% high if you really needed exactly 54,000 MB) ... I stand corrected; "24-bit" rather than "halfword". But, still, why can't the computer perform the algebra for me? And what is the etymology of "AVGREC"? It seems to suggest "Average Record Size"; it means nothing of the sort. (Or, even, "SPACE=(1,54000M)", or even, "SPACE=(1,54G)"?) (Is there a default for the SPACE unit? If not, I'd suggest "1", as in "SPACE=(,540)".) -- gil The keyword creator was apparently thinking about the intended changed interpretation to the SPACE parameter caused by the presence of the AVGREC parameter and not about the meaning of the AVGREC value itself, which is inconsistent with the way all other parameter keywords appear to have been chosen, and makes even less sense when one uses "1" instead of avgreclen for the first SPACE sub parameter. I also agree it would be more intuitive if direct suffixes such as B, KB, MB had been used in the quantity values instead of the separate AVGREC - and while they're adding that enhancement to JCL have the suffixes represent the correct powers of 10. When having an application programmer or an end-user discuss implementation design limits on number of entries or number of records, I never encountered anyone said 10 thousand and meant 10,240 rather than 10,000!! -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Archaic allocation in JCL (Was: Physical record size query)
On 02/13/2012 02:43 PM, Paul Gilmartin wrote: On Mon, 13 Feb 2012 11:38:44 -0600, Joel C. Ewing wrote: It should be possible to just specify data set limits in terms of data bytes expected or records/average-record-length expected without regard for tracks, cylinders, extents, or volumes. ... And the user interface should be simplified. I should be able to code SPACE=(1,540) and let SDB infer an average block size and allocation spare me the algebra of factoring the total space into halfword chunks. This might require a new alternative TU: a 64-bit (for future growth) extent size in bytes. (I'd prefer, for legibility, SPACE=(1,54.000.000.000) with European thousands separators.) -- gil Not sure about your reference to half-word chunks. Although there are 16Mi limits on max numerical value for primary-qty parameter, "SPACE=(1,54000),AVGREC=M" is perfectly legitimate for allocating 54,000 MiB (only about 5% high if you really needed exactly 54,000 MB) for a sequential data set, provided you have a DATACLAS that also specifies EXTENDED, Storage Constraint Relief, and a high enough volume count. Allocation will spread the dataset over as many volumes as necessary and up to as large a number (127?) of extents per volume as necessary to allocate the requested total space. Although the first value in SPACE is described as average record length, it is really only used as a multiplier for primary-qty or secondary-qty and AVGREC for calculating total data bytes needed, under the assumption the actual BLKSIZE will give efficient track utilization. The "gotcha" used to be that if you grossly over-requested space, got space dispersed over umpteen volumes, only used a little of the space, that "RLSE" would then only release the unused space on the last volume actually written and leave all the unneeded, unused space on subsequent volumes allocated until the data set was deleted. One would hope that issue would eventually be resolved and the concept would be extended to data set types other than DSORG=PS. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Archaic allocation in JCL (Was: Physical record size query)
On 02/13/2012 09:19 AM, Chris Craddock wrote: On Mon, Feb 13, 2012 at 8:56 AM, Paul Gilmartinwrote: On Mon, 13 Feb 2012 07:21:11 -0600, McKown, John wrote: Or, as the programmers at our shop would do: SPACE=EAT-EVERYTHING-IN-SIGHT-AND-CAUSE-OTHER-JOBS-TO-ABEND-BECAUSE-MY-STUFF-IS-IMPORTANT-AND-YOUR-STUFF-ISNT. In many other systems, such as Winblows, everybody gets their own personal "space". And if it is "used up", it doesn't impact others. z/OS shares DASD space. ... The z/OS cultural norm for HFS and zFS is to give each user a dedicated filesystem for HOME. This is similar to the behavior of personal instances of those "other systems". I think it is fair to say that JCL and space management are areas where z/OS truly is archaic. The "other" world manages to get by just fine without having to figure out how much resource to give. There's no reason z/OS couldn't do the same other than slavish adherence to legacy. IMHO it is about time the system itself took care answering its own incessant "how big?", "how many?", "how often?" questions. It's 2012 ferpetesakes. I'm all in favor of making sure that existing applications continue to work. I am far less impressed with continuing to impose 1960s thinking on new ones. Requiring application programmers to think in terms of tracks and cylinders and to understand interaction between physical block size and track capacity is indeed archaic, as are artificial restrictions on number of extents or volumes. Prior to emulated DASD and large DASD cache sizes, space/allocation sensitivity to tracks and cylinders was frequently necessary for performance reasons, but that is no longer the case. It should be possible to just specify data set limits in terms of data bytes expected or records/average-record-length expected without regard for tracks, cylinders, extents, or volumes. And given some simple mechanism for specifying such limits, z/OS should also provide support for monitoring whether application growth is causing data sets to be at risk of exceeding their limit. Restricting sequential data set allocation to record allocation of SMS Extended Sequential data sets with space constraint relief and SD block size comes close, but is an incomplete solution and only works for sequential files. The MVS allocation strategy, which generally requires dynamic secondary extensions to data sets when the size exceeds what can reliably be obtained on a single volume, has always been flawed. Even when the exact size of a large data set was known in advance, there was never a guarantee that space for required secondary extensions would be available on the selected volumes. In effect there was no easy way to convey to z/OS via primary/secondary specifications what the true limit of the data set should be because the actual maximum number of secondary allocations was always an unknown, with no guarantee at the beginning of step execution that even one dynamic secondary could be allocated on any of the chosen volumes. Perhaps an awareness of total data bytes involved in a data set is non essential for data sets below some (installation-dependent) total-byte threshold; but at some point for larger data sets those developing the batch application should have an awareness of approximate data set size and records involved so that concurrent space requirements for a job step may be at least approximated up front; and so application programmers don't choose an approach that might be appropriate for a toy application of 1000 records but totally inappropriate for a production application with a million records. If a required batch application is going to consume a significant percentage of the total DASD farm, there also needs to be some means for awareness of that, as it will impact job scheduling and capacity planning. The z/OS fixation on requiring data set SPACE specification for allocation rather than using some totally dynamic approach is no doubt an outgrowth of the desire for MVS to reliably support unattended batch and, as others have mentioned, to prevent one looping batch job from causing termination and denial-of-service of other unrelated jobs by exhausting available DASD space. Properly designed JCL SPACE parameters (which admittedly takes some effort) can also ABEND a batch job step up front if sufficient DASD space does not exist for successful completion -- much more desirable than allowing a batch job step to run for hours and consume valuable resources, and then blow up because space for further secondary allocation is unavailable. Operating systems that don't require space estimates for large file allocation are implicitly saying that reliable running of unattended "batch" processes is of lesser importance. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -
Re: Turning on additional CPs
But would you really expect an effect of more than a few percentage points in the efficiency of one CP from only going from 3 to 4 engines? If you are seeing a large difference in run time, perhaps you should also look for an explanation that can produce a much larger difference than the MP effect. Maybe by removing a CPU bottleneck you have moved your major system constraint elsewhere, either to real memory or to DASD throughput, or some logical interlock. Perhaps the longer running jobs are now doing significant paging because of greater contention for real memory, or are having to wait on physical I/O to DASD more -- because other things that used to be too starved for CPU to compete are now running and using resources other than CPU that used to be more plentiful in a CPU-starved environment. JC Ewing On 02/09/2012 12:04 PM, Hal Merritt wrote: I suppose that is reasonable for a single threaded, CPU bound job as a little is lost from each engine as another is added. However, you should be able to run more concurrent work giving a better over all through put. Another benefit of another engine is that, if not needed for anything else, z/os likes to direct I/O interupts to just one engine. This allows the other engines run a little smoother and again should increase your overall thoughput. -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of gsg Sent: Thursday, February 09, 2012 10:39 AM To: IBM-MAIN@bama.ua.edu Subject: Turning on additional CPs We have a z10 with 4 engines. Since upgrading to this box, we were only running 3 engines. However, we recently turned on the 4th engine. We noticed that several jobs started running longer, which we didn't expect. Could turning on additional engines actually make a job run longer? Also, where can I find any read material on the affect of turning on/off engines. TIA ... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: WLM Capping
One of the subtle misconceptions in the design of WLM was the implicit assumption that you would always have enough hardware resources to do the "required" workload with the "required" response time, and if not would quickly remedy the situation. In the real world, this is not always the case and "sharing the pain" is at times more acceptable than an immediate expenditure to add horsepower (even if it's just a logical change that increases software cost). I would submit that in the business world, while there are frequently workloads that are discretionary in the sense that they may be delayed, most of these are not "discretionary" in the sense that they may be delayed indefinitely or have no required completion window. In a system that is approaching or at saturation, one finds that typical WLM definitions effectively convert workloads that to the user were "discretionary" as to WHEN they might be run into workloads that are not run at all, which is rarely the intent. WLM now has better tools than initially for defining "loved ones" that are "conditionally loved" and which can better address some of these situations. When financial or political considerations force you to periodically run near system saturation for an extended time, you will invariably find that some of the workload priorities have to be rethought to allow discretionary, but non-optional, work to complete within required windows under that environment. Conversely, if you rarely have to run in a resource constrained environment, it probably doesn't make sense to expend the effort to distinguish among discretionary workloads that are truly discretionary and those that are non-optional and only discretionary up to a point (when that point is never being crossed). JC Ewing On 02/08/2012 03:15 AM, Martin Packer wrote: So that told you some of your batch WASN'T (in business terms) truly discretionary. Glad you (by the sound of it) pulled the stuff that mattered if it never ran out of SYSOTHER. Martin Martin Packer, Mainframe Performance Consultant, zChampion Worldwide Banking Center of Excellence, IBM +44-7802-245-584 email: martin_pac...@uk.ibm.com Twitter / Facebook IDs: MartinPacker Blog: https://www.ibm.com/developerworks/mydeveloperworks/blogs/MartinPacker From: David Andrews To: IBM-MAIN@bama.ua.edu, Date: 08/02/2012 00:36 Subject: Re: WLM Capping Sent by: IBM Mainframe Discussion List On Tue, 2012-02-07 at 15:51 -0500, Gibney, Dave wrote: I don't want to imagine what WLM stomping on the brakes looks like in your shop. Biggest hassle for me when I started softcapping was that most of my batch had been discretionary - I always liked the MTTW algorithm. But when we softcapped all that discretionary workload went to the meat locker, and we couldn't have that. Had to do some triage and creative stuff with velocity goals and performance periods to make things right again. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Physical record size query
Under z/OS, "device type" implies the architecture and usage protocols for the device. You can't call a device a 3390 and not have it accurately emulate the characteristics of a 3390, nor can z/OS communicate with a device defined to it as a 3390 as if it were something else and expect success. The blocks per track, tracks per cylinder, and CKD characteristics of a 3390 were fixed when real 3390's were created and are indeed embedded within the z/OS understanding of what is a 3390. To change that definition after the fact would break all sorts of things which currently depend on knowing how to calculate cylinder/track/record locations of internal blocks in a dataset and allocation requirements of a dataset, and data capacity of a track. The only way to change those rules without breaking existing data requires defining a new device type to z/OS, connecting devices of that type to z/OS, and placing datasets on those devices. In the old days before emulated DASD, DASD upgrades not infrequently involved changes in device types and migration/copying of data to convert it to the new device-type rules. Migration could involve considerable effort, and in some cases one would even find applications that wouldn't work on the new architecture because of some embedded assumptions about maximum records per track or cylinder that were violated on higher capacity devices. When emulated DASD became the norm, freezing the device type at 3390 eliminated the need for non-productive DASD device-type migration activities, even if it did also preserve the peculiar block-per-track rules of the 3390. Emulating the 3390 architecture does not necessarily require throwing away "overhead" bytes on physical DASD. Some emulations (e.g. Iceberg RVA) not only compressed the logical data but didn't even store "empty" tracks. This did however complicate things in other areas and create other performance issues, which is why cheaper physical DASD seems to have encouraged fixed mappings between logical tracks and physical DASD, even though that means track "overhead" bytes represent lost capacity. Perhaps in light of the FBA architecture that typically underlies the today's emulated 3390 DASD it might make sense to consider a new emulated DASD device type that is somewhat similar to a 3390 but with no "overhead" bytes lost between emulated physical blocks. This would perhaps be "cleaner" than a permanent 3390 solution, but it would have to be perceived to have enough benefit to justify the migration effort. My guess is that we have stuck with the 3390 type for so long precisely because so far no one has been able to cost-justify such a change. My personal long-term DASD ideal would be some kind of new DASD architecture that would require one to think only in terms of the logical data, not in terms of tracks, cylinders, or even FBA-device blocks; but this would be a massive change of many things in z/OS including the internal structure of a number of dataset types, also difficult to cost-justify. JC Ewing On 02/07/2012 10:01 AM, Dana Mitchell wrote: That leads to a question that I've been thinking about for some time. Since the 3390 geometry is emulated by modern storage control units, why then are the inefficiencies of small blocks emulated also? There are not SLEDs actually storing the data, why are IBG's, sectors, and all the other CKD nastiness emulated that makes 80 byte blocks such a bad idea? IOW, why can't the control unit simply store 708 * 80 byte blocks on a 56,664 byte 3390 track? Does zos's calculations take these inefficiencies into account and only write 78 of these blocks per track? Inquiring minds want to know Dana On Tue, 7 Feb 2012 16:28:33 +0100, R.S. wrote: RAID has very little to do with half-track blocks. Nowadays 3390's are emulated, usually on RAID protected disk arrays (*), but the 3390 geometry remains unchanged from z/OS point of view. So half-track blocks (**) are still the most effective in terms of storage utilisation and performance. ... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: different tape media for ML2 copies in HSM
On 01/27/2012 10:42 AM, Judith Nelson wrote: Hi Don, I wonder if CopyCat works like that as well. I know it stacks many logical tapes to the 3590's, but I will have to check to see if it would update the MCDS as well. Thank you, Judith ... Very seriously doubt it. CopyCat has hooks into CA Tape Management systems, but I don't remember any hooks into dfhsm. Trying to replicate dfhsm TAPEREPL functionality would take some serious dfhsm CDS record twiddling - not really sure I would want a non-IBM copy utility playing at that level. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: MCDS Dataset Help
On 01/26/2012 01:30 PM, George Rodriguez wrote: I'm running the export/import process almost daily. It used to run once a week. * * *George Rodriguez* *Specialist II - IT Solutions* *Application Support / Quality Assurance* *PX - 47652* *(561) 357-7652 (office)* *(561) 707-3496 (mobile)* *School District of Palm Beach County* *3348 Forest Hill Blvd.* *Room B-251* *West Palm Beach, FL. 33406-5869* *Florida's Only A-Rated Urban District For Seven Consecutive Years* On Thu, Jan 26, 2012 at 1:19 PM, Schwarz, Barry A< barry.a.schw...@boeing.com> wrote: -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of George Rodriguez Sent: Thursday, January 26, 2012 8:18 AM To: IBM-MAIN@bama.ua.edu Subject: MCDS Dataset Help Hi MVSListerv, I'm confused about the following information, that's display from command F DFSMShsm,F CDS: ARC0101I QUERY CONTROLDATASETS COMMAND STARTING ON ARC0101I (CONT.) HOST=1 ARC0947I CDS SERIALIZATION TECHNIQUE IS RESERVE ARC0148I MCDS TOTAL SPACE=648000 K-BYTES, CURRENTLY ARC0148I (CONT.) ABOUT 81% FULL, WARNING THRESHOLD=80%, TOTAL ARC0148I (CONT.) FREESPACE=49%, EA=NO, CANDIDATE VOLUMES=0 ARC0948I MCDS INDEX TOTAL SPACE=0010237 K-BYTES, ARC0948I (CONT.) CURRENTLY ABOUT 025% FULL, WARNING THRESHOLD=080%, ARC0948I (CONT.) CANDIDATE VOLUMES=0 snip I guess what has me confused is the 81% full with the 49% freespace... Makes no sense to me! Read the description of the message. The % full is based on the last used RBA while the % free includes all the space after that point PLUS any unused space before that point. It is similar to the situation in a PDS where new data will always be added at the end but there can be gas in the interior. Can someone tell me how to fix this problem? What problem do you think exists? Chapter 3 of the HSM Implementation and Customization Guide has a section on monitoring the CDSs. It tells you how to reorganize them if you want to recover the unusable free space. ... Home of Florida's first LEED Gold Certified School Under Florida law, e-mail addresses are public records. If you do not want your e-mail address released in response to a public records request, do not send electronic mail to this entity. Instead, contact this office by phone or in writing. ... You don't necessarily have to reorganize as soon as the threshold warning occurs, it all depends on growth rate of %full. Check the %full immediately after reorganize and then watch growth pattern. It will grow most rapidly the first day, then slow down as CA/CI splits build up in the most active parts of the MCDS. If it slows down enough that you are still below 90% by end of a week (there's nothing magic about 90% either if growth is slow enough), you choices are either to raise the threshold so it won't complain for a week, or increase the size of the MCDS by a large enough ratio so that the inverse ratio applied to the %full after one week would put that value below your 80% threshold. As the size of the MCDS gets larger with time, odds are a smaller percentage of records will change in the course of a week and larger %full thresholds may be appropriate. You can also track the MCDS %full after reorganize to get some idea of the actual long-term data growth. I always had enough stuff to do without worrying about dfhsm CDS's, so my goal was to be able to ignore them except for once or twice a year: based on empirical data, size them so the initial unused space is adequate for the desired interval with a month or so of fudge factor, and then set the threshold limit to warn you when you reach your fudge factor. And if you over-estimate size, you can just ignore them for a longer interval. Just a nitpick about your VSAM cluster definitions in general: (1) explicitly specifying "NAME" for DATA and CLUSTER components when you want the standard default ".DATA" ".INDEX" suffixes has been pointless for decades, just more stuff to mistype or forget to update; (2) specifying "SPACE" for an INDEX component is redundant because IDCAMS should be able to calculate exactly what it needs based on the number of CA's in your data SPACE definition and explicit values tend invariably to be gross over estimates (e.g.,less than 5 cyls actually needed for your particular 900 CA file on a 3390); and (3) changes to VSAM INDEX CISZ default calculation in the last decade made the rare occurrence where the default was too small into an extremely rare occurrence. INDEX CISZ really shouldn't be specified any more unless you have a known case where the default has been proved too small and part of each CA is unusable, or some explicit requirement for a specific INDEX CISZ is built into an application (which seems unreasonable to me since no application code should be
Re: 3390s on SAN?
On 01/26/2012 09:52 AM, O'Brien, David W. (NIH/CIT) [C] wrote: There is an internal proposal to carve several TB of dasd from one of our non-mainframe depts. And use it to replace our aging HDS DASD. Question: How easy/difficult is this to accomplish? We re-configured an array from 3390 mod-3s to mod 27/50s but the entire array needed to be cleared of data. I'm assuming the same will be true in this case. I'm also assuming that the disks will need to be re-modeled (I am probably not using the correct terminology) to be mainframe compliant. Are my assumptions correct? Thank You, Dave O'Brien NIH Contractor With nothing having been said about the existing type of non-mainframe SAN storage... It's not just a question of raw DASD storage, but whether the existing SAN hardware has the smarts to drive mainframe ESCON or FICON channel interfaces and whether it is able to support the 3990 controller and 3390 CKD disk device geometries and protocols, as these are all radically different from typical non-mainframe disk interfaces. Assuming that the hardware can support these requirements, if it was never originally configured with a mainframe in mind the odds are it will not have any mainframe channel interfaces installed, and at a minimum some hardware upgrade will be required for that. I'm pretty sure IBM disk subsystems that support mix of SAN and mainframe storage require each entire physical array to be allocated to only one of those functions. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: CU Resources Exceeded
On 01/25/2012 12:26 PM, Mike Schwab wrote: The error message appears on printed page 283 in http://www.redbooks.ibm.com/redbooks/pdfs/sg246497.pdf and some diagnostic options are displayed. You might have something defined wrong in the FICON connectors, or cables switched. On Wed, Jan 25, 2012 at 11:44 AM, SUBSCRIBE IBM-MAIN Tom Trainor wrote: After adding devices and two (2) LPARS to IOCDS on 2066, unable to IPL the last two(2) of ten (10) LPARS. The last two (2) are NOT necessarily LPAR numbers "9" and "A" but are the last two (2) of the ten (10) that are activated. Within "Channel Problem Determintation" on 2066 SE the message in the "Analyze Serial Link Status" is: "CU Resources Exceeded - Init Failure - No Resources Available". Has anyone seen this or know what the problem might be. This almost sounds like you might be exceeding the maximum number of logical paths supported by your control unit. Check the control unit documentation for its limit. Each channel interface on the CU requires one logical path for each active LPAR sharing that physical path on a connected processor. Only active LPARs play, so you would always see any problem shift to the last LPARs to be activated. For example, If you have 16 physical paths from the processor to the control unit and share all those physical paths with each LPAR, then each LPAR you activate using those paths requires another 16 logical paths to the CU. Once you reach the supported logical path max of the CU, no more paths can be configured on line to that LPAR or any other LPAR. 8x16 = 128, so if the limit happened to be 128, with 16 logical paths/LPAR none would be left after 8 LPARs were active. If that's your problem, the only choice is to either configure fewer paths per LPAR so you can run more LPARs, or maybe upgrade to a CU that supports more logical paths, or change from ESCON to FICON so you don't need as many paths for same data bandwidth. I think you can release a logical path from an active LPAR by dynamically configuring the physical CHPID offline to that LPAR. I doubt if varying the logical path offline is sufficient. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Going from mod-3 to mod9
On 01/24/2012 07:28 AM, Dennis McCarthy wrote: Hi Joel, We do NOT have any PAV's. We are a pretty small shop. One production LPAR and one test (sandbox for me). The VSAM file in question is open to a single CICS region. Give that additional information, can I expect a negative impact on response time going to the MOD-9's? Dennis You have the potential for a negative impact, but it all depends on the related transaction rates and physical I/O rates in the CICS region. If you have RMF or some other measurement tool that will give average device busy on the current 3330-3's used by the file, you can get some idea in the extreme cases: If average device busy is consistently below 5%, then I would think odds are pretty good that the negative impact would be in acceptable ranges; if you frequently see some average device busy of 20% or higher on multiple drives I would consider the odds high for significant negative impact. In between, things are less clear. RMF device measurement interval relative to arrival pattern of transactions may be an issue -- if transactions are typically clustered in a small part of the RMF measurement interval rather than uniformly distributed, you may need a smaller interval to see if device usage could be a problem. If you have CICS measurement tools of some kind, you may be able to see other things of use, like number of logical I/Os against the file that don't even result in physical I/O because of in-memory buffers in CICS, and proportion of reads versus writes (which must eventually do a physical write). With the major load coming from CICS, one of the possibilities if you have real memory to spare is that you may be able to compensate for any bottlenecks in physical I/O by throwing a large number of additional LSR buffers at the file to raise the odds of finding records in memory and reducing the number of logical reads that require a physical read. Considering the size of the file, you obviously can't have a significant amount of the file in buffers, so the success of this strategy would depend on there being some pattern of clustering in the way records are typically accessed. So, if you're lucky, you may be in a zone where rule of thumb may suggest a Yea or Nay; otherwise, there may be no simple way to determine without taking more risk and trying it. Assuming your device busy rates are in the grey zone, if real memory and adding LSR buffers is an option and I had what I considered really valid reasons why I really want to get to the 3390-9, I think I would try a temporary experiment to significantly increase LSR buffers and see if that significantly reduces I/O rate to the file. (You wouldn't want to run in this mode too long on 3390-3's, because if it does help response, the end users could get used to it and expect it). This way you could at least prove in advance whether an LSR buffer increase could be used to offset a negative impact if the migration to 3390-9 has more impact than acceptable -- and that might impact your decision on whether to attempt the migration. That's the long answer. The short answer is as always "it depends". -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Going from mod-3 to mod9
On 01/23/2012 01:21 PM, Bill Fairchild wrote: In<45e5f2f45d7878458ee5ca679697335502e25...@usdaexch01.kbm1.loc>, on 01/23/2012 at 09:08 AM, "Staller, Allan" said: From the viewpoint of the Operating System, you now have 3 times as much data behind the actuator on Mod-9's as Mod-3's. If the Operating system *thinks* the device is busy, the IO is queued off the UCB and never even tried until it comes to head of queue. You MIGHT have up to three times as much data behind the actuator. That depends on how fully loaded the three mod-3s are which are to be merged onto the same single mod-9; i.e., it depends on which three mod-3s you choose to merge together. If all data sets on all volumes are equally and randomly accessed, then you will have three times as much requirement to access the new mod-9 as any of the three mod-3s had which were merged. However, most data centers have highly skewed access patterns. 80% of the actuators might have only 20% of the total I/O workload. Which means your volumes are almost certainly NOT equally and randomly accessed. You have some volumes that are almost never accessed and some others that are accessed all the time. When z/OS starts in I/O on DASD device , z/OS turns on a flag bit in the UCB for that device that indicates that this particular z/OS image has started an I/O on that device. But if the device is shared, then another z/OS image may have already started an I/O on the same device, turned that same device's UCB flag bit on in its copy of the UCB for the device (which might be device on the other image), and not informed any of the other sharing z/OS images that it is now doing I/O on that shared device. So when image A tests its private copy of the flag bit and finds it off, that does not necessarily mean that the device is unbusy. Image A doesn't care, however. It starts the I/O and turns the bit on. If the shared control unit attached to this device is not an IBM 2105 SHARK (vintage ca. 2000), plug-compatible equivalent, or some successor technology, then image A's I/O will not really be started until image B's already started I/O ends. This will show up on im a! ge A as a spike in device pending time, not in IOSQ time. The 2105 and newer technology have the ability to let multiple I/O requests from multiple sharing systems run simultaneously against the same device as long as there is no conflict between any of the simultaneous I/Os involving both reads and writes for the same range of tracks. The only way to know what will probably happen is to do I/O measurement on your current mod-3 workload. If you don't see much IOSQ time now, then you will see "not much" multiplied by three after merging. How much is not much and/or is negligible is up to you to decide. You might also get an idea as to how to merge volumes together based on their individual IOSQ times; e.g., merge the one with the highest IOSQ time now with the two mod-3s that now have the lowest average IOSQ times. After merging them, measure again for IOSQ time. Only if you have "excessive" IOSQ time, where how much is excessive is up to you to decide, would you need to consider using PAV devices. Currently z/OS's I/O Supervisor has no knowledge of the real RAID architecture backing the virtual SLED, so many of the classic performance- and space-related bottlenecks can theoretically still occur. Bill Fairchild Note the original question from Dennis McCarthy (Jan 20) was not an arbitrary 3390-3 to 3390-9 migration but specifically moving a VSAM file occupying 27 3390-3's to 10 3390-9's, so except for last volume we ARE definitely talking about three times the data behind a logical volume, but the usage and activity rate of the dataset were not specified. IOSQ time and related response time elongation is highly non-linear as device utilization approaches 100%. You could see negligible IOSQ time on each of three 3390-3's running at 34% utilization become astronomical if you merge data from those to a single 3390-9 without PAV, trying to run a load that can't even be satisfied at 100% device busy. Given PAVs and assuming there is enough cache and different physical drives and internal bandwidth on the EMC backing the logical volume, you can in effect exceed 100% logical volume busy (have average number of active I/Os to the volume exceed 1.00) and still get acceptable response. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Going from mod-3 to mod9
On 01/23/2012 10:09 AM, Staller, Allan wrote: Generally speaking...Yes IOW, the response time is more likely to increase than to decrease, and increased response time is a Bad Thing PAV, Cache, and RAID all have impacts that mitigate the 1:3 (or worse) actuator to data ratio on mod 9 vs. 3 mod 3. These impacts may not be part of the intended design. As with most things, YMMV. Few large datasets are less impacted than many small datasets. ... With PAV, if you use PAVs to give the same total number of addresses to the mod 9 as were previously on the three mod-3's it replaces, you could conceivably even get better response from lower IOSQ time from the merging of available PAVs of three volumes into a single pool (unless you are already getting this benefit from Hiper-PAV support) Without appropriately configured PAV, the result is an unknown. Depending on how heavy the concurrent loading of the original mod-3's, you could have anywhere from severe IOSQ delays and response problems to inconsequential IOSQ delays and minimal increase in response time. While it may be easier to conceive of many concurrent users on a volume with many small datasets, large multi-volume sequential datasets can easily have high concurrent usage on different volumes from readers in multiple critical batch workloads; and large VSAM datasets, as in this case, could easily have a large amount of concurrent access on multiple volumes from concurrent on-line transactions and/or from multiple batch workloads. In such cases, merging the independent loads from three drives to one without adequate PAV addresses could easily introduce significant to severe device busy usage and IOSQ delay times. Without knowing something about the concurrent 3390-3 volume usage patterns, response time on 3390-9 without PAVs is a crapshoot even with only a single large VSAM dataset. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: DFHSM: Backup tape lost due to errors
On 01/23/2012 02:46 AM, R.S. wrote: W dniu 2012-01-22 05:21, Joel C. Ewing pisze: I've said it before, but will say it again: modern tape media has such a large capacity that a single dfhsm cartridge can contain an incredibly large number of datasets. It is almost inevitable that loss of a single dfhsm ML2 or Backup cartridge will impact something you care about or can't afford to lose (or are legally required to retain), and it is also inevitable that single cartridges will occasionally be physically damaged. Over the long haul, you really can't afford NOT to duplex all dfhsm carts (and for that matter, non-dfhsm carts that contain data you can't afford to lose). 90% Agreed, some remarks: 1. Cart capacity is not so relevant. The problem exists with any cart capacity, the difference is in a likehood, but the likehood is quite similar (same order of magnitude). 2. It worth to separate two cases: a) Backup copy loss. It's more or less like loss of spare in a RAID group, you did not lose your data, you lost redundancy. Additional event must happen to make backup copy needed for recovery. Of course there could be some legal requirement to have such copy, in such case he requirement is broken. b) Archive copy, ML2 (without backup). In such case lost of cartridge simply means lost of data. I always vote fo duplexing data on tapes, for the same reasons why we use RAID protection on DASD. I mean real media, for example virtual tapes are usually on RAID-protected disks. At least based on our experience, I would disagree on point (1). Over the long haul (after drives and cartridges begin to age) we saw about the same unreadable cartridge rate (inability to later read a cartridge that was written successfully) among 3490 and 3590 - more than one and under five losses per year (typically physical damage to cart or loss of data on cart from some drive failure). The difference in capacity (from 3490) was significantly more than an order of magnitude, so I believe our risk of significant data loss also went up. The newer tape technology tends to be more reliable, with fewer physical volumes and fewer physical mounts, but this can also be offset by a much larger number of physical mounts and opportunities for physical risk for the fewer media volumes that do contain active data. Unquestionably, the potential data loss per cartridge loss (and likelihood of visibility of a single incident to management and end users) is directly proportional to the data capacity of a cartridge. Tape media capacity over the last several decades has increased by over 3 orders of magnitude. Any installation that may have dismissed duplexing when on 3480 or earlier tape technology and hasn't since re-evaluated that decision should do so. Today no one would (I hope) consider running z/OS production on DASD that wasn't covered for single media failure by RAID or mirroring. Similar respect is needed today for data retained on real tape media. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: PDSE
On 01/21/2012 04:21 PM, retired mainframer wrote: :>:-Original Message- :>:From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On :>:Behalf Of Joel C. Ewing :>:Sent: Saturday, January 21, 2012 8:02 AM :>:To: IBM-MAIN@bama.ua.edu :>:Subject: Re: PDSE :>: :>:So the only functionally-equivalent, officially-sanctioned way to :>:accomplish this goal is still to :>:(1) create a new dataset with a different name and copy the data to it, :>:(2) modify PARMLIB LNKLST defs to replace the old library in linklist :>:with the new at next IPL, :>:(3) IPL. :>:And if for some reason you really must have the original dataset name, :>:repeat the process to get back to the old name. If you need the original DSN, after populating the new dataset, uncatalog the original and rename the replacement (both actions not restricted by the current enqueue). This will eliminate the need for the second IPL. And after the IPL, you can rename the original dataset and delete it even though the DSN is enqueued by LLA (assuming you the correct RACF access). This technique is only possible if it is acceptable for the replacement DSN to be on a different volume, AND the original DSN volume is not an SMS volume, but I too have used this in the past when system datasets were divided between two non-SMS volumes and volume residency didn't matter. One must also recognize that there is a slight risk here: that between the "uncatalog" step and the final "rename" there is a window during which the system is in a state where a successful IPL may not be possible should the system crash; so to not tempt fate, you don't want to do this during a storm when the UPS is down or elongate the window by allowing yourself to be interrupted in the middle of the process -- and having an alternative recovery system available for the unlikely worst-case scenario is goodness. I was never hit by a system failure in the middle of one of these sequences, but I am convinced it is only because I was sufficiently paranoid and Murphy knew I had recovery alternatives at hand.:) If the object is to find, if possible, a procedure which passes through no window of risk where a system outage could leave you unable to IPL, then to preserve the original DS name I think you are stuck with two IPL's -- an unpleasant enough prospect that most SysProgs quickly learn alternatives (like the one above) that accept some tolerable level of risk. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: DFHSM: Backup tape lost due to errors
On 01/21/2012 04:06 PM, retired mainframer wrote: If the entire volume is unreadable, it would seem you need to use DELVOL to tell SMS that everything that was supposed to be on that volume is lost. If some of the files on the tape are usable, then FREEVOL may be a better choice or possibly RECYCLE (either combined with the appropriate BDELETEs to skip over the unreadable files). AUDIT does not process the media. It processes the various CDSs and catalogs to ensure consistency but not necessarily accuracy. As with all datasets, anything of which you have only one copy that is unreadable means you have zero copies. -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of af dc Sent: Saturday, January 21, 2012 11:29 AM To: IBM-MAIN@bama.ua.edu Subject: DFHSM: Backup tape lost due to errors Hello, suppose you have an 3592-JB dfhsm cart (native, not a vts stacked cart) that can't be read due to physical damage, no tape backup duplication. Several backups are unique, no additional versions. How to correct bcds ?? Run AUDIT DATASETCONTROLS (BACKUP) with NOFIX (to check the amount corrections) and then with FIX ??? Z/Os V1.12.?? I've said it before, but will say it again: modern tape media has such a large capacity that a single dfhsm cartridge can contain an incredibly large number of datasets. It is almost inevitable that loss of a single dfhsm ML2 or Backup cartridge will impact something you care about or can't afford to lose (or are legally required to retain), and it is also inevitable that single cartridges will occasionally be physically damaged. Over the long haul, you really can't afford NOT to duplex all dfhsm carts (and for that matter, non-dfhsm carts that contain data you can't afford to lose). All it should take is the potential loss of one critical, irreplaceable dataset to justify to management the cost of the extra cartridges and tape drives required. Tape duplexing in some form with off-site storage is also a requirement for any reasonable installation Disaster Recovery plan, so it ought to be cost-justified on those grounds alone, with a side benefit that DR duplex copies can also save your backside from single media disasters. If management lacks the wisdom to support duplexing, file the recommendation away and bring it back out when the next inevitable media failure and massive data set loss occurs. The duplex support in dfhsm makes duplexing ML2 and BACKUP carts trivially easy to do as long as you have enough drives. Similar support should also be available in software/hardware solutions that stack multiple non-dfhsm virtual volumes on physical volumes (e.g., CA-VTape), where similar exposures exist and where duplexing in some form is also highly recommended. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: PDSE
On 01/21/2012 10:25 AM, Paul Gilmartin wrote: On Sat, 21 Jan 2012 10:02:25 -0600, Joel C. Ewing wrote: So the only functionally-equivalent, officially-sanctioned way to accomplish this goal is still to (1) create a new dataset with a different name and copy the data to it, (2) modify PARMLIB LNKLST defs to replace the old library in linklist with the new at next IPL, (3) IPL. And if for some reason you really must have the original dataset name, repeat the process to get back to the old name. Can LINKLIST contain aliases? If so: (0) Place the alias name in PARMLIB LINKLIST defs; IDCAMS DEFINE ALIAS to the real data set name. (1) create a new dataset with a different name and copy the data to it, (2) IDCAMS DELETE ALIAS; DEFINE ALIAS to identify the new data set. (3) LLA REFRESH to identify members in the new data set. Why not? (I know; I've been corrupted by UNIX symbolic links, and imagine aliases are similar.) -- gil If an alias name is acceptable in linklist, that still wouldn't solve the problem. Any system enqueues are always done on the real name at the time of initial allocation for linklist, and the physical location and size of the dataset becomes fixed to the linklist once the dataset is initially allocated and is not changed by a REFRESH. If an active linklist dataset is renamed, cataloged elsewhere, or even deleted, linklist still points to the same original DASD tracks. The only way to physically change the location and/or size of a linklist library is to deallocate and reallocate it, which requires activating a new linklist definition and eliminating any usage of the prior active linklist. The latter is the difficult part because it can only be done by forcing a linklist update on all long running address spaces, some of which cannot be stopped/restarted without an IPL and others of which cannot be restarted without disruption to end users. As has been discussed in the past on this list, forcing a linklist update on an arbitrary running address space is an "unnatural" act that involves risk, and could in the worst case cause z/OS system failure and force an IPL. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: PDSE
On 01/21/2012 07:54 AM, Peter Relson wrote: how does IBM suggest doing a compress on a Linklist lib that needs compressing, inquiring minds would love to know There is no suggestion. This is simply not an operation that is supported or can be supported in general. Peter Relson z/OS Core Technology Design So the only functionally-equivalent, officially-sanctioned way to accomplish this goal is still to (1) create a new dataset with a different name and copy the data to it, (2) modify PARMLIB LNKLST defs to replace the old library in linklist with the new at next IPL, (3) IPL. And if for some reason you really must have the original dataset name, repeat the process to get back to the old name. All the other techniques that have been described here in the past to achieve this and bypass or defer the need for an IPL either don't guarantee the new library will be seen by all address spaces or carry some risk. While those of us who have been around long enough are fairly certain of specific cases at our own installation where the risks of alternative methods are small enough and acceptable, it is understandable that IBM does not wish to "endorse" techniques whose success depends on SysProg competence and judgement and also in many cases upon the tacit cooperation of Murphy in keeping unrelated system failures from occurring in a narrow transition window during which libraries and PARMLIB might be in a state where successful IPL and recovery from system failure is not be possible (without an independent z/OS recovery system). -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: PDSE
On 01/20/2012 08:54 AM, Paul Gilmartin wrote: On Thu, 19 Jan 2012 17:37:12 -0600, Joel C. Ewing wrote: On 01/19/2012 02:55 PM, Paul Gilmartin wrote: On Thu, 19 Jan 2012 11:09:27 -0800, Schwarz, Barry A wrote: Someone might be failing to issue a DISConnect or RELEASE. (It's not at all clear to me why this problem shouldn't occur similarly with a classic PDS.) ... How could this be an issue with the classic PDS? Surely the "connection to members" concept doesn't exist for a traditional PDS, as old member data can only be eliminated by completely reorganizing, compressing the PDS. With no re-use of deleted-member space, there is no need for a special mechanism to prevent over-write of a deleted member while some other task is still reading the now-deleted version of the member. The issue reported by the OP wasn't that a member was overwritten; the issue was that the data set ran out of space. Not really. The context of "this problem" in "It's not clear to me why this problem..." was a discussion of member disconnect or release, and the original problem that started the thread was more along the lines of why isn't deleted space for a LNKLST PDSE library being freed and and available for reuse, even after an LLA REFRESH, as one would expect, rather than why is the data set out of space (which would have been obvious for a PDS that hadn't been compressed). JC Ewing Conceptually, any user who has done a BLDL and saved the TTR has a connection to the member. PDSE merely formalized the process. -- gil ... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Catching a Phantom
On 01/20/2012 05:50 AM, Jan MOEYERSONS wrote: On Thu, 19 Jan 2012 10:21:01 -0600, Chase, John wrote: Occasionally, that MQ-draining transaction will be spawned nearly 1,000 times per second, and apparently "do nothing": No "I'm starting" message, no "I'm finished" message, no ABEND, no nothing. SMF records for these "phantom" transactions, formatted by CA-Sysview, show that each instance starts the program, which then issues two MQOPENs (not back-to-back), two MQCLOSEs (again, not back-to-back), one MQINQ, one MQGET and one SQL SELECT, and then it ends "without comment". Average task lifetime for these "phantoms" is around 8 milliseconds. I think this may be related to the fact that an MQ triggering message is made available to the initiation queue at the very moment the sending application puts an MQ message on the queue, before this application message is committed. Depending on how long it takes the batch program to commit its unit of work, this may well result in the draining transaction firing up, finding its input queue empty and ending again. Because there are still messages on the queue (not yet available for draining, but they are there...) MQ then immediately re-triggers the same. Over and over and over... until a message is actually committed and thus becoming available to the drainer. Simple solution to this problem is to have the draining transaction do its first MQGET with a wait of say 250 to 1000 ms. Cheers, Jantje. This sounds like a plausible explanation, but if true, isn't this a design flaw in MQ that needs fixing? It makes zero sense for MQ to in effect create a "busy" waiting loop that uses resources by having a triggering mechanism repeatedly trigger a transaction for "available message" before the message can actually be obtained, for reasons that the observed transaction loop should make obvious. The resources unnecessarily consumed by such "ghost" receiver transactions could even potentially delay the sender from reaching the point of message commit in a resource constrained environment. If this design was a deliberate one to potentially allow faster response by allowing a "long" receiver-transaction start up time to overlap with a "short" sender time-to-commit, then at most MQ should trigger when the message first enters, and if the corresponding MQGET arrives prior to the commit and gets nothing (which obviously indicates a case where time-to-commit is not "short", relatively speaking), trigger once more only after the commit has occurred and the message can actually be retrieved. The suggested "wait" circumvention requires an application programmer to be aware there is a problem that needs fixing. It would be better for MQ design to prevent the problem in the first place. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: PDSE
On 01/19/2012 02:55 PM, Paul Gilmartin wrote: On Thu, 19 Jan 2012 11:09:27 -0800, Schwarz, Barry A wrote: IEBCOPY compress generates an error message for a PDSE. Become familiar with: Title: z/OS V1R12 DFSMS Using Data Sets Document Number: SC26-7410-10 http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/dgt2d490/3.8.7.1 3.8.7.1 Establishing Connections to Members Someone might be failing to issue a DISConnect or RELEASE. (It's not at all clear to me why this problem shouldn't occur similarly with a classic PDS.) -- gil ... How could this be an issue with the classic PDS? Surely the "connection to members" concept doesn't exist for a traditional PDS, as old member data can only be eliminated by completely reorganizing, compressing the PDS. With no re-use of deleted-member space, there is no need for a special mechanism to prevent over-write of a deleted member while some other task is still reading the now-deleted version of the member. The referenced 3.8.7.1 specifically applies to PDSE libraries; but it also implies that if LLA REFRESH is not sufficient to DISConnect or RELEASE all the old member connects on a PDSE in the LINKLIST and allow the deleted PDSE space to be re-used (don't know for sure that this is the case, but Juergen's experience that started this thread at least raises that as one possibility), then LLA REFRESH would seem to be failing to do something it really ought to be doing. Someone surely is in a position to set up a simple test case, or may have already tested this: A PDSE library in LINKLIST/LLA which no one is actually using with but members and minimal free space; delete all members; check whether there is enough free space to re-create the members; if not, do an LLA REFRESH on the library; re-check whether there is now free space for the new members. If the deleted space is never freed, I would say we have a problem that needs fixing. I gather from past warnings on ibm-main about forcing dynamic LINKLIST updates on running address spaces that there are some cases where only a partial load of a PDSE program object member has been done and a potential of additional future loading activity still exists. If z/OS is smart enough to recognize this situation, then one would want it to preserve a PDSE member connection in such a case and continue to make the old member version available to the old A/S even across an LLA REFRESH; but I would also suspect this technique unlikely to be in use on a typical Installation library added to LINKLIST. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: PDSE
or of keeping table libraries OPEN even after all referenced tables have been closed. ISPF users who were no longer in the application might still have had hanging references to the application PDSE library - but I would not have expected this to tie up connections to specific table members that were no longer in use, or to have the major impact on space reclaim that it had. It's almost as if some PDSE member "connection" issues never get fully resolved until the applications are also forced to de-allocate the PDSE - which is certainly not what I would have expected from my reading of the quoted DFSMS manual. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: ITIL Mainframe Terminology
I Googled ITIL and found http://www.itil-officialsite.com, looked at several of the introductory documents, and found them very heavy on MBA-like generalizations and abstractions. If you didn't know anything about corporate Information Technology to start with, you could read one 58-page intro and still not have the vaguest idea that the main object of IT is to solve corporate business problems by conceptualizing/designing/acquiring/maintaining computer application programs and selecting/acquiring/maintaining/operating the necessary computer hardware on which to run those applications. The documentation I saw seemed oblivious that the basic management issues in corporate IT deal with computer hardware and computer applications, but instead just talked about managing "resources". I wasn't terribly impressed. There have been times when teacher training in this country spent entirely too much time with "Education" courses learning how to teach and not enough time mastering the subjects they were supposed to teach. Corporations run into trouble when they have too many MBA managers who think they can manage the manufacture of "widgets" without understanding "widgets". What little I have seen of ITIL so far reminds me of those approaches applied to IT. Just calling a mainframe a "server" or a "resource" encourages a management mindset that doesn't take into proper account its unique qualities that continue to distinguish it from lesser platforms and continue to justify its existence. I doubt if there are any pure "mainframe organizations" in today's corporate IT world, rather "Data Processing", "Information Technology", "Information Services", or some similarly named corporate Division, which may maintain a mainframe as part of its much larger corporate computer hardware inventory. A large corporate IT division will typically be subdivided into many functional subdivisions which include in some fashion Application Development and Maintenance, Workstation support, End-User IT Help Desk, Technical Support, Operations, Production Control, etc.. Many of these functional sub-areas must deal with or manage across multiple platforms, not just mainframe or non-mainframe. IBM mainframes have also traditionally been called "Processing Systems" (the box(es)that house the central processing elements, processor memory, and I/O channels, and related control), to distinguish them from the separate box(es) that house DASD Disk Storage Subsystems, Tape Subsystems, etc., all of which may also be collectively thought of as "the mainframe"; but that term (mainframe) is less used by those closest to actual management of one. Specific mainframes systems are much more frequently referred to by their model family or type, as in IBM z9, or IBM z10 rather than by some useless generic IBM Marketing name like "Enterprise Server". The model family at least tells you the general functional capability, but little about the specific processing capacity (and cost), which has a wide variation even within the same processor family. IBM Mainframes are probably more frequently known by Operations and Applications Development personnel by the Operating System that runs on the box (e.g., z/OS) rather than the hardware name; because this, rather than the hardware, most affects what interfaces those people see and must use on a daily basis. End users (and non-IT managers) typically only see interfaces provided by Application systems, so in many cases they may not even be directly aware of the underlying Operating System or hardware platform used by those applications. Joel C Ewing On 01/11/2012 03:16 PM, Henrichon, Ryan wrote: A common phrase used in ITIL to refer to mainframes in my shop is "Enterprise Server". However the term "Mainframe" gets used more than "Enterprise Server" does by techies. What is true about any "best practices" or "new process" that a company uses; it is only as good as how involved the employee's are that are using it. -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Bill Fairchild Sent: Wednesday, January 11, 2012 3:08 PM To: IBM-MAIN@bama.ua.edu Subject: Re: ITIL Mainframe Terminology This reminds me of ISO 9000 about 20 years ago. Bill Fairchild -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Jonathan Goossen Sent: Wednesday, January 11, 2012 1:29 PM To: IBM-MAIN@bama.ua.edu Subject: Re: ITIL Mainframe Terminology Peter, You are correct in what ITIL stands for. The British started it. It migrated to the US when companies wanted to cut costs. Several years ago I was required to go through training and passed my certificatio
Re: Clarification on Sysres Volumes(RMF)
I believe JES2 uses physical reserves on entire volume when accessing HASPCKPT datasets and depending on your JES PARMLIB member may hold it for a significant time -- not a good dataset type to put on a SYSRES shared with other systems. Others may be able to suggest JES PARMLIB changes to shorten the hold time, but I would move all HASPCKPT datasets to volumes that have minimal shared access with another system, or at very least off volumes like SYSRES that contain many performance-critical datasets needed by all systems sharing that SYSRES. I suspect the usual z/OS manual recommendations are to put HASPCKPT on a volume by itself, but a sufficiently quiet volume is good enough. JC Ewing On 01/03/2012 02:13 AM, jagadishan perumal wrote: Hi Lizette, Apology for not being precise. Datasets found in SYSRES are : are some BCP related datasets , assembler datasets,language environment datasets,IBM book manager datasets First failure Support technology(FFST) datasets are found. During This Situation when I do : /D GRS,C It does really produces any information about Volume or Dataset Contention. As per Mark advise I did TSO RMFMON and Found some SYS1.HASCKPT datasets used by JES2, some User Job using their Assigned volumes. Jags On Tue, Jan 3, 2012 at 1:21 PM, Lizette Koehlerwrote: Hi All, One thing I have noticed from the report is that most of the user ID are the reason for volume contentions. Name ReasonCritical val. Possible cause or action DOV0053DEV -Z18RS193.0 % delay May be reserved by another system. DOV0061DEV -Z18RS192.0 % delay May be reserved by another system. DOV0065DEV -Z18RS191.0 % delay May be reserved by another system. Is it something I need to look into user's Proc ? Could anyone please throw some light on this. Jags Jags, Have you tried to see what datasets are open and enq'd during this time? What research have you done? What datasets are on your SYSRES volume? You have not answered these qestions. Have you issued a D GRS,C to see if anything is obvious? Lizette .... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: IBM Manuals
On 12/30/2011 10:30 AM, Shmuel Metz (Seymour J.) wrote: In<4efdc973.3050...@acm.org>, on 12/30/2011 at 08:23 AM, "Joel C. Ewing" said: First thing I do with new equipment/appliance manuals is either find and save an on-line pdf version Why? I know that in some cases PDF is all that's available, but certainly not in all. I want a form that preserves for my personal use, and has the ability to recreate if necessary, the original multi-page documents for human viewing. PDF was designed with precisely that in mind and does it very well. I want a format with a proven track record of continued support over an extended time period on multiple hardware platforms and operating systems. The PDF specs are openly available, free PDF readers are available for multiple environments from multiple independent sources. Ditto for free print-to-PDF converters, and direct PDF creation support is now built into many applications as well. There are other formats that are useful or even "better" in specific environments, but nothing at this point is as ubiquitous as PDF, and I have no idea what systems I may be running ten years from now. If you can get a text-based PDF document from the original source, that would certainly be preferable, as that allows text searching capability. But, if all you have is a hard copy, none of the current freely-available OCR tools come close to preserving the original document as accurately as image-based PDF, unless you have the time for extensive manual editing. Bitsavers.org uses a modified archive approach that uses higher resolution to allow possible future OCR; but compensates for higher resolution by using black/white threshold images that sacrifice quality of embedded document illustrations. I prefer to go with lower resolution adequate for human reading and preserve gray scale, and even color, where its use is significant. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: IBM Manuals
First thing I do with new equipment/appliance manuals is either find and save an on-line pdf version or scan them in as a pdf image document. By keeping all such in digital form, I can either find them later if needed as reference or make them large enough to read easily. Wife's iPad2 booklet this December was so small I had to scan at 300 ppi instead of usual 150 ppi to get decently formed font and could barely read the original hard copy even with my reading glasses! Guess we should be thankful IBM chose bookmanager and pdf to save trees rather than reducing z/OS manuals to 3x5 and microprint. JC Ewing On 12/29/2011 11:38 PM, Ed Gould wrote: Scott: Last year I got an brand new IPAD. The installation instructions were on a 3 X 5 in a 4 page "booklet". The font size was 4 and I could not read it to save my life. I had to get a friend to come over to read them so I could do the "install". Bah humbug so much APPLE being user friendly. Ed ... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Question on adding an SVC routine dynamically to a running system
On 12/29/2011 11:06 AM, Dave Day wrote: Answering my own question, just for the record. I just tried it, and there is no need to actually have the module in the LPA chain. SVCUPDTE does indeed update the SVC table, and the SVC can be executed with no problems. --Dave Day - Original Message - From: "Dave Day" Newsgroups: bit.listserv.ibm-main To: Sent: Thursday, December 29, 2011 9:00 AM Subject: Question on adding an SVC routine dynamically to a running system The auth services guide in the chapter on user SVCs, states a type 3 needs to be in LPA. Yet the SVCUPDATE will accept, and update the svc table, an address in ECSA. Do I need to add the routine to LPA using CSVDYLPA prior to SVCUPDATE? --Dave Day If the intent was a permanent change, just be sure you DO get appropriate parmlib and/or module location changed before next IPL, so you don't get surprised by unexpectedly losing your updated SVC at next IPL. I always tried to do both updates at about the same time under the assumption that Murphy might arrange for unscheduled IPLs at his convenience, not mine. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: removing nulls from a file
On 12/29/2011 09:19 AM, Brad Wissink wrote: We have a client that is trying to transfer us a file and it is loaded with nulls. They say that is the way it comes from the purchased software they have on their workstation. The file has a null character inserted after every character so it looks like this 1 12/29/2011 becomes F1004000F100F2006100F200F9006100F200F000F100F100... Has anyone seen anything like this before? Is there a quick and easy way to remove all the nulls? It's almost as if they are using software that somehow mistakenly thinks EBCDIC is a two-byte character encoding, or the process for conversion to EBCDIC was written by a neophyte programmer who treated each character as a null-terminated string and some how managed to include the null termination for each character when outputting the converted data. Maybe they have their software package mis-configured in some way, or since they are talking about workstation software producing the data, the software package generating the data may be ignorant of EBCDIC and the fault may be in what technique they are using after-the-fact to either convert the data to EBCDIC or transmit the data with conversion. If at all possible, try to get clarification of exactly what software and options they are using to produce the data and exactly what technique and options they are using to transmit the data. If possible, get them to transmit via other methods (Email, FTP, etc.) samples of any intermediate files as binary data so you can see exactly what data format they are really dealing with on their workstation (as opposed to what the client may think he has). If they are really transmitting twice the number of needed bytes, fixing the problem at their end would be the better solution. Perhaps given enough background on what they are doing, a solution would become obvious, or you would be able to search on-line for a solution if the client lacks the technical expertise to do so on their own. If everything is kosher on their end and the byte doubling is somehow occurring just on your receiving system, then hopefully you will have enough to be able to recreate and fix the problem on your end. Just cleaning up bad data after the fact will work, but is not as desirable as eliminating the creation of the bad data in the first place. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: cpu / machine identification
On 12/27/2011 07:55 AM, Mark Zelden wrote: On Mon, 26 Dec 2011 16:11:02 -0500, zMan wrote: OK, I gotta ask -- what's the problem you're trying to solve? You don't trust your customers? In over a quarter century in the mainframe software business, I've come across ONE customer running software on an unlicensed box, and it was an oversight -- and a nice full-price bluebird for the sales rep. I don't believe "CPUIDs" are worth the hassle. Obviously the point of view of someone who doesn't make a living by selling their software. I can tell you from personal experience (one of my clients) who I helped write CPU protection for about 10 years ago that there were many instances of unauthorized use and in at least one case I know about the abuse was rampant. I know a lot of the unauthorized use wasn't intentional, but a lot of it was also or shops just didn't care since there was no checking. Some of the companies that used this software outsourced their IT, and ended up using it on different machines than those that were licensed. Or the outsourcer copied it to other machines / environments / clients. My client must have lost hundreds of thousands of dollars in licensing / maintenance fees and fees from related litigation. In my own personal experience as a sysprog, I know some of the same things have happened unintentionally. Consolidations, moving LPARs around, creating / cloning new LPARs can lead to this and when the software doesn't check it's easy for the "techies" to make mistakes since they often (usually?) don't know the T's and C's of all the software contracts. So even though it can be a pain, I actually prefer that any vendor that cares, checks the CPU id for authorization. If the software has a "site license" option, then have a method for a non-cpu specific key to generate for the client. Provide an easy way to change the key and a grace period that won't put the shop's business in jeopardy because of a missing / wrong key after a CPU upgrade or engine add. Regards, Mark -- Mark Zelden - Zelden Consulting Services - z/OS, OS/390 and MVS mailto:m...@mzelden.com Mark's MVS Utilities: http://www.mzelden.com/mvsutil.html Systems Programming expert at http://expertanswercenter.techtarget.com/ Although most of my career was spent with configurations with no more than two independent mainframe systems, I can appreciate the additional confusion of trying to track inconsistent vendor license terms in a much larger environment and how that would raise the probability of unintended violations. Thankfully I've never had direct experience with a company that was in process of outsourcing their mainframe operations, so I hadn't considered that aspect. Obviously a company with no reservations about shafting its own IT department for short-term profit would probably have even less compunction about doing the same to its former software vendors. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: cpu / machine identification
temporary keys, that could be read/emailed/FAXed by support at 3AM if you really screwed up and were dead in the water despite the above. Now, this also meant that there were folks carrying beepers and temp keys, so they could do that after-hours support. Are you prepared to deal with all this? Is it worth it? As you can tell, I'm not a fan of such mechanisms. But it's not my decision (doh), so I'm trying to help :-) -- zMan -- "I've got a mainframe and I'm not afraid to use it" -- zMan -- "I've got a mainframe and I'm not afraid to use it" ... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Imagine having to deal with THIS in production
On 12/21/2011 02:02 AM, Shmuel Metz (Seymour J.) wrote: In , on 12/20/2011 at 09:46 PM, John Gilmore said: Others are observing this sky too, and no cleric wishes to be made a figure of fun by denying what is obvious to many others. Shirley biblical literalism is an exception to that. I can't speak to the Christian scriptures, but the Tanakh is absolutely laden with obvious metaphor. And since the 2/3's of the Christian scripture that is the Old Testament is essentially the Tanakh, the same is true of it. That doesn't seem to stop a majority of fundamentalist Christians from taking some of the obvious metaphors, like the multiple, conflicting creation stories in Genesis, literally. In the Christian New Testament the parables of Jesus are also obvious metaphors and essentially the entire book of Revelations as well, which again doesn't keep many fundamentalist Christians from trying to apply the later literally to our times rather than the Roman context in which it was written -- with the result that several times a decade in this country some religious nut will use his "inspired" interpretation of Revelations to confidently predict the exact date of the end of the world, attract a small following, and then vanish into richly-deserved obscurity when the date passes without incident. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Imagine having to deal with THIS in production
On 12/19/2011 05:39 PM, John Gilmore wrote: Joel C. Ewing writes (of Julian days): | That makes much sense for astronomers that work through | the night and sleep during the day, but is a terrible fit for people | and businesses that have to deal with "normal" work hours and | who would never tolerate the same period of daylight being | called by two different dates What people find tolerable is a function of their experience. When my wife and I lived in Iran we rapidly came to terms with the convention that the day ends at sundown and even, with only a little more difficulty, with idea that a dinner invitation for Tuesday night was an invitation to have dinner following sundown on Monday. However that may be, this objection has another, much more important defect. It confounds internal representations for machines with external representations for people, which need to be interconvertible but should seldom--I had almost written never--be the same. John, If you had read all of the included previous thread context in my previous response, the context was "the world's eventual conversion to ", not a discussion limited to internal date usage by machines. I would say that makes the tolerance of people for the date format highly relevant and an asset to the objection, not a defect. There are of course other strong arguments against universal usage of JD for dates any time in our lifetime. As long as we remain an Earth-centric and not a space-centric culture, that makes it unlikely most people would favor an ordinal-based standard date format like "Star Date" or "Julian Day" which has no obvious relationship to Earth's annual seasons, when awareness of those seasons is so important to our physical comfort and survival. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Imagine dealing with THIS in production
On 12/19/2011 11:53 AM, Mike Schwab wrote: How about the Julian Day as used by astronomers? http://en.wikipedia.org/wiki/Julian_day Julian day is used in the Julian date (JD) system of time measurement for scientific use by the astronomy community, presenting the interval of time in days and fractions of a day since January 1, 4713 BC Greenwich noon. Julian date is recommended for astronomical use by the International Astronomical Union. Almost 2.5 million Julian days have elapsed since the initial epoch. JDN 2,400,000 was November 16, 1858. JD 2,500,000.0 will occur on August 31, 2132 at noon UT. (Often .leading 2.4 million is assumed and the low order 5 digits is used.) Time is expressed as a fraction of a day. 0.1 day = 2.4 hours, 0.01 = 14.4 minutes. 0.001 = 1.44 minutes, 0.1 = 0.864 seconds. x.000 is Noon UT 1200Z Modified Julian Date subtracts 0.5 so x.000 is Midnight UT Z. On Mon, Dec 19, 2011 at 10:01 AM, Paul Gilmartin wrote: On Mon, 19 Dec 2011 08:44:00 -0600, Chase, John wrote: Perhaps the world's eventual conversion to "Star Date" (or similar) will be less confusing and disruptive :-) Ummm... NVFL. See: http://en.wikipedia.org/wiki/Stardate -- gil Usage of Julian Day will never catch on with non-astronomers for one simple reason not yet mentioned - its formal definition requires the date change to occur at noon, not midnight. That makes much sense for astronomers that work through the night and sleep during the day, but is a terrible fit for people and businesses that have to deal with "normal" work hours and who would never tolerate the same period of daylight being called by two different dates. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: One Less Mainframe Shop
On 12/16/2011 04:06 PM, DKM wrote: Just over seven years ago, I was hired as the Financial System Administrator at my place of emplacement. In my first interview, I was told how they were getting ready to pick a new ERP and get off their “archaic” mainframe. After I was hired, the IT director at the time told me with glee how they would be shutting down the mainframe in six months. This shocked me a bit it was going to take at least a year to go live with the new ERP solution. It turned out maintenance on the 20-year-old software was going to end in six months. The mainframe was actually scheduled for shutdown six months after we went live on the new software and platform. Well we did go live on the new ERP within a year, but the mainframe at one time had run the entire business of the company and while the financial suite was the last large part to go off it, there were still several “smaller” but just as important systems still running on it. Consequently, it took seven years, and two other IT directors, before access to the now 11-year-old System/390 was finally cut this week. At some point after the New Year, a ceremony is being planned to let the Chairman flip the final switch to turn off the system. He has been a “Champion of Modernization” to get us off the mainframe for almost 10 years. I’m sure speeches will be made about how far we have come. Yet, as I look around at the countless servers, real and virtual, and think about the major software platforms hosted by outside vendors, all to replace the one S/390 that was divided in to four virtual systems I can’t help but wonder if we are really better off. DKM Since this sounds like management by ideology and 1990's airplane mags, I don't suppose they were honest enough to compare their current total IT costs now with their prior mainframe IT costs, which could likely have been reduced by simply upgrading to modern z boxes and DASD and a more modest migration strategy. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Is there an SPF setting to turn CAPS ON like keyboard key?
There is no "code" to send to the terminal to display lower case as caps in 3270 architecture - it was a manual switch on 3278's, to make the local display show both upper and lower case as upper case at the terminal even though dual case was transmitted both ways. It was global for the entire display, not field specific. If Hummingbird is an accurate 3270 emulator, it should have an option to emulate this feature, but I doubt this is what you are seeking, as it can get extremely confusing in a hurry with applications that actually support or require dual case in some contexts. This display option made sense in the late 1970's when all applications still required upper case and all 3270 applications forced all input to upper case. This feature was also for the benefit of users familiar with prior 3277 mono-case display limitations, who found 3278 dual-case behavior distracting when ENTER was pressed and the application echoed data back to the terminal in upper case and changed the visual appearance. I used 3277's for a relatively short time before 3278's became available and quickly adapted to preferring dual-case mode even for mono-case applications, as it made it obvious whether you had actually sent the data to the application or not. Some of the "old timers" of that day disliked the hardware change and continued to use the mono-case setting on 3278's for years. JC Ewing On 12/15/2011 02:29 AM, Cris Hernandez #9 wrote: thanks to all for the feedback. it's a hummingbird terminal emulator but I have no idea how to program it by any other means than using the options panels provided. the consensus here seems to be that if it can be done, the emulator would need to be sent code to turn caps on. appreciate everyone's fond memories of dumb terminals as well, they may have been featureless, but I never had to shut down TSO because the system was hosed... a lot less distractions too. --- On Wed, 12/14/11, Paul Gilmartin wrote: From: Paul Gilmartin Subject: Re: Is there an SPF setting to turn CAPS ON like keyboard key? To: IBM-MAIN@bama.ua.edu Date: Wednesday, December 14, 2011, 8:37 PM On Wed, 14 Dec 2011 17:21:11 -0800, Cris Hernandez #9 wrote: yeah, TRANSLATE works to change it after the user hits enter, but I want characters to show in caps as soon as the character is typed, for visual purposes more so than anything else, and do so in both panel displays and spf editing. Data entry at the terminal is completely invisible to ISPF. No setting can affect it. Hmmm. Of course you may have the source to x3270. You could modify it. If there were way to send an escape sequence to the terminal, sort of complementary to the pt3270 that oedit/obrowse use, or akin to the mode switching done by IND$FILE, a nonsensical cursor addressing sequence, an emulator could capture that and switch the case. Your motivation is good. It seriously violates WYSIWYG to type a page of text; press ENTER; and stare in dismay as the whole page changes. The worst design ever was the old 3277 (IIRC) which displayed text in CAPS and sent it to the host in lower case. And some independent terminal vendors actually provided a switch to enable compatibility with this horrendous misfeature. also wondering if panels have the ability to enlarge font size by row. Does your terminal hardware (emulator) have this capability? -- gil ... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: JCL "sheesh!" for today
On 12/10/2011 12:19 AM, Ed Gould wrote: Joel, My reading of your reply is frankly confused. Any DD card that needs to be continued must end with a comma and start (in the continuation card with a // and a blank) in CC 4 or at least up to CC 16 (no continuation in CC 72 is needed). A parm on the exec card in order to be continued had different rules and IIRC must start in CC 16 *AND* cannot be any longer than 100 characters (thats a double restriction) *AND* must have a continuation in 72 (thats three) which IIRC there is no other DD parameter has any restrictions (that I can remember of). Ed - Original Message - From: Joel C. Ewing To: IBM-MAIN@bama.ua.edu Cc: Sent: Friday, December 9, 2011 10:19 PM Subject: Re: JCL "sheesh!" for today On 12/09/2011 11:57 AM, Ed Gould wrote: JC, My memory indicates that the parm on the exec card is really the only real column sensitive "Nasty" left in JCL. Although there is some debate whether JCL is a language (or not) any language that I am familiar with does have column restrictions of some type. Ed All JCL statement continuation records are column "peculiar", not just those involving PARM values or quoted strings. A continuation which splits a quoted string is column sensitive on where the first record ends (column 71) and on where you must resume the string on the continuation (column 16). But, all other statement continuation records are also column sensitive in that continued parameters must resume in columns 4 - 16. This complicates manual verification of JCL, because visually it is difficult to distinguish between any parameter continuation that resumes in column 16 versus one that resumes in column 17, but of course the latter fails. It is true that there are languages (e.g., COBOL, FORTRAN, Assembler) with column restrictions that are an integral part of the language syntax; but a number of other languages (e.g.,PL/I, C, REXX) are essentially free-form with a syntax that is column insensitive. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org ... Ed, As you indicated in your response, all DD statements (or any other JCL statement for that matter) that need to be continued at a point not in a quoted string require that the continuation must resume on the following "card" in columns 4 - 16. This is an *arbitrary* JCL column sensitivity that would be regarded by many as "peculiar", especially when contrasted with the unrestricted location of the parameter field on the first "card" of the statement. This restriction is is easily violated accidentally by JCL programmers who attempt to align subsequent parameter continuation cards with a parameter field that just happens to start on column 17 or later on the first "card" of statement. I would call this a "nasty" column sensitivity in the JCL syntax as well, because I have seen so many cases over the years where this caused JCL syntax errors when the coding intent was obvious to a human. That the continuation rules for a quoted string are even worse doesn't mean that normal JCL continuation rules are adequate or column insensitive. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: JCL "sheesh!" for today
On 12/09/2011 11:57 AM, Ed Gould wrote: JC, My memory indicates that the parm on the exec card is really the only real column sensitive "Nasty" left in JCL. Although there is some debate whether JCL is a language (or not) any language that I am familiar with does have column restrictions of some type. Ed All JCL statement continuation records are column "peculiar", not just those involving PARM values or quoted strings. A continuation which splits a quoted string is column sensitive on where the first record ends (column 71) and on where you must resume the string on the continuation (column 16). But, all other statement continuation records are also column sensitive in that continued parameters must resume in columns 4 - 16. This complicates manual verification of JCL, because visually it is difficult to distinguish between any parameter continuation that resumes in column 16 versus one that resumes in column 17, but of course the latter fails. It is true that there are languages (e.g., COBOL, FORTRAN, Assembler) with column restrictions that are an integral part of the language syntax; but a number of other languages (e.g.,PL/I, C, REXX) are essentially free-form with a syntax that is column insensitive. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: JCL "sheesh!" for today
Fred Brooks was the one who made side remarks in a 2004 talk about preferring program time scheduling to be embedded in the programming languages rather than having a separate JCL language, NOT me. I'm guessing the idea was it would seem more natural to the programmers and they wouldn't be required to learn a separate JCL language; but Fred was only talking in generalities with no references to specific designs or examples. I personally don't see how that would work, or even be an improvement, and time scheduling of a program addresses such a small part of the scheduling problem in the typical MVS production environment. It's the ordering of program steps, other inter-step dependencies, specifying all the input and output data sets for the programs, and correctly handling all this when the system is maxed out and everything is running late, that introduces the complexity, not just some simple matter of specifying what time(s) to run a program. I have never had time to sit down and seriously consider a redesign of JCL, other than getting far enough to recognize it is non-trivial and that no existing language (like REXX) is a good fit. While it would be appealing to have a more functional JCL language, trying to make JCL into a full-fledged procedural language greatly complicates, perhaps in general makes impossible, calculating data set names early enough to handle job-level enqueues or providing simple mechanisms for partial reruns. It would be a big improvement if one could just clean up all the syntactic garbage in the current JCL language, like dependencies on specific columns and a seemingly random and irrational intermixing of keyword, positional, and even positional-keyword parameter types; and bring more consistency to the language. But, any changes that weren't somehow upward compatible and which required migration effort would meet heavy resistance from existing users. Any major migration would be a hard sell, even if it allowed new functionality that was difficult to resist. JC Ewing On 12/08/2011 03:17 PM, Lindy Mayfield wrote: Would you mind, please, Joel (or anyone else), giving me an example of some sort of JCL substitute language you and Fred speak of? In the real/wild world? I personally do not see any benefit from a JCL-like language being in the "style" of another language. But I think I got this wrong. This is something I think - and I said I think - that I've learned a bit over the past few years. That when you take a general purpose type of language and make it specific for one purpose only... then just go from there. All the stuff you need to know you need to know from the assembly language. But in my case, and perhaps many others, I wrote and modified over 10's of thousands of Cobol and CSP (not to mention JCL, Rexx, SQL, etc) lines of code during my lifetime up to a certain point. And I never, ever knew any MVS instructions at all, other than, by accident IEFBR14. I always knew the BR14 was somehow special. Took me forever to find out, but I did. What I would say is this: If you made JCL more C like when running C programs, could I use it for running other programs like Cobol or Rexx or IDCAMS? Then why not make it more C-like or Rexx-like or macro-like? It all boils down to the same thing, doesn't it? And if I hate C as much as some hate JCL, then what? Didn't we already learn this lesson from shell's and shell scripting languages? Today, for 45 more minutes is Sibelius' birthday. Happy birthday. This is a nice Finlandia, if I didn't mistakenly post it already. I've been having fun on both IBM-MAIN and ASSEMBLER-LIST for the past few days, so I get mixed up. http://www.youtube.com/watch?v=ci3RPAOFok4 Regards Lindy -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Joel C. Ewing Sent: Thursday, December 08, 2011 10:45 PM To: IBM-MAIN@bama.ua.edu Subject: Re: JCL "sheesh!" for today I don't think Fred Brooks was really thinking things through when he said he would have preferred language-specific support within higher-level programming languages for scheduling program execution rather than OS/360 JCL ( http://lilliana.eu/downloads/jcltalk.txt ). Can you imagine the operational chaos of trying to manage production environments where each high-level language had its own unique and incompatible way of specifying program execution sequencing and specifying data flow for each execution step? As bad as JCL is, it at least does isolate most of the Operating-System specific and installation-specific details of program execution from programs, so that a program in a higher-level language has a reasonable chance of running without modification in many different environments; and Operations personnel only have to have JCL expertise, not unders
Re: JCL "sheesh!" for today
On 12/08/2011 09:08 AM, Tom Marchant wrote: On Thu, 8 Dec 2011 07:56:52 -0600, Paul Gilmartin wrote: On Thu, 8 Dec 2011 07:35:36 -0600, Tom Marchant wrote: One difficulty with trying to replace JCL with ReXX code is that the dynamic allocations can lead to more deadlocks between jobs. The initiator avoids them by ENQing all the data sets defined in the JCL before the job starts. That function could, and should, be provided to Rexx in a function package. I can even imagine an extension to BPXWDYN that would allow allocating multiple data sets in a single call. Yes, that would be easy. More difficult would be to maintain the ENQ across deallocation and reallocation to different DDNAMES for use in different steps. Would you want to allocate all of the data sets for a hundred steps at the same time in the beginning of the job? Even for those steps that are skipped? What happens if you need to add or remove a step? The enqueue and restart issues to me are major reasons why you really need a specialized language designed to support program flow control. You don't want to have to manually specify all the enqueue details, and you do want a language that makes it easy to monitor for success of complex program sequences and easy to perform partial job restart when failures occur. I don't think Fred Brooks was really thinking things through when he said he would have preferred language-specific support within higher-level programming languages for scheduling program execution rather than OS/360 JCL ( http://lilliana.eu/downloads/jcltalk.txt ). Can you imagine the operational chaos of trying to manage production environments where each high-level language had its own unique and incompatible way of specifying program execution sequencing and specifying data flow for each execution step? As bad as JCL is, it at least does isolate most of the Operating-System specific and installation-specific details of program execution from programs, so that a program in a higher-level language has a reasonable chance of running without modification in many different environments; and Operations personnel only have to have JCL expertise, not understand every programming language used in the shop. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: ROOT file system is out space
On 12/06/2011 04:35 AM, John McKown wrote: Total agreement. Updating the running system is dangerous. I did so once by mistake (luckily it was my sandbox). I suffered an outage because an update to a LINKLIB module required a corresponding update to an LPALIB module. When somebody did an LLA REFRESH, the system died. I don't know if this can happen with UNIX, but at the least, if the SMP/E job fails, there could be incompatible files in the filesystem. "Your gun, your bullet, your foot" On Mon, 2011-12-05 at 20:16 -0500, Shmuel Metz (Seymour J.) wrote: In<4edd3dab.5070...@bremultibank.com.pl>, on 12/05/2011 at 10:54 PM, "R.S." said: Why do you think so? Because it's a ticking time bomb. But it's not my dog. Double Ditto. Besides the issue of a PTF having pre- or co-requisites, there are many instances of a single PTF changing multiple elements at the same time. There is always the implicit requirement that all such interrelated changes be made concurrently; but if you install on live libraries or live file systems all changes will in fact be made serially, and the potential results of any task that might be using any of that code during the transition is unpredictable. That introduces the possibility for random Reliability/Availability/Security failures. Doing things in a way that is known to be a RAS exposure may be acceptable in the PC world, but not in the z/OS world. There are some rare instances where a PTF may only affect a single element, or you may know and be able to restrict all the potential users of affected code so that an apply of a few PTFs can be done to live libraries without significant exposure; but the chances of that being the case when applying all PTFs for a new RSU level are zero. In the worst case scenario, a full library or some other failure during the APPLY to live libraries may leave some of the very tools you need to fix the failure in an unusable state, and even an IPL at that point may not help or may even break additional things. The only safe installation technique in general is to install to alternate libraries and alternate file systems and use some technique (such as IPL with suitable parmlib changes) to install all changes concurrently, with some way to revert to old libraries if there are any significant issues with the new code level. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: SETTING CONDITION CODE
And another fairly obvious alternative (but with possible political issues) is to provide on-line documentation for every production job which Operators are required to consult before reporting a problem, which could tell them that RC=0002 on that step is acceptable (something simple like text members in a PDS for each production job can be used here). That rule would solve the problem even without a job scheduler. And you could migrate to that mode of operation easily by defining default rules for Operations to follow if there is nothing explicit for a job and then providing explicit documentation for those jobs that require different rules for success as bogus calls by Operations point out which jobs need different rules documented. If you have a production job scheduler (OPCS, CA-7, ZEKE, etc.), as Peter has already mentioned you can have Operations define for an individual job what maxcc constitutes successful execution, and Operators should be trained to only call if the scheduler says the job is "unsuccessful". It is a semi-standard of IBM utilities that conditions that are most likely fatal have cc>=8, and that having cc>=4 is required to indicate some condition that might be a problem. It would not be unreasonable then to define an installation standard that cc<4 from a step is OK, and if there were any cases from applications programs that violated that convention, one could require a following conditional step that forced a more obviously-fatal error if one of those exceptional steps had a "bad" cc < 4. If the job failing is not a production job but a test job, then reporting failures should not even be the responsibility of Operations. Test jobs should be monitored by the submitter. JC Ewing On 11/28/2011 04:47 PM, Farley, Peter x23353 wrote: One alternative is to use your scheduling software to permit RC=0002 or less from that job to be marked "successful". OPCS can do this, we use that facility quite regularly. Certain jobs are OK if any step returns RC=0004 or less, but 0008 or higher is flagged as an error in the OPS lists. Only jobs flagged as errors in the scheduling system need to be investigated. I'm reasonably certain that other commercial-grade scheduling software has similar capabilities. HTH Peter -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of John Dawes Sent: Monday, November 28, 2011 1:16 PM To: IBM-MAIN@bama.ua.edu Subject: SETTING CONDITION CODE G'Day, Can I override a condition code so as to force it to a ? For example the job-step executes successfully and puts out a COND CODE=0002. The job continues on to the next step which is what we want. The reason I need a instead of a 0002 is because I get paged because the operator sees a 0002. I tried the following by using IDCAMS: //STEPCOND EXEC PGM=IDCAMS //SYSPRINT DD SYSOUT=* //SYSIN DD * IF MAXCC=0002 - THEN - DO - SET MAXCC=0 - END //* However this did not force a on the previous step. Is there a way I can add the IDCAMS IF-THEN-DO logic in the step which is giving me the headache? Or is there a way of forcing a on that step. Here is what the (headache) step is doing: //PRINTDS EXEC PGM=PRTDS, // PARM=('DDNAME(INPDS)', // 'SEA(.) REP(X''40'')', //'SYSOUT(X)') //INPDS DD DSN=SYS2..INFO.CONFIG,DISP=SHR //SYSPRINT DD SYSOUT=* //* -- ... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: Licence to kill -9
er worked in such an environment, rather in ones where systems programmers were frequently expected to be extremely creative and were continually involved in developing, not just following, standards, -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: PS file record Delete - Query
If by "without disturbing any other lines" you mean leaving data of remaining lines intact, then "yes"; if you mean not effecting even the record ordinals or edit session line number of following lines, or not having to physically re-write all the undeleted records to DASD at the end of the edit session, then "no". If you have an edit macro that first identifies multiple line numbers to delete and you next want to delete those lines using original line numbers, one approach would be to delete from highest numbered lines to lowest. Other approaches would be have your macro use ISREDIT commands to eXclude only the lines to be deleted as they are found and then have the macro delete all excluded lines, etc. 1000 records is a very small file in the MVS world. If you had a large file and needed to delete a single record without having to physically re-write the entire file, then you would not choose a PS file for storage, but probably either a VSAM file or a DB2 table that directly supports deletions of individual records or rows. But then you would also need something more sophisticated than the ISPF Editor to do the delete. JC Ewing On 11/28/2011 04:14 AM, jagadishan perumal wrote: ISREDIT macro to delete a desired Line. On Mon, Nov 28, 2011 at 1:50 PM, Binyamin Dissen wrote: On Mon, 28 Nov 2011 10:22:03 +0530 jagadishan perumal< jagadish...@gmail.com> wrote: :>We have a PS file record where we have more than 1000 records. Just of of :>curiosity is it possible to delete a selected line without using "D" tso :>command. Objective behind this deletion is just to make sure the deletion :>is just done without disturbing anyother Lines. Tried Googling to find if :>there are any TSO tricks to do the same but not able to get the one. Your request doesn't make any sense. Obviously if you delete a line, the later lines will shift up. If, instead, you merely want a non-ISPF way to delete a line, there are many options - though each, like ISPF, will temporarily make some lines unavailable. You can use IEBGENER/SORT to copy the lines before and the lines after to a temporary and then copy it back. You can do the same with REXX. Etc. -- Binyamin Dissen http://www.dissensoftware.com Director, Dissen Software, Bar& Grill - Israel ... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: Java based Web Emulator
On 11/22/2011 03:15 AM, jagadishan perumal wrote: Hi, Are there any opensource (Java based) software mainframe emulator which can run inside Internet Explorer . I am just trying to enable the mainframe connection for the remote users. Regards, Jags ... If you are talking about remote access to a mainframe, what you more likely mean is a telnet3270-capable 3270 terminal emulator, not a mainframe emulator. A java-based mainframe emulator would be totally gross in terms of performance - not to mention that you couldn't legally run any current IBM mainframe software under such an environment. I have never used browser-based 3270 emulators because in the past they haven't gotten very good reviews. Telnet 3270 terminal emulators for native MS Windows environments tend to not be free, but some are not that expensive (< $50). If you have to have opensource and free, there is always X3270 under some flavor of Linux, or somewhat less spiffy, X3270 under Cygwin under Windows. Works, but tends to be a little kludgy under Windows. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: ADRDSSU Compatibility
On 11/18/2011 04:15 PM, Ted MacNEIL wrote: I'm glad to know "n-2" compatibility is now supported. If that were true long ago when the previous dfdss 64KiB change was made, it certainly wasn't advertised then. Funny, I thought n-1 was supported for a long time! - Ted MacNEIL eamacn...@yahoo.ca Twitter: @TedMacNEIL Obviously this depends on what is meant by "compatibility" PTFs. New dfdss versions have always been compatible in reading dumps produced by previous version(s) - changes that invalidated archived dfdss tapes would not be tolerable.. I also can't recall a case other than the 64K block size change where the n-1 version wasn't able to read tapes produced by version n as long as you didn't explicitly use some new features introduced in the new version. The problem with the 64K block change was that you got the new feature by default. I may have missed it, but I don't remember there being a PTF to the old version to allow it to read 64K blocks, at least not at the time we migrated. In some cases the best you can hope for is toleration support for the n-1 version so that it will try to do something reasonable or at least give meaningful warnings or errors if there is some new construct in the dump file related to a new feature in version n that the old version can't fully handle. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: ADRDSSU Compatibility
On 11/18/2011 08:18 AM, Norbert Friemel wrote: On Fri, 18 Nov 2011 07:42:34 -0600, Joel C. Ewing wrote: ... But, take note that this backward compatibility does not always exist. It seems that about every ten years or so as tape technology advances IBM decides to raise the block size written by dfdss, first to 64KiB, relatively recently to 256KiB. Be sure to pay attention when this is mentioned in migration notes, because it invariably means that new dump tapes CANNOT be read by back-level versions of dfdss, and you might not have complete control over or knowledge of the update level of dfdss at a recovery site. IBM supports "n-2" releases. The 256K blocks were new in z/OS 1.12. There are compatibility PTFs for 1.10 and 1.11 (OA30822). Norbert Friemel I'm glad to know "n-2" compatibility is now supported. If that were true long ago when the previous dfdss 64KiB change was made, it certainly wasn't advertised then. The fact that compatibility PTFs for back-level releases are available still doesn't mean you can assume without checking that they are actually installed at some DR site not under your direct control, or that you don't need to be extra careful at such junctures that you have updated all your stand-alone dfdss tapes and have resolved compatibility issues on any back-level systems on site that could conceivably become an issue for local recovery. It's easy to get complacent about issues that are so rare they at most exist for a few months every decade. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: ADRDSSU Compatibility
On 11/18/2011 07:11 AM, Norbert Friemel wrote: On Fri, 18 Nov 2011 10:26:32 -0200, Carlos Bodra - Pessoal wrote: If I have a volume backed up using ADRDSSU 1.11 (DUMP DATASET) is possible to restore it using ADRDSSU 1.10? (backward compatibility) -- Yes, if the coexistence and fallback PTFs are installed on z/OS 1.10 http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/E0Z2M171/2.1.2.1 Norbert Friemel ... But, take note that this backward compatibility does not always exist. It seems that about every ten years or so as tape technology advances IBM decides to raise the block size written by dfdss, first to 64KiB, relatively recently to 256KiB. Be sure to pay attention when this is mentioned in migration notes, because it invariably means that new dump tapes CANNOT be read by back-level versions of dfdss, and you might not have complete control over or knowledge of the update level of dfdss at a recovery site. Although we never had to work around a dfdss failure, our DR tape set always included our current version of stand-alone dfdss, just in case - so we knew we could always restore our emergency restore system (at same software level as production) and didn't have to constantly remember to verify DR site dfdss compatibility that only rarely would be an issue. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: TSO SCREENSIZE
On 11/10/2011 02:58 PM, Rick Fochtman wrote: --- Remember how old the 3270 architecture is. Wikipedia says about 1972. Think 1 Mhz 8080 as "top of the line" micro processor. The original 3277 and its controllers were STUPID. Rather than put a more powerful processor in the controller, IBM decided to offload the "complicated" function of calculating the position of the data into the host. Made of discrete transistors and resistors! Very primitive. So, the host just sent a simple to understand "buffer address" (a single number) to the 3274. Not without a time machine. The 3274 came later. The original 3270 controller lineup was 3271, 3272 and 3275, the latter combining controller and display. -- Wasn't there also a 3276, with a display and controller that would handle the integerated display, plus 7 more display-only devices? Rick Yes indeed, for remote BSC or SDLC operation. The other 7 devices could also include 3270 printers. Deceptively the same external size as a 3278, but with enough additional steel and components inside to be appreciably heavier than a 3278, which was already borderline for one person to carry. Trying to carry a 3276 by yourself was not wise if you wanted to avoid back problems. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: Segmenting an output spool file in z/OS - 2nd attempt
Leslie, Off topic, but did you do anything different on 2nd posting to get it to work? Like something we could advise for others with similar base64 posting issues? I received some additional headers from ibm-main with your 2nd posting attempt, most notably a "X-MIME-Autoconverted: from base64 to 8bit by bama.ua.edu id p9KEf61i001292" which implies Email was still received by bama in base64 format, but for some reason it was able to convert the 2nd post to 8bit and handle it reasonably but failed to do so for the 1st. From the "Received" headers, it looks like the 2nd posting was received by a different ua.edu mail server than the first post (mailapp-2.ua.edu versus mailapp-1.ua.edu), so perhaps the difference is an issue at ua.edu. Maybe they're experimenting with a circumvention/resolution of the original problem and you just hit a "fixed" path on 2nd try. (The routing of Email though mo.gov servers was different as well, but base64 encoding apparently got to bama in both cases) JC Ewing On 10/20/2011 09:41 AM, Turriff, Leslie wrote: Hi, I know that in z/VM and z/VSE I can break a large output spool file into several segments. I've Googled, and looked through the z/OS JCL Reference and User Guide and the JES2 Introduction and Commands manuals, but I can't find an equivalent mechanism for z/OS. Perhaps a different term is used in z/OS, and I'm looking for the wrong thing? Leslie Turriff State of Missouri Information Technology Services Division -- 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 -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: DFSMS and HFS/ZFS Datasets.
On 10/15/2011 12:29 AM, Ravi Gaur wrote: 2 cents should have seperate storage group..it's going to easier life with the issues/problem comes and also defining various options which not needed with basic file system... Look in DFSMS : Implementing System-Managed storage 7.9 - Defining SMS Construct for HFS Data.. ... In an SMS shop those differences are easily handled via DATACLAS and MGMTCLAS assignments by ACS routines. I would think whether you would want to isolate HFS/ZFS to a separate STORGRP or STORCLAS would depend a lot more on other factors unique to your shop. If you have a significant number of developers using HFS/ZFS who think DASD is infinite, that could be an issue in favor of a separate DASD pool, but you can have exactly the same issues with standard MVS data sets used for testing under either batch or TSO, so HFS/ZFS is hardly unique in its potential for conflict with DASD space that might be needed by production work loads. With modern highly-cached DASD subsystems with PAV support, performance conflicts really shouldn't be a motivation for a separate STORGRP either. Different backup criteria for HFS/ZFS files can also be handled by MGMTCLAS, unless you choose to use a full-volume backup strategy as one user described, or some totally independent backup strategy that gets down to the level of changed files within an HFS/ZFS file system. If most of your DASD is still non-SMS and you have a non-trivial system, you have other much more serious issues. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: Sad News: [IP] Dennis Ritchie dies
On 10/12/2011 09:11 PM, David Boyes wrote: ��z{S���}�ĝ��xjǺ�*'���O*^��m��Z�w... Attempt at reposting after manually decoding the garbaged base64 utf8 block in David's post: Begin forwarded message: From: Tim Finin Date: October 12, 2011 9:32:06 PM EDT Subject: Dennis Ritchie has died Sad news. Rob Pike reports on Google Plus that Dennis Ritchie died at his home this weekend after a long illness. Ritchie created the C programming language and was a key contributor to Unix. In 1983 he received the ACM Turing Award with his long time colleague Ken Thompson for the development of operating systems theory and the implementation of the UNIX. He was elected to the National Academy of Engineering in 1988. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: WLM Classification Rules question
On 10/07/2011 01:25 PM, Shmuel Metz (Seymour J.) wrote: In<4e8f1ad5.5020...@acm.org>, on 10/07/2011 at 10:29 AM, "Joel C. Ewing" said: If the list server does not support broadcasting multi-part MIME messages, I don't see any easy way it could possibly handle an incoming message that is UTF-8 base-64 encoded without first decoding the message to UTF-8 to allow adding the list footing without requiring a 2nd part in the message body. Why? The footer is ASCII. Any valid ASCII is valid UTF-8. Just BASE64 encode it and append. I don't claim to be an expert on MIME Email syntax (only to know enough to be dangerous), but there is a definite end-line for a base64 encoded block (short line with trailing "="'s). Is it legit to follow one base64 encoded block immediately with another in the same message part without any additional headers? Appending two base64 encodings in this way at the line level is not equivalent to a single base64 encoding as there is still a distinct division between the two blocks. But if allowed, that would be certainly be a simpler solution than merging the two blocks into one (which requires at least a partial decoding of the last line of the first block to fill out that line) But, in the last case I received, this wasn't what the list server sent. It separated the second UTF-8 base64 footing from the previous base64 encoded block with an intervening multi-part separator line sequence. I thought once you added a multi-part separator, that required a new set of Contents headers, not to mention the other multi-part headers that were missing. In any event, what I'm receiving is definitely not making my Thunderbird client happy and I've generally had good success with it; although, I can well believe other Email clients may behave differently on questionable Email formats. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: WLM Classification Rules question
On 10/07/2011 09:35 AM, Roberts, John J wrote: Anyone have any idea what is going on here? This looks to me like the email I am receiving is ill-formed. Some installations have special handling for email sent to addresses outside the home organization. The email is intercepted and stored in a database. What gets sent to the destination email address is a rich text page containing hyperlinks the recipient can use to access the original message, but it works only for a limited period of time (like 30 days). Of course, when these rich text pages are sent, they appear as hexadecimal garbage to the list servers. I am plagued with this at my installation. I always have to watch that my posts are small since big messages trigger the special secure handling. John But in this particular case, the data definitely is just UTF-8 text not RTF, only it is base64 encoded (which shouldn't be necessary these days - see 8BITMIME, circa 1994). The message body appears to be quasi-multi-part MIME with some missing multi-part headers (specifically the initial ones that are required to indicate the message body is multi-part and the multi-part separator string) and also with missing Content headers for a 2nd base-64-encoded part in the message body (apparently a base64-encoded ibm-main footer added by the list server), and finally there is no final multi-part header to indicate the end of the 2nd part. If the list server does not support broadcasting multi-part MIME messages, I don't see any easy way it could possibly handle an incoming message that is UTF-8 base-64 encoded without first decoding the message to UTF-8 to allow adding the list footing without requiring a 2nd part in the message body. From the data I am seeing, it looks like the list server attempted to add the footer message in a format consistent with the original message data (UTF-8, base64), but was unable to handle the generation of all the headers required to make this really work. If that is the case, then either the arrival of UTF-8 base64 rather than UTF-8 Text must somehow be prevented, or the server must be smart enough to convert the UTF-8 base64 to UTF-8 Text before attempting to process it. Or, perhaps this failure was triggered by an incoming message that was actually multi-part (someone with email configured to send both text and html formats all the time?) -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: SPOOL move
And usually the only reason one would need to do this migration IS a new DASD subsystem and the removal of an old subsystem, which typically requires moving many volumes, not just SPOOL. Having a product like TDMF or FDRPAS makes such moves non disruptive and so painless, that every one of our DASD subsystem acquisition contracts for over a decade has included temporary availability of such a product to do the migration! Considering the capacity of current DASD subsystems, to do it any other way may require a month or more overlap between old and new DASD with at least some brief down time, unless you are willing to incur substantial down time for mass DASD moves. With one of the above migration tools, the moves can be done non-disruptively over at most several days. Joel C. Ewing On 10/07/2011 09:17 AM, Mark Jacobs wrote: If you have a product such as IBM's TDMF or FDRPAS it will happily move active spool volumes to another device. Mark Jacobs On 10/07/11 10:10, R.S. wrote: I have to move JES2 SPOOL volumes from one control unit to another (another dasd box). IMHO the simplest method is to stop JES2, move the volumes using DSS copyvol and voila. However this method require an outage which can be long, because SPOOL is large. I just looked at the documentation - it seems such movement could be done by addition/deletion volumes to spool. However deletion of spool volumes is not so funny - when some long running task keep output on the volume, I can only wait. Oboviously closing tasks like CICS or DB2 means production outage, so closing JES2 is not big issue in such case. Q: Is there any method to move spool volumes (or at least majority of them) ? Any clue? Regards -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: WLM Classification Rules question
This is another of several recent messages that displays as total garbage to me (Thunderbird 6.0.2 on Fedora 15). Anyone have any idea what is going on here? This looks to me like the email I am receiving is ill-formed. Using "view source" I can see that the main message body is formatted with content Content-Transfer-Encoding: base64 Content-Type: text/plain; charset=UTF-8 and what immediately follows those headers does appear to be legit base64 down to a point. Feeding just those initial lines into web base64 translators gives valid results: "how about *SI*, Subsystem Instance? ... (etc. including some characters that definitely require UTF-8) But immediately on the next line following this valid base64 content is: --bcaec547ca51d8417704aeadc3a4 DQotLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tDQpGb3IgSUJNLU1BSU4gc3Vic2NyaWJlIC8gc2lnbm9mZiAvIGFyY2hp dmUgYWNjZXNzIGluc3RydWN0aW9ucywNCnNlbmQgZW1haWwgdG8gbGlzdHNlcnZAYmFtYS51YS5l ZHUgd2l0aCB0aGUgbWVzc2FnZTogR0VUIElCTS1NQUlOIElORk8NClNlYXJjaCB0aGUgYXJjaGl2 ZXMgYXQgaHR0cDovL2JhbWEudWEuZWR1L2FyY2hpdmVzL2libS1tYWluLmh0bWwNCg== which looks like a multi-part Email separator (there were no earlier multi-part headers to make this legit), followed by more base64 content minus the required Content headers. Interestingly enough the final encoded piece when manually decoded appears to be the ibm-main footer: "-- 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"; Is this problem possibly being caused by some people sending multi-part MIME email to the list and totally confusing the list server which doesn't allow multi-part? Or is there some bug in the list server's way of appending the list footer when the post is base64 encoded? Or, is this just another nasty side effect of sending email through down-level email clients or servers that still after all these years fail to properly support 8-bit email and force auto-conversion to base64 encoding at some point during the transmission? JC Ewing On 10/06/2011 11:20 PM, Cobe Xu wrote: ��z{S���}�ĝ��xjǺ�*'���O*^��m��Z�w!j��Ɔ�r&�WB�4���7V'7�7FV���7F�6S�#��R y-y=y� yy� ... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: DFSMShsm DUPLEX tapes.
If your object is just to cease duplexing and get down to a single HSM volser from two, and it doesn't really matter whether the one left used to be the primary or the duplex volume, then one possibility is to do an HSM TAPEREPLACE to convert DUPLEX volumes into the primary volume, leaving no duplex volume. TAPEREPL will still take some time as it must do updates to the HSM Control Datasets to replace all primary volser references with the duplex volser, but less time and no tape handling compared with a RECYCLE. If you plan to then turn off duplexing and run in this mode for any length of time, I trust your tapes are not real cartridges but virtual cartridges that are duplexed in some other way. A single physical cartridge is always at much greater risk from physical damage than two, especially with one at an alternate site, and running without HSM data duplexed in some fashion is ill-advised. I'm also unaware of any command to just delete the duplex volume. I would be inclined to suspect that the only references to a duplex volser is a field in the HSM OCDS record for the primary volume, a member in the RACF HSMHSM tapevol profile, and volume status information in your tape management system; but I would still be hesitant about trying to purge that data manually in the absence of a direct DFHSM command to do the delete. It doesn't seem like an HSM duplex delete function should be that difficult to implement: Perhaps it should be considered as a possible candidate for a SHARE requirement. JC Ewing On 10/05/2011 11:02 AM, Barry Jones wrote: Hi Listers, Does anybody know if there is someway to do a DELVOL of just the duplex copy of a tape. IE: ML2 tape VOLSER 123456 has a duplex copy on VOLSER 654321. I want to keep 123456, but delete the association with 654321 and return it to scratch. I know I can do it by SETSYS DUPLEX(MIGRATION(N)) and then RECYCLE EX VOL(123456), but I have thousands of volumes and dont want to have to recycle all of them. I've scanned the fine manual, but cant see anyway to do it. B. ... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: Suggestion for a Job running Under a Loop
On 09/20/2011 07:45 AM, Steve Comstock wrote: On 9/20/2011 4:22 AM, Jake anderson wrote: I am not doing a COMPILE,BIND,LINKLED. Its the application trainee users who perform this task. Not the same user but different users performing same task as part of training. Just curious to know if there are any limitation on TGNUM or CPU time to know if a Job is taking more resources. Jake On Tue, Sep 20, 2011 at 6:04 AM, Lizette Koehlerwrote: Jake, I am a little confused. Are you doing a COMPLE,BIND,LINKED every time you run your program? Lizette Ah. For trainees running training jobs, you really should put TIME= on JOB statements and OUTLIM= on all SYSOUT DD statements. And if this is a training exercise, the instructor in charge should already have a solution for the exercise and know what is a reasonable upper bound for OUTLIM on the SYSOUT datasets and TIME for the execution step of the exercise. Forcing such limits in some way, at least on the test execution step, is infinitely simpler than trying to dynamically determine if a running job step is doing something reasonable or not and still achieves the desired end result of not allowing bad test jobs to waste huge amounts of system resources. A default job step TIME value may be easily assigned by having a separate job class in JES for the test/trainee jobs and specifying the default there. Another simple possibility would be to require the trainees to use a supplied PROC for their execution which includes the desired limits for the test execution job steps. A more difficult solution would be to use JES exits to enforce the limits for specific test job classes or job names. If those limits also catch a few trainee programs that might produce correct results, but in a terribly inefficient way, that is also a good training experience. Considering the approximate MIPS ratings of typical processors today, any program that uses more than a few seconds of CPU time must of necessity contain program loops. Distinguishing productive loops from inefficient loops and from non productive loops for an arbitrary program is a non trivial exercise. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: Jabberwocky was:PTF question
Try "`Twas brillig, and the slithy toves Did gyre and gimble in the wabe: ..." http://www.jabberwocky.com/carroll/jabber/jabberwocky.html Must be careful to preserve correct wording or you could distort the entire meaning of the poem :) JC Ewing On 09/05/2011 10:14 PM, Ted MacNEIL wrote: Did gyre& gymbol on the wabe. PS: I don't think troth is the 'correct' spelling. - Ted MacNEIL eamacn...@yahoo.ca Twitter: @TedMacNEIL -Original Message- From: Ed Finnell Sender: IBM Mainframe Discussion List Date: Mon, 5 Sep 2011 22:16:35 To: Reply-To: IBM Mainframe Discussion List Subject: Re: PTF question 'Twas brillig in the slimy troth... In a message dated 9/5/2011 8:55:03 P.M. Central Daylight Time, jo.skip.robin...@sce.com writes: KFRoaXMgaXMgaG93IGl0IGxvb2tzIGluIG15IFNlbnQgZm9sZGVyLi4uKQ0KDQpBcyB0aGV5IHNh -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: PTF question
On 09/05/2011 05:39 AM, גדי בן אבי wrote: R200, R300 and R500 are the versions of TWS that each ptf applies to. Since I don't have TWS, I don't know what the current versions are. If you can, use SMP/E RECEIVE ORDER command. You can then specify the APAR (PK85334), and the correct PTF, and it's prerequisites will be retrieved from IBM's servers. Gadi -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Alvaro Guirao Lopez Sent: Monday, September 05, 2011 1:26 PM To: IBM-MAIN@bama.ua.edu Subject: PTF question Hi listers, I have a question related with PTFs releases download: When I see the next table at IBM web: Applicable component levels - R200 PSY UK49618<https://www14.software.ibm.com/webapp/set2/ordermedia/shopCart?ptfs=UK49618> UP09/10/05 P F910 - R300 PSY UK49649<https://www14.software.ibm.com/webapp/set2/ordermedia/shopCart?ptfs=UK49649> UP09/10/06 P F910 - R500 PSY UK49655<https://www14.software.ibm.com/webapp/set2/ordermedia/shopCart?ptfs=UK49655> UP09/10/06 P F910 What is 'R200', 'R300' and 'R500'? How can I know what PTF I must download an apply to my systems? Thanks for your help. -- Un saludo. Álvaro Guirao If you can view any one of the PTFs, it will most likely have "IF FMID" and "VER" clauses in the control information which will clearly tie all the various versions of the PTF to the FMID for which they apply. For most (all?) product areas there will be an obvious correspondence between the four-character indicators and the product version FMIDs , like based on last characters of FMID. Once you know that correspondence holds for the product area in question you won't even need to check the internals of some PTF. I never have understood why after all these years IBM hasn't bothered to change their convention and use the full FMID in the applicable component level description and avoid all the unnecessary confusion for new users! If you don't know for sure which FMIDs are on your system, that can generally be determined by displaying the appropriate global zone gzone data from the ISPF SMP/E dialog. Or alternatively, you can use the SMP/E dialog to search the target zone for a specific product FUNCTION FMID sysmod and see which is in your target zone. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: ISMF QUESTION
On 08/30/2011 10:58 AM, esmie moo wrote: Good Morning Gentle Readers, I am trying to compile a report of dsns which use a specific MANAGEMENT CLASS via ISMF. The fields I choose are 3& 26. For some reason I does not show the MANAGEMENT CLASS in the report. It has under the MANAGEMENT CLASS NAME. I tried selecting STORAGE CLASS (27) the same thing happens. Is there something special I have to do when selecting these fields if so please advise me? Thanks in adance. Since SMS class data is in the catalog, I would suspect that means you need to generate your dataset list "from catalog" rather than "from VTOC". If any dsns are migrated, you also need to specify "Y" for "Acquire data if DFSMShsm migrated". Any displayed DATA or INDEX components of VSAM datasets will show null class values, as class values belong to the CLUSTER; so if your object is to list allocated space associated with VSAM files (which is only on DATA and INDEX) for a specific MGMTCLAS, you may not be able to do this easily with ISMF. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: FTP JESLRECL Limit
On 08/29/2011 05:22 PM, Paul Gilmartin wrote: On Sun, 28 Aug 2011 11:38:29 +0530, amit wrote: i believe your answer points to the Value and range in JES sysparms...can you set that value to 999:)? I'm mystified that a JES sysparm should appear in the publication cited, z/OS V1R12.0 Communications Server IP Configuration Reference SC31-8776-18 Which is a TCP/IP publication, not a JES publication. I find no reference to JESLRECL in: JES2 Initialization and Tuning Reference Document Number SA22-7533-10 which would seem a more appropriate publication. Where should I be looking? What's the name of the PARM? why not? Can't find where to set it. On Sun, Aug 28, 2011 at 11:09 AM, Paul Gilmartin wrote: On Sat, 27 Aug 2011 16:05:12 -0500, Mike Schwab wrote: http://publib.boulder.ibm.com/infocenter/zos/v1r12/index.jsp?topic=/com.ibm.zos.r12.halz001/cjeslre.htm Why? On Sat, Aug 27, 2011 at 9:40 AM, Paul Gilmartin wrote: ftp> quote site jeslrecl=999 200-Jeslrecl parameter (999) must be between 1 and 254. Jeslrecl ignored. 200 SITE command was accepted Why? -- gil The "jeslrecl" parameter is not described in the JES manuals because it is not a JES parameter. It is a sub-parameter of the "site" command of the MVS FTP server, hence it is described in the manual that describes the FTP server commands, as one should expect. My reading of the IP Configuration Reference manual indicates that when submitting jobs to JES via FTP, jeslrecl plays the same role as the LRECL in a DD statement that allocates an INTRDR SYSOUT to spool a job to JES. One shouldn't expect to find "jeslrecl" in the JES manuals any more than one would expect to find the DD LRECL parameter described in the JES manuals. I think the restriction on JES INTRDR LRECL was raised at some point to 32KiB. If so, the FTP jeslrecl maximum value is a restriction imposed by FTP, probably for consistency with older versions of MVS. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: alter dataset blocksize
On 08/29/2011 12:56 AM, MONTERO ROMERO, ENRIQUE ELOI wrote: Hi, Why to preserve or change the date of the dataset? An user had a library (PDS) with the incorrect BLOCKSIZE, so the HSM backup/migrate did not work for this dataset, we asked to the customer to fix the dataset using standard procedure (create new one, copy, delete old& rename the new one). But... The user told us, it was fixed, so i went to check it, found i could perform an hsend backds ..., it was correct, and check the library info, found it with the correct blocksize, and also found that the creation date was the same, "it was not changed". It make me surprised, couse i have no explanation why this happens. Thanks, Enrique Montero The far most likely cause of the original problem is that the user ran a batch job step to create a member in the PDS and either the program DCB or the DD statement for that file in that job step explicitly specified the inconsistent block size. Perhaps when informed of the problem the user recognized what he had done and simply ran another job step to store a new PDS member specifying the correct block size to reset the PDS block size attribute and deleted any members that were written with incorrect block size. That doesn't alter the PDS create date and may be much more efficient than creating a new copy of the PDS if you know this is what caused the problem. In fact, you likely would have to fix the original PDS in this fashion before you could read the old members to copy them to a new PDS, and once it is fixed the new PDS becomes redundant (this is not to say that there might be other ways to trash a PDS where a new PDS creation would be desirable). Or, a less sophisticated user might have found he couldn't read any old members in the PDS and just restored a back-level version of the PDS with DFHSM, which as Allan noted also preserves the creation date. DFHSM auto backup uses the data set "changed" bit, not the creation date, to determine if the dataset has been modified since the last successful backup; but an explicit "hsend backds" will create a potentially redundant backup even if the data set was just restored by the user from a DFHSM backup and the "changed" bit is off. Analyzing SMF records and DFHSM logs can give some clues about what was done, but if you are curious sometimes the simplest approach is to just ask the user what he did to resolve the problem. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: Cryptography Processor
unreasonable. RACF, or an equivalent security package, is a major player in the protection control design, but z/OS internal design and the hardware are what make it work. In a normal z/OS installation, only a z/OS system programmer could introduce code into the system that would affect z/OS operation globally, and maintenance philosophies and the restricted number of system-level z/OS vendors and the fact that code distribution comes direct from those vendors and requires manual installation action to install make it very unlikely that malware would come via that route (it would probably destroy a vendor's business if it ever happened). Application programmers can in general only introduce code that would cause failures within their own limited application area, not compromise z/OS itself, and maintenance philosophies, multi-party testing, and installation standards would make it highly unlikely that any non-approved code from outside (which could invite copyright or patent violation exposures) could be introduced, much less malware. The largest body of z/OS users, end users of applications, can at best only alter data within their application area and would not have any interfaces or access to alter any code that manipulates the data; so none of their actions can introduce a virus or malware. It's possible some subset of end users might also have access to TSO or the Unix Shell, where there are tools to create and execute code; but to use those tools to create or install malware would be a deliberate act with multiple steps, not an automated process a virus could easily exploit, and any potential damage would be restricted to those limited non-system directories/files/datasets to which that user had UPDATE access and to his own user session. z/OS simply does not allow end users to exchange data in forms that potentially have hidden embedded executables or allow user "data" objects to be automatically executed unintentionally because of hidden type codes, much less allow such code to alter arbitrary files or system control information unrelated to the original data or user. Those "features" in MS applications and MS Windows environments open many paths that can be and have been exploited by viruses. Other virus exploits have attacked specific server flaws on specific platforms to introduce executable malware code, but for many reasons (different memory usage patterns, memory isolation/protection, incompatible hardware code, z/OS code and dataset isolation/protection, internet firewall protection, fewer TCP/IP server types to exploit), those approaches are very unlikely to succeed on z/OS. The cryptographic processor is optionally used during the installation of z/OS maintenance, but merely as a tool to validate that no failure occurred during an electronic transmission of the software. The reliability/security of the z/OS system software is established by validating the source of the code to a trusted vendor (IBM), not by encrypting it. The primary significance of the cryptographic processor on z-architecture is in the offloading of the encryption of application data for securing application data and for maintaining application encryption keys in a physically secure manner. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: UCAT allocation to CATALOG task
On 08/21/2011 02:09 AM, Anthony Fletcher wrote: Does anyone on the list remember advice that a change occurred with z/OS 1.11 that might result in UCATs being allocated to the CATALOG task under z/OS 1.11 early on in IPL, when this almost certainly as not happening under z/OS 1.10? We have been upgrading a 3-LPAR non-SYSPLEX system from z/OS 1.10 to 1.11, and didn't have any problems with the first two, but in the last one ended up in a VARY OFFLINE PENDING situation when a volume was varied offline. The system is not a sysplex but does share several volumes across the LPARS. A F CATALOG,UNALLOCATE(catname) command did unallocate the catalog and the allocation against the volume that we wanted t take offline. ... At least during normal system operation a UCAT is only allocated if there is a reference that requires it to be opened, and it tends to stay open and allocated at that point even if not needed. If you are certain it is not needed, try closing the UCAT first (F CATALOG,CLOSE(catname) and then see if you can UNALLOCATE it. You can't un-allocate while it is still open. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: ABEND - 637-04
The requirement was that a physical block not be less than 18 bytes. Older tape technology could not distinguish blocks shorter than 18 bytes from tape noise. Since an FB dataset with LRECL < 18 could easily end up with a final block containing only 1 record (possibly even in the middle of the dataset if extended with DISP=MOD) I would think the physical block restriction would also have to imply a minimal LRECL of 18 as well. Similar considerations would also apply to VB files after taking the BDW and RDW bytes into account, and in this case there would have had to be an minimum actual record length of 14 (including RDW) for the last record (LRECL for VB is only the max allowed) in order to guarantee success. I'm pretty sure all those restrictions no longer apply, since all tapes subsystems from 3490 and later compact the logical blocks known to the operating system into physical "super blocks" and the physical block size on the tape is no longer under the control of operating system code. Joel C Ewing On 08/19/2011 06:48 AM, Shmuel Metz (Seymour J.) wrote: In <1240651948-1313637071-cardhu_decombobulator_blackberry.rim.net-95694995-@b12.c1.bise6.blackberry>, on 08/18/2011 at 03:11 AM, Ted MacNEIL said: I thought there was a minimum of 18 bytes for LRECL. No. CIG -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: Last card reader?
On 08/17/2011 06:05 PM, Ed Gould wrote: Rick, My vague memory was a 407 read cards and you had a board you could place wires what you wanted to o print add subtract and take the results and print it out on the printer(132? Positions?). The wires were collided. So you could manipulate the data if needed and move it to the print buffer . I don't think you could divide just add subtract and maybe multiply. ( not sure about multiply). Ed I used to have a xerox copy of a 407 board wiring configuration that was supposed to do multiplication. I never tried it, as no longer had access to a 407 at the time I first saw this, but preserved it just because it was fascinating that it could be done at all. The 407 had the ability to suppress card advance and re-read the same card multiple times. Someone figured out how to use that feature to perform multiplication by repeated addition. No doubt would have been a slow as heck, but amazing that it could be done at all. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: ABEND - 637-04
Knowing John's background, I would suspect he may have intended to state "For RECFM=U, 32760 MAY indeed BE optimum", or perhaps intended to restrict the remark just to load libraries. I believe all the arguments in the archives for using block size 32760 with RECFM=U on 3390 were specifically directed to datasets that are used for Load Libraries. While this is probably the most common usage for RECFM=U, there is nothing to disallow other usage of RECFM=U; and those other applications, unlike program management, might not be smart enough to utilize shorter blocks to effectively utilize a DASD track. I agree that it has been established that BLKSIZE 32760 makes sense for a RECFM=U load library on a 3390. But, a RECFM=U data set that is not a load library would not necessarily meet the additional requirement of containing a combination of long and short blocks, which as John states is a necessary condition for a block size greater than half-track to have any chance of decent track utilization on a 3390. Joel C. Ewing On 08/17/2011 12:54 PM, John Eells wrote: That really depends on the RECFM and, for variable blocked RECFMs, the distribution of block sizes. For RECFM=U, 32760 is indeed optimum (search the archives for why), and for some (not necessarily common) distributions of variable blocks it might well use space better than half track blocking. Anything that writes a combination of long and short blocks can yield surprising results. (We found that z/OS fonts, for example, use the least space at a counterintuitive block size.) For plain old sequential FB data, though, it's quite right that 32K is a poor choice of block size. Bill Fairchild wrote: 3390 full track size is ca. 56K, and 3380 is ca. 47K. But you were right - 32K is still not optimal for 3390. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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: Soft Capping an LPAR
If you have RMF, you can run RMFPM (free download) on a workstation and have it track both the LPAR MSU and the LPAR 4-hour rolling MSU in near real time and periodically check the values in RMFPM, perhaps even take a few extra seconds to have RMFPM display a graph of the values to make trends more obvious. If the object is just to get advance warning of capping, it isn't necessary to continually monitor the values, since you are dealing with a 4-hr average. If you hit your cap, you usually have had a load problem (or application performance problem) for an extended period. Of course the potential warning time this gives depends on whether the LPAR is configured in a way that allows for a peak MSU that is higher than the soft cap MSU by a significant factor: A peak LPAR MSU capability greater than 4 x cap MSU means capping could occur from less than 1 hour of peak load. More modest peak/cap MSU ratios restrict how quickly the 4-hour average can rise and also make it more likely that abnormally high loads will hit that peak and generate some response complaints from end users even before capping occurs, alerting that there is a load problem that needs to be resolved before capping makes it much worse. Using something like RMFPM to get a general idea of what constitutes "normal" patterns of MSU usage for an LPAR for different hours of the day and different days of the month can also teach you when you need to monitor things more closely and also make it more likely that changes that introduce application performance bugs will be spotted as an abnormal pattern and fixed sooner rather than later. JC Ewing On 08/12/2011 09:05 PM, David Middlebrook wrote: I have a z10 processor running multiple customer systems where we are using VWLC, one of the systems was soft-capped during peak processing for the day which caused multiple CICS regions to go max task resulting in time outs, retries, transaction back out etc which just added to the demand on CPU resources. Is there a monitor that will raise an alert when a system is approaching being soft capped? TIA ... -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- 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