Keyboard Shortcuts
Likes
- H390-MVS
- Messages
Search
GETDTE - MVS 3.8j TK4
Hi
I am trying to implement GETDTE on my TK4 UPD 8 system. I get an error in JOB4 in the installation process. I suspect this is related to missing usermods. Where are the usermod library located? *? Prerequisite: User Modifications? ? ? ? ? ? ? ?
*? ===================================================================
*
*? ? ?Two user-mods, ZP60014 and ZP60038, are REQUIRED to process
*? CLIST symbolic variables via the IKJCT441 API on MVS 3.8J before
*? using GETDTE.? Otherwise, GETDTE will function but will NOT? ? ?
*? create CLIST variables.
*
*? ? ?More information on the above user-mods can be obtained from
*? the following website:
*? ? ? ?http://www.prycroft6.com.au/vs2mods/
*
*? ? ?Check your system to determine if one or both user-mods are
*? required.? ZP60038 requires ZP60014.
GETDTE ¨C System DSECTsbest regards Dagfinn D. Hammar LOC OBJECT CODE ADDR1 ADDR2 STMT SOURCE STATEMENT 000000 3125 TIOT DSECT 3126 IEFTIOT1 Task *** ERROR *** 00000 3127 TIOTLEN EQU *-TIOT Lengt GETDTE ¨C System DSECTs LOC OBJECT CODE ADDR1 ADDR2 STMT SOURCE STATEMENT 000000 3129 JSCB DSECT 3130 IEZJSCB Job/S GETDTE ¨C System DSECTs LOC OBJECT CODE ADDR1 ADDR2 STMT SOURCE STATEMENT |
Re: HELP! How to return to MVS Console
Hello Fish,
Am 26.02.2024 um 19:25 schrieb Fish Fish <david.b.trout@...>: Yes, because I've not seen any adverse effects.Just for the record, for years I'm ending OS/390 by just issuing a quit to the Hercules console to make it close the disk files properly.So you DO care about Hercules closing its disks properly but DON'T care about your OS/390 guest closing its disks properly? In contrast, sending a SIGTERM to Hercules does leave the disk files in a state where they need repair. You're a strange person, Patrik. :)Thanks for the compliment I hereby return. ;-) See my original comment. No adverse effects seen, do only on a quiet system.Of course I do this while the system is idle from user-submitted jobs.But to hell with any system processes who's disk buffers might not have been flushed yet? You don't care about those? They're not important, right? For *me* it works. Your mileage may vary. Do I smell some personal offense of yours?Too lazy to manually wade through the shutdown sequence,Of course! No one wants to mess with any stinking automation after all. Just pull the plug! Apparently.Had not noticed any adverse effects.Well, each guest is different I guess. OS/390 must obviously behave quite a bit differently than other guest operating systems. Exactly.Your mileage may vary.Indeed. If it works for you? Fine. But I personally would *never* publicly recommend such an irresponsible highly risky procedure to others.As I said, it works for me without visible or even adverse effects. But then as I guess you say, your mileage may vary. Right? >;-)Correct. p.s. I'm curious: do you power off you Linux host operating system the same way? Just pull the plug?No. Because at next boot, I'll get journal recovery notes. I'm not aware of any WTO messages regarding my habit on OS/390. :wq! PoC |
Re: problem meet when run dfsmsdss restore
Sorry, I had assumed you had a running system already.
The absolute simplest way to get a driving system up is to use Customized Offerings Driver (COD).? It's a single volume z/OS at V2.4 (or 2.3), that you can IPL direct from a USB drive.? You can order it from ShopZseries and I think there is no cost so long as you already have a mainframe and access to ShopZ. ? It's a full fledged system, but is only meant to be used to install a new OS when you have a new box that doesn't support your old z/OS, or you have no previous z/OS to start with.? The process is actually jsut IPLing DFDSS and creating the volume, then when you are done, you IPL the single volume system you just created.? From that point you can use it to install your volume copies or install from a z/OSMF installation package. You could also build your backup with AWSTAPE on your mainframe platform and use that tape for the stand alone install, but COD is way easier.? There are also unterse products that run on the PC platform which would get you to where you need to be also. |
Re: problem meet when run dfsmsdss restore
Hi Shixinlong,
Have a look at wwwunterse, it runs on windows platforms and Linux also There is a Download button on the web page, which will ask you for some name and email details. You must not publish the files to any forum or public download, if you wish to tell someone about wwunterse then please only pass the link Regards, Tom |
Re: problem meet when run dfsmsdss restore
Hi Brian,
Thanks so much for your time and detail steps. Two more questions per your reply: The first? is about pre-allocate datasets on hercules. Can i use dasdinit64 command to create it? or some command else? The second is from where can i run the unpack procedure as i don't have a z/os system up on hercules. I have only a standalone ickdsf and a standalone dfsmsdss restore.? My dasd device is 3390-9,original z/os version is 2.2. the hardware is z15. I need to restore the resident volume and catalog volume to hercules first so that i can ipl my system and restore the rest volumes via JCL. Thanks again for your kindly and patient guidance. |
Re: problem meet when run dfsmsdss restore
This procedure works 100% of the time.
On the source system, (the real mainframe in your case): This first step uses DFdss to dump the volume to a temporary dataset that we will process with TRS to make it portable //DUMPDASD EXEC PGM=ADRDSSU????????????????????????????????????????? ? //SYSPRINT DD SYSOUT=*?????????????????????????????????????????????? ? //DISK1??? DD UNIT=3390,VOL=SER=Z25PTS,DISP=SHR??????? <-- this is the volume that you want to copy to Hercules (or any system) ????????????? //TAPE1??? DD DSN=SMSTAPE.Z25PTS.VOLUME.BACKUP,????? <-- this is the output file, which for me is on DASD, but can be on tape but the next step runs faster on DASD ?????????????? //??????????? DISP=(,CATLG,DELETE),????????????????????????????????? ? //??????????? UNIT=3390,SPACE=(TRK,(6000,6000),RLSE)???????????????? ? //SYSIN??? DD *????????????????????????????????????????????????????? ? ?PARALLEL??????????????????????????????????????????????????????????? ? ?DUMP FULL INDDNAME(DISK1) OUTDDNAME(TAPE1) CANCELERROR -??????????? ? ?? COMPRESS OPT(4) ADMINISTRATOR???????????????????????????????????? ? //*??????????????????????? This step makes the temporary dataset portable.? It can be sent via FTP to anywhere in BINARY format.? ?????????????????????????????????????????? //PACK0 EXEC PGM=TRSMAIN,PARM=PACK??????????????? <-- this step compresses the data in IBM's TRS format ?????????????????? //SYSPRINT DD SYSOUT=*?????????????????????????????????????????????? ? //INFILE DD DISP=(OLD,DELETE),DSN=SMSTAPE.Z25PTS.VOLUME.BACKUP??????? <-- this is the dataset created in the first step //OUTFILE DD DSN=BRIANW.Z25PTS.TERSED,DISP=(NEW,CATLG),?????????????? <-- this is the output that will be FTP'ed to your PC MAKE sure to transfer in BINARY format //??????? UNIT=TAPE,LABEL=(1,SL),RETPD=3,??????????????????????????? ? //??????? DCB=(RECFM=FB,DSORG=PS,LRECL=1024)????????????? The OUTFILE above needs to be FTP'ed in BINARY format to wherever you are sending it.? If to your PC, you will need to then FTP it to Hercules, but you can also FTP it directly to your hercules instance thereby removing the need to FTP it twice.? I use the FTP to the PC so that I have a copy of the dataset that I can fall back to if something happens.? ??????????? On the target Hercules system, first create a new dataset to FTP the file from the previous step into.? You need to manually create it because FTP doesn't like sending really large datsets without pre-allocation, but you could certainly use SITE commands to control the space and LRECL etc.? I jsut pre-allocate it because it's quicker for me. That dataset on your Hercules instance should ALWAYS be created as FB lrecl=1024 You then need to run another TRSMAIN to UNPACK the dataset that you just FTPed (same as previous step, but with parm=UNPACK make sure to use binary transfer in the FTP, preallocate as DSNTYPE=LARGE, because if you are sending 3390-9 or larger they tend to be big Most are very big pri=4000,sec=4000 in cylinders.? Most volumes are much smaller, but I have 8 3390-54 volumes that are just for transferring volumes in one big SMS pool.? I delete them and copy the compressed CCKD 64 volumes every time I bring up hercules, so they are always empty when I start up. //UNPACK0 EXEC PGM=TRSMAIN,PARM=UNPACK??????????????? <-- this step decompresses the data from IBM's TRS format ?????????????????? //SYSPRINT DD SYSOUT=*?????????????????????????????????????????????? ? //INFILE DD DISP=(OLD,DELETE),DSN=BRIANW.Z25PTS.TERSED??????? <-- this is the output that will be FTP'ed to your PC //OUTFILE DD DSN=BRIAN.Z25SYS.VOLUME,DISP=(NEW,CATLG),?????????????? <-- This is the input to the next step TRS will set the DCB back to the original in the first step //??????? UNIT=3390,VOL=SER=TRAN01,?????????? <-- one of my 3390-54 volumes used as a temp holing area. ???????????????? //??????? SPACE=(CYL,(4000,4000)),DSNTYPE=LARGE Then run DFDSS restore on hercules system? (if needed, this can be a SArestore): ?? //STEP1? EXEC? PGM=ADRDSSU, REGION=0M ???????????????????????????????????? //*??????????????????????????????????????????????????????????? ? //*??????????????????????????????????????????????????????????? ? //SYSPRINT? DD? SYSOUT=*?????????????????????????????????????? ? //OUTDD1?? DD??? VOL=SER=G25SYS,UNIT=3390,DISP=OLD????????? <--- the volume you are making ?? //INDD1 DD DSN=BRIAN.Z25SYS.VOLUME,???????? <-- this is the dataset you ftped ?????????????????? // DISP=(OLD,KEEP,KEEP)??????????????????????????????????????? ? //SYSIN??? DD??? *???????????????????????????????????????????? ? ?RESTORE FULL -??????????????????????????????????????????????? ? ???????? INDDNAME(INDD1) -???????????????????????????????????? ? ???????? OUTDDNAME(OUTDD1) -?????????????????????????????????? ? ???????? CANCELERROR -???????????????????????????????????????? ? ???????? WAIT(2,2) PURGE ADMIN COPYV?????????????????????????? ? /*????????????????? This process needs to run for each volume you want to use.? It seems like a lot of work, but actually it's very straightforward. There are two other methods that I know of that also always work, but this is the one I use the most. There is no limit to the volume size, I have moved terabyte EAS 3390 volumes with no issues. Brian ???????????????????????????????????????? |
Re: problem meet when run dfsmsdss restore
CCKDDUMP program on actual system. On Sat, Mar 9, 2024 at 7:58?AM <shixinlong@...> wrote:
--
Mike A Schwab, Springfield IL USA Where do Forest Rangers go to get away from it all? |
Re: problem meet when run dfsmsdss restore
Hi Brian, yes i followed the step on the hercules home page. i converted the backup tape to aws format. and want to load it to the dasd inited.
after i restored the resident volume i can ipl the zos system in hercules. Could you kindly provide some step-by-step instructions on?copy a volume from a mainframe to Hercules. Your help is greatly appreciated. Thank you |
Re: Newbie question about REVIEW/RPF etc
For serious entertainment I recommend running a Cobol compile on a ICT 1501 (with tape drives only - as no disk available). Even a smallish program say 2000 cards took over 30 minutes but the tape units was a joy to behold, even more so than the later kit.
toggle quoted message
Show quoted text
Luckily only had to do so a very few times. This I think was at George Wimpy, the builders based in London, back in the late 60's early 70's. They also had a ICT 1900 system which was the one I usually worked on under George 2 ICT became ICL some years later and at there site in Queensway, London when I worked for them they had a LEO III which I also work on that sat in their basement and had zero idea how they managed to get it in as next to no doors to outside and the lift was passenger size only and small. May be they built it the way they do ships by putting in the engine after the keel is constructed but without any roofs or second floors, etc. That said the building was old though so may be not :) Vince . On 09/03/2024 04:21, Clement Clarke wrote:
|
Re: Newbie question about REVIEW/RPF etc
The LEOs.? Fabulous machines.? Multiprogramming way, way back. ? We used a compiler language called CLEO.? It was a bit like COBOL but came out years before COBOL, I think. ? I can remember playing around with the first COBOL compiler on an ICL 1902 with 4 tape drives. No disks (neither did the LEOs).? To see the ICL COBOL compiler at work was interesting.? All 4 tape drives going to compile a simple program! ? Of course, the Mainframes then only had a few hundred K of main storage. ? Even the IBM 65 that Shell Oil got in Melbourne Australia only had 3/4 of a meg.? And yet we managed to send out accounts to many thousands of people - many using Shell Heating Oil for their homes.? And there was a program that used the temperature to calculate who might need more heating oil in the next few weeks, and schedule the trucks to deliver it. ? Clem |
Re: Multi-Volume Dataset
Hi Daniel, I am not sure of the original purpose of your project. If it is to generate a workload for your system or to create large data sets then you may find the IVP jobs, IVP4 and IVP5 provided in SORT.MVS38.CNTL in the TK5 system as useful examples. |
Re: IEHMOVE fails with IEH453I DATA SET SYS3.TESTPDS.ASM NOT MOVED/COPIED BECAUS TO DATA SET PREVIOUSLY RACF DEFINED
Hi Clem,
I'm not sure if the dd option LIKE is available at your mvs level, if it is then an iefbr14 step before your copy would allocate the new dataset with minimal jcl Step 1 exec iefbr14 //newds? dd dsn=SYS3.TESTPDS3.ASM, // like=SYS1.MACLIB,disp=(new,Keep), // vol=ser=work01 Step2 ?exec iehmove apologies for the font etc, typing on my phone Regards, Tom
|
Re: Multi-Volume Dataset
Joe: Indeed, using your recommended blocking sped up the generate step by about 1/3 and the sort step by about 1/4. Thanks. On Fri, 2024-03-08 at 05:27 -0800, Joe Monk wrote:
|
Re: Multi-Volume Dataset
How about fixing your blocking to a more efficient blocksize? For a 3390, the tracksize?is 56,664 bytes per track, 15 tracks?per cylinder, and the optimal blocking is half-track. However, there is some overhead, so the max optimal usage on a 3390 track is (27998 * 2) = 55996. 27,998 / 80 = 349 records per half-track 349 * 80 = 27920 blocksize At a blocksize of 8,000 bytes, you can only fit 3 blocks on a half-track, or 6 blocks on a track, so, you waste about 8K per track! (55,996 - 48,000 = 7996) At a blocksize of 27920, you fit 2 blocks per track, with a waste of ?156 bytes per track! (55,996 - 55,840 = 156) - much more efficient So, change //SEQOUT DD UNIT=WORK,DCB=(LRECL=80,RECFM=FB,BLKSIZE=8000),? // DSN=&&SORTTEMP,DISP=(,PASS),SPACE=(CYL,(500,100),RLSE) ?? to //SEQOUT DD UNIT=WORK,DCB=(LRECL=80,RECFM=FB,BLKSIZE= 27920),? // DSN=&&SORTTEMP,DISP=(,PASS),SPACE=(CYL,(500,100),RLSE) ?? Not only will your job run faster, but the disk space will be better utilized and reduce waste. Joe On Thu, Mar 7, 2024 at 6:48?PM Daniel L. Srebnick via <dan=[email protected]> wrote: Here's the actual job.? If I change SEQOUT to UNIT=(3390,59) the JCL |
Re: Eliminating duplicate records.
Hi Alejandro,
I have E35 exit code that can drop duplicates. Please correspond with me offline too avoid to many messages in the group and I can most likely amend it to fit your requirements. At the moment it simply takes ONE offset and length in the record and uses this as the duplicate data test. Best regards, ¦Ó¦Ò³¾ ( e: botongrui@... ) |
Re: Eliminating duplicate records.
Hi David.
AFAIK (I tried anyways) the implementation of SORT present in Tk5 (I guess in Tk4- too, along with the one referenced by Jay Moseley in his website... both with all that 60's 2311/2314 DASD dependency) seems to be a very old one (I guess by its name, it is from OS/360...so 60's stuff), that pre-dates the one that usually folks refer to on blogs, guides, etc.. when someone googles around. There's a very nice, modern, pdf documentation about OS/360 SORT/MERGE for MVS 3.8 included (which, as Tom, pointed before, basically allows to use more recent DASDs), and it is consistent with the behaviour... So I think the popular SUM FIELDS=NONE comodity found around appears to be a facility of later versions and/or operating systems (I guess possibly on the 80's...who knows :-P). Cheers. |
Re: Eliminating duplicate records.
Hi:
I'm probably missing something, but on SORT program under JCL you can use SUM FIELDS=NONE to remove duplicates. For instance SORT FIELDS=(1,10,CH,A) SUM FIELDS=NONE Will sort and delete any duplicates under same criteria (having same value on positions 1 to 10). Also on COBOL it can be made, I can search for an example I made like 15 years ago if you want. Regards.? |