HyperPAV devices (was: PLPA and COMMON PAGESPACE Size)
Jim Mulder wrote: > It checks to see > if you specified WLM PAV when you defined the device in HCD. Since > WLM doesn't need to manage PAVs when the control unit has been told > to use HyperPAV mode, our intention was that the specification of > WLM PAV in HCD would be irrelevant for HyperPAV devices Well, that "would be irrelevant" thing brings me to the following question: If we want to migrate from PAV to HyperPAV, do we need to change the device definitions in the IODF from WLMPAV YesDevice supports work load manager to WLMPAV No Device supports work load manager and our WLM policy from Dynamic alias management . . . . . . . . YES (Yes or No) to Dynamic alias management . . . . . . . . NO (Yes or No)? -- Zaromil
Re: ISPF Display eXcluded lines (was: zOS Unix DOS2UNIX)
Gil, > While we're on editor techniques, can I do with ISPF > something I've found useful in XEDIT: momentarily and > reversibly display excluded lines. For example, after doing > the analogue of EXCLUDE ALL and a few FINDs, I can do > > DISPLAY 0 > ... which displays only the remaining excluded lines > > DISPLAY 0 1 > ... which displays all lines, whether excluded or not > > DISPLAY 1 > return to displaying only the nonexcluded lines. You could use "FLIP" command for "DISPLAY 0" / "DISPLAY 1". Never tried to do something like "DISPLAY 0 1". -- Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: z9 DRIVER 67
Sam, > Where did you see that? Is this something you ran into using > SA/390 or some other product that exploits an interface to the HMC? For last few years we have used exclusively GDPS Standard actions to manage all our z/OS systems/lpars (IPL, stop, reset,...). An SA/390 command using BCP ii is INGHWCMD STATCOM -- Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: z9 DRIVER 67
Sam, Has anyone else migrated to the z9 microcode level DRIVER 67? We are planning on moving to DRIVER67L this weekend. This is concurrent as we have all the requisites installed for it to be but seems to be a fairly large bundle so I curious if anyone has had any problems. we have problems with BCP internal interface - it is extremely slow now. -- Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: HMC - Load type SCSI
Radoslaw, > Regarding document mentioned by Brian (thank you BTW): this document is *very* interesting, but it describes technical details of IPL, but not how can I have it enabled, or why I don't have it. Redbook "Linux for zSeries: Fibre Channel Protocol Implementation Guide" ^http://www.redbooks.ibm.com/redbooks/pdfs/sg246344.pdf > 3.2 Enabling the SCSI IPL feature To IPL Linux from an SCSI attached disk (either from VM or in an LPAR), the SCSI IPL feature (FC 9904) must be installed. When FC 9904 is active, Linux images on zSeries are enabled to perform IPL from FCP-attached SCSI disks. z/VM 4.4 is a prerequisite to IPL Linux under z/VM. Installation of the SCSI IPL feature is non-disruptive on the z800, z900, and z990. However, a power-on reset (POR) is required to enable it. When installing the SCSI IPL feature, ensure that the Enable SCSI IPL check box is marked as shown in Figure 3-2. From the Reset profile, select the Options tab. and (more or less stating the same) VM63348: NEW FUNCTION - GUEST IPL FROM SCSI ENABLEMENT http://www-1.ibm.com/support/docview.wss?uid=isg1VM63348 > NOTE: For 2064/2066 only: A microcode upgrade is required to support Guest IPL from SCSI. When installing the upgrade be sure that the "Enable SCSI IPL" checkbox is selected on the IML Options panel. If the checkbox is not selected to enable SCSI IPL, VM will display the following message whenever a Guest IPL command is issued: HCPLDI2813E "CP is running on a processor or in a mode that does not support Guest IPL from SCSI disks". For 2084 An upgrade is also necessary but the feature is automatically enabled. There is no checkbox to select. For specific details see page 5-16 of the "Support Element Operations Guide Version 1.7.3" (SC28-6818-02) available via Resource Link -- Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: FW: Deleted PDS - K - kg - Kg
<--snip--> > Usually they mean "binary" kilo and mega, although it is not proper in > aspect of IS standard. However this is different story: for example > every RAM manufacturer use MB and GB instead of correct MiB and GiB. Probably because the "i" qualifier is relatively recent? I do remember Seagate and others using decimal capacities because it made their disks appear bigger (e.g., 65K tracks rather than 64K). <--snip--> I would say that it is still so (the 160GB Samsung disk I bought has about 149 GiB capacity) and have never thought of "it made their disks appear bigger" as a reason. Isn't that so because disk capacity is not base 2 (i.e. base 1024) dependent? <--snip--> > BTW: In Europe we *say* "kilo" as abbreviation of "kilogram", but we > *write* it properly: "kg", so assumption about kilograms is bad shot. Perhaps in your neck of the woods, but I spent fourteen years in Austria and saw plenty of stores with prices with just a K. And technically, it should be Kg, not kg, because the convention is to capitalize multiplicative prefixes. <--snip--> K instead of kg: prices written manually with a chalk on small paper or wooden surfaces? Or maybe it is normal to me and I just did/do not perceived that? And I don't know what "technically" means in this context, but it is kg: http://physics.nist.gov/cuu/Units/units.html > The smallest capitalized prefix is M for mega: http://physics.nist.gov/cuu/Units/prefixes.html > Binary multiplicative prefixes are all capitalized. -- Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Dynamically Disabling ISV Products
You may be right. Someone elses post(after mine) said a search with 1 space yielded the same results as 2 spaces, so it appears the blanks may be getting eliminated. Not only mainframes have documentation - there is a help for each input field and it is just a click away: "Help for String Enter the string you are searching for. To search for messages about John Kennedy, simply type John Kennedy in the search box. This will show all the messages that contain the words "John" and "Kennedy" close to each other. " Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Job Stream via FTP
> Paul > > It's now a bit of a puzzle how you got your "copy and paste" from the z/OS client. Solaris (or some other Unix) FTP server, saying that "filetype=jes" command is not implemented or not allowed there? Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Job Stream via FTP
> Apparently z/OS treats "SITE" as "quote SITE". According to RFC959 (4.1.3) "SITE" is an FTP service command. So only clients not having "SITE" command implemented have to use 'quote "SITE"' form. You have to use the "quote" command to forward the whole command string to server. Compare the replies to help site and quote help site for a client that supports "SITE" (z/OS or Solaris, according to your example). Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Job Stream via FTP
Gil, And, ironically, from a z/OS client: EZA1701I >>> SITE filetype=jes 502 SITE command not implemented. I have never seen this (it has worked on MVS, OS/390,and works on z/OS 1.8): site filetype=jes >>> SITE filetype=jes 200 SITE command was accepted Isn't the message you get an answer from the FTP server you logged on (maybe an Unix and not z/OS FTP server)? Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: ISPF list data set attributes?
I just tried to print a data set list from "ISRUDLP Data Set List Utility". I got: .--- --. | The listing of data sets was placed in the ISPF list data set. Lines have | | been wrapped because the logical record length is less than 127 for RECFM F | | or 131 for RECFM V. | '--- --' o There's no code associated with this message. Is it supposed to be "self-explanatory"? I know the modal reaction of this group to self-explanatory messages. Or do I need to choose an option to display the code? To see the message ids, go to Settings --> Identifier --> Message Identifier .-- ISPF Settings --. |Message Identifier | | | | Enter "/" to select option| | / Display message identifier | | | | Default setting for message identifier| | 2 1. Off | |2. On | | | | | .---. and you'll get .--- -. | ISRU342A The listing of data sets was placed in the ISPF list data set. | | Lines have been wrapped because the logical record length is less than 127 | | for RECFM F or 131 for RECFM V. | .--- -. According to "z/OS V1R8.0 ISPF Messages and Codes" http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/ISPZMC50/SPTWQ820 >: 2.2414 ISRU342A ISRU342A The listing of data sets was placed in the ISPF list data set. Lines have been wrapped because the logical record length is less than 127 for RECFM F or 131 for RECFM V. Explanation: This is an informational message. The option 3.4 SAVE command was successfully processed and the requested information has been written to the ISPF list data set but has been wrapped. User Response: Use option 0 - List data set characteristics - to increase the logical record length. System Programmer Response: Increase the value of the LIST_LOGICAL_RECORD_LENGTH configuration table keyword. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: DISP=(MOD,DELETE)
Paul, The JCL fragment: //STEP EXEC PGM=IEFBR14 //* //SYSUT1DD DSN=user.FOO.BAR.SMPLOG,DISP=(MOD,KEEP), // VOL=SER=TSO026, //UNIT=SYSALLDA where 'user.FOO.BAR.SMPLOG' does not pre-exist on the volume produces allocation messages: IEF236I ALLOC. FOR MODKEEP STEP IEF237I 3ECA ALLOCATED TO SYSUT1 IEF142I MODKEEP STEP - STEP WAS EXECUTED - COND CODE IEF285I user.FOO.BAR.SMPLOG KEPT IEF285I VOL SER NOS= TSO026. IEF373I STEP/STEP/START 2007142.1554 ... but VTOC listing does not show the data set. As I read the " 12.19.11 Summary of Disposition Processing" in the JCL RM, I believe it should be kept. according to "12.19 DISP Parameter" in the JCL RM (z/OS 1.8): If the system obtains unit and volume information for an OLD, MOD, or SHR status, the data set is treated as if it exists, whether or not it is physically on the device. and the "4.6.2.1.8 Disposition Processing of Data Sets that Do Not Exist" in the JCL UG (z/OS 1.8) says: When VOLUME and UNIT Are Coded: When you code VOLUME and UNIT on the DD statement, a JCL error will occur if the problem program attempts to open the data set. Otherwise, the data set disposition depends on the DISP normal termination disposition: When the normal termination disposition is KEEP, the job log will show that the data set was kept. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Sysplex timer
Victor, Yesterday we tried to synchronize our Sysplex Timers with an ETS, but we couldn't get it. My problem is I didn't find any manuals with help. It shows an error window, but I don't know where to look for information about it. Could you tell me where can I look for? have you seen the IBM Techdocs Flash 10561: "IBM 9037 Sysplex Timer - External Time Source Problems" http://www-03.ibm.com/support/techdocs/atsmastr.nsf/WebIndex/FLASH10561 > revised on 19.03.2007: with the following -- Abstract: Problems getting an ETS External Time Source (not modem) to go operational may be due to the format of the time data being transmitted to the 9037 by the time code receiver/generator. The 9037 accepts three specific formats from External Time Sources (time code receiver/generator). These formats are described in the Maintenance Information for the 9037 Model 2 Sysplex Timer, SY27-2641-02. When an ETS problem is discovered, ensure that the ETS device being used adheres to the correct structure that you have configured on your 9037. -- and more details later in the text? -- Zaromil P.S. I sent this mail through ibm-main web interface and got the answer: "Your posting to the IBM-MAIN list has been rejected because it only contains material in a format disallowed by the list configuration. The list is configured to remove unwanted material and process the remaining text; however, in this case, there would be nothing left after removing the disallowed material. You may want to resend your message in plain text, which is always allowed." -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Who uses IEE899I for Automation?
>I'm pretty sure the AO folks use it if there is a buffer shortage. >They parse the output to see what console has the shortage and change it to >roll mode or perhaps even VARY it OFFLINE. >If any console gets more that 200 messages, automation does a K Q,L=consid >to remove them. Won't the Message Flooding SPE (OA17514) take care of this problem? I don't believe it can solve problems in the situations described in the "MVS Planning: Operations": Recovery Actions for a WTO Buffer Shortage 1. Determine why the buffers are full and correct the problem. Possible reasons are: * A console is not ready because: - An I/O error occurred. - One or more consoles have a slow roll time. - The console is in roll deletable (RD) mode but the screen is filled with action messages. * The buffer limit is too low to handle the message traffic in the system. Either the MLIM value in the CONSOLxx parmlib member is too low, or the system default is not sufficient. I've also seen consoles in the held mode causing WTO buffer shortage. -- Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Where is the NED set
But the configuration data (NED et al.) is stored in a control block that has not been publicly documented. I believe that it, too, can be displayed by an operator command. If an in-storage NED is hosed, a vary online or reIPL may be necessary. Also a PTF may need to be applied. As I understand it, last four digits in "SCP DEVICE NED" identify CU logical address and device UA. Are the good and bad device in the same logical CU? What does the DEVSERV QDASD command returns for both devices? Are the SCU-SERIAL and DEVICE-SERIAL the same in both cases or you see the anomaly there too? Have you tried the VALIDATE parameter of the DEVSERV command (I don't know if it can help, we used it when moving to new devices without an IPL): DS QDASD,,validate -- Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: CDIMG format?
Thanks. They are __supposed__ to be ISO9660 filesystems that can be burned. I say __supposed__ to be because they are somehow corrupt. That's why my software (k3b to be exact) didn't like them. Have you tried to mount them as loop devices, something like mount -t iso9660 -o loop /path/to/image.iso /mnt/isoimage? -- Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Print Dataset List with Last Date Referred
in 1.7 with pdf3.4, the dsn list does have the reference date. the problem may be the spflist lrecl going to 121 and truncating the last few fields. Look in "ISPF Services Guide" for LMDLIST ... STATS(YES) ... ===> variable ZDLRDATE or "TSO/E REXX Reference" LISTDSI ... ===> variable SYSREFDATE -- Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: TRSMAIN question
<- snip -> > Once you're in gzip you can decompress it on a PC I'm sure. > > Oh - you might have to pipe it through iconv to get it into ASCII / > Unicode before zipping it. Thanks for the idea. It may be easier than trying to reverse engineer TRSMAIN. Assuming that I had the talent to do so and it is not forbidden by IBM. <- snip -> What about infozip ? Knowing nothing about RACF IRRADU00 reformatted records, a question: Is it possible to transfer less data using DFSORT/ICETOOL to extract what you need or even to found the problematic records (or the number of them)? Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Interrupting DSLIST [resent with the correct Reply To]
Of course most TSO users now have keyboards with SysRq, Scroll Lock, and Pause/Break keys wasting space on their PCs. Do they get used at all? We use Pause/Break key as Clear key. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: moving files with file extensions from windows to mainframe
<- snip -> If I have a windows directory containing hundreds of files that all have the same file extension eg filname.DEF, how can I drop the file extension and get them into a pds on the mainframe. <- snip -> What about: 1. step DOS command: rename *.DEF *. 2. step ftp with mput * -- Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: SMS question - sharing DASD but not SMS environment
We want to share a few DASD among our four Lpars and have SMS doing allocation. Each Lpar runs z/OS as a monoplex seperate from the others. Two are 1.4 and two are 1.7 right now. We do not have any kind of sysplex. Nor do we have an SMSplex, whatever that is. We would update SMS rules in every Lpar to indicate these volsers in an SMS storage group and set up storageclass, dataclass & mgmtclass. Each high level qualifier would be setup in every Lpar's master catalog using an existing ucat on a non-SMS dasd that is online to all lpars. Would SMS for lpar-P get confused if SMS for lpar-D was allocating a file on those dasd at the same time she was? Does SMS from one lpar need to know what SMS from the other lpar is doing? Our example: - sysplex A (production) with 5 systems z/OS 1.7, GRS star - sysplex B (test) with 1 system z/OS 1.7, GRS star The volumes used to transfer data from sysplex A to sysplex B are SMS managed, defined in both SMSes. Allocation of datasets containing data to be transfered is possible only in the sysplex A (source). In the sysplex B (test) shared volumes are defined with DISABLENEW attribute, so it is not possible to allocate them. The usercatalog is on one of the shared volumes (SMS managed). It works like this for some years now. If you want to be able to transfer data in different directions, you could control the allocation the way we do, or using security definitions, or ASC routines. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Is it possible to prevent a structure into a particular CF (C oupling Facility)?
Jan, Site-2 the recovery site with the PPRC-ed disks & with the controlling K-sys (The advantage of this is that the Controlling system will continue to be available even if a disaster takes down the whole production site.). I would say you also need alternate couple datasets to be in the site-2 if you want your K-sys to survive a failure in the site-1. Q2. We are not that sure at all that we are able to keep our (production) structures at the site-1 CFs with PREFLIST & ENFORCEORDER(YES). For example, what happens with a duplexed structure (at production Site-1) when one of the site-1 CFs goes down? I have no experience with this, we had got rid of the third CF long before structure duplexing was available. But maybe an assymetrical connectivity could solve your problems: If you don't go to use the site-2 CF in the production with systems running in the site-1, you could connect it to site-2 only (k-system and backup systems). In that case only the structures used by k-system running alone could be allocated / rebuilded into site-2 CF. (Well, that is a theory. I've not seen it in praxis.) Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Is it possible to prevent a structure into a particular CF (C oupling Facility)?
Re. Paolo Cacciari saying: >Obviously, I assume that your Data Centers are not involved in a Dispersed >Parallel Sysplex. It *IS* about a GDPS config, Paolo, with the K-sys controlling system in the C/R site. What kind of D/R you want to have? What do you plan for the case that you have disk problems only? You move all your systems to the site-2 after recovering the secondary volumes or you want to be able to use the secondary volumes running the systems in the site-1? The way I understand Paolo's question is: you don't plan to disperse the application systems over both sites, some of htem running in the site-1, the others in the site-2? Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Is it possible to prevent a structure into a particular CF (C oupl ing Facility)?
You answered yourself. PREFLIST. In case of further investigation: you can have multiple CFRM policies with different PREFLISTs. Assuming it is a GRS star: if he loses site1 with a CFRM policy without a site2 CF, he loses the controlling system, too. That is surely a disadvantage. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Is it possible to prevent a structure into a particular CF (C oupling Facility)?
Is there a way to *prevent* that structures will be put into this 3-rd CF at DR site-2 in *NORMAL OPERATION mode*, i.e. when the whole production is running in site-1? We think it can't be accomplished with just specifying the PREFLIST, listing the preference order of CFs that one wants structures to be allocated to. * Any experience and/or good avices with this? To exploit the PREFLIST, try ENFORCEORDER(YES). With REALLOCATE you can always correct the allocations (although I don't see how the structures come in the site-2 CF with both site-1 CFs active, with maybe the exception of DUPLEXed structures). Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: ftp time problem
The STDENV dataset (which has to be a RECFM=VB and possibly LRECL=256). I support a number of systems which are country specific but share the same configuration files so I decided to use an MVS symbol for TZ. The symbols correspond to the military/aviation timezone codes for Central European Time (A) and Greenwich Mean Time (Z). The FTPA member contains the following: _BPX_JOBNAME=FTPD TZ=CET-1CEST,M3.5.0/01:00,M10.5.0/02:00 I would expect the (A) to contain the following in the second line: TZ=CET-1CEST,M3.5.0/02:00,M10.5.0/03:00 -- Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Data set ENQueues and DEQueues in Jobs
<- snip -> Are you sure that a single volume can't be known by two different names? Consider: surely it's possible to rename a volume (the verb "clip" comes to mind). Can one system dismount and rename a volume while it remains mounted on another system, then remount it by the new name while the other system still knows it by the old name? <- snip -> Oh yes. Just replace "dismount" and "remount" with "vary offline" and "vary online" respectively. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Looking For A Search Engine
Some time ago I made some changes in a rexx I found on the CBT tape (I'm sorry not to know the author's name). It searches for a given string in all datasets that match a given dataset name pattern and/or allocated on volumes that match a given volser pattern. I use this JCL: //T1AI$RCG JOB (CA-IT,50150,36110),'BSM: Tisler',CLASS=Y, // MSGCLASS=Q,MSGLEVEL=(1,1),NOTIFY=&SYSUID //* - //STEP010 EXEC PGM=IKJEFT1A, // PARM='ISPF CMD(%FINDSRCL)', // TIME=1440 //STEPLIB DD DISP=SHR,DSN=SYS1.REXX.SEAGLMD //ISPPROF DD LRECL=80,BLKSIZE=0,RECFM=FB,DSORG=PO, // DISP=(,DELETE), // SPACE=(TRK,(1,1,4)) //ISPMLIB DD DISP=SHR,DSN=ISP.SISPMENU // DD DISP=SHR,DSN=ISF.SISFMLIB //ISPPLIB DD DISP=SHR,DSN=ISP.SISPPENU // DD DISP=SHR,DSN=ISF.SISFPLIB //ISPSLIB DD DISP=SHR,DSN=ISP.SISPSENU // DD DISP=SHR,DSN=ISP.SISPSLIB //ISPTLIB DD DISP=SHR,DSN=ISP.SISPTENU // DD DISP=SHR,DSN=ISF.SISFTLIB //SYSEXEC DD DISP=SHR,DSN=myuserid.REXX.CEXEC //SYSTSIN DD * sx03 2 xr%8%% sys1.** /* //SYSTSPRT DD SYSOUT=* // 3 sys1.p*r*lib*.** //* sys4.p*r*lib //* ISP.V1R2M3 --- to execute the rexx procedure (there could be problems with "!" instead of "|"): /* - Rexx - */ /* Function : Search for a string in PO, PO-E, PS data sets */ /* using ISRSUPC (SearchFor ISPF utility) */ /* with following parameters: */ /* SRCHCMP - search (??) */ /* ANYC - any case*/ /* NOSUMS - no summary listing section */ /* LMTO - list group member totals only */ /* */ Version = "0.02.01 / 19.01.2005" say say 'Search for a string, version' Version say /* */ /* This rexx looks for a string depending upon given volume or */ /* data set pattern. It uses ISPF LM utilities. */ /* */ /* LMDLIST is used to fetch all the dataset getting qualified for */ /* given input. */ /* LMMLIST is used for getting all the members of a partitioned */ /* dataset. */ /* */ /* It takes a string and 'pattern of dataset or volume' as input*/ /* and lists all datasets matching the pattern or given volume. If */ /* organisation of a dataset is 'PO' it list all the members and*/ /* find given string in them . It searches all sequential and */ /* partitioned data set matching a given pattern or volume. */ /* */ /* Dataset pattern can be given like userid.*.* or anyname.p*.q**/ /* */ /* The following data sets will be searched:*/ /* */ /* PS, PDS, PDSE dataset organization and */ /* F, FB, FBA, V, VB, VBA record format */ /* */ /* */ /* */ /* Initialization of variables */ /* */ String = '' VolName = '' DSName = '' SAY ' --' SAY ' This rexx searches for a given string in all PS, PO, AND PO-E data' SAY ' sets matching a given data set name pattern and/or residing on the' SAY ' volumes matching a given volser pattern.' SAY ' --' SAY ' ' SAY 'Enter the string to be searched for (without quotes)' PULL String SAY ' ' SAY 'Please enter one of the following:' SAY ' 1 - search all data set(s) on volume(s)' SAY ' 2 - search data set(s) on volume(s)' SAY ' 3 - search data set(s)' SAY ' 4 - quit' PARSE PULL Option SAY ' ' StartTime = sysvar('SYSCPU') call time('R') /* */ /* Get search parameters */ /* */ SELECT WHEN Option = 1 THEN DO SAY 'Search volume(s)' PARSE PULL VolName SAY 'Searching "'VolName'"' IF
Re: MPF message suppression
A few years ago we conseidered implementing a message-flooding prevention table. We have never used it, but maybe it could be interesting for you. "z/OS V1R7.0 Comm Svr: SNA Resource Definition Reference" http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/F1A1B650/5.7?DT=2 0050720144034 > "5.7 Message-flooding prevention table The message-flooding prevention facility identifies and suppresses duplicate messages that are issued in rapid succession. This reduces the possibility of duplicate messages flooding the operator console and concealing critical information. VTAM bases its suppression on the time interval since the message was last issued and on the similarity of variable text in the original and subsequent message. For each candidate message, the message-flooding prevention table contains the criteria that must be met before VTAM suppresses duplicate messages and whether suppressed messages are sent to the hardcopy log. The suppression criteria include the amount of time between the original and subsequent messages, and an indication of which variable text fields are to be compared. If the message is reissued within the specified time interval and the specified variable text fields contain the same information, VTAM suppresses the message" etc. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: FTP Translate parm
<- snip -> //STEP0010 EXEC PGM=FTP, // PARM='( EXIT TRACE TRANS PECTAPP.BLVTOBSP.TCPXLBIN' EZA1451E Cannot load translate table specified by TRANSLATE parameter PECTAPP.BLVTOBSP.TCPXLBIN CX0278 main: Exit with error: 14 >>> File DOES exist, it is in the correct format. I works with other FTP(LOCSITE SBD) commands. <- snip -> I haven't used it for years, but I believe you must not specify the whole name. According to manual SC31-8780-05 http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/F1A1B950/3.2?SHEL F=F1A1BK61&DT=20050708142126> - TRANslate data_set_name Specifies the data set name of a nonstandard translation table. If you specify this parameter, FTP uses the translation table in the user_id.data_set_name.TCPXLBIN data set, rather than the standard translation table provided with TCP/IP (hlq.STANDARD.TCPXLBIN). The hlq.STANDARD.TCPXLBIN data set is never used if you specify the TRANSLATE parameter. If user_id.data_set_name.TCPXLBIN does not exist, FTP uses hlq.data_set_name.TCPXLBIN. If neither data set exists, or if they were incorrectly created, FTP ends with an error message. Since the TRANslate parameter also dictates the search order for DBCS translation tables, you might wish to use a customized DBCS translation table but not require a modified SBCS translation table. If this is the case, copy hlq.STANDARD.TCPXLBIN into the nonstandard TCPXLBIN translation table data set to ensure FTP will start. - I believe it is only the data_set_name part that you need to specify. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Help : FTP A PDS
Shmuel, > That's nice, but "192.168.243.49" is not a URL. Sorry and thank you for correcting me. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Help : FTP A PDS
<- snip -> >C:\>FTP 12.34.56.78 [or whatever your mainframe's IP address or URL is] Host name, not URL. FTP will not be able to parse a URL. <- snip -> I am not sure if I have tried FTP with URL instead of host name, but linux FTP clients (ftp, lftp, ncftp), default Windows 2000 FTP, and z/OS 1.7 FTP allow the use of URLs ( just checked again the last two). In Windows "H:\>ftp 192.168.243.49" and in z/OS "TSO FTP 192.168.243.49" both establish a connection. Help / man page states "host" and not "host name". Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Help : FTP A PDS
<- snip -> C:\>FTP 12.34.56.78 [or whatever your mainframe's IP address or URL is] userid password lcd "whatever Windows folder you want to put the members in" cd my.PDS.name mget * [hit enter for every prompt - alternatively use the prompt command first to turn off prompting] <- snip -> You can also disable the prompting during mget starting the FTP transfer if you enter the following command: "ftp -i 12.34.56.78" To see more FTP-client options and their descriptions enter "ftp --help" Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Adding LPARs without POR
> No, until they need the memory, you can give it to other LPARs. We haven't reconfigured the storage element offline in an active lpar for a long time, but years ago we had a need once and it did not work, because that was a CICS allocated storage and stopping it was the same as IPL. > Alternatively you can add memory concurrently before activating the new LPARs. I don't believe there are many fans of it here. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Adding LPARs without POR
I don't understand two things: 1. Who needs hundreds of linux servers on a single IFL, even if it is possible? 2. Inactive partitions allocate no physical memory, but if I want to activate them I must have memory available to accomodate their needs. That means I must have the capacity although the lpars are still inactive. So, indirectly they do eat memory. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: dynamic PAV
<- snip -> >Where is the switch to turn I/O Priority Management on ? IIRC on the same screen as where you define the SDC's for CPU, SRB, MSO & IOC. <- snip -> In WLM, Option 8. Service Coefficients/Options Panel "Service Coefficient/Service Definition Options": . . . Enter or change the service definition options: I/O priority management . . . . . . . . YES (Yes or No) Dynamic alias management . . . . . . . . YES (Yes or No) -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: dynamic PAV
<- snip -> In the initial setup we defined in HCD base (3390B) and alias (3390A) volumes with wlmpav=yes (for the aliases). <- snip -> Do you have WLMPAV=YES for the base volumes, too? Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Barbaras (mini-)rant
<- snip -> I'm with Jim on this. I was a contractor in the mid to late '90's and came across the early TCPIP stack, written in PASCAL and ported from VM. As I recall it performed OK, and had some quite advanced features like VIPA which was the subject of another recent thread. <- snip -> I installed TCPIP 2.2 or 2.2.1 in November 1992. It was written in PASCAL(/VS?), there were two subsystems: TNF and VMCF (virtual machine communication? facility) because it was a port from VM. Dataset names were hardcoded, so I had problems with our dataset naming standards, messages had no message ids. Somehow I do not remember VIPA at that time, I think it came in 3.3 or 3.4. I believe to remember it was possible to change the MAC address on the 3172. In the release 3.1 or 3.2 there were two FTP servers, one written in C and one in PASCAL(/VS?). Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Enhanced Catalog Sharing
> What is the gain or the advantage of using ECS. >From an old redbook (Enhanced Catalog Sharing and Management, SG24-5594-00, July 1999): --- When CAS opens an eligible catalog and the system is in catalog sharing mode, as described in "Activating ECS" on page 18, the integrity VVR is copied into the Coupling Facility structure. Once this occurs, the systems that share the catalog do not have to read the VVR to detect change, as they will receive notification from the Coupling Facility. Also, the requirement to ENQ on the resource is removed because the Coupling Facility architecture incorporates the necessary locking. --- and --- This is an elegant solution to the performance issues. The elapsed time is reduced because GRS is no longer used, and the number of I/Os is reduced because the VVR does not have to be read. Also, the new function fits smoothly alongside the existing reliable and robust code, which will still be used for non-ECS catalogs. Non-ECS catalogs are now described as being in VVDS mode, which indicates that the VVR is being read from the VVDS. In the rare circumstance of a connectivity or Coupling Facility structure failure, the ECS catalogs will revert to VVDS mode, allowing the integrity of the catalogs to be maintained across such a failure. --- Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: SSL Certificate Signing
<- snip -> So I had them redo the request and I made sure to FTP it back to z/OS in ascii mode. I still get the same error. <- snip -> Why ascii? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Info-ZIP and blksize
Last few days I used Info-ZIP quite a lot (z/OS 1.6 and 1.7). The problem I have is that the blocksize of zipped files equals 80, so the zipped files need a lot of space. E.g. adding: 'N500223/D#060301/CHAN#IOQ/RMF#BC01.TXT' (in=5893937) (out=523379) (deflated 91%) total bytes=5893937, compressed=523379 -> 91% savings and the dataset allocations: Command - Enter "/" to select action Tracks %Used -- N500223.D#060301.CHAN#IOQ.RMF#BC01.TXT 120 90 N500223.D#060301.CHAN#IOQ.ZIP#BC01 90 93 Is there any possibility to influence the block size (without modifying the code)? Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: CF usage at DR location
<- snip -> I have a full test this weekend, so hopefully will have some of this ironed out. I am going with a newly formatted CFRM dataset and retaining all the other COUPLE datasets. If I use a new CFRM dataset, where is the CF policy information stored? I was thinking it was stored in the XCF COUPLE dataset as how would z/OS know how to format the CFRM dataset when coming up? I hope to come up in the new environment and the structures are re-created in the new CF, but all other COUPLE datasets are intact, e.g. DASD Logger files will retain their state. <- snip -> Why wait? If you want to use a newly created CFRM couple dataset, allocate it on a mirrored volume, write the policy you want into it, and that should be it. Just don't use it before the test. But, if you are already testing, you could first try to IPL with the old CFRM couple dataset, having a policy with all CFs in it. If it does not work, take the new CFRM dataset. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Accessing The Archives
Darren wrote: > That problem has been rectified. Thank you, Darren. Sam wrote: > You can also search them through the Google Groups Archive http://groups.google.com/groups?oi=djq&as_ugroup=bit.listserv.ibm-main Thank you, Sam. Using the mailing list exclusively, I just forget the possibility to check the news group. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Accessing The Archives
> Read the trailer inserted by the list serve! I have problems when searching the archives today. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: CF usage at DR location
<- snip -> We were early users of XRC circa 1998. In our first few tests, we had lingering problems because, with mirrored CFRM data set, the newly IPLed systems could never get over the loss of the two CFs left back in the smoking hole. They were convinced that some magic would bring the lost CFs back to life. I put the problem to Curt Jews at SHARE--in SCIDS of course--and he recommended *not* to mirror the CFRM data set. Voila. The crux is to IPL the first system with an empty, freshly formatted CFRM data set. There is now no trace of the lost CFs ever having been in use. <- snip -> I remember a lot of problems in case of a failing CF or migration to new CF(s) in the 90's. But, the functionality is much better now. Recently we moved a sysplex from 2084 to 2094 (four systems and two ICFs on the same CEC). We did not use a freshly formatted CFRM couple dataset. Instead, before the upgrade we activated a new CFRM policy, the old policy containing CFs CFOLD1 and CFOLD2 and the new policy containing CFs CFOLD1 and CFNEW2. After the successfull upgrade we activated the actual CFRM policy, containing CFs CFNEW1 and CFNEW2. That worked without problems and we had a failback possibility all the time. And as you wrote, there is now no trace of the lost CFs ever having been in use. Neither we used freshy formatted (new) XCF couple datasets nor freshly formatted (new) CFRM couple datasets. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: HDS backup process
Ron, > Are you using the BCM (Business Continuity Manager) Host software, or the archaic PPRC commands? As a GDPS PPRC site I believe we are stuck to PPRC commands. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: HDS backup process
This thread has begun with a PPRCopy question. I am not sure if this two parts of Bruce' mails (not to cite John and Ron) talk about PPRC (Truecopy) or Flashcopy (Shadowimage). I wasn't able to see any of these features in our z/OS 1.7 PPRC environment using HDS 9980. <- snip -> You don't need to lose the relationship to do the backup. You probably want to backup a point-in-time copy of the volumes, so you can just SUSPEND the PPRC relationship. This freezes the copy and makes it read/write so you can relabel (CLIP) it. After the backup, you can RESYNC the PPRC copy, which just copies tracks updated since the SUSPEND. <- snip -> Supposing we do it in a z/OS 1.7 system I don't understand how is it possible to clip the secondary volume. I get the info about the device being a PPRC secondary device. Even if I take the primary volume offline, I get the same kind of message if I try to take the secondary volume online. Bruce wrote: <- snip -> I believe that RESYNC will copy any tracks which have been modified on the source OR target. <- snip -> The other thing that I don't understand: To be able to write on the secondary volume, I suppose I have to use CDELPAIR or at least CRECOVER. If I use CDELPAIR, I can not RESYNC anymore. If I use CRECOVER the secondary volume switches in the SIMPLEX status, so how comes its bitmap is updated? Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: newbie qdasd question
<- snip -> when I issue the console command DS QDASD,5B00 I receive the following output: RESPONSE=MJR0 IEE459I 14.41.05 DEVSERV QDASD 452 UNIT VOLSER SCUTYPE DEVTYPE CYL SSID SCU-SERIAL DEV-SERIAL EF-CHK 5B00 $X5B00 2107921 2107900 3339 A014 0175-BNCZ1 0175-BNCZ1 RDFEAT? 1 DEVICE(S) MET THE SELECTION CRITERIA 1 DEVICE(S) FAILED EXTENDED FUNCTION CHECKING Can you please explain me what 'RDEFEAT?' does mean? <- snip -> I don't know, but I suppose it hase something to do with features of the box. In our case in the last column I see EF-CHK **OK** If you try DS QDASD,5B00,rdfeats do you get FEATURE CODES AT V followed by an 8 rows x 4 columns with four doublewords per column, all that after the output you already sent? Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: CF usage at DR location
<- snip -> >these datasets will be PPRC "mirrored" One of the two sets of datasets IBM recommends to not mirror is the CDS set. The other is PAGE. I have another set that I never mirror'd: SCRATCH. The overhead of synchronous mirroring kills the first two, for sure. <- snip -> The GDPS hyperswap is impossible without mirroring the PAGE datasets. We don't mirror couple datasets (we are GDPS environment), but mirroring the LOGR couple datasets is a must if you want to have the information about the logger staging datasets. As I understand it, the reason not to mirror the couple datasets (XCF couple dataset in the first place) is the possibility of freeze on the primaries. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: CF usage at DR location
<- snip -> as I wrote, you actually CAN mantain your CFRM couple datasets, with your policy, provided that your policy comprises both primary and DR CF; what you need to have for DR purposes only, is an empty and initialized XCF COUPLE dataset (easy to do once and to forget) and a PARMLIB COUPLExx member, to be used in DR situations. ...snip ... I was hoping to use the existing COUPLE dataset and the existing policy. Since these datasets will be PPRC "mirrored" I was hoping to find a way to not maintain 2 separate policy definitions so that the day-to-day policy definitions would be automatically used at the DR location. ...snip ... <- snip -> You could try the following: If you do not mirror the primary XCF couple dataset you could dedicate a volume for it and use the same dataset name and volser in the DR site (the volume - and so the primary XCF dataset - is not mirrored). That way you can use even the same COUPLExx. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: HDS backup process
<- snip -> I was told that in order to do this the secondary volume labels' must be clipped before I can bring them online but by clipping them I also loose the primary-secondary volume relationships. Besides using another lpar or other 3rd party utilities to run the backups, how does one handle this situation? <- snip -> >From "Hitachi Freedom Storage™ Lightning 9900™ V Series Hitachi TrueCopy – S/390® User and Reference Guide" The RCU does not allow a TC390 R-VOL to be online and rejects all host-requested read and write I/O operations for a TC390 R-VOL. The TC390 R-VOLs must be offline during normal TC390 operations. Note: TrueCopy – S/390® provides a special R-VOL read option (see section 2.2.4) which allows read-only access to the R-VOL while the pair is suspended. If you need write access to a TC390 R-VOL, you must delete the pair. 2.2.4 R-VOL Read Option For additional flexibility, TrueCopy – S/390® offers a special R-VOL read option. The Hitachi Data Systems representative enables the R-VOL read option on the RCU (mode 20). The R-VOL read option allows you to read a TC390 R-VOL only while the pair is suspended, that is, without having to delete the pair. The RCU will allow you to change only the VOLSER of the suspended R-VOL, so that the R-VOL can be online to the same host as the M-VOL while the pair is suspended. All other write I/Os will be rejected by the RCU. The MCU copies the M-VOL VOLSER back onto the R-VOL when the pair is resumed. When the R-VOL read option is not enabled and/or the pair is not suspended, the RCU rejects all read and write I/Os to a TC390 R-VOL. Note: If you need write access to an R-VOL, you must delete the pair. Note: For 2105 controller emulation, the CSUSPEND command to the R-VOL of a suspended TC390 pair will be rejected when the TC390 R-VOL read option is used. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: GDPS and DASD Vendors
<- snip -> I thought that Hiperswap was a software only feature ( after all it only swaps UCB's ) . So i am curious .. why would hardware matter ? <- snip -> I would not say that Hyperswap (GDPS) only swaps UCB's. It must communicate the hardware to change the mirroring direction at the same time. I remember we had to wait for new microcode levels to use Hyperswap or at least some of its new features. E.g. one of the features is (was) changing the PPRC link direction. The other example was the migration from hw model x to hw model y (both models from the same vendor). Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Clist problem: character [ is converted to : at output
We've got a problem with clist. The character [ and ] (left and right square brackets) are stored ok in the dataset (x'AD' and x'BD') but when clist displays them at terminal they appear as : (colon). This is true even if the clist displays any dataset with [ or ]. Non-displayable characters in TSO(?). Output from a rexx exec: 00 01 02 03 37 2D 2E 2F 16 05 25 0B 0C 0D 0E 0F : : : : : : : : : : : : : : 10 11 12 13 3C 3D 32 26 18 19 3F 27 1C 1D 1E 1F : : : : : : : : : : : : : 40 4F 7F 7B 5B 6C 50 7D 4D 5D 5C 4E 6B 60 4B 61 ! " # $ % & ' ( ) * + , - . / F0 F1 F2 F3 F4 F5 F6 F7 F8 F9 7A 5E 4C 7E 6E 6F 0 1 2 3 4 5 6 7 8 9 : ; < = > ? B5 C1 C2 C3 C4 C5 C6 C7 C8 C9 D1 D2 D3 D4 D5 D6 : A B C D E F G H I J K L M N O D7 D8 D9 E2 E3 E4 E5 E6 E7 E8 E9 63 EC FC 5F 6D P Q R S T U V W X Y Z : : : ^ _ 79 81 82 83 84 85 86 87 88 89 91 92 93 94 95 96 : a b c d e f g h i j k l m n o 97 98 99 A2 A3 A4 A5 A6 A7 A8 A9 43 BB DC 59 07 p q r s t u v w x y z : : : : : 68 D0 51 42 C0 44 47 48 52 53 54 57 56 58 4A 67 : ü : : ä : : : : : : : : : Ä : 71 9C 9E CB 6A CD DB DD DF E0 5A 70 B1 80 BF FF : : : : ö : : : : Ö Ü : : : : : 45 55 CE DE 49 69 9A 9B AB AF BA B8 B7 AA 8A 8B : : : : : : : : : : : : : : : : 2B 2C 09 21 28 65 62 64 B4 38 31 34 33 B0 B2 24 -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: How to quickly shut down system
<-- snip --> Let's assume the following scenario: "Disastrous" power outage occured, no alternate power source is available, only UPS battery. Time for battery discharge is approx. 20-30 minutes. <-- snip --> Why do you believe you have 20-30 minutes to do something? I would look how long can the weakest component stay available: switch or any other interface on the path to devices and then to the mirror devices. How can you be sure that the connectivity between cpu and devices or primary and secondary is still there during 20-30 minutes? E.g., in the case that there is a connection through a switch and it loses power, how long is its time for battery discharge? Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: IEASYM override in LOADXX
<-- snip --> NUCLST Z0 PARMLIB SYS1.LPLEX.PARMLIB PARMLIB SYS1.PARMLIB PARMLIB SYS1.IBM.PARMLIB SYSCAT CATL00133CCATALOG.LPLEX.MASTER *-* * SL20 SPECIFIC VALUES* *-* LPARNAME SL20 IEASYM (01,Z0,L) IODF E0 SYS0 LPLEXFPB B0 Y SYSPARM (Z0,20) *-* * General valid (if you have more CECs, repeat this section: * *-* HWNAME IEASYM (00,Z0,L) SYSPARM (Z0) <-- snip --> I just tested it. It does not work the way I thought it should. Looking again at the examples, I would say that you can not override statements defined in an HWNAME section with statements defined in the LPARNAME section. It looks like you have to define a statement always on the same level, like if you have 10 lpars on the CPC, you need ten LPARNAME sections to define IEASYM statement different for a single system (I do not believe I am going to test this). Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: IEASYM override in LOADXX
<-- snip --> Example: IEASYM (00,Z0,L) NUCLST Z0 PARMLIB SYS1.LPLEX.PARMLIB PARMLIB SYS1.PARMLIB PARMLIB SYS1.IBM.PARMLIB SYSCAT CATL00133CCATALOG.LPLEX.MASTER SYSPARM (Z0) *-* * SL20 SPECIFIC VALUES* *-* LPARNAME SL20 IEASYM (01,Z0,L) IODF E0 SYS0 LPLEXFPB B0 Y SYSPARM (Z0,20) <-- snip --> Your IODF statement is only valid for SL20? Looking at the description of LOADxx in the "MVS Initialization and Tuning Reference", I don't see a possibility to override a statement. There is no such combination in the examples. But, you could try the other way around: -- NUCLST Z0 PARMLIB SYS1.LPLEX.PARMLIB PARMLIB SYS1.PARMLIB PARMLIB SYS1.IBM.PARMLIB SYSCAT CATL00133CCATALOG.LPLEX.MASTER *-* * SL20 SPECIFIC VALUES* *-* LPARNAME SL20 IEASYM (01,Z0,L) IODF E0 SYS0 LPLEXFPB B0 Y SYSPARM (Z0,20) *-* * General valid (if you have more CECs, repeat this section: * *-* HWNAME IEASYM (00,Z0,L) SYSPARM (Z0) -- Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Using symbolic in JCL
<-- snip --> Different local offsets in the CLOCKxx member. Or somebody does a weird T CLOCK= command. The TOD clocks must be in sync. But the local times may differ. <-- snip --> Well, I know that. We don't use it, but I think I have no problems understanding it. <-- snip --> For the truly strange (like me): Imagine a sysplex where there is a different z/OS image for every possible "time zone" in which the company does business. The users in that time zone are directed to their z/OS image. This "solves" the local time issue! <-- snip --> Shouldn't "local time" be just a representation (like in FTP or mail clients - you sent your mail at 10:59:58 -0600, it arrived at 18:00:49 +0100); now let's look at resources shared in a sysplex: - what time should be written in the operlog? - what time should be written in the sysplex coupling datasets, logger coupling datasets? - what time use HSM or equivalent product for migration/restore/backup? - what time is used in DB2 logs, CICS logs, assuming DB2 and/or CICS instances run in images in different timezones? Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Allocated volumes after the power loss in the storage subsyst em
John, <-- snip --> I've lost track here! You're running GDPS (I assume with PPRC and Hyperswap), and you lost power to your secondary site. I assume that those devices were also the secondary Duplex devices. <-- snip --> Yes, GDPS with PPRC and Hyperswap. But, we have simplex (i.e. not mirrored) devices too in both sites. Applications that had problems use these devices. Secondary devices are not allocated by applications, they are offline and allocated by Hyperswap address space (status F-SYS). <-- snip --> Is it possible that you were experiencing a Long Busy situation in your production site? (admittedly - very long!) <-- snip --> No, we have not run in this situation yet. Freeze was done, the systems in the primary site experienced no problems. All issues had to do only with allocated simplex volumes not being available at the time. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Using symbolic in JCL
Walt, <-- snip --> Unfortunately, since different systems in a sysplex (or sharing the spool) could have very different times and dates, they could get very unexpected results if we let those symbols be used in batch job JCL. Suppose the job converted on a system where the date is today, and ran on a system where the date is several days, or weeks, ago. (The normal case is to have them very close, of course. And the more likely unusual case is to have them simply a few hours off. But nothing requires that the dates/times be anywhere close across systems, that I know of.) <-- snip --> According to "Setting Up a Sysplex": "Similarly, for MVS to monitor and sequence events in a sysplex as you intend, time must be synchronized across all systems." and "MVS compares time stamps from the TOD clock with time stamps recorded in the sysplex couple data set. If time stamps in the couple data set are later than those specified by the incoming system, MVS determines this is an error and does not allow the system to join the sysplex." How is then possible to have different times and dates in a sysplex? I always assumed there is a single clock and a single time in a sysplex, and all sysplex-wide applications use this concept: all logs, data sharing, couple data sets... Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: REXX Execs to help in String SRCH
A few times I had to check the system datasets (or sysres volumes) for a specific string occurence. I found it very expensive (CPU and elapsed time) to search a PDS dataset using rexx (allocating every single member as a sequential dataset). To reduce the number of allocations I called program ISRSUPC from rexx with ADDRESS ISPEXEC "SELECT PGM(ISRSUPC)" , "PARM(SRCHCMP,ANYC,NOSUMS,LMTO)" Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Allocated volumes after the power loss in the storage subsyst em
Kees, > are you testing disaster situations We are a GDPS site, 3 production parallel sysplexes, 2 - 3 sites, PPRC and Hyperswap, with unplanned Hyperswap in test parallel sysplexes (two of them). A smaller part of production volumes is not mirrored (customer's decision). Because of GPDS we do test outages quite regularly in our test sysplexes, simulating an outage mostly through generation of messages. The real storage subsystem outages are quite rare, because we need the help of the vendor's technician (field engineer?). In the production there are regular annual disaster tests for a whole site, but not with power outage (lot of non-mainframe servers could have problems establishing normal work). > or did you really have a complete power outage on a storage device and was the device configured sufficiently redundant and if so, how could this happen? There was a UPS problem, it registered(?) anomaly in the electricity and itself and cut the electricity on the site (that was my understanding). I came later in and had to do only with the problems on the mainframe side. The simplex devices (that is, not mirrored) were not critical for the production (the applications could go on with the half of the capacity). The only problem was that applications hanged as result of these allocations. Luckily, it happened in the night, and as the electricity came, we had the access to all the volumes again - no problems. Well, as Ed Finnell wrote: "when the subsystem gets IML'd (takes forever under normal circumstances)". Normally, it is a primary GDPS site for one production sysplex, but at the time that was our secondary site, so we lost only the systems running there. Actually we had seen this behaviour a few years ago, as we had done a lot of basic GDPS functionality testing: I/O intensive job writing to two different volumes on two subsystems, one is powered off, you get only the IOS002A message, the job is (if I remember correctly) all the time "in" in SDSF DA, no CPU-time change, then we activate the second box, job writes on as nothing happened. At that time we thought it was a nice feature . Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Allocated volumes after the power loss in the storage subsyst em
> You normally don't see this, but this still does not mean it cannot occur. Were the devices really "offline"? I do not believe they were "offline" in the system (another thing to check during the test!), but they were offline in the real world. > The UCB Online, Alloc, Busy etc. indicators are independent bits, that do not have a fixed relationship, however z/OS ensures that things happen in a certain order and so there are combinations of bits that you will probably never see. However, you are in an abnormal situation, so you might see rare combinations now. > OLTEP e.g. was one application that allocated an offline device and so the UCB status was really and correctly offline/allocated. GDPS Hyperswap does it, too. > Did you try a V xxx,ONLINE as indicated by IOS002A? This should simulate the I/O interrupt the system is waiting for. I don't believe anyone has done it. Another thing to check. Thank you for your input. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Allocated volumes after the power loss in the storage subsyst em
> So, missing interupt messages should pop up after 45 seconds. Don't you > see any IOS messages in the syslog? Yes, IOS002A NO PATH AVAILABLE ... at the moment the electricity was gone. By the time we tried to fix the problems in DB2, we found that the status of "alloc" for those devices somehow misleading. Anyway, we did not found a solution how to free the applications out of this. I've already learned that offline devices can have "alloc" status, but it still confuses me. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Allocated volumes after the power loss in the storage subsyst em
Hello Peter, > What does a "D IOS,MIH,DEV=/cuaa" for one of the hanging devices tell > you? 00:45. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Allocated volumes after the power loss in the storage subsyst em
Kees, thank you for the reply. > I wonder if the I/O does not time out and recovery procedures will be driven. In that case the timeouts must be extremly long (more than a couple of hours) . > You could try the V ,OFFLINE,FORCE to Box devices. This will terminate all outstanding I/O's with an I/O error and this should trigger some action in the applications. I've never thought of this possibility. I'll test it. > Hopefully, this storage box does not contain volumes to keep you z/OS running. No, no, no, only DB/2 and CICS . Regards, Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Allocated volumes after the power loss in the storage subsystem
Hello! Scenario: A power loss for an storage subsystem. Situation: The applications having allocated volumes on that subsystem hang. Neither system nor applications know that these volumes are no more accessible. There are no messages in syslog / operlog / application logs about any problems concerning the allocated volumes. Question: Is there any possibility to get out of these problems except IPL / applications restart / waiting for the electricity + IML of the subsystem? Have a nice day! Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: how to lose a sysplex in 30 seconds
Bill, <- snip -> The existence of signalling connectivity created a race condition, in which MVSA and MVSB were competing to detect and report the loss of access to the CDS at their respective sites. MVSB won the race, detecting and signalling the loss of the primary CDS before MVSA detected loss of the alternate. MVSA got MVSB's signal, initiated removal of the primary, and then detected the inaccessibility of the alternate. <- snip -> What would happen if MVSA got the race? Would MVSB try to access the primary during acknowledgment of the removal of the alternate CDS? If not, could it survive the loss of connectivity to primary if it (connectivity loss) would be short enough? Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: how to lose a sysplex in 30 seconds
<- snip -> We have 2 LPARs in a sysplex, running on 2 different machines in 2 different sites. What happened was we lost connectivity between our 2 sites for a few seconds. As a result, MVSB (running in site B) lost its connectivity to the primary SYSPELX couple data set residing on dasd in site A, and issued the following message: IXC253I PRIMARY COUPLE DATA SET 953 XCF.COUPLE01 FOR SYSPLEX IS BEING REMOVED BECAUSE OF AN I/O ERROR DETECTED BY SYSTEM MVSB ERROR CASE: PERMANENT ERROR The above message was then issued by MVSA as well. Sadly enough, our alternate SYSPLEX couple data set resides on dasd in site B. So MVSA had no connectivity to it, which led to a Disabled Wait 0A2 RC 20 in MVSA. <- snip -> Did you lose the XCF signalling connectivity at all? If you have got the above message on the MVSA as well, the connection was still there at the time. <- snip -> After that, MVSB issued the following message: IXC256A REMOVAL OF PRIMARY COUPLE DATA SET 463 XCF.COUPLE01 FOR SYSPLEX CANNOT COMPLETE UNTIL THE FOLLOWING SYSTEM(S) ACKNOWLEDGE THE REMOVAL: MVSA Of course, MVSA could never acknowledge since it was in a disabled wait. <- snip -> Your automation tool could vary offline system(s) that are already in disabled wait in the case of IXC256A. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Allocation Messages
< snip > //* //STEP1EXEC PGM=IEFBR14 //TEMP DD DISP=(,PASS),SPACE=(TRK,0),UNIT=SYSDA //* //STEP2EXEC PGM=IEFBR14 //TEMP2 DD DISP=(OLD,PASS),DSN=*.STEP1.TEMP //* I get the following allocation messages for STEP2: ** IEF373I STEP/STEP2 /START 2005324.0926 IEF374I STEP/STEP2 /STOP 2005324.0926 CPU0MIN 00.00SEC ... IEF237I 3D55 ALLOCATED TO SYS1 IEF285I SYS05324.T092605.RA000.KEEPTEMP.R0100540 KEPT IEF285I VOL SER NOS= MVS3WK. IEF285I SYS05324.T092604.RA000.KEEPTEMP.R0100539 DELETED IEF285I VOL SER NOS= MVS3WK. IEF375I JOB/KEEPTEMP/START 2005324.0926 IEF376I JOB/KEEPTEMP/STOP 2005324.0926 CPU0MIN 00.00SEC ... BOTTOM OF DATA ** < snip > On our system (1.6) I see the following messages: IEF237I 9970 ALLOCATED TO SYS1 IEF285I SYS05325.T080809.RA000.T1AI$AMB.R0202035 KEPT IEF285I VOL SER NOS= CTST09. IGD105I SYS05325.T080809.RA000.T1AI$AMB.R0202034 DELETED, DDNAME=TEMP2 and, if I omit the //TEMP2DD statement: IEF237I 9971 ALLOCATED TO SYS1 IEF285I SYS05325.T080945.RA000.T1AI$AMB.R0202037 KEPT IEF285I VOL SER NOS= CTST10. IGD105I SYS05325.T080945.RA000.T1AI$AMB.R0202036 DELETED, DDNAME=TEMP -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: ABC of Z/os
> I'm experiencing some problem on finding the books. A serach in the redbook site shows no results for ABCs V3/4/6/7/8. Any idea ? Searching for "abcs of z/os" I get 9 hits. Volume 3 is there too: 4. ABCs of z/OS System Programming Volume 3, SG24-6983-01 Redbook, published 7 November 2005, last updated 7 November 2005 Volumes 4, 6, 7, and 8 are not published yet (at least as far as I know). Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: How much real is required for 64-bit
Not to mention with a wait stated system, since you cannot 'attach a zillion page packs'. You are limited to 253x4GB page space, about one TB. Having so many page space makes one want to use dynamic PAV volumes, it means 253x2 device devices addresses, and it means maximum 128 base volumes in a logical CU. If we take a 5-way sysplex, we need 5x253x2 device addresses for page datasets. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: LMDLIST, strange behaviour or WAD?
Sorry. - second part: only the dataset name should be: - second part: only the last dataset name Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: LMDLIST, strange behaviour or WAD?
The differing datasets are temporary, allocated dynamically by processes running on your system, and you're listing a volume that's designated in your temp pool. Between the time your rexx reads that dataset list, and then reads it again, these easily could have changed. You're not scanning a static environment. Not on our system: - first part: hundreds of dataset names - second part: only the dataset name Three runs, three times the same pattern. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Geoplex performance considerations
> It depends on what you call CF mirroring: System Managed Coupling Facility Structure Duplexing is supported now. This requires direct links between the CF's and these links seem to be limited to 5 km to prevent unacceptable delays. According to "System-Managed CF Structure Duplexing", page 32 GM13-0103-05 (June 2004) http://www-03.ibm.com/servers/eserver/zseries/library/techpapers/pdf/gm13010 3.pdf > "Recommendation: In a GDPS/PPRC multi-site configuration, do not duplex CF structure data between coupling facilities located in different sites; rather, if desired, duplex the structures between two coupling facilities located at the same site. CF structure data is not preserved in GDPS site failover situations, regardless of CF Duplexing." There is no newer information on "Coupling Facility and the Parallel Sysplex" http://www-03.ibm.com/servers/eserver/zseries/pso/coupling.html > Regards, Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Geoplex performance considerations
> When I left IGS in June 2004, CF mirroring was not supported. > Is it now? No. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: ESS pprc question
> We have two ESS800 in two locations with PPRC v2. We plan to power-off > ESS_A (primary) for realocation (1-2 hours). For this period of time > system and application will be stopped. > During this time we don’t plan to start system in secondary location, so > there will be no update on ESS_B > I’m looking for a safe procedure to power-off ESS_A with existing PPRC > and I don’t want to perform ‘initial copy’ after procedure. My opinion, assuming all systems are down during the relocation: Method 1: - Stopping all systems leaves the volumes in sync. You don't have to do anything. Method 2: - before you power down a box, "Delete pair" - after the relocation "establish nocopy" Question: Do you make any changes in the GDPS DASD configuration (e.g. pprc links)? Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Geoplex performance considerations
<---snip---> To all who have helped with our research into GDPS with SRDF/A, THANKS! What we've found is that virtually no one is running a GDPS with syncronous data replication, and we fould only one instance of active data sharing across sites. We did find a handful of people running dual sites with application routing giving the appearance of a GDPS, and with failover between the sites based on asyncronous data replication. If you'd like more info on what we've fould, check with Robert Catterall at IDUG - Robert will be publishing our results. <---snip---> If I understand you correctly, your research and results have to do with the GDPS with SRDF/A only. If you mean GDPS, PPRC, multisite (dual) workload, planned/unplanned hyperswap, then your results are false. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: XMIT problem
- XMIT nwic.myuhas dsname('ij.source') outdsname('ij.xmit') - : -- PDSFAST/ZOS VER 5.1CC CPUID 2084-00116205-1110 COPYRIGHT (C) 2003 SOFTWARE ENGINEERING OF AMERICA, INC. PDF0101I START PDSFAST EXECUTION PDF0103CCOPY OUTDD=SYS00087,INDD=((SYS00083,R)) PDF0111I UNLOAD: SYS00083 DSN=N500223.GDPS.CNTL VOL=CTST27(PDSE DCB=(80 3,120 FB PO) PDF0112ISYS00087 DSN=SYS05237.T105403.RA000.N500223.R0300610 VOL= TST25(DISK) DCB=(3,136 3,120 VS PS) PDF0131M SYS00083: MEMBERS: 41 COPIED 0 NOT COPIED 41 TOTAL PDF0189I END OF COPY STEP 1 - RC(0) , 198 CALLS TO BSAM PDF0199I END OF PDSFAST EXECUTION - RC(0) INMX000I 0 message and 70 data records sent as 1668 records to WIENX.N500223 INMX001I Transmission occurred on 08/25/2005 at 10:54:03. *** -- Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: IFIND
> They are functionally equivalent. If I try IFIND in SDSF under SE or SJ, the cursor jumps to the first (non-blank?) character in the next line, not to the next occurrence of the string. if I try IFIND under SB nothing happens (the cursor does not move). Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Setting the Creation Date for datasets
Also there's the concept that with copy commands the file already existed on the source (and target) environment; with FTP, by nature it's a 'foreign' file which didn't previously exist on teh target system. It's therefore a new creation at the time it was sent. To me it looks like there are two different approaches used when copying files in MVS: - dataset level: old copy creation date&time < new copy creation date&time - PDS(E) member level: old copy creation date&time = new copy creation date&time In the second case it looks like someone found it to be important to propagate the original creation and change date & time (XMIT does it between systems). Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: DUMMY and BLKSIZE=0
-Original Message- From: john gilmore On the other hand //SYSPRINT DD DSN=NULLFILE,BLKSIZE=0,LRECL=137,RECFM=VBA can be altered readily by simply supplying a different licit value of DSN=. This difference was the basis for my statement that DSN=NULLFILE was superior to DUMMY when used inside procedures. I believe we used the same technique for input datasets in the concatenations. It was easy for our users to call a procedure specifying //STEPXX EXEC PROC,PROJLIB1=XXX,PROJLIBY=YYY,USERLIB1=ZZZ,USERLIBZ=WWW that is, editing only the dataset names, which generated something like //SOMELIB DD DISP=SHR,DSN=GENERAL.LIB1 <-- hardcoded (while mandatory) . . . // DD DISP=SHR,DSN=GENERAL.LIBX <-- hardcoded (while mandatory) // DD DISP=SHR,DSN=XXX <-- substitution PROJLIB1 . . . // DD DISP=SHR,DSN=YYY <-- substitution PROJLIBY // DD DISP=SHR,DSN=ZZZ <-- substitution USERLIB1 . . . // DD DISP=SHR,DSN=ZZZ <-- substitution USERLIBZ // DD DISP=SHR,DSN=NULLFILE <-- default parm value . . . // DD DISP=SHR,DSN=NULLFILE <-- default parm value Generallly, the users were not supposed to be good in JCL, and there were maybe 3 copies of JCL manuals for all of us (system programmers, aplication programmers, and users). That was the time before we moved to ISPF panels and skeletons. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: DUMMY and BLKSIZE
It's easy to suspect that at some point in time IBM resolved all such problems by adopting a tactic of converting DSN=NULLFILE to DUMMY early in the JCL conversion process, and that the few (only?) contributor[s] to this thread that maintain (without presenting evidence) that the behaviors differ perhaps have lingering memories of the status quo ante. It is a way back (MVS SP1 ???), but I think there was a functional difference: it was not possible to use DD DUMMY for dummy input datasets, you had to use DSN=NULLFILE. Testing today gives no difference. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: PDF vs BookManager
Depending on how the .PDF was generated, you might not even be able to do that. I was reading one a couple of days ago and needed to go to section 2-4 or some such, according to the index. With some PDFs, you can type "2-4" into the page number field and jump directly there. Not so with whatever manual I was reading at the time. Thus, I pretty much had to plug in a random page number (64? 51? 123?) and hope I was in the vicinity of 2-4. Thankfully, that situation is fairly uncommon. I would not call it "fairy uncommon". All OS/390 and z/OS manuals and Redbooks I've seen had book page numbering different from pdf page numbering. Žaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: PDF vs BookManager
PDFs, IMO, are wonderful to print from. They are nice for serial reading (oh, how do I bookmark so that I come back to where I left off?). I have Acrobat Reader 6.0.2 in german language, so there will be some problems with vocabulary: - Ctrl-K (go to General settings or something like that) - select Program start (or something like that) - instead of "eBooks only" (or something like that) select "All files" (or something like that)in drop-down menu - Check Use page cache (or something like that) It works for me, at least for "Diagnosis Tools and Service Aids" :-) Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Re: Simultaneous system and CF failure and ISGLOCK REBUILDPERCENT
Kees, --- --- I suppose you select the icons on the HMC and do a deactivate for them. We concluded that our CF's were activated before the z/OS LPARS and in our case, DB2 immediately started recovery for its lost structures. Shortly after that, the z/OS LPAR was deactivated and on a restart the found situation was not as we expected, because DB2 had already started recovery and was interupted in that process. --- --- I payed attention to this during earlier tests (systems and dasd failure) but I am not sure how we have done it this time. I'll try again. I haven't look in the operlog (operlog structure was in the lost CF and was never rebuilt), just in the syslogs. --- --- We found out that we had to deactivate the z/OS LPAR first and had to deactivate the CF LPAR within the XCF timeout value, to prevent other systems in the sysplex from starting recovery actions. This really simulated a full machine outage as expected. --- --- Thank you, Kees. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
Simultaneous system and CF failure and ISGLOCK REBUILDPERCENT
I made some testing to see what happens in a case of a CF failure or CF and system(s) failure. Our test parallel sysplex configuration: 4 systems (S1 - S4) z/OS 1.6 2 coupling facilities (CF1 & CF2) CFLEVEL 14 SMF policy with ISOLATETIME(0) CONNFAIL(NO) GRS STAR configuration (ISGLOCK in CF2) I was said that loosing one or more systems and the ISGLOCK structure at the same time could bring the whole sysplex down if we use SFM weight defaults of 1. Our ISGLOCK structure has a default REBUILDPERCENT (according to documentation it should be 1, but in the message IXC360I displayed as "REBUILD PERCENT: N/A") Test scenario 1: - set SFM weight for S3 to 40 (that was supposed to be the minimal weight for the important systems), other systems have weight = 1 - deactivate lpars S3, S4, and CF2 at the same time Result: - systems S1 and S2 partitioned the S3 and S4 from the sysplex and rebuilt the ISGLOCK. Everything works as expected (except operlog, etc.) After that I tried to build a scenario where the ISGLOCK rebuild will not happen. I took the explanation from the "Setting Up the Sysplex" (SA22-7625-09 because of the APAR OA05860). Test scenario 2: - set SFM weights for S2, S3, S4 to , S1 to 1 - deactivate lpars S2, S3, S4, and CF2 at the same time I expected the system S1 not to be able to start a rebuild process because of the explanation and formula in the manual. Results: - system S1 partitioned the S2 - S4 from the sysplex and _then_ it rebuilt the ISGLOCK. Everything looks normal (as in results in scenario 1) Now I have some questions: 1. Does a system / connector rebuild a structure with default rebuild percent in a case of a connectivity loss no matter what SFM weights are defined? 2. To rephrase the question 1: Is the default rebuild percent the same as REBUILDPERCENT(1)? 3. What happens in a case that I lose one or more systems and the primary couple data sets at the same time? Is the partitioning of the lost system(s) still possible or the sysplex hangs because the CDS switches to alternate cannot be acknowledged by failed systems? That is, what is the order how it is done: first partition of the failed system(s) and than cds switch or the other way around? (Such a test is not easy to organize.) I hope i have not make the mail to complicated. Zaromil -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html