Re: Batch job elapsed time baseline and alerts
With "several hundreds if not thousands jobs running at any give time" you must be using an automated job scheduler of some sort. Does your scheduler maintain 'average run time' in its database? Does the scheduler itself, allow notification if that average time is exceeded? If not - you can use MXG, or other means, to calculate statistics about run time. If you have to get notification in real time, you're going to need to code some stuff in exits. One approach (I haven't coded this so bear it in mind): Store the average time somewhere (MXG, wherever). For each job put that time into a comment card in the JCL. Use a fixed, easy to parse format. In exit IEFUJV, parse that comment card to make the average run time into a binary, add store it into a storage area you GETMAIN for each job (make sure to store the pointer into the user area for the job provided by the SMF user exit interface). Mark a flag in that area as 'no msg issued'. In exit IEFUJI, get the current time, add the average run time to it, and store the result in your job 'control block' In IEFU83 (shouldn't need IEFU84/85 for regular batch jobs, I wouldn't think?) get the current time, and compare it to what you stored. If it is larger than what you stored, issue a message via WTO and mark the flag as 'message issued'. This is to prevent issuing the message more than once for the job. IEFU83 gets called a lot since it is called for every SMF record created for a job so make sure your routine is coded efficiently. This is off the top of my head. I used IEFU83 to provide the ability to notify as soon as you exceeded what you expected; if you don't need to be that granular use IEFACTRT for step end, or IEFUSI at step initiation, and check between steps. The 'hard part' is getting the information into the system with the job. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: report writer alternatives
Is this for users, "power users", or programmer/developers? For some of those groups, SyncSort (not sure of the new names yet) or DFSORT provide pretty good reporting capabilities and great performance. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: DFSORT and SMF date selection
Thank you, I think this is a better solution for me right now. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: DFSORT and SMF date selection
I am not using IFASMFDP I am doing other things in the DFSORT application. I'm trying not to do all of the I/O to select the records with IFASMFDP, then read the selected records a second time into DFSORT to process them. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: DFSORT and SMF date selection
Thanks for the solution. I am wondering, though: since I am not interested in dates in the 20th century, could I use COND=(12,3,Y2U,GE,) ? The rightmost 3 bytes of the SMF date would seem to satisfy Y2U and as long as my offset of -7 doesn't take me to a previous century I should be OK, right? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
DFSORT and SMF date selection
The brain is just not working today I guess. I want to use DFSORT to select some SMF records for a week; I'm getting myself confused in the DFSORT application programming guide Do I want to use INCLUDE COND=(11,4,PD,GE,) or INCLUDE COND=(11,4,Y4U,GE,) or have I missed something entirely. The point is to have the job be scheduled weekly, after our last SMF dump at midnight and subsequent accumulation of SMF records so today's date minus 7 gives us yesterday and the 6 days previous. I'm just not sure of the right way to do this with SMF dates which are right now x'0121165F' -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
ldapsrch/ldapsearch and netrc
I have been unsuccessful so far in finding out the answer to this in the IBM documentation on the LDAP client utilities. Maybe my search-fu is lacking today, but here's the question: When running the ldapsrch utility in z/OS (IKJEFT01 in batch or in TSO) can a NETRC file be used instead of the -D and -w parameters? If anyone on the list knows, please reply. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: TWS replacements
ASG Workload Scheduler (formerly Beta42, formerly from Pecan) has "agents" for Windows and *nix and z/OS; I would assume their Zeke schedule would have those capabilities too. The shop I retired from used the z/OS agents to coordinate scheduling of data gathering for MXG and chargeback so the 'Admin LPAR' would kick off the MXG and chargeback jobs only after the jobs on each LPAR that created the daily SMF datasets had run. Also tested using the Windows agent to kick off executing commands on a Windows server and that worked but was never implemented in production. If I recall the source was not shipped for the agents so probably the *nix one was compiled for x86 as well as the Windows one, but never needed to test that . Our schedulers like Beta42 better than Zeke when they tested it and found CA-7 too complex, but our schedulers were mostly upgraded data entry folks, they were extremely focused on 'easy to understand' and not so much on 'great features'. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Starting a started task from a started task
this _may_ work: Include a dummy dataset with DISP=OLD in the JCL for both started tasks. Start them one after the other in the order you want them to run or if you have an automated operations tool, start them a second apart or more - whatever you are comfortable with. The DISP=OLD will cause an enqueue and prevent the second one from starting until the other one finishes. Depending upon your shop's parameters the operators may or may not see a message about the second started task waiting for dataset. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Inspecting and extracting from /OS transportable files on other platforms?
The best tool to transform XML is XSLT. It's a little funky to learn, because it's a "functional programming language" but it's designed for XML. I used it this way ] > The stylesheet does all the work. You'll find information about XSLT around the 'net. I don't claim to be good at it, I hack away until I get what I need. Note that the contents of do not _have_ to be XML, I have used this to transform calendar text files into XML and played with extracting info from COBOL source. Also note: I have not done this with the XML you're talking about. I don't remember (Semi-retired) if z/OS contains a 'batch' XML processor; the one I have on my PC at home was XT, written in Java. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Using PCOMM with SSL ?
If you can export the cert in z/OS then you ought to be able to use one of the RACF certificate displays to determine who signed it. I _think_ you can export this in Base64 format, move it to Windows as text, and import it into the Windows certificate store. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Finding JES2 node name
I guess I'm asking if IEFSSREQ is somehow better than AMASYMBF for retrieving the node value. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Finding JES2 node name
The code is not just to _find_ the node, but we want to use the node name for other purposes in this program. So SDSF isn't going to help here. As for other methods of using IEFSSREQ and/or control block chain chaining why are they better than the ASASYMBM/F program interface? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Finding JES2 node name
The program is already in Assembler - so I'm guessing I would use the AMASYMBM or AMASYMBF service to resolve SYSNODE -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Microsoft using IBM was Re: IBM splitting into two companies
The Mono project ran .NET stuff on Linux, and I have seen it running on z/Linux -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Finding JES2 node name
I have some code that follows a control block chain to the HCCT to get the JES2 node name, but that requires going to key zero (or so the source says, it's been years ). Is there a way to retrieve the JES2 node name in a program running in batch, without being APF-authorized? We'd like to remove the need for this program to be in an APF-authorized library. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: z/OS use of "legacy" programming languages
I know there's still a VS FORTRAN compiler on a machine I work on - is it still supported, or does it just "work". Haven't seen updates for it in forever; we are running down what was compiled with it that may still be running. All the important stuff moved to a FORTRAN supplied by another vendor that runs on PCs or other platforms; what's left would be dumb little utility-type programs that some user is loathe to turn loose of (and those users can probably tell you how they used to punch the input cards for the program. ) -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Possible new function for DFSORT?
I could try to do that but I was hoping to use just the tools I have. The nslookup command on z/OS (TSO and "TSO in batch" returns more than one line, which means writing a Rexx exec to parse it into something to match to the output of the other stuff -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Possible new function for DFSORT?
I don't think this function exists, and I'm thinking about writing up a SHARE requirement for it (which I guess these days becomes an RFE but I'm a member of SHARE so I think I'll go that way). I thought I'd ask whether it would be used enough to justify going through that process. What I'm thinking about is a function (well, two) to take a binary IP address and do DNS lookup, retrieving the host/domain name. At the shop where I'm currently working, many logs don't do the lookup when logging, to save some overhead.Yet, it's always nice, when creating a report such as 'who is connecting to us the most', to be able to assign a name rather than an IP address; if the lookup can be done _after_ the data is summarized it would be a lot more efficient. Hence the thinking about making this a function for SORT. I think this would involve two functions - one for IPV4 and one for IPV6. We might also like two formats similar to DT1 and the like, for formatting IPV4 and IPV6 binaries into strings. Although I think that can be managed now by handling individual parts of the binary and interspersing separator characters "manually" it would be much cleaner if it were handled like date/time fields are. Do you think this capability would be used enough to make it an RFE? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CL/ / Supersession
The password might be sent in the clear - but that is an SNA SLU (the TN3270 LU for the terminal session) to PLU (CL/Supersession) session and as such would require looking into methods for encrypting SNA sessions in the Comm Server manuals. I'm pretty sure it's not flowing over IP if TN3270 and Supersession are on the same system. If you're going between systems it might be Enterprise Extender? Which flows over IP and I'd guess Communication Server manuals would tell you how to encrypt that as well. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
ISPF 3.4 DSLIST questions
I have some questions about the ISPF 3.4 utility. 1. Why does 'Referred' show on the "total" display for datasets, but if you print the dataset list, you don't get it? 2. Are there ways to extend what is displayed? For one example: I would like to have column for 'Your Access' that would show me what RACF says my access is, rather than having to do LD DA(/) ALL GEN on a line, and "suffer" through the TSO command output (as I've rarely worked with ACF2 and never with Top Secret I don't know if such a request can be done for 'generic security system') -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: dfdss equivalent to fdr map
Question: does it really matter with a volume that's a virtual thing implemented on a RAID array? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CL/ / Supersession
TN3270 supports secure sockets. Just make your TN3270 terminals "auto logon" to CL/Supersession and you're good. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: IDCAMS not freeing file
Questions: 1. Is this the same VSAM file shared between the two partitions? 2. Is the VSAM file referenced in JCL anywhere, or just in the commands in IDCAMS? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: tape drice or tape cart???
However you are specifying a device, or your task is selecting the device, the device is either already allocated to something else, or it is boxed and has no I/O paths.I'm not a big VM guy, but is it possible you have a device defined to multiple guests, and you have it online to more than one, without some sort of device sharing synchronization software to make sure only one is using it at a time? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Mainframe user ID length
As many have stated you are limited to 8 upper case characters, 7 if you still use UADS; however if the MQ user is off-platform, perhaps one of the various tools for mapping other IDs to a RACF ID could be used? These are all part of RACF (not sure about ACF2 or Top Secret) but as I see it they boil down to authenticating using some non-RACF method, then taking that ID and mapping it to a RACF ID, which would be the ID used in the ACEE (and probably other places), but the user wouldn't necessarily need to know it. There are probably people on this list far more expert in those than I am, having never had to implement one of them, but I know they exist. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: DB2 / DFSort question
I was not aware that the UNLOAD utility invoked DFSORT, I figured DSNTIAUL did. I was wondering it that was as efficient as invoking an SQL query in E15, but with the other responses I think it's probably not worth coding another one to find out. As for creating the symbols - I can probably find ways to do that from the DDL created by DSNTIAUL. Thanks for all the responses! -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
DB2 / DFSort question
Before I start chiseling a piece of stone into a wheel :-) - I've been thinking about ICETOOL and DB2. Is there already in existence something that would provide a SELECT (similar to what is used in DSNTIAUL or DSNTIAD) to select DB2 data as an E15 exit for DFSORT? My ideal would be (Maybe I should write an RFE) -- In ICETOOL the ability to code SQLSEL FROM(SQL) USING() CNTL DD * SELECT -- It would be nice if this could execute before SYMNAMES is read, so the SELECT handler could create symbols for all of the columns in the result set. Maybe it is just as efficient to use DSNTIAUL with ORDER BY ? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: LOGR Offload Dataset deletion problem
Maybe it's because you have start date in the DATE parameter? In the documentation at there are specific rules about ignoring the start date when using ARCHIVE or DELETE https://www.ibm.com/support/knowledgecenter/SSLTBW_2.4.0/com.ibm.zos.v2r4.ieag200/dumplopt.htm There are these statements (those not applying to DELETE have been skipped to make this post shorter) 3. ARCHIVE and DELETE options: When you specify the ARCHIVE or DELETE option in LSNAME, pay attention to these items: - The start date and time of the IFASMFDL is always the start date and time of the log stream. The start time that is specified by the START parameter is ignored. The start date that is specified by DATE or RELATIVEDATE parameter will also be ignored. - If you specify the DELETE option, everything in the specified log stream is removed, from the start of the log stream up until the end date that is specified by the DATE or RELATIVEDATE parameter and the end time that is specified by the END parameter. The OUTDD statements are not permitted when the DELETE option is used. - When using the DELETE and ARCHIVE options, you must ensure that only one instance of the IFASMFDL program is running at a time for a given log stream. Running multiple instances of the dump program against the same log stream leads to unexpected results. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
FOCUS / FOCLOG
Do any of you run FOCUS for Mainframe from IBI? If so, do you also license their FOCLOG feature to record information about what programs (called focexecs) have run and what resources they used? [Without that log, it's pretty opaque - you can see in SMF that FOCUS was run as a batch program, but not which of its "programs" has been invoked. The usual SMF statistics for a batch job are there, but you can't tie it back to a FOCUS program name. ] If you used FOCLOG, how did it work for you? It writes to an MVS dataset (I believe with DISP=MOD) rather than recording via SMF. IF we were to use it, I foresee issues like B37 abends on the log dataset, or having to manage many GDG generations every day, and the requirement to allow _everyone_ to update it because this is a product that can be used to create ad-hoc reports. I'd like to hear of any experiences anyone has had with this component. Thanks in advance -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Clist dataset
We submit a batch job with two steps: first step IEFBR14 with the dataset name allocated DISP=OLD, second step IDCAMS to do ALTER dsname NEWNAME(newdsname). The job ENQs on the dataset name so new TSO allocations for it fail, and the job waits until all TSO users using it have freed it or logged off. Once it has the dataset it finishes very quickly. I think we tried having some 'dummy' DD in the IDCAMS step with the dataset name and DISP=OLD (eliminating the IEFBR14) but I can't remember if that worked or not -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SMF for cataloged procedures and INCLUDEs. WAS Re: "Trapping" messages written to JESYSMSG?
I don't check this list regularly but I think you slightly misunderstood. I'm not looking for who is using what procedure _library_. I'm looking for the use of individual members. RACF doesn't (unless it happened in a recent release and I missed it) allow us to audit individual members of a library. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
SMF for cataloged procedures and INCLUDEs. WAS Re: "Trapping" messages written to JESYSMSG?
As mentioned before, I've created a SHARE requirement and an RFE to have the system record the information from the IFC001I (cataloged procedure was expanded from... msg) and IFC002I (INCLUDE was expanded from... msg) message in an SMF record.If this would be important or useful to your shop, you should vote this up. I know we've wanted to find a way for years to determine what cataloged procedures are in use, and eliminate the ones no one uses from the non-IBM libraries. This will also help determine if someone is using JCLLIB to run their own version of a procedure. Here's the RFE link: http://www.ibm.com/developerworks/rfe/execute?use_case=viewRfe_ID=133491 Here's a link to the SHARE discussion page but please not it might move. I put it in JES in the RFE system and the SHARE requirements system, and IBM moved it to BCP in the RFE system. https://www.share.org/p/is/vi/=173=21 -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Tape Mount delay notification
Can you, if you automate the message, in that automation code issue something that gets invoked after on a timer basis? If so you can issue a message 'n' minutes after the mount message. I know AF/Oper from IBM/Candle has that capability but I don't know much about OPS/MVS. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Personal Communications Printer Session
I don't know of a list to discuss PComm but if these printers support IP printing (either direct socket or LPD protocol) - and most do these days - you can define the IP address or host/domain name in something like VPS or Infoprint and eliminate the need for the 3270 printer emulation session entirely. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Volume compare utility
I don't think we had the number of volumes Skip has, but we did several migrations using short-term leases of FDRPAS and they all went extremely well; the great thing about FDRPAS is that you don't need to do most of it during "down time". -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: How to grant access to CONSPROF - Conclusion
If this is running in batch, can you make the ID have PROTECTED status, as long as the ID is propagated from something that submits the job? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: "Trapping" messages written to JESYSMSG?
I have created a SHARE requirement for this (which I think was submitted to RFE) _and_ an RFE ( 133491 ) to write IEFC001I and IEFC002I information to SMF. If voting is still open and this information is important to your shop, please vote. My suggested solution was to make any solution available to all JES systems that are currently GA. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: How to Identify PROC name in SMF
Not the simplest, but Write an IEFUJV exit. When you find EXEC on a statement, and you cannot find PGM= on that statement, you have a proc. Write a simple SMF record with a user record type (128 -> 255) . Post process with SORT to summarize usage counts. This won't pick up JCLLIB, etc. but even with JCLLIB there's no easy way to tell which proc came from a JCLLIB and which from a proc in the JES concatenation; you might need to get into JES2 exits or source to look at that. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: FTPLOGGING not working
If you do find out what they did to resolve it, would you please post it here for the education of all? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: "Trapping" messages written to JESYSMSG?
Is there an exit that can "see" these messages? Maybe an MPF exit? If so, you could always write an exit to echo the messages to some routing code that doesn't go to the console, process it with automation, and hopefully prevent it from going to your operlog or syslog. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Concatenating VB and FB ?
I believe concatenation now uses the largest blksize, not the first one. Not sure about LRECL -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: ind$file - omvs copy equivalent
I don't understand why the OP wants to receive it to USS and then copy it to an MVS dataset when they could run FTP under TSO and receive the XMI file directly? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SMS ACS Routine Decompiler
I would have assumed that "SELECT WHEN" would be processed into different things internally than "IF/THEN/ELSE", but your evidence would imply otherwise... it implies that "SELECT WHEN" is perhaps pre-processed into "IF/THEN/ELSE" statements before "compiling". Either that, or someone tried to write one themselves and it wasn't very good I wonder if IBM has such a tool? If they do, perhaps your client can convince IBM of the need to help them out? I know we got some APF rebuilt by IBM one time, after libraries were lost. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Concatenating VB and FB ?
I seem to be finding different answers on this. A vendor used to ship some files as PDSes with RECFM=FB and LRECL=80 (BLKSIZE 23440). User-customized members at this shop were put in a different PDS, with the same attributes, and concatenated in cataloged procedures, ahead of the vendor's libraries. Pretty standard practice I'm sure most are familiar with. Suddenly, because (I'm told) of a merging of code bases at the vendor, their PDSes are now RECFM=VB and LRECL=2044 (BLKSIZE 27998) ! My instincts tell me this isn't going to work well, but with changes in concatenation of libraries over the course of my career I'm not sure.Here's what I think: because of the "new" rule where the largest BLKSIZE sets the buffer size, we'll be OK for reading the blocks (23440 fits into 27998) but when we try to read a member from the VB library, the RDWs are going to mess things up. I have tried searching for the answer, but haven't, apparently, found the right source yet. What say you all? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: ASG Workload Scheduler?
"the big diversion"? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
ASG Workload Scheduler?
I'm curious about how many here run ASG's Workload Scheduler, which was Beta-42, and before that was created by Pecan. The shop I'm working for still runs it and is still happy with it for years though they've asked me at times to do some specialized reporting about tasks and schedules (which I manage to do with SORT for the most part) that aren't part of the product. Just looking to see if there are others to SHARE stuff with. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Concatenation and UNIT=(,n)
Figuring that this is 'working as designed', I have started a SHARE requirement, where discussion could take place. My suggested improvment is that for multi-unit, multi-volume datasets (and I realize this probably mostly applies to tape) a consolidated list of volumes for the concatenated datasets be constructed, and the mounts happen from that list. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Concatenation and UNIT=(,n)
For a number of years I've let the catalog and/or SMS supply the unit, but I've supplied unit count (ex: UNIT=(,3) ) to overlap I/O and mount processing. I even use it with virtual tape systems that have physical tape backup - because it can allow a recall of a virtual volume to be done while I/O is processing. But I never noticed until know that if you concatenate two DDs like this: //SORTIN DD DISP=SHR,BUFNO=60, // UNIT=(,2), // DSN=dataset.one // DD DISP=SHR,BUFNO=60, // UNIT=AFF=SORTIN, // DSN=dataset.two and 'dataset.one' only has one volume, the system will issue a mount for that volume, and then one for volume 'PRIVAT', it won't mount the first or only volume of the second dataset. Is this behavior something that can be modified with some obscure JCL or system setting, or is this a candidate for an RFE? I'm making the assumption that physical tape hasn't reached the point where it isn't used anywhere, of course. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SMS Compression measurment?
Not worried about the efficiency of compression. The issue is showing to someone the CPU saved by using zEDC instead of SMS compression. IBM's zBNA tool estimates it, and we may just use that, but I was hoping someone had a way to measure the CPU used just for SMS compression in a job, without having to run it twice. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: TCPIP IP address for current LPAR
These may be dumb questions but: 1) I'm assuming you connected to the remote application first? Else how could it connect without the IP address somehow? 2) Why isn't the remote application content with the host/domain name (since IP addresses _can_ change)? 3) Why isn't the remote application using DNS (via nslookup or other means) to get the IP address? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Processing Partition organization type record
You can do this with ISPF in batch, and Rexx. I've used this when I had to
change the one thing in hundreds of PDS members. Of course you have to balance
your development and debugging time versus just doing repetitive edits via
RCHANGE (PF6)
To invoke ISPF in batch. Note BDISPMAX very large to prevent failures from
exceeding an ISPF message limit
//IKJEFT01 EXEC PGM=IKJEFT01
//SYSEXEC DD DISP=SHR,DSN=your.exec.library.with.the.exec
//ISPPLIB DD DISP=SHR,DSN=ISP.SISPPENU
//ISPMLIB DD DISP=SHR,DSN=ISP.SISPMENU
//ISPSLIB DD DISP=SHR,DSN=ISP.SISPSENU
//ISPTLIB DD DISP=SHR,DSN=ISP.SISPTENU
//ISPPROF DD DISP=SHR,DSN=your.ISPF.profile.library
//SYSTSPRT DD SYSOUT=(,)
//SYSTSIN DD *
ISPSTART BDISPMAX(9) -
CMD(%Rexx_exec dataset_name variable_value)
This is the Rexx I used with some changes to protect the innocent..
/* REXX */
/* This is the exec named Rexx_exec in the batch job */
/* This exec will invoke ISPF EDIT with a macro for each member */
/* of a PDS
*/
PARSE ARG DSN variable
ADDRESS ISPEXEC
/* variable comes in at invocation, must VPUT to use it in the edit macro. If
you don't need to pass a variable you can eliminate this */
"ISPEXEC VPUT variable SHARED"
"ISPEXEC LMINIT DATAID(DATA1),
DATASET('"DSN"') ENQ(EXCLU)"
/* sets the DATAID for the PDS passed in
DSN */
"ISPEXEC LMOPEN DATAID("DATA1") OPTION(INPUT)"
/* OPEN FOR INPUT */
EDITRC = 0
LMRC = RC
DO WHILE LMRC = 0 & EDITRC < 8
/* BUILD MEMBER LIST unless an error
*/
"ISPEXEC LMMLIST DATAID("DATA1")
OPTION(LIST) MEMBER(MEMBER)
STATS(NO)"
LMRC = RC
/* CAPTURE RETURN CODE */
IF LMRC = 0 THEN
/* IF A MEMBER NAME */
DO
/* RETURNED OK
*/
SAY 'PROCESSING MEMBER' MEMBER 'OF MASTER AT' TIME()
"ISPEXEC EDIT DATAID("DATA1") MEMBER("MEMBER") MACRO(edit_macro_name)"
/* edit the member using edit macro */
EDITRC = RC
/* return code from edit
*/
END
END
"ISPEXEC LMMLIST DATAID("DATA1") OPTION(FREE)"
/* FREE DATA LIST */
"ISPEXEC LMCLOSE DATAID("DATA1")"
/* CLOSE DATASET related to DATAID */
"ISPEXEC LMFREE DATAID(DATA1)"
/* Free the DATAID */
Skeleton of the edit macro
ADDRESS ISPEXEC
"ISPEXEC VGET variable SHARED"
variable = "'"||variable||"'"
/* had to add single quotes for
some reason */
/* may not be
needed */
"ISREDIT MACRO"
"ISREDIT (CURDSN,,) = DATASET"
"ISREDIT (CURMEM) = MEMBER"
IF MEMBER = ' '
Re: long alias names question
These are all good questions - in this instance I don't know the internals of the product that might be creating these long names. It's a 'developer studio' type of product which runs on Windows and communicates to a started task from the same vendor which can also run on other platforms such as a Windows service or a Unix process (I think it's a daemon but haven't researched this product running on *nix). So, predictably the developers are grousing about the 8-character limit. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: SMS Compression measurment?
Yes, I've looked at Scott Barry's presentation. It tells me a lot of good info about zEDC measurements, and it does document compression information in type 14/15 but it appears as though IBM didn't provide any measurement of CPU used by SMS compression. I guess the fallback is to try to run the same job twice, once with compressed output and once without compression (and also read each back in). -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
SMS Compression measurment?
I can't remember if I posted this before, but it's come up again so I'm asking. Is there any way, short of running a job with SMS turning compression, and running it a different time with compression off, to measure the CPU time used by SMS compression for a dataset? I know I can get zEDC CPU savings estimates from zBNA and I'm pursuing that path too, but it seems to me it's something that ought to be reported; especially since recent versions of z/OS have zEDC measurements reported to SMF. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Rexx not processing email formatting
If those quotation marks are part of the output of the Rexx routine, that's the issue. E-mail program would not treat those lines as the commands they purport to be, because the lines don't begin with a recognizable command they begin with a string that consists of the double quote PLUS the command. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: DASD-only logging
Skip - we run IEAMDBLG to offload OPERLOG to sequential datasets, named with the date, once a day. You could offload your three systems to temporary sequential datasets and combine them via SORT (DF/ or Sync) into something sorted by date and time. One caveat - even if you use HCFORMAT(CENTURY) the IEE042I SYSTEM LOG DATA SET INITIALIZED messages seem to have 2-digit years so I'd recommend omitting them . -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
long alias names question
Is there a way to use long alias names in PDSE for members that are not program objects? I've seen symbolic links to a PDSE (or am I seeing one for each member?); we're asking if we do that to share a _data_ PDSE with tools that only understand directories, can we create names longer than 8 bytes ? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Another IBM tape not for z/OS
I moved a lot of tape datasets to SMS managed disk, also, but one group of application people fought the idea of using unique, catalog-able names for their datasets. Pure laziness - they didn't want to go through the work of making the changes and testing. They did eventually have to do it though - AFTER a scheduler/operator looked up the dataset name in CA-1 and put the _wrong_ set of volume numbers into the production JCL for job involving mailouts for jury duty.The ensuing brouhaha over citizens getting jury notices two months in a row caused inquiries from levels of management they hadn't heard from in years! -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Exit Calls or Using
Let us not forget one of the fun cases: exit code which exists, but because of external circumstances, is ineffective: for example account number checking when the idea of chargeback has been dropped and the check now allows anyone to use any code, or an exit which uses RACF to check on something and the profile has been changed so that UACC allows everyone to pass. So your exit is invoked, MAY not be needed, but requires further work to check. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: zos program to supply web based client
I haven't seen the Apache web server mentioned for some reason. I haven't used Apache but on the previous Domino-based web server you could write a Rexx CGI (Common Gateway Interface, not Computer Generated Images) program that could be invoked via URL and retrieve whatever you want. When I fooled with it I chose to return XML but you can return whatever the client program needs I am sure. I don't know if Apache still support CGI but it definitely supports _some_ way to invoke an application program via the web. The Apache server is "free" with z/OS, too, I believe. Disclaimer: I am retired and consult part-time, so if some z/OS 2.3 thing has changed circumstances I might not be completely up to date. Hopefully to be remedied at a future SHARE someday. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: How to capture LOGS of system running tasks (was Job submit using REXX )
I am not sure why you want to capture this with Rexx, but I will second the idea of using the SPIN= keyword to spin off log entries on a regular interval. If you just want to capture to disk, you can use one of the products mentioned, or if you don't have such a product, you can create an external writer STC (external writer comes with z/OS) to capture that output class. If you are looking for particular messages in the log of the task or job, a better tool is to use an automated operations tool, many of which can invoke Rexx when trapping a specific message. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Allocating an SMS managed dataset in TSO with space in bytes
This is, actually, a request for records - I was using the 1-byte average record size because I thought the original poster wanted to allocate in megabytes. If you want to request records and you know the LRECL or the average LRECL, use AVGREC=U/K/M, and SPACE=(avgLRECL,(pri,sec)) where 'pri' and 'sec' are in records (AVGREC=U), thousands of records (AVGREC=K), or millions of records (AVGREC=M). My only problem with this was convincing application programmers to use it. I used it all the time, but still found application programmers trying to calculate efficient block sizes (they did, eventually, see the BLKSIZE=0 light) and use various 'space calculators' to come up with a number of cylinders. I proposed, but never got to do, a 'what you don't know about zOS' class to touch on some of these things. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CICS and DB2
I believe there is a DB2 JDBC driver. How, exactly, you get it installed and running under CICS should be documented by IBM. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Allocating an SMS managed dataset in TSO with space in bytes
AVGREC=M,SPACE=(1,(1,1)) allocates one MB with a 1MB secondary extension in JCL. AVGREC(M) AVBLOCK(1) SPACE(1,1) will do the same thing in TSO. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: IBMVSM,ZOSMIGV2R3_NEXT_VSM_USERKEYCOMM
I'm catching up late here, but in connection with the original poster's statement, couldn't the health check "go hunting" and report the offenders? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
IEE042I in OPERLOG - why no century?
First let me state we are copying the hardcopy log to disk in two ways: 1. External writer proc, SLOGWTR, which writes to a GDG 2. OPLOGCPY job which copies from system Logger OPERLOG to disk I haven't convinced others to rely on OPERLOG yet :-(. In any case, I have a small issue. Wonder if it is "working as designed", or a bug (but it's been going on for years so probably not)? We copy OPERLOG to disk, there is one message that does not have a 4-digit year. - IEE042I (see below) === 2018096 00:00:00.10 STC01917 0290 W L 2018096 00:00:00.10 STC01917 0090 AFTRAP *** AFOPER SLOGWTR SLOGWTRX: S S 2018096 00:00:00.10 STC01917 0290 S SLOGWTR.L 2018096 00:00:00.10 0281 IEF196I IEF237I JES2 ALLOCATED TO SYSLOG4 2018096 00:00:00.11 0281 IEF196I IEF285I +MASTER+.SYSLOG.STC0190 2018096 00:00:00.10 STC3 0090 IEF453I PPSJOB01 - JOB FAILED - JCL ERROR 2018096 00:00:00.11 0281 IEE043I A SYSTEM LOG DATA SET HAS BEEN QU 18096 00:00:00.11 SYSLOG IEE042I SYSTEM LOG DATA SET INITIALIZED 2018096 00:00:00.11 STC3 0090 $HASP395 PPSJOB01 ENDED === HCFORMAT is not coded in PARMLIB so defaults to the 2-digit year for SYSLOG. Nothing in the manual that I could find about HCFORMAT for OPERLOG (I always assumed OPERLOG was 4-digits always, until noticing this anomaly). Has anyone else run into this? Is HCFORMAT(CENTURY) the only solution? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: How to speed up IFASMFDP SMF Dump process
Let me add to my post - if you read abut SMF CI Size in member ACHAP03 of the MXG source/documentation library, you will see a very good discussion of the trade-off between DASD space and CISIZE in MANx datasets. Apparently SMF doesn't do VBS in the MANx datasets in the same way "regular" VBS is done, and can waste space because of how it does it. The ANALSMF program includes what they call an "SMF Simulator" that attempts to help you find the optimum CISIZE for the SMF records you produce. There are also a lot of other SMF tips in ACHAP03 that are worth reading. If you don't have MXG perhaps you could contact Dr. Merrill and ask if he would share that information as a separate document or if he has done a SHARE presentation about it. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: How to speed up IFASMFDP SMF Dump process
Do you have MXG? They have a program, ANALSMF, which will analyze your current SMF. In general, what I've done and which seemed to help: Set MANx CISIZE to half track (26624). In your IFASMFDP JCL to dump a full MANx dataset (we use IEFU29 to issue a start for SMFDUMP JCL to dump one whenever it is full, your method may vary), use this for the input, so represents the MANx dataset name. //SMF DD DISP=SHR, //AMP='BUFND=60', // DSN= 60 buffers = 2 cylinders if you are using half track CI sizes. On the output sequential datasets on disk we use RECFM=VBS,LRECL=32760,BLKSIZE=0 and let the system choose the blocking. The last time we did this we had SMS turn on compression for these so it shows BLKSIZE=32760 but I believe if you don't use compression the system will choose half-track blocking. Keep in mind that compression uses CPU time unless you have zEDC. The important part: we use BUFNO=60 to match the input. We never tested whether output needed to be larger than input, or not. If you have huge amounts of data you might consider using striping for the output if it is going to disk; I don't know how much it will help elapsed time for the dump but it will help the elapsed time of whatever uses the data. I tested it once but we didn't implement it. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: [EXTERNAL] Re: Delete dataset catch-22
According to the message the IGGCL0FM-2 shows the DADSM Scratch status code. According to DFSMS DFP Diagnosis, this status code could mean: 2 (X'02') One of the following conditions occurred: - The data set could not be scratched because the console operator or TSO/E terminal operator did not specify the correct password in the two attempts allowed. - The user tried to scratch a VSAM data space or a catalogued VSAM data set. - The user tried to scratch the VTOC index data set. - An SMS validation failure occurred. - Verification of the last referenced date failed. The first one sounds the most likely (to me). I seem to remember there is a utility somewhere to turn off the 'password protected' but it might in fact be AMASPZAP to the DSCB. I'll have to do a deep dive to see if I can find where we did that. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Measurement discussion
I created an SHARE requirement (if you're a member you can find it at this link: https://www.share.org/p/is/vi/iid=154=12 ) about this, but I'd love to hear some discussion so I'm posting here too. My idea is that when IBM offers some new "outboard" capability such as zEDC, they need to provide SPE PTFs to add measurement of the "old way" so we can use data to decide whether to order, and of course pay for, the new feature. For example, in the zEDC case it would have been nice to have a measurement of the CPU time used for SMS compression. If you haven't voted on the requirement, please do so, but whether you do or you don't I'd like to hear opinions and ideas. A second part of the "outboard" measurement issue occurs to me and it isn't _just_ hardware related. While we have a huge amount of information in the SMF type 30 records, CPU use is still scattered among many records. For example, anyone who uses DB2 at their shop knows that the DB2 CPU time is part of the CPU time recorded in the SMF type 30 - but how _much_ of the CPU time is due to DB2? The record doesn't tell us. I think it's time for IBM to think about some sort of CPU measurement recording "architecture" - maybe a section of type 30, maybe a new SMF record - that gathers in one place the CPU measurements of all 'subsystems' used by a job/transaction/enclave where 'subsystem' can be a software piece _or_ a hardware piece. Record the CPU time for use GP, zIIP, crypto engine, zEDC engine, IDAA card, in-DB2 time, time in CICS code, time in IMS DL/1 processing, whatever. I'm no expert and I know there are going to be nuances (what if my job uses zEDC _and_ my in-DB2 time uses zEDC? for one thought) - but we have a lot of nuances now, too. Again, opinions and ideas are what I'm looking for. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Managing Archival Tape DSN for $AVRS
When I last worked on this, we used CA-1's catalog retention (RETPD=99 if I recall) for the GDGs, but because of the 255-entry GDG limit we had to do some 'merge jobs' to combine various archive "tapes" (ours were redirected to disk by Tape Mount Management) so we could have a year or two retention. Catalog retention for GDGs means that when the GDG entry "rolls off" the tape should expire during your next CA-1 cycle. I don't know what level of z/OS you are on but at z/OS 2.2 the ability to set the GDGEXTENDED parameter to YES (default is NO) in IGGCATxx means you can go to 999. What, specifically, are you trying to do with $AVRS? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: JOBGROUP in JCL
Has anyone documented the JOBGROUP interaction with existing schedulers? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: DFSORT for HTTP logs - RECFM and BLKSiZE?
These are on z/OS in Unix Systems Services, the logs for the IMWHTTPD web server. I tried without DCB but I believe some DCB must be supplied , as someone said the access method will deal with the Unix-to-record stuff. I thought my LRECL was 290, but things did not work until I increased it. It is working now. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: DFSORT for HTTP logs - RECFM and BLKSiZE?
Maybe I had LRECL too small? LRECL=512,BLKSIZE=32760 worked -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: DFSORT for HTTP logs - RECFM and BLKSiZE?
I tried this, and ended with S001-1 abend //SORTEXEC PGM=SORT //SYSOUT DD SYSOUT=* //SORTWK01 DD UNIT=DASD,SPACE=(CYL,(500,500)) //SORTIN DD PATHOPTS=ORDONLY, // PATH='/usr/lpp/internet/server_root/logs/httpd-log.Jan052018', // FILEDATA=TEXT, // RECFM=VB,LRECL=290,BLKSIZE=27998 //SORTOUT DD SYSOUT=* //REPORT DD SYSOUT=* //* //SYSINDD * OPTION VLSHRT,VLSCMP INREC PARSE=(%01=(ABSPOS=5,ENDBEFR=BLANKS,FIXLEN=15), %02=(STARTAFT=X'AD',ENDBEFR=BLANKS,FIXLEN=20), %03=(STARTAFT=X'BD407F',ENDBEFR=BLANKS,FIXLEN=6), %04=(STARTAFT=BLANKS,ENDBEFR=X'7F',FIXLEN=200), %05=(STARTAFT=BLANKS,ENDBEFR=BLANKS,FIXLEN=3)), BUILD=(1,4,%01,%02,%03,%04,%05,X'0001') SORT FIELDS=(46,200,CH,A) SUM FIELDS=(249,4,BI) OUTREC BUILD=(1,4,46,200,X,249,4,BI,M10,LENGTH=9) //* -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: DFSORT for HTTP logs - RECFM and BLKSiZE?
it appears to be 'V' but I don't see where I could get BLKSIZE from -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Release calendar format
If you need to publish calendar information for others to use, I suggest using your PC/Mac calendar tool and publishing the calendar in .ics format (icalendar, RFC 5545). In Outlook you can publish to a WebDAV server or export to a file (or share via e-mail).Put that file on your web server, and send people the URL to the file. In most calendar programs (including on phones!) you can subscribe to a 'Internet calendar' (Outlook's term). When you've subscribed, your application periodically downloads a new copy of the file. I use Outlook. We create an icalendar file using some SAS code for some things, and store it on our mainframe web server, and Outlook subscribes to it (for what you're doing, the mainframe might be optional). Outlook downloads a new copy every time it checks the mail. These files are pretty much text only so download cost is small. The Internet calendar appears as a separate calendar in Outlook, I can choose to view it as a separate tab, or in 'overlay mode' where Outlook shows me a combined view. The main advantage of doing it this way is that you don't have to format anything - you have a calendar UI right there in Outlook. You can choose to look at a week at a time, a month, or whatever your calendar UI supports. You can search the calendar too. I guess my main point is, don't do a lot of work inventing a calendar format, when the "unformatted" data can be viewed in so many ways by existing tools -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
DFSORT for HTTP logs - RECFM and BLKSiZE?
I want to do some simple reporting on my HTTP logs (httpd-log.MMMdd). I'm coding PARSE= to get the information I want, and I think I understand that part, but when I run my job SORT complains about RECFM, LRECL, or BLKSIZE. Viewing the logs, it looks like the record length is 290 (that's what ISPF view tells me) but where do I determine RECFM and BLKSIZE? How do I code the JCL for these beside the PATH= and PATHOPTS= values? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: VSAM Performance - CPU reduction
Unless things have changed, a 26K CI size is better than 32K because two CIs fit on a track. If I'm understanding this application it: 1. reads everything from the first file 2. for each record in the first file, it attempts to do a random read to find a match in the second file 3. IF it finds a match in the second file, then it might need to read one or more _other_ records in the second file - positioned by some value in the record that matched? Seems to me you want a 26K CI size and at last BUFND=60 on the first file, to maximize sequential read of that file. Going too large on BUFND might be a waste, since your processing of the records in each CI are limited by what you're doing in the second file. You definitely want a large BUFNI= for the second file because you don't want to do I/O for the key values. I'm not sure how much a large BUFND buys you - it depends how the related records (the ones you are going to retrieve depending on program logic) are stored and how many of them you need to read. As someone else said, reading too many of them will cost you. Are the related records read using some alternate key? If so you might want a third 'file' that also reads that file, with index buffers for that alternate key. One other thing - are these updated a lot? If updates are infrequent, or if performance can stand it, reduce the free space in the VSAM files, so that you're not reading in a lot of "empty" space in the CIs. These are just suggestions, of course, without know a lot about your data. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Preliminary query about mainframe "environmentals"
I think it'd be interesting. It's been a while since I had access (now retired / consulting) but the HMC allowed you to pull up something (a spreadsheet if I recall) of kWH by some interval, so you could (probably only on z13 and above) have some real data. I don't know if they ever externalized it to SMF - I thought I submitted a requirement for that at one time, but maybe I just drafted it and forgot to add it. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Restful Call
z/OS 2.2 has a set of callable routines that provide HTTP client capabilities. I saw a presentation at SHARE San Antonio. I think it is part of the 'web enablement toolkit' - see this link https://www.ibm.com/support/knowledgecenter/en/SSLTBW_2.3.0/com.ibm.zos.v2r3.ieac100/ieac1-cwe-http.htm -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
SCRT and ISVs?
Forgive me if I posted this before... If so, life intervened and I don't remember. Is there a list documenting which ISVs participate in Subcapacity Reporting and which will supply an ISVLIB file for such use? I know CA Common Services will, if you set the option, produce type 89 records but I haven't yet found where they produce ISVLIB (but I'm searching today, just adding this here for additional information). -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CA1 to RMM conversion
I can vouch for 'modern CA1' - I went through merging tape databases for several LPARs into one shared tape database, then after that was able to do maintenance taks without outage. re: BSAM vs. VSAM vs. whatever.It would be nice if products which used "databases" used a standard API (such as SQL) and we could use whatever database we had already licensed for the underlying storage of the information. I mention SQL because it is extremely common, ODBC would be another choice - as long as it is _standard_ . Tools would then not have to include 'database management' (although I realize that's a 'sunk cost for CA1) - saving some development cost for vendors and customers could manage the databases with familiar tools. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Sort Question
I know I am late to this, but I see no BLOCK CONTAINS 0 RECORDS in the COBOL. I'm not totally current on COBOL releases, but A) is that still required to use block size from the JCL B) is it even relevant in this instance - COBOL will use the block size value of the dataset that's input, yes? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: TRTCH=NOCOMP and SMS Compression
I think I want to set up a data class without compression, and assign certain data sets to it. I was _hoping_ that I could override the SMS compression via TRTCH=NOCOMP so that I could show people the difference in CPU time with and without SMS compression. I will go ahead and create the non-compression data class and an ACS routine to assign "test" datasets to it. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: EREP Symptom and/or Software Records
It's been a few years (retired / consulting now) since I had the duty of looking at EREP records but as with what many others have said, it's about what seems "normal" in your environment, although I've had instances of 0C4 that was occuring regularly, though handled by the piece of software in question without killing anything. As for searching for APARs, you used to be able to copy those symptom strings into the search field such as: PIDS/5695DF105 RIDS/IGG0CLA9 RIDS/IGG0CLX0#L PRCS/00F6 PRCS/ and it was a pretty reliable way to zero in on something. I don't know if IBM's support site search still works that way, but if it does it saves quite a lot of time. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
TRTCH=NOCOMP and SMS Compression
We have a job which creates "tape" files in TMM (Tape Mount Management) via SMS redirection. The data class assigned to the output dataset has SMS tailored compression turned on. I would like to see how much that compression is costing in CPU time by running the job without compression. Will TRTCH=NOCOMP override the data class setting, or do I need to create a special data class with no compression just for this test? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: IEFU8* exit points and RMF type 70
Bringing this back up again because I still haven't found the answer yet. Are SMF type 70s written by RMF via branch-entry, requiring me to use IEFU85 exit point to examine them, or can I use just IEFU83 or IEFU84 ? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: IPLing z/OS 2.1+ with no NIP console
Do these chpids connect to any fiber switches not controlled by you? I got called very early one morning due to a bunch of channel errors. At the data center where I worked, fiber switches were not owned by our group. The brainiacs who did own responsibility for the switch decided to take half of one down for some sort of change / maintenance and didn't bother to tell anyone who had equipment connected to it. Needless to say, no mainframer after that wanted to configure tape or DASD through those switches. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Syncsort step running too long
Did you determine what the problem was? My input would have been to compare Syncsort default parameters in case someone forgot to migrate them to the new release. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Real Userid when specifying USER=
SMF type 26 (job purge) has the submitting user ID - but _not_ task identification (job name/job id/reader time). We have for some reason jobs submitted by the scheduler which run IEBEDIT to "submit" another job and that job in turn "submits" another - all with user ID propagation. They are all submitted under the ID in effect when the scheduler submitted the first job - which means we can't easily tell which job did the submission. I intended to put a SHARE requirement or RFE in on it, but I don't remember if I did it. I'll check that soon. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: AW: Re: RFE to add new sizes to AVGREC and IDCAMS Define
I see and understand some people's desire to directly allocate in KB, MB, GB (or, coming soon to an array near you, probably TB, and PB). What I like about AVGREC is the ability to allocate in _records_, eliminating tedious space calculations when I have a good estimate of the size of a dataset. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
IEFU8* exit points and RMF type 70
Someone wants a WTO that shows SMF70LAC (the 4-hour rolling average of MSUs value) every time the RMF type 70 is written. IEFU8* is the best exit point to do that, as near as I can tell, however what I don't know is whether type 70s are written the "normal" way, via branch-entry, or via cross-memory, require the record be examined at the respective 83, 84, or 85 exit point. At this point, I'd probably write the exit and name it as one of the modules at all of the exit points - as long as I do my WTO in a reentrant manner I should be OK, but I'd still kind of like to know the answer. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: OS ENQ from COBOL
IF the batch job is scheduled, and IF you have an automated operations tool, and IF your scheduler has an external command interface, THEN: What we (in Pecan/Beta42/ASG Workload scheduler) was define a variable (example $CICS1), and make a predecessor for the batch job with the value $CICS1 = UP. In the automated operations too, we had two message traps: one for the HASP message indicating CICS1 started, which would set the variable to UP, and one for the HASP message indicating CICS1 ended, setting the variable to DOWN. The scheduler would then only start the job when the variable was set to UP. I believe you can do a similar thing with scheduling environment (SCHENV= on the job card), too. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
DFSORT SQZ question
I've created an FTP client report in comma-delimited format from SMF 118s (yes, I know they should be capturing 119s but they aren't yet, and this isn't about that). To make the IP address readable, I used an edit on each byte of the address and the SQZ function to put them back together. However, to do this I had to do the edit on an INREC statement, and then treat the four edited fields as one on an OUTFIL statement. Here are the control statements, sorry for the length but it's more readable this way: === OPTION VLSHRT,VLSCMP INCLUDE COND=(6,1,BI,EQ,118, AND,23,2,BI,EQ,3) INREC FIELDS=(1,24, 37,1,BI,M10,LENGTH=4, 38,1,BI,M10,LENGTH=4, 39,1,BI,M10,LENGTH=4, 40,1,BI,M10,LENGTH=4, 189,8, 25,4, 49,8, X'0001') SORT FIELDS=(25,16,CH,A,41,8,CH,A,49,4,CH,A,53,8,CH,A) SUM FIELDS=(61,4,BI) OUTFIL FNAMES=REPORT, REMOVECC, HEADER1=(C'IP,Local ID,Command,Remote ID,Count'), BUILD=(1,4, 25,16,SQZ=(SHIFT=LEFT,MID=C'.'), C',', 41,8, C',', 49,4, C',', 53,8, C',', 61,4,BI,M10,LENGTH=6) = Given that this could have as input millions of records depending upon how long you want to summarize, I'd like to be able to use the 4-byte IP address in the sort, and _then_ make it readable. Is there a way for SQZ to work on a group of fields? I'm envisioning something like this where the grouping of fields in parenthesis before the SQZ function means sort does those actions, then applies SQZ to the entire result... === OPTION VLSHRT,VLSCMP INCLUDE COND=(6,1,BI,EQ,118, AND,23,2,BI,EQ,3) INREC FIELDS=(1,24, 37,4, 189,8, 25,4, 49,8, X'0001') SORT FIELDS=(25,4,CH,A,29,8,CH,A,37,4,CH,A,41,8,CH,A) SUM FIELDS=(49,4,BI) OUTFIL FNAMES=REPORT, REMOVECC, HEADER1=(C'IP,Local ID,Command,Remote ID,Count'), BUILD=(1,4, (25,1,BI,M10,LENGTH=4,26,1,BI,M10,LENGTH=4,27,1,BI,M10,LENGTH=4,28,1,BI,M10,LENGTH=4),SQZ=(SHIFT=LEFT,MID=C'.'), C',', 29,8, C',', 37,4, C',', 41,8, C',', 49,4,BI,M10,LENGTH=6) = Seems like it would be a performance improvement for this use case anyway... -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Real purpose of hipersocket
I would like to add one other advantage that I took advantage of. When you define your hipersocket network, you can define a very large MTU size for the hipersocket network. This can eliminate a lot of segmentation/re-assembly overhead because the data transmission is broken up into fewer large pieces. This really can help elapsed time for transmission of large things like FTP of large files, NJE transmission of large reports, or perhaps DRDA between database systems, and I never noticed any downside for smaller things. As mentioned previously, both networks can co-exist, and you can define the routing so that hipersockets are preferred. I don't have my definitions handy but we used VIPAs and so we could reach the zOS image via either of two IP addresses. We used 192.168.x.x addresses for the internal (hipersocket) network; I recommend making the last octet or the last two octets the same as those of the OSA IP address, but it's not required. You can test the difference in speed - find a large dataset (or create one), and FTP it using the OSA network and the other system's non-hipersocket IP address, then FTP the same file using the hipersocket IP address. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Resource Group Limits as a cost containment method?
There's a system where their 4-hour rolling average maximum is always during their batch window, when they routinely reach their defined capacity limit, which in turn affects what they pay in a chargeback system. Because of scheduling dumbness (another story) they have several gaps in their batch window where nothing runs, giving some leeway for jobs to increase their elapsed time - not the usual thing you want to do, but bear with us. Would it be worth investigating setting a resource group limit for the batch service class(es) to hold the total 4HRA down as a cost-saving measure, or are resource group limits going to cause more trouble than they are worth? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: RFE for new SMPP support in zOS
Thanks for the info... Using a RESTFUL interface is probably the way to go for the use cases I had in mined - primarily having an automated operations tool send notifications to support personnel. We used to have a way to do it for pagers but never got one built when cell phones became the standard issue for on call people. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
RFE for new SMPP support in zOS
After discussion at SHARE, I created an RFE (back in March... sorry for the late notice, but life happened and I forgot about it) to add support for the SMPP (Short Message Peer-to-Peer) protocol, for outbound messages. This is the protocol used to communicate the the servers that support the short message service (SMS) also known as text messaging. As stated in the RFE this would allow applications to send notifications to mobile devices via text message. This would use the same port that they're using for the HTTP support. If this sounds like something you'd be interested in you can vote for the RFE at http://www.ibm.com/developerworks/rfe/execute?use_case=viewRfe_ID=85091 -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
