Re: Are there tasks that don't play by WLM's rules
>Well, zIIPs are cheaper than CPUs, but in this case maybe CuOD is the >best choice. If you pays more, then probably under powered anyway... How would zIIPs help here? -- Peter Hunkeler Credit Suisse -- 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: Are there tasks that don't play by WLM's rules
>Well, that's what I am attempting to do. Did I mention that the >vendor is IBM? And that the product is developed in the same >location as WLM? In Germany, no less? Well, unfortunately, there are still too many people out there in software development that don't have a clue what z/OS UNIX really means. Porting software from one OS to another implies a certain level of knowledge on both OSs, IMHO. When porting to z/OS UNIX, there is yet another key player that needs to be factored in: MVS! Most ports probably are more or less easy to do. However, there are techniques that work perfectly in a "pure" UNIX but will fail miserably or become CPU hogs on z/OS UNIX because MVS was not designed to work that way. When porting big applications or servers to z/OS UNIX the developers must be willing to change the design here and there to cope with z/OS UNIX and MVS behaviour. This makes the difference between a successful port and a not so successful one. I'm *not* meaning to give offence to anybody! Since there *is* z/OS UNIX which *is* XPG compliant, there seems to be no need to know about the MVS part, so people don't get educated on MVS and, worse, are not even told to get advice from MVS people. -- Peter Hunkeler Credit Suisse -- 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: Are there tasks that don't play by WLM's rules
Thanks to all who responded. I find it really interesting that just about everyone agrees to put these things into sysstc, an idea that got vehemently vetoed by the 'head WLM developer' when I suggested it in a lengthy conference call. Thanks to Peter Hunkeler for the good stuff about Unix. I will take me some time to absorb that! :-) >In Barbara's case, I would like the product has to be re-engineered to >signal completion of its transactions so that she could use response time >goals. Exactly. That's what I would like, too. >Then get your boss to go scream at the vendor to put proper >instrumentation into their product. There's not a lot else you can do. Well, that's what I am attempting to do. Did I mention that the vendor is IBM? And that the product is developed in the same location as WLM? In Germany, no less? Unfortunately, my boss adopts the attitude not to do anything until there is a problem. And then we get yelled at because we did not take 'pro-active' measures (to use a buzzword). So I am yelling right now (at IBM, via complaint). >(Did I mention velocity was a bad idea?) Even the developer said that, of course not as bluntly as you! :-) Best regards, Barbara -- Der GMX SmartSurfer hilft bis zu 70% Ihrer Onlinekosten zu sparen! Ideal für Modem und ISDN: http://www.gmx.net/de/go/smartsurfer -- 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 read source from a PDS member
John/OP: You might have overlooked an interesting gem of a feature of z/OS JCL: FREE=CLOSE. If you need to provide several alternative tables, you might be able to do so strictly in the enveloping JCL by doing something like this: //SYSLIB DD DISP=SHR,FREE=CLOSE,DSN=USERS.INPUT(TABLEX) //SYSLIB DD DISP=SHR,FREE=CLOSE,DSN=USERS.INPUT(TABLEY) //SYSLIB DD DISP=SHR,FREE=CLOSE,DSN=USERT.MUMBLE(TABLEZ) Note that the DDNAME is replicated three (3) times above, for example. For the cost of a full OPEN/GET/CLOSE you get to read the record from the first member and be positioned to the second member on your next OPEN/GET/CLOSE. Not high performance but pretty high function and very easy to maintain (just change the JCL). z/OS does not limit DDNAMES to a single appearance in a step (although you'd be surprised how many think otherwise). By crafty use of multiple DDs with identical names that resolve to different datasets (even to completely different data set names). Since any DD can use traditional data sets or PATH= it is also quite general. Nearly every high level language supports file opens and closes so nearly every high level language can perform this "trick". All you need to do is stop looping through the open/close once the open fails. (You ought to provide for error recovery if the open fails anyway so that might not be asking too much.) -- Tom Schmidt Madison, WI -- 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 read source from a PDS member
On Thu, 18 Oct 2007 18:24:12 -0500, Paul Gilmartin wrote: >On Thu, 18 Oct 2007 11:24:19 -0500, Tom Schmidt wrote: >> >>z/OS has the venerable OPEN-J service that might be what you are looking for >>here. Check the z/OS publications for the OPEN macro's TYPE=J operand. >> >Those fixated on performance will object to performing an OPEN >for each member rather than a single OPEN and multiple FINDs. OP did not state any performance requirements; he was interested only in alternatives to DYNALLOC. I provided him with one. Furthermore, OPEN-J has a different path length than traditional OPEN. Run a few benchmarks. I agree with you & others that BPAM is yet another possible alternative for OP to consider. Different strokes for different folks. -- Tom Schmidt Madison, WI -- 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
JRNAD and compressed data in extended format mode
Does anyone know how the data is presented to the JRNAD exit when the VSAM file has been defined with SMS with COMPACTION=Y (compression is on) ? Is the record image compressed or decompressed? Thanks, Steve -- 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 read source from a PDS member
Paul Gilmartin wrote: On Thu, 18 Oct 2007 17:27:02 +0200, Binyamin Dissen wrote: BPAM is not THAT unfriendly. If it truly were "not THAT unfriendly" you'd not have troubled to write a wrapper to hide the hostility from the programmer; nor would numerous other inventors who independently replicated your effort. I don't want to start a mud slinging contest, but the reason I use macros and subroutines is to isolate my code from system changes, and to free time to do useful work, rather than repeat the same drudge work over and over. Thus over the decades, some BPAM code has been replaced by DESERV, some LOCATE code by the catalog interface, and OBTAIN SEEK/SEARCH by CVAF Functions, etc. Friendliness of a function becomes irrelevant when you work with it only once every few years. Gerhard Postpischil Bradford, VT new e-mail address: gerhardp (at) charter (dot) net -- 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 read source from a PDS member
On Thu, 18 Oct 2007 17:27:02 +0200, Binyamin Dissen wrote: > >BPAM is not THAT unfriendly. > If it truly were "not THAT unfriendly" you'd not have troubled to write a wrapper to hide the hostility from the programmer; nor would numerous other inventors who independently replicated your effort. >A while back (probably before many lurkers on this list were born, but I >digress) I wrote a set of subroutines called QPAM to hide the tricky stuff >from the programmer (don't know if I can give it out). > >QPAM GET - get next record >QPAM PUT - put a record >QPAM TRUNC - force a write of all queued blocks >QPAM RLSE - stop reading current member > >You still hard coded the FIND and STOW, but they are rather simple. -- gil -- 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 read source from a PDS member
On Thu, 18 Oct 2007 11:24:19 -0500, Tom Schmidt wrote: > >z/OS has the venerable OPEN-J service that might be what you are looking for >here. Check the z/OS publications for the OPEN macro's TYPE=J operand. > Those fixated on performance will object to performing an OPEN for each member rather than a single OPEN and multiple FINDs. >If you are looking from the perspective of a higher-level language such as C >(although I might otherwise argue that point) then you can use BPXDYN >service to switch both member name and dataset name. > Ah, I forgot; as you note BPXWDYN also has a CALL interface (and in one convoluted case I encountered TSO's 100-character PARM limit; I resorted to several ALLOCATEs and a BPXWDYN CONCAT). But it isn't suited to the task of accessing various members from a given DDNAME (Suppose the DDNAME is a mixed concatenation of PDS[E]s and HFS directories). -- gil -- 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: Auxiliary Storage Usage - History Data?
On Oct 18, 2007, at 3:15 PM, Patrick Lyon wrote: Thanks to all who have responded. We went through this all again today and DB2 crashed. After restarting it, it doesn't take up near the storage as just before it went down. After doing some research, I found some storage related PTFs that were not on describing not freeing storage. I will apply these ASAP. But back to my previous question, is there an RMF report that I can use via batch to see history on how many AUX frames were allocated to what address space via the REPORTS or SYSREPORTS statement, or via some other report feature that I am not familiar with? I need to go back weeks of kept SMF data, not just hours of real time online data. Thanks again for all who have responded. Best Regards, Patrick Lyon Patrick, There is an RMF report (you feed SMF recs for the time period you want reported) that gives you numbers on paging slots used/available. Its been 10 years since I ran it but I am sure thats its still there. The last time I ran it I went with several months worth of data to prove that we didn't need more page spaces. Ed -- 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
Enterprise PL/I issue on z/OS 1.8
I don't know if anyone else is affected by this but I thought that I would post a "head's up" just in case. We have encountered an issue with existing Enterprise PL/I programs when going to z/OS 1.8 from z/OS 1.6. PTF PK17116 seems to be incorporated in the LE base and it requires a recompile with compiler PTF PK17575 installed for any previously-compiled Enterprise PL/I programs if a FINISH ON unit in the source code is intended to be entered during termination. Otherwise, the FINISH ON UNIT may not be entered. We didn't have either PTF installed in our z/OS 1.6 environment. -- 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: Auxiliary Storage Usage - History Data?
Thanks to all who have responded. We went through this all again today and DB2 crashed. After restarting it, it doesn't take up near the storage as just before it went down. After doing some research, I found some storage related PTFs that were not on describing not freeing storage. I will apply these ASAP. But back to my previous question, is there an RMF report that I can use via batch to see history on how many AUX frames were allocated to what address space via the REPORTS or SYSREPORTS statement, or via some other report feature that I am not familiar with? I need to go back weeks of kept SMF data, not just hours of real time online data. Thanks again for all who have responded. Best Regards, Patrick Lyon -- 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 read source from a PDS member
www.cbttape.org there is a subroutine that is callable from Cobol that will let you read a member of a PDS -- 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: More SSL/TLS and FTP woes
Well known port 990 was removed from the RFC by the IETF. However, once an application has been accepted by IANA to assign a well known port to the list, it is not removed just because IETF made changes to the RFC. I have had to fight this battle more than once since I have to deal with the army. They are very strict when it comes to using IANA well known ports. Someone needs to get these two organizations together so eveyone is on the same sheet of music. -- 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: Healthcheck CHECK(IBMCS,CSVTAM_CSM_STG_LIMIT)
For the archive I was told there will be an APAR against VTAM. No more details so far >Ray, >thanks for the vrey helpful explanation. I try to build a list >instead of answering in general. > >1. Our IVTPRM contains (IBM defaults) >FIXED MAX(120MB) >ECSA MAX(120MB) >I now know why IVT5539I stay the MAX ECSA is 92638K but >IVT5538I shows the correct value. You're right the >ECSA in IEASYSxx for this sandbox is 100MB. This match the >90% you mentioned. However the healthcheck or the response should take care >about this. I also seen no IVT msgs which indicates the >ECSA in our IVTPRMxx setting doesn't fit for our ESCA setting >in the IEASYSxx. So the ECSA in IEASYSxx should be also an indicator >for this check or time for new check "Ignoring the IVTPRMxx setting" >because of low ECSA in IEASYSxx. > >2. System Programm Response still stay the default values in IVTPRM00 are >100MB for both FIXED and ECSA. >This is no longer true since z/OS R8 > >3. If 120MB is recommended (default) by IBM why does the check still checks >100MB? > >4. I still wonder about the DISPLAY CSM command in the response even >such a command doesn't exists. It's called DISPLAY NET,CSM but it also >appears in the VTAM books that way. Confusing > >5. Does a HWM really makes sense for this check? Shouldn't it be better to use >the current HWM for ECSA or FIXED since last IPL values as given in msgs IVT5594I? > >6. If I monitor the system for a week and decide a value of 50MB would >be fine for our environment will this drive the healthcheck exception every >time as the parm is MAXECSA(100M),MAXIFX(100M)? > >Roland -- 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
Hardware slowdown sours IBM numbers
MIPS sales down 21% year-on-year is indeed a "difficult comparison". Nine months is a long time to run at standard expense levels until the cash starts to flow again. I don't share the view that the turndown is due to people waiting for new processors. The rumour mill has not been that active. Some is undoubtedly due to economic circumstances, and some due to the fact that just about every e-commerce supplier you find on the web is promoting another platform. As a measure, I was amazed to find someone I was helping with a web site was selling a Series i solution - first IBM VAR I've bumped into at random for a year or more. More and more mid-sized and even large companies are turning to Open Source. http://www.oscommerce.com and http://www.joomla.org being prime examples. Not only do they work - they're free, which means no budgetary barriers to growth. There are also huge skill pools out there. It reminds me very much of the early days of OS/360, HASP and ASP - when everyone had the source code and a huge community debugged and developed for free. APARs were often sent in with suggested (and tested) code changes attached. Various IBM field types were commenting as early as April that they simply couldn't see sales opportunities. To quote one: "My customers are all full or gone - I have nowhere left to place machines." z6 is going to be more fun than anyone wants - whilst I believe it will meet its general performance targets, I suspect there will be great variation between workload types - possibly greater than we've seen before. We might see a much greater demand for capacity planning and tuning skills than we've seen for many years. "If this machine is 50% faster than the z9, why does my application run slower?" Isolated cases, certainly, but we all know how much effort has to go into dealing with each one before the user is placated. One unhappy user gets more management time than 99 happy ones. As always, my attention is drawn to smaller offerings. It would be nice to think some of the fraying at the edges could be stopped. There is a significant downside for Software Division rolling forward into a brick wall over the next couple of years - users paying software charges on Flex-ES systems who could never dream of affording a real box. I believe the potential among current Flex-ES users is around 15 to 20 "real" boxes. That's a LOT of software revenue to lose for such a small return. It really is quite incredible. I've heard of executives regarding 5% installed base shrinkage per year as "acceptable". I wonder what Thomas J. Watson would have said if anyone had said that to him. IMO IBM works too much to its own internal goals and not enough to industry benchmark goals. Many expect the Baby z6 to have much finer granularity with an entry level system having much lower product costs than a z9 BC. But the dear old HP41 says that by the end of 2008 it won't actually help. -- Phil Payne http://www.isham-research.co.uk +44 7833 654 800 -- 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: UCB Wikipedia Article
In a message dated 10/18/2007 2:02:16 P.M. Central Daylight Time, [EMAIL PROTECTED] writes: >The captured UCB isn't a copy of the UCB - it is simply another virtual view of the UCB created via the IARVSERV page level sharing services. Thanks for clarifying that. I haven't yet used or studied IARVSERV, so I didn't make the connection. Bill Fairchild Plainfield, IL ** See what's new at http://www.aol.com -- 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: Are there tasks that don't play by WLM's rules
In a message dated 10/18/2007 12:40:25 P.M. Central Daylight Time, [EMAIL PROTECTED] writes: Then get your boss to go scream at the vendor to put proper instrumentation into their product. There's not a lot else you can do >> Well, zIIPs are cheaper than CPUs, but in this case maybe CuOD is the best choice. If you pays more, then probably under powered anyway... ** See what's new at http://www.aol.com -- 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: Zapping HFS files
> -Original Message- > From: IBM Mainframe Discussion List > [mailto:[EMAIL PROTECTED] On Behalf Of Tony Harminc > Sent: Thursday, October 18, 2007 12:00 PM > To: IBM-MAIN@BAMA.UA.EDU > Subject: Re: Zapping HFS files > > > On Thu, 18 Oct 2007 07:48:01 -0500, McKown, John > <[EMAIL PROTECTED]> wrote: > > >IIRC, the original question was not necessarily about zapping a UNIX > >resident program object. So, how do I zap a binary file, residing in > >UNIX, which is not a program object? > > I doubt there is any way, beyond writing a simple program of your own. > AMASPZAP historically used EXCP and its own channel programs > to update a > record in place in any kind of disk dataset. It has a > load-module-friendly > user interface, but of course can also be used to ZAP > anything on disk. When > AMASPZAP is used to ZAP a program object in either a PDSE or > a UNIX file, it > uses the Binder API to do the work, and the update is not > done in place. > (The first time you see an IEW message from running AMASPZAP > can be a bit of > a surprise...) > > In other words, AMASPZAP has become just a user interface to > the Binder API > for ZAPping program objects, but I don't think IBM has also > made it into a > user interface for updating UNIX files generally. > > Tony H. I asked over on the Linux-390 forum about this. GNU patch has a --binary option for changing binary (non text) files. Unfortunately, IBM once again "lags behind" GNU in functionality. -- John McKown Senior Systems Programmer HealthMarkets Keeping the Promise of Affordable Coverage Administrative Services Group Information Technology The information contained in this e-mail message may be privileged and/or confidential. It is for intended addressee(s) only. If you are not the intended recipient, you are hereby notified that any disclosure, reproduction, distribution or other use of this communication is strictly prohibited and could, in certain circumstances, be a criminal offense. If you have received this e-mail in error, please notify the sender by reply and delete this message without copying or disclosing it. -- 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: Are there tasks that don't play by WLM's rules
>Date:Wed, 17 Oct 2007 10:46:45 +0200 >From:"Vernooy, C.P. - SPLXM" <[EMAIL PROTECTED]> >Subject: Re: Are there tasks that don't play by WLM's rules >The problem here is, that WLM is not capable of managing its children >when they are a little more exotic than the usual kids in class. or at >least, it is not capable of reporting meaningfully about them. E.g. if I >organize 100 TSO users to repeatedly enter a transaction at the same >time, I bet WLM will fully panic and report PI's in the 100's, although >performance might be good. Not so. If the 100 TSO users have response time goals, then there is a good chance that many, if not all of them will achieve their goal. The issue is with velocity goals. John Arwe wrote a paper many years ago that is still true. http://www-03.ibm.com/servers/eserver/zseries/zos/wlm/documents/velocity/velocity.html If the MPL of a given service class period can change significantly during the day, then the velocity achieved can also vary significantly, and a velocity goal is not very effective. In the PeopleSoft case I referenced earlier, the PI with a velocity goal for DDF ranged from .1 to 10.0 across 24 hours, even though the response time was always acceptable. The fix there is to use response time goals, as Martin also said. In Barbara's case, I would like the product has to be re-engineered to signal completion of its transactions so that she could use response time goals. As this is unlikely, I would suggest running the product in SYSSTC. I don't like making that recommendation, as it makes me feel like a Unix weenie, but it would at least ensure the transactions were given the best service possible and avoid any possible finger pointing from users. regards, Tom Tom Russell "Stay calm. Be brave. Wait for the signs." -- Jasper FriendlyBear "... and remember to leave good news alone." -- Gracie HeavyHand -- 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: UCB Wikipedia Article
IBM Mainframe Discussion List wrote on 10/18/2007 07:32:53 AM: > I have never delved into I/O prevention, so I don't know what it does, how, > or why, but I believe it is used in preventing an I/O's being executed rather > than positioning the request in the queue. I/O Prevention was introduced in SP2.1.6 in support of HARP a.k.a XRF (it was a hot standby thing). It was used for stopping I/O on the dying system which was being switched to the hot standby. It could be considered to be a software ancestor to what we now do via Fencing (a.k.a Channel Subsystem Iosolation) in the machine when XCF/SFM partitions a system out of a sysplex. I think I/O prevention is currently used to implement the IOACTION STOP command. > The purpose of the UCB Capture service is to allow anciently compiled > programs using ancient access methods that require the DCB, DEB, > IOB, etc., to be > in storage below the ancient 16MB line to do I/O with a thoroughly modern UCB > that is above the line. The big problem was that the DEB has room only for a > 3-byte address of the UCB. Rather than redesign the ancient access methods > that were based on EXCP, IBM provided a way for old access methods to get to > new UCBs above the line with the system service IOSCAPU. IOSCAPU creates a > copy (called "capturing" the UCB) of an above-the-line UCB in the user's > private storage and puts that copy below the line. The various > ancient control > blocks that need to point to a UCB will point to the private > storage copy with > a 24-bit address. The captured UCB isn't a copy of the UCB - it is simply another virtual view of the UCB created via the IARVSERV page level sharing services. > Leading edge access methods are continually updated by IBM > to support every new wrinkle that the IOS developers apply to UCBs in order > to support system growth. Many of the original constraints in the software > and hardware architecture of S/360 have been relaxed, with occasional seismic > jolts to the appearance of the UCB; e.g., UCBs can now be above the 16MB > line, devices can have five hexadecimal digits in the device number(formerly > known as "device address"), there can be more than 65536 devices onone LPAR, > UCBs can be dynamically created in order to support dynamically installed new > I/O hardware, and more than one I/O can be simultaneously active > with the same > device. Technically, a device number is still only 4 hex digits, but the same device number can exist in more than one subchannel set. For convenience in things like GTF and System Trace, we concatenate the subchannel set id to the device number, and sometimes loosely think of that as a 5 digit device number. Jim Mulder z/OS System Test IBM Corp. Poughkeepsie, NY -- 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: Zapping HFS files
On Thu, 18 Oct 2007 07:48:01 -0500, McKown, John <[EMAIL PROTECTED]> wrote: >IIRC, the original question was not necessarily about zapping a UNIX >resident program object. So, how do I zap a binary file, residing in >UNIX, which is not a program object? I doubt there is any way, beyond writing a simple program of your own. AMASPZAP historically used EXCP and its own channel programs to update a record in place in any kind of disk dataset. It has a load-module-friendly user interface, but of course can also be used to ZAP anything on disk. When AMASPZAP is used to ZAP a program object in either a PDSE or a UNIX file, it uses the Binder API to do the work, and the update is not done in place. (The first time you see an IEW message from running AMASPZAP can be a bit of a surprise...) In other words, AMASPZAP has become just a user interface to the Binder API for ZAPping program objects, but I don't think IBM has also made it into a user interface for updating UNIX files generally. Tony H. -- 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: TPX and ACLE Code (RESOLVED)
Well, the problem was identified with Super Users in TSS. They get an additional message I was not accounting for. So the new ACLE for TPX looks like this USERID SEARCH' DS600306:' BRANCHEQ,DISPATCH STOP DISPATCH MSG ' LOOKING FOR DISPATCH SIGNON FIELD' * CLEAR KEY'LOGON' ENTER ... SEND KEY'&USERID' ENTER ... SEND KEY'&PSWD' ENTER ... SEND SEARCH 'VRDMI000'<=== NEW BRANCH EQ,DONE <=== NEW ENTER ... SEND KEY'CA' ENTER ... SEND * * FINISHED WITH THIS LOGON * DONE STOP Lizette > >Hi Liz, >We have never had Dispatch defined in our ACT. We only have a profile >definition for Dispatch to supply the ACB name. Each user is granted access to >the profile via ACF2 and most all users are not allowed to to use a "/A" >command. > >Our ACL is very simple and looks a bit like yours. > >* OPTIONFLOW,ON *** DEBUG ** > OPTIONMAXI,2000 >* OPTIONTERM,ON *** DEBUG ** >* OPTIONRESIDENT,ON >* > WAIT 2 WAIT 2 SECONDS >LOOK1 SEARCH'TASK CODE' SEARCH FOR THE PROMPT > KEY 'HI &USERID'ENTER LOGON COMMAND > ENTER > WAIT 1 WAIT 1 SECOND > SEARCH'PASSWORD:' SEARCH FOR PSWD PROMPT > BRANCHNE,DONE > KEY '&PSWD' ENTER PASSWORD > ENTER >* >DONE STOP WE'RE LOGGED ON NOW >* > >Is this problem happening to a user that successfully gets in sometimes but >not always? > >Could this be an issue with the pool of virtuals or the logmaode that they are >picking up or possibly the term type on their TN3270 session (assuming they >comming in via TN3270)? > -- 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: More SSL/TLS and FTP woes
>>> On Thu, Oct 18, 2007 at 9:21 AM, in message <[EMAIL PROTECTED]>, "Chase, John" <[EMAIL PROTECTED]> wrote: -snip- >> Can one ftpd >> process listen on both 21 and 990 at the same time? > > I had "ass.u.me"-d so, having become accustomed to most "mainframe" > software having the ability to do many things concurrently. Bit I > notice in the ftpd.log that each individual FTP request runs with a > different PID than the server itself, so maybe the FTP software actually > has only one "ear" with which to listen for connection requests. Not having had access to USS for some time now, I can only ask if there is the equivalent of the "netstat -lnp" command in Linux to show you what processes are listening on what ports. If there is, that will tell you what's what. Mark Post -- 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: TPX and ACLE Code
Hi Liz, Our CA-Dispatch is not defined in the ACT. Never has been, and has been working like this for over 16 years. Our users are not permitted to use the "/A" command. ACF2 provides the users with access to the PROFILE for CA- Dispatch. Our ACL is similar to yours; * OPTIONFLOW,ON *** DEBUG ** OPTIONMAXI,2000 * OPTIONTERM,ON *** DEBUG ** * OPTIONRESIDENT,ON * WAIT 2 WAIT 2 SECONDS LOOK1 SEARCH'TASK CODE' SEARCH FOR THE PROMPT KEY 'HI &USERID'ENTER LOGON COMMAND ENTER WAIT 1 WAIT 1 SECOND SEARCH'PASSWORD:' SEARCH FOR PSWD PROMPT BRANCHNE,DONE KEY '&PSWD' ENTER PASSWORD ENTER * DONE STOP WE'RE LOGGED ON NOW * Could this issue be a problem with the pool of virtual terms you are using or maybe related to the termtype that the user is selecting on a TN3270 session. Is this problem hapening to a user that can get on sometimes but not every time? -- 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: Are there tasks that don't play by WLM's rules
Peter Hunkeler writes some good stuff about z/OS Unix; > PMFJI, but the last few remarks about being "ported from open world" > made me think. So here are a few thoughts related to z/OS UNIX: > Exactly. Execution velocity as a goal mechanism has been a really bad idea since the very beginning, but it would be tough to get that horse back in the barn! WLM is just a B-average accountant with a minor in statistics. It has to see a sufficient number of samples in order to even guess at what's going on and unless the application is properly instrumented to signal its own internal transaction begin/end/delay, all you can get from WLM is exactly that: a guess. For transactional applications you need a scalpel and execution velocity is just a blunt object. What you have is a somewhat volatile transactional application masquerading as a STC and _NOT_ using the reporting interfaces that would allow WLM to make any sense of what is going on. Unfortunately for such a situation there is no such thing as a meaningful execution velocity that WLM _COULD_ "manage" this type of work the way you want. For that reason, I agree with Mark Z, Shane and others; plunk that bad-boy in SYSSTC. It won't be explicitly "managed" but it will get serviced when it needs to and that's really what you're trying to accomplish. Then get your boss to go scream at the vendor to put proper instrumentation into their product. There's not a lot else you can do. CC (Did I mention velocity was a bad idea?) -- 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 read source from a PDS member
>On Thu, 18 Oct 2007 09:43:51 -0400, bit.listserv.ibm-main wrote: > >I need to read source code from a PDS member whose name I don't know >until my program runs, much the same as the COBOL >compiler handles "COPY memname" statements. Is dynalloc the only way to >go about it or does zOS provide a nice friendly >interface that lets me ask for source member data a line at a time? >Something comparable to VSE's LIBRM would be ideal. z/OS has the venerable OPEN-J service that might be what you are looking for here. Check the z/OS publications for the OPEN macro's TYPE=J operand. If you are looking from the perspective of a higher-level language such as C (although I might otherwise argue that point) then you can use BPXDYN service to switch both member name and dataset name. -- Tom Schmidt Madison, WI -- 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: TPX and ACLE Code
Hi Liz, We have never had Dispatch defined in our ACT. We only have a profile definition for Dispatch to supply the ACB name. Each user is granted access to the profile via ACF2 and most all users are not allowed to to use a "/A" command. Our ACL is very simple and looks a bit like yours. * OPTIONFLOW,ON *** DEBUG ** OPTIONMAXI,2000 * OPTIONTERM,ON *** DEBUG ** * OPTIONRESIDENT,ON * WAIT 2 WAIT 2 SECONDS LOOK1 SEARCH'TASK CODE' SEARCH FOR THE PROMPT KEY 'HI &USERID'ENTER LOGON COMMAND ENTER WAIT 1 WAIT 1 SECOND SEARCH'PASSWORD:' SEARCH FOR PSWD PROMPT BRANCHNE,DONE KEY '&PSWD' ENTER PASSWORD ENTER * DONE STOP WE'RE LOGGED ON NOW * Is this problem happening to a user that successfully gets in sometimes but not always? Could this be an issue with the pool of virtuals or the logmaode that they are picking up or possibly the term type on their TN3270 session (assuming they comming in via TN3270)? Alan Scott -- 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: TPX and ACLE Code
That is something I will work towards. I just found out that TPX V5.2 (just out) will allow batch updating to the ACT table. Lizette > >You should normally add all of your applid's to the TPX ACT table. There are >options in the TPX command authorization class panels that determine >whether TPX users can activate sessions dynamically (the "A" command). If >dynamic sessions are allowed the user can activate sessions that have not >been predefined in the ACT and user profiles. Hopefully that ability is only >permitted to the "power" users in your organization. For the normal user they >should only be allowed to activate sessions that are defined in the ACT and >allowed by their TPX profile. -- 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 read source from a PDS member
On Thu, 18 Oct 2007 08:21:44 -0600 Howard Brazee <[EMAIL PROTECTED]> wrote: :>On Thu, 18 Oct 2007 09:43:51 -0400, in bit.listserv.ibm-main you :>wrote: :>>I need to read source code from a PDS member whose name I don't know until my program runs, much the same as the COBOL :>>compiler handles "COPY memname" statements. Is dynalloc the only way to go about it or does zOS provide a nice friendly :>>interface that lets me ask for source member data a line at a time? Something comparable to VSE's LIBRM would be ideal. :>CoBOL compilers handle COPY commands at compile time, not run time. :>But you didn't indicate that your source code is CoBOL, it may be some :>real-time language.Could you provide more detail? BPAM is not THAT unfriendly. A while back (probably before many lurkers on this list were born, but I digress) I wrote a set of subroutines called QPAM to hide the tricky stuff from the programmer (don't know if I can give it out). QPAM GET - get next record QPAM PUT - put a record QPAM TRUNC - force a write of all queued blocks QPAM RLSE - stop reading current member You still hard coded the FIND and STOW, but they are rather simple. -- Binyamin Dissen <[EMAIL PROTECTED]> http://www.dissensoftware.com Director, Dissen Software, Bar & Grill - Israel Should you use the mailblocks package and expect a response from me, you should preauthorize the dissensoftware.com domain. I very rarely bother responding to challenge/response systems, especially those from irresponsible companies. -- 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: TPX and ACLE Code
You should normally add all of your applid's to the TPX ACT table. There are options in the TPX command authorization class panels that determine whether TPX users can activate sessions dynamically (the "A" command). If dynamic sessions are allowed the user can activate sessions that have not been predefined in the ACT and user profiles. Hopefully that ability is only permitted to the "power" users in your organization. For the normal user they should only be allowed to activate sessions that are defined in the ACT and allowed by their TPX profile. -- 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 read source from a PDS member
On Thu, 18 Oct 2007 16:34:36 +0200, Itschak Mugzach wrote: >Do you REXX? This is the best language to deal with strings, allocations and >dynamic variables on the fly. It is also run natively under TSO (and TSO >batch) and thus, supports ALLOC commands. > And under IRXJCL and Unix System Services, as well as TSO and TSO batch, BPXWDYN is available, which is in some respects superior to ALLOCATE; in others inferior. >-Original Message- >From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf >Of Howard Brazee >Sent: Thursday, October 18, 2007 4:22 PM > >On Thu, 18 Oct 2007 09:43:51 -0400, in bit.listserv.ibm-main you >wrote: > >>I need to read source code from a PDS member whose name I don't know >>until my program runs, much the same as the COBOL compiler handles >>"COPY memname" statements. Is dynalloc the only way to go about it or does >zOS provide a nice friendly interface that lets me ask for source member >data a line at a time? Something comparable to VSE's LIBRM would be ideal. > Dynalloc and reading records from the file are not alternatives to each other; first you dynalloc, then you read. Precious few languages support statically allocating a DDNAME to a library then reading a selected member from that library, as: "open(DDNAME(MEMBER))"; likely because that protocol is not well supported by QSAM. >CoBOL compilers handle COPY commands at compile time, not run time. > So what's the point? It demonstrates the basic feasibility of the operation. -- gil -- 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 read source from a PDS member
-Original Message- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Howard Brazee Sent: Thursday, October 18, 2007 9:22 AM To: IBM-MAIN@BAMA.UA.EDU Subject: Re: How to read source from a PDS member On Thu, 18 Oct 2007 09:43:51 -0400, in bit.listserv.ibm-main you wrote: >I need to read source code from a PDS member whose name I don't know >until my program runs, much the same as the COBOL compiler handles >"COPY memname" statements. Is dynalloc the only way to go about it or does zOS provide a nice friendly interface that lets me ask for source member data a line at a time? Something comparable to VSE's LIBRM would be ideal. CoBOL compilers handle COPY commands at compile time, not run time. But you didn't indicate that your source code is CoBOL, it may be some real-time language.Could you provide more detail? I didn't get the original message. But the question is, what language are you using, and how do you find out what the member is that you need to read? >From this, a better answer can be given. If you are writing an ALC program, you will need to start with BLDL and a DCB to match. Your DD will have to point to the PDS (not member specific). Regards, Steve Thompson -- Opinions expressed are my own. -- -- 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: Secure FTP Config: CIPHERSUITE statements?
Chase, John wrote: Hi, All, Is it "better" to order the CIPHERSUITE statements from weaker to stronger, or from stronger to weaker? Why? TIA, -jc- Logically I would go from stronger to weaker since you want the strongest encryption that both sides can understand. -- Mark Jacobs Time Customer Service Tampa, FL -- A desire not to butt into other people's business is at least eighty percent of all human wisdom...and the other twenty percent isn't very important. Jubal Harshaw (Stranger in a Strange Land) -- 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 read source from a PDS member
Do you REXX? This is the best language to deal with strings, allocations and dynamic variables on the fly. It is also run natively under TSO (and TSO batch) and thus, supports ALLOC commands. Itschak -Original Message- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Howard Brazee Sent: Thursday, October 18, 2007 4:22 PM To: IBM-MAIN@BAMA.UA.EDU Subject: Re: How to read source from a PDS member On Thu, 18 Oct 2007 09:43:51 -0400, in bit.listserv.ibm-main you wrote: >I need to read source code from a PDS member whose name I don't know >until my program runs, much the same as the COBOL compiler handles >"COPY memname" statements. Is dynalloc the only way to go about it or does zOS provide a nice friendly interface that lets me ask for source member data a line at a time? Something comparable to VSE's LIBRM would be ideal. CoBOL compilers handle COPY commands at compile time, not run time. But you didn't indicate that your source code is CoBOL, it may be some real-time language.Could you provide more detail? -- 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 -- 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
Secure FTP Config: CIPHERSUITE statements?
Hi, All, Is it "better" to order the CIPHERSUITE statements from weaker to stronger, or from stronger to weaker? Why? TIA, -jc- -- 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 read source from a PDS member
How about just passing the member name as a parameter in a PROC? //MYPGM PROC MBR=NOMBR //RUNIT EXEC PGM=MYPGM //INPUT DD DISP=SHR,DSN=MY.PDS(&MBR) Tim Hare Senior Systems Programmer Florida Department of Transportation Tel: +1 (850) 414-4209 -- 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 read source from a PDS member
> -Original Message- > From: IBM Mainframe Discussion List > [mailto:[EMAIL PROTECTED] On Behalf Of Howard Brazee > Sent: Thursday, October 18, 2007 9:22 AM > To: IBM-MAIN@BAMA.UA.EDU > Subject: Re: How to read source from a PDS member > > > On Thu, 18 Oct 2007 09:43:51 -0400, in bit.listserv.ibm-main you > wrote: > > >I need to read source code from a PDS member whose name I > don't know until my program runs, much the same as the COBOL > >compiler handles "COPY memname" statements. Is dynalloc the > only way to go about it or does zOS provide a nice friendly > >interface that lets me ask for source member data a line at > a time? Something comparable to VSE's LIBRM would be ideal. > I didn't see the original post. What language? Enterprise COBOL makes this very simple because it can hide the details of dynamic allocation from the programmer. In Java, use JZOS and the zFile class to do all the difficult work for you. In assembler, you use BPAM. That entails OPENing the PDS as a PDS, doing a FIND for the member name, then doing a READ for each physical block of the member. Deblocking into individual records is done by the application program, not the access method. If you run your program under ISPF, then you can use the LM* services to do all the "fiddly" work for you. I don't know z/VSE, but I did a quick look at it. I don't know of anything as nice and easy as that in z/OS. -- John McKown Senior Systems Programmer HealthMarkets Keeping the Promise of Affordable Coverage Administrative Services Group Information Technology The information contained in this e-mail message may be privileged and/or confidential. It is for intended addressee(s) only. If you are not the intended recipient, you are hereby notified that any disclosure, reproduction, distribution or other use of this communication is strictly prohibited and could, in certain circumstances, be a criminal offense. If you have received this e-mail in error, please notify the sender by reply and delete this message without copying or disclosing it. -- 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 read source from a PDS member
> -Original Message- > From: IBM Mainframe Discussion List On Behalf Of Howard Brazee > > On Thu, 18 Oct 2007 09:43:51 -0400, in bit.listserv.ibm-main you > wrote: > > >I need to read source code from a PDS member whose name I don't know > >until my program runs, much the same as the COBOL compiler handles > >"COPY memname" statements. Is dynalloc the only way to go > about it or does zOS provide a nice friendly interface that > lets me ask for source member data a line at a time? > Something comparable to VSE's LIBRM would be ideal. > > CoBOL compilers handle COPY commands at compile time, not run time. But compile time *is* the run time of the compiler, which is doing the COPYing.. -jc- -- 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 read source from a PDS member
On Thu, 18 Oct 2007 09:43:51 -0400, in bit.listserv.ibm-main you wrote: >I need to read source code from a PDS member whose name I don't know until my >program runs, much the same as the COBOL >compiler handles "COPY memname" statements. Is dynalloc the only way to go >about it or does zOS provide a nice friendly >interface that lets me ask for source member data a line at a time? Something >comparable to VSE's LIBRM would be ideal. CoBOL compilers handle COPY commands at compile time, not run time. But you didn't indicate that your source code is CoBOL, it may be some real-time language.Could you provide more detail? -- 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: Flashcopy and ADR369D
Greg, You just need to use the ADMINISTRATOR keyword. You must have RACF (or equivalent) authorization to use the ADMINISTRATOR keyword. The RACF requirements are documented in Chapter 6 of the DFSMSdss Storage Administration Reference. You should also review the following URL to automate the flashcopy, backup and restore process. http://www.opentechsystems.com/dbs.php Don Bolton Director Technical Services www.OpentechSystems.com -Original Message- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Greg Shirey Sent: Wednesday, October 17, 2007 3:28 PM To: IBM-MAIN@BAMA.UA.EDU Subject: Flashcopy and ADR369D Listers, We've recently acquired a Shark and are experimenting with Flashcopy on z/OS 1.7. When we do a COPY FULL (with DUMPCONDITIONING specified and with or without FCNOCOPY), message ADR369D is generated, causing the operator to have to respond U for "write access a VTOCIX data set" on the target volume. Is there something we can do to either eliminate or automate the reply? TIA, Greg Shirey Ben E. Keith Company -- 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 -- 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: Are there tasks that don't play by WLM's rules
>>- I am a little worried about your remark about the (short) life of the >>address spaces: creating a new thread in Unix is just simply adding an >>entry to a process table; if it requires address space creation in z/OS, >>this is a heavy overhead part added to kicking off the task. Can you >>play around with BPXPRM parameters to keep these address spaces (BPXAS I >>suppose) alive for a longer period, so they can be reused? >An idea to keep in mind. I am not really all *that* fluent in unix/bpx >stuff, I didn't even know that that possibility exists. >All our USS workloads tend to write job end iefactrt messages and iefuji >job begin messages into syslog at a very high rate, indicating short >lived address spaces, and obscuring the really important information.:-) PMFJI, but the last few remarks about being "ported from open world" made me think. So here are a few thoughts related to z/OS UNIX: - z/OS UNIX process are run in UNIX initiator address spaces, aka BPXAS. WLM starts one if the kernel needs while creating a new process. Once a process ends, the BPXAS is going to sleep for 30 minutes, then terminates. (No, you can't change that interval) So, address space creation is only needed if there is no idle BPXAS waiting for work. - In a busy system with quite some UNIX work going on, you'd have a good chance to find an idle BPXAS. However, when the UNIX workload goes down for more than 30 minutes, all those idle BPXAS will terminate. A sudden spike in UNIX work will then lead to a bunch of new BPXAS being created, very expensive, indeed. - Do you see a lot of "BPXAS starting" syslog entries when your workload arrives? If so, this might be part of the problem. - z/OS UNIX offers the possibility of so called "local processes". If your workload is calling spawn() to create the new processed, you might try to set _BPX_SHAREAS=YES and see if this helps. - You might want to try to implement a simple program that does a couple of simultaneous forks() to create a bunch of child processes. They don't have to do anything meaningfull. They can return as soon as all childs have been created. This would ask for so many BPXAS as there are childs. Then when your workload spikes up, the BPXAS are ready. - Finally regarding the iefactrt and other syslog messages: Remember a UNIX process is kind of like a job: SMF records will be cut whenever the process does an exec() and finally when the process ends. SMF exits are driven unless you suppress them for SUBSYS OMVS im SMFPRM. So, seeing all those messages does not mean the address spaces the processes ran in have terminated (see BPXAS above). As long as there are no "BPXAS has ended" messages no UNIX initiator AS has terminated. Hope that helps somewhat -- Peter Hunkeler Credit Suisse -- 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: WLC and Netview
- Original Message From: Mark Neal <[EMAIL PROTECTED]> Hi Mark, > I am not aware of any Job Scheduling capabilities in Netview. maybe I used the wrong term. We use the CHRON capability of Netview to submit batch jobs when we like to: every day, only at w-e, only at certain days of the months, etc. > So for instance, you could submit a job at 01:00, check messages and wait > for successful completion and submit the next job. If AF/OPER is capable of doing this, well, it's enough for us. Walter ___ Want ideas for reducing your carbon footprint? Visit Yahoo! For Good http://uk.promotions.yahoo.com/forgood/environment.html -- 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
More SSL/TLS and FTP woes
John, In Z/OS 1.7 the Implicit SSL/TLS is no longer supported for FTP. You need to use Explicit (Auth SSL). The support for Explicit has changed, so you should upgrade to BlueZone Secure FTP 4.2. Z/OS 1.7 and BlueZone 4.2 support RFC-4217. http://rfc.net/rfc4217.html We have separated BlueZone Secure FTP from the BlueZone package and give it away free for personal, commercial, or government use. This replaces BlueZone FTP (non-secure) and Seagull Free FTP. It installs in a different folder, so you can test it without breaking for upgrading your current BlueZone Desktop or Web-to-Host install. BlueZone Secure FTP still ships with BlueZone and installs/upgrades the same way. BlueZone Secure FTP support SSL/TLS, SSH, IPV6 and uses FIPS 140-2 encryption for SSL/TLS. You can download it at http://www.bluezonesoftware.com/products/secure-ftp . No login or email address is required. You can choose to register it or not. Steve Bireley Vice-President Product Development BlueZone Software 1-404-364-1731 www.bluezonesoftware.com www.rocketsoftware.com Date:Wed, 17 Oct 2007 09:03:17 -0500 From:"Chase, John" <[EMAIL PROTECTED]> Subject: More SSL/TLS and FTP woes Hi, All, I couldn't find anything relevant to the "problem du jour" in the archives or the CS for z/OS 1.7 TCPIP Implementation Volume 2 Redbook, so.. I'm able to employ SSL/TLS for FTP using the Bluezone FTP client, but only if I configure it to use port 21 and "AUTH_TLS". I cannot get it working via "implicit" secure FTP using port 990; the z/OS 1.7 FTPD replies "connection refused". AFAICT, I have "all the ducks lined up in a row", with one possible exception: I don't explicitly "reserve" port 990 (and 989?) in the PORT configuration statement of PROFILE.TCPIP. The IP Configuration Reference manual "suggests" it's not necessary to do so. Might this be the "missing link" after all? Do I need to (additionally) explicitly specify the statements in FTP.DATA that the manual says are defaults for TLS_PORT, etc.? TIA, -jc- Steve Bireley Vice-President Product Development BlueZone Software 1-404-364-1731 www.bluezonesoftware.com www.seagullsoftware.com www.rocketsoftware.com -- 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: Any way to duplex SMF data?
Mark Zelden wrote... > Isn't Tivoli Decision Support the new name for what used to be SLR? No, it's a replacement product, with zero code base in common. Cheers,. Martin Martin Packer Performance Consultant IBM United Kingdom Ltd +44-20-8832-5167 +44-7802-245-584 [EMAIL PROTECTED] Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU -- 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: More SSL/TLS and FTP woes
> -Original Message- > From: IBM Mainframe Discussion List On Behalf Of Joel Ivey > > ?? Should you not have a separate ftpd for 990 traffic? Hmmm.. You may have a point there.. In fact, you may have "the" point there. > Can one ftpd > process listen on both 21 and 990 at the same time? I had "ass.u.me"-d so, having become accustomed to most "mainframe" software having the ability to do many things concurrently. Bit I notice in the ftpd.log that each individual FTP request runs with a different PID than the server itself, so maybe the FTP software actually has only one "ear" with which to listen for connection requests. Certainly, starting a second ftpd to listen on port 990 should be simple enough. -jc- -- 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: Are there tasks that don't play by WLM's rules
On Thu, 18 Oct 2007 11:00:45 +0200, Barbara Nitz <[EMAIL PROTECTED]> wrote: >My biggest concern with the high PI is actually that WLM will only try to help the service class every 3rd trip through WLM (so I was told). Exactly why I think SYSSTC is the best place for this. It sounds like you really need to let this work run when it wants to. It doesn't matter that MQ is in a service class with a goal and will have a lower DP. Obviously WLM changes are easy to put in and take out... so I would really try to convince your co-worker(s) to try it and then monitor via RMF III to see what kind of CPU delays other important work is getting (compare before and after). Mark -- Mark Zelden Sr. Software and Systems Architect - z/OS Team Lead Zurich North America / Farmers Insurance Group - ZFUS G-ITO mailto:[EMAIL PROTECTED] z/OS Systems Programming expert at http://expertanswercenter.techtarget.com/ Mark's MVS Utilities: http://home.flash.net/~mzelden/mvsutil.html -- 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: Any way to duplex SMF data?
On Thu, 18 Oct 2007 09:02:00 +0100, Martin Packer <[EMAIL PROTECTED]> wrote: >Seymour wrote: > >> The closest I've seen was SLR, whatever they're calling it these days. > >They're calling it "unsupported for 10 years" :-) though I still run it >and rely on it for my day job (and am enhancing our record mappings and >reporting even as we type). :-) > Isn't Tivoli Decision Support the new name for what used to be SLR? http://www-306.ibm.com/software/tivoli/products/tds-390/ Mark -- Mark Zelden Sr. Software and Systems Architect - z/OS Team Lead Zurich North America / Farmers Insurance Group - ZFUS G-ITO mailto:[EMAIL PROTECTED] z/OS Systems Programming expert at http://expertanswercenter.techtarget.com/ Mark's MVS Utilities: http://home.flash.net/~mzelden/mvsutil.html -- 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: WLC and Netview
Hi Walter, I am not aware of any Job Scheduling capabilities in Netview. I suspect you are using automation to run batch jobs. If that is the case, you should be able to do the same thing in AF/OPER. AF/OPER has nice message and time automation. So for instance, you could submit a job at 01:00, check messages and wait for successful completion and submit the next job. Mark -- 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: Zapping HFS files
> -Original Message- > From: IBM Mainframe Discussion List > [mailto:[EMAIL PROTECTED] On Behalf Of Paul Gilmartin > Sent: Wednesday, October 17, 2007 5:29 PM > To: IBM-MAIN@BAMA.UA.EDU > Subject: Re: Zapping HFS files > However: > > #16.2.1 "z/OS V1R7.0 MVS Diagnosis: Tools and Service Aids" > > ___ > > 16.2.1 Inspecting and modifying a load module or program object > >To inspect or modify a program object that is in a z/OS UNIX file >system, use the PATH parameter on the SYSLIB DD statement > instead of the >DSNAME parameter. > > -- gil IIRC, the original question was not necessarily about zapping a UNIX resident program object. So, how do I zap a binary file, residing in UNIX, which is not a program object? -- John McKown Senior Systems Programmer HealthMarkets Keeping the Promise of Affordable Coverage Administrative Services Group Information Technology The information contained in this e-mail message may be privileged and/or confidential. It is for intended addressee(s) only. If you are not the intended recipient, you are hereby notified that any disclosure, reproduction, distribution or other use of this communication is strictly prohibited and could, in certain circumstances, be a criminal offense. If you have received this e-mail in error, please notify the sender by reply and delete this message without copying or disclosing it. -- 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: ALLOCAS High cpu utilization - Solution
After speaking with IBM support, we ended up applying the ptf for OA20727 and IPLing the system. BTW we are at z/OS 1.7 and JES2 Gadi -- 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
dfsmshsm csalimit best practice
I am getting message ARC0058I CSA USAGE BY DFSMSHSM HAS REACHED THE ACTIVE THRESHOLD OF 90K BYTES, ALL BUT BATCH WAIT REQUESTS FAILED I am running a batch HMIGRATE against GDG's mask. "sys3.listvtoc.**' This command generates 3537 ARC1007I MIGRATE REQUESTS. All but the current GDGs are already migrated from the last HMIGRATE that ran the previous night. The step abends with a rc=16. The current SETSYS CSA is specified at 100kb of CSA storage for MWEs. DFSMSHSM V1R6. I understand that HSM now gets it's MWEsttorage from ECSA and I should be able to increase the setting without impacting other applications. Is this true? What is the best practice for setting the CSALIMIT? Regards, Judy Ellis -- 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: UCB Wikipedia Article
In a message dated 10/18/2007 1:21:44 A.M. Central Daylight Time, [EMAIL PROTECTED] writes: >1. There are still UCBs *below* the line, so while parts are always in ESQA, parts can still be in SQA. The poor UCB has undergone cruel and unusual punishment in the evolution of OS/360 to today's z/OS. I think in OS/360 there was only one piece, which is now called the Common Segment to differentiate it from the many other segments. Since there was only one piece way back then (I think), it would not have been called the Common Segment but just the "UCB", as no one knew there would be beaucoup segments later. By the time MVS/XA arrived, there was also a "fake" 512-byte prefix just below the Common Segment. Only the last 56 bytes of this prefix were used. Some parts of this original prefix are now in a Prefix Extension. UCBLOOK is one way to find all the various pieces. Whenever an unauthorized application needs to do an I/O, it invokes some kind of access method that is supplied by IBM. The access method knows how to "access the UCB." An authorized application can, of course, do the necessary accessing directly. >2. There was a way to change the FIFO queuing of IOQs to an UCB starting way back when. This was never my area (so I am murky on this), but I think the IOPID (IO prevention identifier) did the trick, somewhere around the bend in the knee. I have never delved into I/O prevention, so I don't know what it does, how, or why, but I believe it is used in preventing an I/O's being executed rather than positioning the request in the queue. But at least since MVS/XA there has been an I/O "priority" associated with the address space, originally used by SRM, and which causes new requests to be placed in the UCB queue for the device in decreasing I/O priority. This was similar to the dispatching priority that determines which ready task gets to execute next on a processor. There is now another technique for prioritizing queued requests that involves WLM and cross-system considerations, and this priority byte is also sent to the DASD subsystem (formerly known as "control unit") to influence the performance of the channel program once it has been started via SSCH. >Coming back to 'application'. Just yesterday we did a ccw trace to 4 devices that aren't even online to that system. The SSCHs were issued under the EMC host component address spaces. From my point of view, this is 'an application', too, but it knows (or should know) *a lot* about the way IOS and the hardware work. That component runs authorized, of course, so it really can point to the actual storage the UCB resides in. It can even change the bits in there. It is no great trick (once you know how, of course) to doing I/O to a device that is offline as long as the system has a UCB for it and at least one operational path to the device. When the last channel path to the device is varied offline, then you cannot access that device. The purpose of the UCB Capture service is to allow anciently compiled programs using ancient access methods that require the DCB, DEB, IOB, etc., to be in storage below the ancient 16MB line to do I/O with a thoroughly modern UCB that is above the line. The big problem was that the DEB has room only for a 3-byte address of the UCB. Rather than redesign the ancient access methods that were based on EXCP, IBM provided a way for old access methods to get to new UCBs above the line with the system service IOSCAPU. IOSCAPU creates a copy (called "capturing" the UCB) of an above-the-line UCB in the user's private storage and puts that copy below the line. The various ancient control blocks that need to point to a UCB will point to the private storage copy with a 24-bit address. Leading edge access methods are continually updated by IBM to support every new wrinkle that the IOS developers apply to UCBs in order to support system growth. Many of the original constraints in the software and hardware architecture of S/360 have been relaxed, with occasional seismic jolts to the appearance of the UCB; e.g., UCBs can now be above the 16MB line, devices can have five hexadecimal digits in the device number (formerly known as "device address"), there can be more than 65536 devices on one LPAR, UCBs can be dynamically created in order to support dynamically installed new I/O hardware, and more than one I/O can be simultaneously active with the same device. Bill Fairchild Plainfield, IL ** See what's new at http://www.aol.com -- 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: Are there tasks that don't play by WLM's rules
On Oct 18, 2007, at 6:00 AM, Vernooy, C.P. - SPLXM wrote: "Barbara Nitz" <[EMAIL PROTECTED]> wrote in message news:<[EMAIL PROTECTED]>... The MVS implementation of BPX 'address space creation' was changed back in the nineties to use RSM shared pages (a virtual overlay of the creator's address space) because most of the parent address space pages were thrown away by the child anyway. That improved performance a lot, similar to what can be seen in the 'distributed world'. Before that, 'getmain at the given address' was implemented to make the parent identical to the child in the beginning. I know the storage part and its efficiency, but creating an address space is something else, that costs time. Somewhat off the topic but I feel thats its important to add this. In the early 80's I was given the task to help out a division get a new system I was talking with an SE (she was a beginner) and we got into a discussion of address space create. She was under the impression that it was no big deal and it was a minor thing to do. In fact she was suggesting that they set up (IMS? its been a while) that whenever a transaction was needed that it go through address create. I suggested to her that she read a little more on the process that is needed to do a address space create and then come back and tell me it should be done all the time. She wondered off and (I think) she posed the question to someone in IBM and they told her that her idea wasn't so hot and how much it would cost. She came back in a week or so and said that I had been right. I know she was a rookie and clearly did not have any idea on the overhead, but she finally got the idea that it was a lot more than just minor overhead to do. Ed -- 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: Are there tasks that don't play by WLM's rules
Kees, thanks a lot for taking the time to discuss this with me! Now that you mention it, it appears so obvious, and I was just too slow to see it. (Guess somewhere along the way I got tangeld up in other tangents.) I'll talk to my colleague and will try to implement it as soon as possible (first in test). Now, why didn't WLM development suggest this to me?!? Best regards, Barbara -- GMX FreeMail: 1 GB Postfach, 5 E-Mail-Adressen, 10 Free SMS. Alle Infos und kostenlose Anmeldung: http://www.gmx.net/de/go/freemail -- 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: Are there tasks that don't play by WLM's rules
"Barbara Nitz" <[EMAIL PROTECTED]> wrote in message news:<[EMAIL PROTECTED]>... > > > The MVS implementation of BPX 'address space creation' was changed back in the nineties to use RSM shared pages (a virtual overlay of the creator's address space) because most of the parent address space pages were thrown away by the child anyway. That improved performance a lot, similar to what can be seen in the 'distributed world'. Before that, 'getmain at the given address' was implemented to make the parent identical to the child in the beginning. I know the storage part and its efficiency, but creating an address space is something else, that costs time. > > >- If the DP's are well positioned between the Sysstc (MQ etc.) and the > >low prio other stuff, what exactly do you expect from WLM? > > a) skip the 'I don't look at you for a while' algorithm for *this* workload. We have these spikes, and 30s are too long if there really is an increase in load and "someone else has the high DP due to WBIFN not needing it/being overlooked for that interval." THIS is what CPU Critical prevents, if 'someone else' has a lower IMP than your WBIFN. Maybe you can look at this again. Kees. ** For information, services and offers, please visit our web site: http://www.klm.com. This e-mail and any attachment may contain confidential and privileged material intended for the addressee only. If you are not the addressee, you are notified that no part of the e-mail or any attachment may be disclosed, copied or distributed, and that any other action related to this e-mail or attachment is strictly prohibited, and may be unlawful. If you have received this e-mail by error, please notify the sender immediately by return e-mail, and delete this message. Koninklijke Luchtvaart Maatschappij NV (KLM), its subsidiaries and/or its employees shall not be liable for the incorrect or incomplete transmission of this e-mail or any attachments, nor responsible for any delay in receipt. Koninklijke Luchtvaart Maatschappij N.V. (also known as KLM Royal Dutch Airlines) is registered in Amstelveen, The Netherlands, with registered number 33014286 ** -- 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: Are there tasks that don't play by WLM's rules
Kees, >- How did this thing run in the 'open' world (or didn't it)? Machines >with dozens of processors to execute these 65 tasks immediately are not >common there either AFAIK. No idea. The architecture in the 'open' world may be completely different. >- I am a little worried about your remark about the (short) life of the >address spaces: creating a new thread in Unix is just simply adding an >entry to a process table; if it requires address space creation in z/OS, >this is a heavy overhead part added to kicking off the task. Can you >play around with BPXPRM parameters to keep these address spaces (BPXAS I >suppose) alive for a longer period, so they can be reused? An idea to keep in mind. I am not really all *that* fluent in unix/bpx stuff, I didn't even know that that possibility exists. All our USS workloads tend to write job end iefactrt messages and iefuji job begin messages into syslog at a very high rate, indicating short lived address spaces, and obscuring the really important information.:-) The MVS implementation of BPX 'address space creation' was changed back in the nineties to use RSM shared pages (a virtual overlay of the creator's address space) because most of the parent address space pages were thrown away by the child anyway. That improved performance a lot, similar to what can be seen in the 'distributed world'. Before that, 'getmain at the given address' was implemented to make the parent identical to the child in the beginning. >- If the DP's are well positioned between the Sysstc (MQ etc.) and the >low prio other stuff, what exactly do you expect from WLM? a) skip the 'I don't look at you for a while' algorithm for *this* workload. We have these spikes, and 30s are too long if there really is an increase in load and someone else has the high DP due to WBIFN not needing it/being overlooked for that interval. b) if that is impossible, lean on the application to *define* what a transaction is so that 'the performance people' have a better way to manage the workload. c) a combination of both. I am well aware that that requires design changes. :-) Best regards, Barbara -- GMX FreeMail: 1 GB Postfach, 5 E-Mail-Adressen, 10 Free SMS. Alle Infos und kostenlose Anmeldung: http://www.gmx.net/de/go/freemail -- 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: Any way to duplex SMF data?
Yes BUT note there is no REAL resemblance between the 2 products... SLR is based on VSAM databases - which is why I stick with it for my mode of usage. TDS is based on DB2 databases - which wouldn't suit my case but WOULD be good for a regular installation. If you wonder why MY usage drives me to VSAM-based it's because I get a tranche of data in from a customer, do my thing with it, and then let it go off to ML2. Rather more complex to do with DB2. Cheers, Martin Martin Packer Performance Consultant IBM United Kingdom Ltd +44-20-8832-5167 +44-7802-245-584 [EMAIL PROTECTED] Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU -- 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: ALLOCAS High cpu utilization
I don't know if it could be the same problem that we had 2 months ago, but, if you have 3 cpu's and the ALLOCAS use the 30%, probably it is in a cpu loop We suffered this problem the last august and the only way to stop allocas was doing IPL We have z/os 1.8, and we found 2 ptf's to solve the problem http://www-1.ibm.com/support/docview.wss?uid=isg1OA20727 http://www-1.ibm.com/support/docview.wss?uid=isg1OA20056 Hope this information can be useful Antonio -- 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: Any way to duplex SMF data?
SLR moved thru Tivoli Performance Reporter that again migrated into: 5698-A07 Tivoli Decision Support for z/OS Birger Heede IBM Denmark Martin Packer wrote: Seymour wrote: The closest I've seen was SLR, whatever they're calling it these days. They're calling it "unsupported for 10 years" :-) though I still run it and rely on it for my day job (and am enhancing our record mappings and reporting even as we type). :-) Cheers, Martin Martin Packer Performance Consultant IBM United Kingdom Ltd +44-20-8832-5167 +44-7802-245-584 [EMAIL PROTECTED] Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU -- 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 -- 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: Are there tasks that don't play by WLM's rules
"Barbara Nitz" <[EMAIL PROTECTED]> wrote in message news:<[EMAIL PROTECTED]>... > Kees, > > >- Is the WLM PI number the real problem, or is the performance actually > >bad? A bad report about a well running application is not the end of the > >world. > We started looking at this when we had complaints about missing throughput and bad response times. Before that, we didn't even realize how bad the PI was. > > >- If they all pop up at about the same time, how would you like to see a > >2 CP Lpar handle 65 tasks at (about) the same moment??? WLM or no WLM, > >here you have a real problem I think, even if the Lpar Weight were high > >enough. > Exactly my point: I think *someone* needs to take a real hard look at how this product was ported to z/OS. (I didn't mention that it was ported from the 'open world', did I?) I also think to 'play nice' in a z/OS world, it needs design changes, one of them being the use of WLM macros to actually define the start and the end of a transaction, just so I can use a response time goal. :-) > In addition, there need to be much better guidelines on how to tune this application with the obvious things - like storage usage in LE, like what effect committing messages has (there were also lots of SSRBs - suspended somewhere in RRS/LOGR processing) and how often to do that. I am sure my 1.5s trace table only scratches the surface. Besides, a trace table is not the right tool to look at a problem like this. > > >did you define them as CPU CRITICAL? > No. On the assumption that I can *see* that the DP of these 65 things is definitely higher than anything else on that lpar (aside from sysstc and the obvious supporters). Also, giving it one more processor didn't help the PI. Should I test that, anyway? > > My biggest concern with the high PI is actually that WLM will only try to help the service class every 3rd trip through WLM (so I was told). WLM doesn't look at it for two intervals. And the intervals are long, anyway (10s?). The WLM developer basically told me that it doesn't matter that WLM will not help the class for that long. This is what really go my hackles up, we just don't have a continuous workload, we have extreme spikes and are supposed to have good response times even in spike situations. > In order to achieve a continuous help for this class, I have to set the velocity artificially to 1% or less, just to influence the computing, if I can even define it that low. That probably gets me a whole lot of other problems. > I have also been told (completely differnt guy in WLM) that it is necessary to define a resource group with a very high guaranteed minimum consumption whenever I go below 30% for an exvel goal. So far I have managed to avoid that by basically only using the importance for classification and the exvel always being 31 or higher. Which doesn't give me much room until I hit the 'unachievable goal'. (Oh, I *was* told that Imp1, exvel 40% is too ambitious!) > > Does that make my dilemma clearer? > > Best regards and thanks, Barbara Yes, the picture becomes clear. I still have 3 thoughts: - How did this thing run in the 'open' world (or didn't it)? Machines with dozens of processors to execute these 65 tasks immediately are not common there either AFAIK. - I am a little worried about your remark about the (short) life of the address spaces: creating a new thread in Unix is just simply adding an entry to a process table; if it requires address space creation in z/OS, this is a heavy overhead part added to kicking off the task. Can you play around with BPXPRM parameters to keep these address spaces (BPXAS I suppose) alive for a longer period, so they can be reused? - If the DP's are well positioned between the Sysstc (MQ etc.) and the low prio other stuff, what exactly do you expect from WLM? WLM can in fact only influence performance of tasks by manipulating their relative DP's. If they are positioned well, it is up to the Dispatcher to dispatch them on the available CP's. What can WLM stil do in this situation? Regards, Kees. ** For information, services and offers, please visit our web site: http://www.klm.com. This e-mail and any attachment may contain confidential and privileged material intended for the addressee only. If you are not the addressee, you are notified that no part of the e-mail or any attachment may be disclosed, copied or distributed, and that any other action related to this e-mail or attachment is strictly prohibited, and may be unlawful. If you have received this e-mail by error, please notify the sender immediately by return e-mail, and delete this message. Koninklijke Luchtvaart Maatschappij NV (KLM), its subsidiaries and/or its employees shall not be liable for the incorrect or incomplete transmission of this e-mail or any attachments, nor responsible for any delay in receipt. Koninklijke Luchtvaart Maatschappij N.V. (also known as KLM Royal Dutch Airlines) is
Re: Are there tasks that don't play by WLM's rules
Kees, >- Is the WLM PI number the real problem, or is the performance actually >bad? A bad report about a well running application is not the end of the >world. We started looking at this when we had complaints about missing throughput and bad response times. Before that, we didn't even realize how bad the PI was. >- If they all pop up at about the same time, how would you like to see a >2 CP Lpar handle 65 tasks at (about) the same moment??? WLM or no WLM, >here you have a real problem I think, even if the Lpar Weight were high >enough. Exactly my point: I think *someone* needs to take a real hard look at how this product was ported to z/OS. (I didn't mention that it was ported from the 'open world', did I?) I also think to 'play nice' in a z/OS world, it needs design changes, one of them being the use of WLM macros to actually define the start and the end of a transaction, just so I can use a response time goal. :-) In addition, there need to be much better guidelines on how to tune this application with the obvious things - like storage usage in LE, like what effect committing messages has (there were also lots of SSRBs - suspended somewhere in RRS/LOGR processing) and how often to do that. I am sure my 1.5s trace table only scratches the surface. Besides, a trace table is not the right tool to look at a problem like this. >did you define them as CPU CRITICAL? No. On the assumption that I can *see* that the DP of these 65 things is definitely higher than anything else on that lpar (aside from sysstc and the obvious supporters). Also, giving it one more processor didn't help the PI. Should I test that, anyway? My biggest concern with the high PI is actually that WLM will only try to help the service class every 3rd trip through WLM (so I was told). WLM doesn't look at it for two intervals. And the intervals are long, anyway (10s?). The WLM developer basically told me that it doesn't matter that WLM will not help the class for that long. This is what really go my hackles up, we just don't have a continuous workload, we have extreme spikes and are supposed to have good response times even in spike situations. In order to achieve a continuous help for this class, I have to set the velocity artificially to 1% or less, just to influence the computing, if I can even define it that low. That probably gets me a whole lot of other problems. I have also been told (completely differnt guy in WLM) that it is necessary to define a resource group with a very high guaranteed minimum consumption whenever I go below 30% for an exvel goal. So far I have managed to avoid that by basically only using the importance for classification and the exvel always being 31 or higher. Which doesn't give me much room until I hit the 'unachievable goal'. (Oh, I *was* told that Imp1, exvel 40% is too ambitious!) Does that make my dilemma clearer? Best regards and thanks, Barbara -- GMX FreeMail: 1 GB Postfach, 5 E-Mail-Adressen, 10 Free SMS. Alle Infos und kostenlose Anmeldung: http://www.gmx.net/de/go/freemail -- 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: Any way to duplex SMF data?
Seymour wrote: > The closest I've seen was SLR, whatever they're calling it these days. They're calling it "unsupported for 10 years" :-) though I still run it and rely on it for my day job (and am enhancing our record mappings and reporting even as we type). :-) Cheers, Martin Martin Packer Performance Consultant IBM United Kingdom Ltd +44-20-8832-5167 +44-7802-245-584 [EMAIL PROTECTED] Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU -- 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: Are there tasks that don't play by WLM's rules
"Barbara Nitz" <[EMAIL PROTECTED]> wrote in message news:<[EMAIL PROTECTED]>... > Mark and Shane, > > >Do you have enough engines and capacity where bursts won't hurt other > >workloads? If so, I might be inclined to run this work in SYSSTC (I > >assume discretionary won't work for this since you mentioned "on time") > >and then WLM doesn't have to manage it. > > thanks for that suggestion. It was vetoed vehemently by the 'head WLM developer' when I suggested it. (And given that the address spaces have a life time from less than a second to forever, there is also no real way to give them a response time goal based on how long they live.) But now that the two of you suggest this, too, I might broach the idea again with my colleague (and ignore the developer). Should things get really nasty, we may want to do that. > > Let me give you some more information on this: > 1. The lpar is not the lpar with the highest weight on that box. There is another sysplex on it that has a significantly higher weight. I mention this because our workload is basically driven by what the stock exchange does (completely outside our control), and in our experience the weights (and number of processors) determine how fast we are once business picks up significantly. During the stock crashes early in the year we ended up taking down and deactivating low importance lpars in advance just to handle the spikes. So I wouldn't really call that 'enough capacity' to spare, given the prohibitive software prices. > 2. Varying the spare processor online to that lpar (giving it 3 cps) doesn't even show up in the PI. The reason being, (according to WLM) that the delays are not caused by cpu, but that's not what RMF tells me. RMF MIII says that about 60% is processor delay, the rest falls into 'other'. > 3. That application has firm ties in MQSeries, so it shouldn't really run higher than that. (I think.) MQS is in STCHIGH (Imp1, exvel 50%), sysplex wide service class. > 4. When the first complaints came, I started looking at STCHIGH, which incidentally was also the SC 'the broker' (these 65 asids on average) run in. 'The broker' only runs on that lpar, together with its own MQS and other infrastructure. The only other stuff there is a websphere application server that does not have that much load and runs in a lower importance SC. > 5. First thing we did was put 'the brokers' into their own SC (Imp1, exvel 40%). And that's when the problem really showed up with PIs generally in the double digits, sometimes even more, and never near 1. (I even did a new SAS report to show all intervals with a PI>1 fro all service classes.) > - > The ETR is still open, and I had sent WLM SMF99 records. The 'head WLM developer' promised to tell us if we can live with this situation when the box gets full (and the lpar gets full). Interestingly enough, the last update in the ETR says " has sent his findings to . Please contact to discuss the situation." Now the question is: Why aren't *we* informed what 'the situation' is? This really has my suspicions up, especially as we still have a complaint open with that ETR. > - > With this information, any more ideas what to do? > Best regards, Barbara > -- Barbara, I have studied the entire thread carefully for the last 30 minutes and I am a little lost. What goes through my mind is: - Is the WLM PI number the real problem, or is the performance actually bad? A bad report about a well running application is not the end of the world. I suppose you really have performance problems with the application. - If they all pop up at about the same time, how would you like to see a 2 CP Lpar handle 65 tasks at (about) the same moment??? WLM or no WLM, here you have a real problem I think, even if the Lpar Weight were high enough. - I thought about SYSCTC too, but your hesitation in relation to MQ seems valid. Something else that can contribute: did you define them as CPU CRITICAL? This will prevent de DP of the address spaces to fall below your other, less important work and also eliminate the delay WLM needs to raise the DP to the necessary value. ExVel goals have always been a problem to me too, even for simple batch. My conclusion is that a exvel=40 only means that everyone gets 40% of what they ask. Big yelling jobs get 40% of what they call for; well behaving silent jobs also get 40% of what they politely request. This makes it difficult to give both jobs a fair share of CPU resources. Regards, Kees. ** For information, services and offers, please visit our web site: http://www.klm.com. This e-mail and any attachment may contain confidential and privileged material intended for the addressee only. If you are not the addressee, you are notified that no part of the e-mail or any attachment may be disclosed, copied or distributed, and that any other action related to this e-mail or attachment is strictly prohibited, and may be unlawful. If you have recei
Re: Flashcopy and ADR369D
- Original Message From: Greg Shirey <[EMAIL PROTECTED]> > When we do a COPY FULL (with DUMPCONDITIONING specified and with or > without FCNOCOPY), message ADR369D is generated, causing the operator to > have to respond U for "write access a VTOCIX data set" on the target > volume. Is there something we can do to either eliminate or automate the > reply? Greg, ADMIN, ALLD(*), ALLX: bypass reading the source VTOC ADMIN, PUR : bypass reading the target VTOC my SYSIN stmt looks like: COPY FULL IDY(dasd01) ODY(Flash1) DUMPCOND FCNC FR(REQ) - ADMIN PUR ALLD(*) ALLX and ADR369D does not pop up. HTH. Walter Marguccio z/OS Systems Programmer Munich - Germany ___ Yahoo! Answers - Got a question? Someone out there knows the answer. Try it now. http://uk.answers.yahoo.com/ -- 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: SYSLOGx1 Task Name in z/OS 1.7
>This is how z/OS emulates UNIX processes. >Some are invoked by fork() which leaves the 'parent' task running. >All invoked this way will terminate when the invoking task ends. >Others (like FTPD) are invoked by spawn() which leaves the process >running even when the 'parent' goes away. These are better managed >by modifying USS address spaces (OMVS) to terminate, or through the http://bama.ua.edu/archives/ibm-main.html