1 WWW.IDUGDB2-L.ORG /home/listserv/home/db2-l December 2008, week 2 2 233 23_Problem with Image Copy13_Robert Knight28_bknight@REMOTEDBAEXPERTS.COM30_Sun, 7 Dec 2008 20:06:17 -0500732_us-ascii Recently had an image copy process abend with the following error message.

TMSSMF06W END OF DSNB CHAIN REACHED



Is this due to a large number of datasets stored on a tape backup?

Can't find any information on the error message.

Any assistance would be greatly appreciated.

Bob Knight





The IDUG DB2-L Listserv is only part of your membership in IDUG. DB2-L list archives, the FAQ, and delivery preferences are at www.idug.org under the Listserv tab. While at the site, you can also access the IDUG Online Learning Center, Tech Library and Code Place, see the latest IDUG conference information, and [...] 236 53 58_Help me to monitoring jdbc application from DB2 zOs angle.14_Renzo razzetti24_renzo.razzetti@GMAIL.COM30_Mon, 8 Dec 2008 15:27:28 +0800462_ISO-8859-1 Hi all.

I have to monitoring a jdbc application using type 4 driver directly to connect with the DB2 zOs server. I would like to know besides the normal indicators such as response time , CPU time, suspension time, which other indicators I have to pay attention on evaluation of JDBC application. Also I will appreciate if you can send me relevant and specific information of how to monitoring jdbc application from DB2 Zos point of view. [...] 290 95 27_Re: Problem with Image Copy12_Isaac Yassin20_yassini@BEZEQINT.NET30_Mon, 8 Dec 2008 09:55:12 +0200380_iso-8859-1 i,

Use google :-)

This is from CA - Brightstor manual:

TMSSMF06W END OF DSNB CHAIN REACHED Explanation: While trying to locate a secondary data set that received a catalog action, the end of the DSNB chain was encountered. User Response: Run TMSPTRS to report on chaining errors. See the BrightStor CA-1 Utilities and Reports Reference Guide. [...] 386 238 32_Re: WLM STORE PROCEDURE QUESTION10_Dave Barry14_dbarry@UPS.COM30_Mon, 8 Dec 2008 03:18:25 -0500491_us-ascii Sounds like you may need to use dynamic subsystem name in WLM application environment definition. Instead of naming the subsystem, try using the special symbol &IWMSSNM.

Start parameters . . . . DB2SSN=&IWMSSNM

________________________________ From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Miguel Del Valle Martinez Sent: Tuesday, December 02, 2008 2:20 PM To: DB2-L@www.idugdb2-l.org Subject: [DB2-L] WLM STORE PROCEDURE QUESTION [...] 625 285 27_Re: Problem with Image Copy10_Dave Barry14_dbarry@UPS.COM30_Mon, 8 Dec 2008 03:33:01 -0500521_us-ascii See http://www.jmit.com/os390/docs/CA/CA1-52-Messages.PDF.

________________________________ From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Robert Knight Sent: Sunday, December 07, 2008 8:06 PM To: DB2-L@www.idugdb2-l.org Subject: [DB2-L] Problem with Image Copy

Recently had an image copy process abend with the following error message.

TMSSMF06W END OF DSNB CHAIN REACHED



Is this due to a large number of datasets stored on a tape backup? [...] 911 44 27_Re: Problem with Image Copy14_James Campbell25_jacampbell@ACSLINK.NET.AU30_Mon, 8 Dec 2008 20:18:11 +1100571_ISO-8859-1 DSNBs are the things that store information about the second and subsequent dataset on a tape in a CA-1 database. CA-1 messages codes start with TMS.

You should talk to the people who manage your CA-1 environment.

James Campbell

On 7 Dec 2008 at 20:06, Robert Knight wrote:

> > Recently had an image copy process abend with the following error message. > > TMSSMF06W END OF DSNB CHAIN REACHED > > > Is this due to a large number of datasets stored on a tape backup? > > Can´t find any information on the error message. > > Any [...] 956 65 27_Re: Problem with Image Copy13_Robert Knight28_bknight@REMOTEDBAEXPERTS.COM30_Mon, 8 Dec 2008 06:59:48 -0500507_iso-8859-1 Greatly appreciate all responses on this issue.

Thanks for the link.

Bob Knight

-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of James Campbell Sent: Monday, December 08, 2008 4:18 AM To: DB2-L@www.idugdb2-l.org Subject: Re: [DB2-L] Problem with Image Copy

DSNBs are the things that store information about the second and subsequent dataset on a tape in a CA-1 database. CA-1 messages codes start with TMS. [...] 1022 54 31_DB2 V8 cm going to ENFM and NFM12_Jeff Frazier31_Jeffrey.Frazier@WENDYSARBYS.COM30_Mon, 8 Dec 2008 10:47:59 -0500491_US-ASCII Hello, We are going to ENFM then NFM this Saturday. Before we run DSNTIJNE I know we have to take an I/C of CM catalog plus before that we will be taking a full volume backup of DB2 subsystem. If for whatever reason we need to go back to CM is the I/C of catalog we took before DSNTIJNE usable to go back to or would it be better just to restore back to full volume backup? This is our test subsystems by the way. Any experiences anyone would care to share? Thanks In advance, [...] 1077 70 37_Re: Recovery and datasharing question0_22_DB2information@AOL.COM28_Mon, 8 Dec 2008 11:00:58 EST474_US-ASCII Hanne, You may wish to visit _www.recoveryknowledge.com_ (http://www.recoveryknowledge.com) and review the GENDB2 software product. The complete recovery process is automated for datasharing and non-datasharing for onsite and offsite Disaster Recovery (DR.

Ed. **************Make your life easier with all your friends, email, and favorite sites in one place. Try it now. (http://www.aol.com/?optin=new-dp&icid=aolcom40vanity&ncid=emlcntaolcom00000010) [...] 1148 58 38_Problem with RTS reorgleafnear/v8 z/os12_Backes, Mike21_Mike.Backes@OA.MO.GOV30_Mon, 8 Dec 2008 10:45:40 -0600482_us-ascii We have serveral indexes that have been reorganized within the last month or so, that are now maxing out the integer type Of reorgleafnear. Not sure what could be causing this and I was wondering if anyone else had any similar experience on what is going on here?

I did not have any utilities or any other type of abends that would cause this or anything like that, and most of these are fairly inactive/small Indexes on smallish tablespaces.(non partitioned). [...] 1207 22 22_DB2 9 - It's Not a Gas10_Jorg Lueke23_jorg_h_lueke@BCBSMN.COM30_Mon, 8 Dec 2008 16:58:53 +0000449_utf-8 DB2 went from V8 to just plain 9. I think because they didn't want the upcoming version to have the unfortunate name of DB2 VX. Now they can just be cool with DB2 X. Will Professor X finally switch from Oracle?

______________________________________________________________________

* IDUG 2009 Denver, CO, USA * May 11-15, 2009 * http://IDUG.ORG/lsNA * ______________________________________________________________________ [...] 1230 44 46_terminate utility equivalent on open systems ?14_Bruce Lightsey25_Bruce.Lightsey@ITS.MS.GOV30_Mon, 8 Dec 2008 10:58:38 -0600514_US-ASCII DB2 v9.5 fp1 on AIX 5.3 64-bit

had a hangup over the weekend where Friday night's backup ( online, include logs ) was hung up waiting on ........ something. Saturday's backup was stalled , waiting on Friday's to finish.

When I went to stop Saturday's, I could not find anywhere on control center to stop it, could not find any command to terminate a particular utility from the command line, and no real help on how to whack an AWOL backup ( some days I really, really prefer z/OS ! ). [...] 1275 27 62_Re: Help me to monitoring jdbc application from DB2 zOs angle.10_Jorg Lueke23_jorg_h_lueke@BCBSMN.COM30_Mon, 8 Dec 2008 17:01:42 +0000410_utf-8 DB2 elapsed times, cpu times, ziip usage, and causes of waits. Nothing really different on a day to day basis than anyother connection method.

The primary issues we end up with is more bad access paths with dynamic SQL either because of the code or because correlated stats are required.

Then someone in the company needs to understand the connections, pooling, timeouts, network etc. [...] 1303 25 24_Re: DB2 pricing question16_Peter Adlersburg22_p.adlersburg@GMAIL.COM30_Mon, 8 Dec 2008 19:23:11 +0000790_utf-8 Gentlemen,

thanks for your input - that's something I can work with.

kind regards,

peter

______________________________________________________________________

* IDUG 2009 Denver, CO, USA * May 11-15, 2009 * http://IDUG.ORG/lsNA * ______________________________________________________________________





The IDUG DB2-L Listserv is only part of your membership in IDUG. The DB2-L list archives, FAQ, and delivery preferences are at http://www.idug.org/lsidug under the Listserv tab. While at the site, you can also access the IDUG Online Learning Center, Tech Library and Code Place, see the latest IDUG conference information and much more. If you have not yet signed up for Basic Membership in IDUG, available at no cost, [...] 1329 130 62_Re: Help me to monitoring jdbc application from DB2 zOs angle.15_Debabrata Ghosh21_the_ghosh@HOTMAIL.COM30_Mon, 8 Dec 2008 15:22:35 -0500407_iso-8859-1

101 data would not give you a slightest clue on what's going on on network, you will probably see a large class1 elapsed time, but you can use tool like sysco ethrol network monitoring tool to see end-to-end performance data. RMF mon III enclave data will reveal not of mvs related delay info. since you are saying that all your setups are ok, I would say that you focus on 2 things [...] 1460 110 62_Re: Help me to monitoring jdbc application from DB2 zOs angle.14_Grainger, Phil20_Phil.Grainger@CA.COM30_Mon, 8 Dec 2008 20:54:04 -0000445_iso-8859-1 or you could look at Wily Introscope (http://www.ca.com/us/application-management.aspx) - I think this gives the level of end to end data you are looking for

Phil Grainger CA

________________________________

From: DB2 Data Base Discussion List on behalf of Debabrata Ghosh Sent: Mon 08/12/2008 20:22 To: DB2-L@www.idugdb2-l.org Subject: Re: [DB2-L] Help me to monitoring jdbc application from DB2 zOs angle. [...] 1571 21 62_Re: Help me to monitoring jdbc application from DB2 zOs angle.10_Jorg Lueke23_jorg_h_lueke@BCBSMN.COM30_Mon, 8 Dec 2008 21:34:31 +0000766_utf-8 True, but the OP asked about monitoring from the DB2 z/OS point of view. The network is a wholly different beast and usually managed by people far hardier than those that deal with the database.

______________________________________________________________________

* IDUG 2009 Denver, CO, USA * May 11-15, 2009 * http://IDUG.ORG/lsNA * ______________________________________________________________________





The IDUG DB2-L Listserv is only part of your membership in IDUG. The DB2-L list archives, FAQ, and delivery preferences are at http://www.idug.org/lsidug under the Listserv tab. While at the site, you can also access the IDUG Online Learning Center, Tech Library and Code Place, see the latest IDUG conference [...] 1593 86 50_Re: terminate utility equivalent on open systems ?16_Gregory Palgrave28_gpalgrave@STAFF.IINET.NET.AU30_Tue, 9 Dec 2008 10:22:25 +0900375_us-ascii Sadly, they still haven't ported the excellent z/os utility handling to the LUW platform.

On LUW you've got a mix of db2 command line options:

LIST APPLICATIONS [SHOW DETAIL] LIST UTILITIES [SHOW DETAIL]

FORCE APPLICATION (...) - to kill the db2 thread you identified. Kill -9 from the *nix OS side if you know the pid it is running under. [...] 1680 207 37_Re: Recovery and datasharing question16_Robert Catterall21_rfcatterall@GMAIL.COM30_Mon, 8 Dec 2008 22:46:37 -0500579_ISO-8859-1 Hanne (and Chris),

I got some additional information on this topic that may be useful to you.

The directive to code the conditional restart control record so as to truncate all DB2 members back to the same ENDLRSN point is based on an assumption that you won't be recovering at the DR site using the current (at the time of the disaster event) active log data from the primary site (you WOULD be recovering using this data if you were mirroring the active log data sets at the DR site, but you've mentioned that you're exploring a scenario in which [...] 1888 361 55_Antwort: [DB2-L] Problem with RTS reorgleafnear/v8 z/os11_Roy Boxwell16_R.Boxwell@SEG.DE30_Tue, 9 Dec 2008 08:38:55 +0100674_iso-8859-1 This is a "known" problem in RTS....here is my little RTS APAR list that I keep handy

RTS APARS from IBM

APAR V7 V8 V9 Description PK35250 UK24354 UK24355 UK24356 Incorrect key numbers possible during log apply PK46792 UK27887 UK27888 UK27889 DBID's with 255 possible when using REPAIR PK49433 UK36450 UK31869 UK31870 REORGPSEDUODELETE is incorrect PK49823 UK28667 UK28668 UK28669 Bad NULL indicator handling PK51197 N/A UK28827 UK28828 SPACE can be negative if more than 524287 pages PK52434 UK29778 UK29779 UK29780 HIPER EXTENT sometimes not updated by shrlevel change reorg PK53341 UK32913 UK32914 UK32915 TOTALROWS or TOTALENTRIES < 0 or > [...] 2250 55 34_AW: [DB2-L] DB2 9 - It's Not a Gas0_26_Walter.Janissen@ITERGO.COM30_Tue, 9 Dec 2008 09:35:03 +0100582_iso-8859-1 Jorg

Can you explain, why VX is an unfortunate name? In Germany DB2 9 is a unfortunate name, because it sounds like: "DB2 no"

Mit freundlichen Grüßen Walter Janißen

ITERGO Informationstechnologie GmbH Anwendungsentwicklung Laufzeitarchitektur Victoriaplatz 2 40198 Düsseldorf mailto:walter.janissen@itergo.com

Vorsitzender des Aufsichtsrats: Jürgen Vetter Geschäftsführung: Dr. Bettina Anders (Vorsitzende), Dr. Christian Nymphius, Dr. Michael Regauer, Wolfgang Schön. Sitz: Düsseldorf, Handelsregister: Amtsgericht Düsseldorf, HRB 37996 [...] 2306 32 35_Re: DB2 V8 cm going to ENFM and NFM12_Gareth Jones19_jonesgth@UK.IBM.COM30_Tue, 9 Dec 2008 09:36:50 +0000561_UTF-8 My advice would be that, as fall back from ENFM or NFM to CM is not supported in V8, you should use the full volume backup, including catalog, directory, logs, BSDS, user table spaces etc. Once you have started updating objects, and then start recovering the catalog/directory to a prior point in time, you make things very complicated for yourself. You need to identify what would cause you to want to fall back to CM, and how long after entering ENFM/NFM this could be. In test this may not be so important, but in production it is as you probably [...] 2339 101 46_AW: [DB2-L] AW: [DB2-L] DB2 9 - It's Not a Gas12_Peter, Georg15_G.Peter@DZBW.DE30_Tue, 9 Dec 2008 11:10:44 +0100718_iso-8859-1 >>>>>In Germany DB2 9 is a unfortunate name, because it sounds like: "DB2 no" <<<

Well said, Walter.

It really sounds like "DB2 nein" ;-))

With kind regards - mit freundlichen Gruessen, G e o r g H . P e t e r ------------------------------------------------------------------- Datenzentrale Baden-Wuerttemberg Development and Product Support (E3) Krailenshaldenstrasse 44, 70469 Stuttgart, Germany, Europe e:mail G.Peter@dzbw.de Phone 0049-711-8108-271 PC-Fax 004971189696071 Internet (only in german language):http://www.dzbw.de ---------------------------------------------------------------------- This e-mail is environment friendly and was made only from recycled electrons. [...] 2441 264 51_Antwort: [DB2-L] AW: [DB2-L] DB2 9 - It's Not a Gas11_Roy Boxwell16_R.Boxwell@SEG.DE30_Tue, 9 Dec 2008 11:26:08 +0100382_iso-8859-1 because V is 5 and X is 10 it could be taken as DB2 15 (Or if you really remember your Roman numeracy rules 5!) which is not what is intended!!





Roy Boxwell SOFTWARE ENGINEERING GMBH -Product Development- Robert-Stolz-Straße 5 40470 Düsseldorf/Germany Tel. +49 (0)211 96149-675 Fax +49 (0)211 96149-32 Email: R.Boxwell@seg.de http://www.seg.de [...] 2706 265 51_Antwort: [DB2-L] AW: [DB2-L] DB2 9 - It's Not a Gas11_Roy Boxwell16_R.Boxwell@SEG.DE30_Tue, 9 Dec 2008 11:27:15 +0100653_iso-8859-1 or look here if you think its really gas.....

http://en.wikipedia.org/wiki/VX_(nerve_agent)



Roy Boxwell SOFTWARE ENGINEERING GMBH -Product Development- Robert-Stolz-Straße 5 40470 Düsseldorf/Germany Tel. +49 (0)211 96149-675 Fax +49 (0)211 96149-32 Email: R.Boxwell@seg.de http://www.seg.de

Software Engineering GmbH Amtsgericht Düsseldorf, HRB 37894 Geschäftsführung: Siegfried Fürst, Gerhard Schubert









Walter.Janissen@ITERGO.COM Gesendet von: DB2 Data Base Discussion List 09.12.2008 09:35 Bitte antworten an DB2 Database Discussion list at IDUG [...] 2972 34 33_unusual Error message in db2diag.16_sheetal munjewar19_smunjewar@YAHOO.COM30_Tue, 9 Dec 2008 04:25:36 -0800830_us-ascii Hi,

unusual message I have seen in db2diag.log, Any pointer will be appreciated.

db2level : db2 V8 FP14 OS : AIX

Error message :

MESSAGE : DIA3220W Client sent data from IP address = "xx.xx.xx.xxx", port = "3796".

sheetal--







______________________________________________________________________

* IDUG 2009 Denver, CO, USA * May 11-15, 2009 * http://IDUG.ORG/lsNA * ______________________________________________________________________





The IDUG DB2-L Listserv is only part of your membership in IDUG. The DB2-L list archives, FAQ, and delivery preferences are at http://www.idug.org/lsidug under the Listserv tab. While at the site, you can also access the IDUG Online Learning Center, Tech Library and Code Place, see [...] 3007 25 28_Re: It depends - Latin Motto12_Gareth Jones19_jonesgth@UK.IBM.COM30_Tue, 9 Dec 2008 14:53:33 +0000787_UTF-8 Hello Bill,

The motto is possibly a bit dull, being one word: Pendet.

Regards,

Gareth

______________________________________________________________________

* IDUG 2009 Denver, CO, USA * May 11-15, 2009 * http://IDUG.ORG/lsNA * ______________________________________________________________________





The IDUG DB2-L Listserv is only part of your membership in IDUG. The DB2-L list archives, FAQ, and delivery preferences are at http://www.idug.org/lsidug under the Listserv tab. While at the site, you can also access the IDUG Online Learning Center, Tech Library and Code Place, see the latest IDUG conference information and much more. If you have not yet signed up for Basic Membership in IDUG, available at no cost, [...] 3033 29 75_[z/OS] Any book available for exam 737 DB2 9 System Administrator for z/OS?15_Jorge Martelanz27_jorge_martelanz@HOTMAIL.COM30_Tue, 9 Dec 2008 14:55:42 +0000763_utf-8 Hi

I would like to know if there is (or will be) available a book for exam 737 DB2 9 System Administrator for z/OS.

I am looking for a book focused on the exam, not DB2 manuals.

Thanks

Jorge

______________________________________________________________________

* IDUG 2009 Denver, CO, USA * May 11-15, 2009 * http://IDUG.ORG/lsNA * ______________________________________________________________________





The IDUG DB2-L Listserv is only part of your membership in IDUG. The DB2-L list archives, FAQ, and delivery preferences are at http://www.idug.org/lsidug under the Listserv tab. While at the site, you can also access the IDUG Online Learning Center, Tech Library and Code Place, see [...] 3063 53 37_Re: unusual Error message in db2diag.13_Ian Bjorhovde23_ian.bjorhovde@GMAIL.COM30_Tue, 9 Dec 2008 08:35:47 -0700615_ISO-8859-1 It would be useful if you posted the entire entry from db2diag.log that contains this message. Otherwise it's nearly impossible to gauge what this is.





On Tue, Dec 9, 2008 at 5:25 AM, sheetal munjewar wrote: > Hi, > > unusual message I have seen in db2diag.log, Any pointer will be appreciated. > > db2level : db2 V8 FP14 > OS : AIX > > Error message : > > MESSAGE : DIA3220W Client sent data from IP address = "xx.xx.xx.xxx", port = > "3796". > > sheetal-- > > > > > ______________________________________________________________________ > > * IDUG 2009 [...] 3117 109 50_Re: terminate utility equivalent on open systems ?13_Ian Bjorhovde23_ian.bjorhovde@GMAIL.COM30_Tue, 9 Dec 2008 08:41:10 -0700339_ISO-8859-1 If you're backing up to disk, force application will usually work. However, if you're backing up to storage manager (TSM, NetBackup, etc) you will generally have to ask the storage admin to kill the session from their admin console.

*Do not `kill -9' a DB2 engine process* -- this will take the whole instance down. [...] 3227 74 79_Re: [z/OS] Any book available for exam 737 DB2 9 System Administrator for z/OS?14_Sevetson, Phil22_PSevetson@FISA.NYC.GOV30_Tue, 9 Dec 2008 11:51:07 -0500686_us-ascii Jorge,

You'd be looking for the study guide? "DB2 9 for z/OS Database Administration: Certification Study Guide (Paperback)" by Susan Lawson and Daniel Luksetich

http://www.amazon.com/DB2-OS-Database-Administration-Certification/dp/15 83470743/ref=sr_1_1?ie=UTF8&s=books&qid=1228841277&sr=8-1

or, if that doesn't work for you, http://tinyurl.com/63td3p



-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Jorge Martelanz Sent: Tuesday, December 09, 2008 9:56 AM To: DB2-L@www.idugdb2-l.org Subject: [DB2-L] [z/OS] Any book available for exam 737 DB2 9 System Administrator for z/OS? [...] 3302 83 38_Re: AW: [DB2-L] DB2 9 - It's Not a Gas11_Bear, Brian23_Brian.Bear@CHARMING.COM30_Tue, 9 Dec 2008 13:29:11 -0500452_iso-8859-1 I'm sticking with ... DB2 A



-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Walter.Janissen@ITERGO.COM Sent: Tuesday, December 09, 2008 3:35 AM To: DB2-L@www.idugdb2-l.org Subject: [DB2-L] AW: [DB2-L] DB2 9 - It's Not a Gas

Jorg

Can you explain, why VX is an unfortunate name? In Germany DB2 9 is a unfortunate name, because it sounds like: "DB2 no" [...] 3386 110 38_Re: AW: [DB2-L] DB2 9 - It's Not a Gas19_Jardine, Lawrence J19_JardineLJ@AETNA.COM30_Tue, 9 Dec 2008 13:39:14 -0500545_iso-8859-1 I like it, but one step further:

I'm guessing they could have prevented all past and future issues by using hex: DB2 x'08', DB2 x'09', DB2 x'0a', etc.

Larry Jardine Production DBA DBA Plan/Product Requests: https://wiki.aetna.com/confluence/display/dbaplan/Requests

-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Bear, Brian Sent: Tuesday, December 09, 2008 1:29 PM To: DB2-L@www.idugdb2-l.org Subject: Re: [DB2-L] AW: [DB2-L] DB2 9 - It's Not a Gas [...] 3497 110 38_Re: AW: [DB2-L] DB2 9 - It's Not a Gas14_Sevetson, Phil22_PSevetson@FISA.NYC.GOV30_Tue, 9 Dec 2008 13:59:42 -0500330_utf-8 DB2 1010 (Base 2)? DB2 101 (Base 3)? DB2 22 (Base 4)? DB2 12 (Base 8)

-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Bear, Brian Sent: Tuesday, December 09, 2008 1:29 PM To: DB2-L@www.idugdb2-l.org Subject: Re: [DB2-L] AW: [DB2-L] DB2 9 - It's Not a Gas [...] 3608 127 79_Re: [z/OS] Any book available for exam 737 DB2 9 System Administrator for z/OS?20_Bruzdzinski, Michael27_Michael.Bruzdzinski@SSA.GOV30_Tue, 9 Dec 2008 14:22:02 -0500381_us-ascii The ""DB2 9 for z/OS Database Administration: Certification Study Guide (Paperback)" by Susan Lawson and Daniel Luksetich is great for administrator test (730? 732?) for z/OS but is not designed for the system administrator test. It does have some information regarding system parameters but does not focus on the activities germane to the system administrator test. [...] 3736 98 38_Re: AW: [DB2-L] DB2 9 - It's Not a Gas10_Jorg Lueke23_jorg_h_lueke@BCBSMN.COM30_Tue, 9 Dec 2008 19:27:53 +0000337_utf-8 Walter,

That is why you should keep German clean and not let all those English phrases and words slip in. My father would be most upset if you said Dee Bee Two Nein instead of Deh Beh Zwei Neun.

VX is often referred to as VX gas. It was World War II chemical agent that is quite unpleasant and nowadays illegal. [...] 3835 150 38_Re: AW: [DB2-L] DB2 9 - It's Not a Gas10_Roger Hecq18_Roger.Hecq@UBS.COM30_Tue, 9 Dec 2008 15:00:53 -0500484_us-ascii VX nerve agent is a nasty product and its use is banned. It was not used in WWII, however. It was developed, in the UK, in the early 1950s.





Roger Hecq MF IB USA DB Support 203-719-0492 / 19-337-0492

-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Jorg Lueke Sent: Tuesday, December 09, 2008 2:28 PM To: DB2-L@www.idugdb2-l.org Subject: Re: [DB2-L] AW: [DB2-L] DB2 9 - It's Not a Gas [...] 3986 54 79_Re: [z/OS] Any book available for exam 737 DB2 9 System Administrator for z/OS?13_Willie Favero21_wfavero@ATTGLOBAL.NET30_Tue, 9 Dec 2008 21:06:55 -0600603_UTF-8 I was told a book is in plan for the 737 Cert test... not sure when it will be available though.

Willie

Jorge Martelanz wrote: > Hi > > I would like to know if there is (or will be) available a book for exam 737 DB2 9 > System Administrator for z/OS. > > I am looking for a book focused on the exam, not DB2 manuals. > > Thanks > > Jorge > > ______________________________________________________________________ > > * IDUG 2009 Denver, CO, USA * May 11-15, 2009 * http://IDUG.ORG/lsNA * > ______________________________________________________________________ > > > > The IDUG [...] 4041 154 55_Re: Antwort: [DB2-L] AW: [DB2-L] DB2 9 - It's Not a Gas12_Isaac Yassin20_yassini@BEZEQINT.NET31_Wed, 10 Dec 2008 07:37:48 +0200599_iso-8859-1 Hi,

VX is a nickname for lethal nerve gas

Isaac Yassin

---- Original message ---- >Date: Tue, 9 Dec 2008 11:26:08 +0100 >From: Roy Boxwell >Subject: [DB2-L] Antwort: [DB2-L] AW: [DB2-L] DB2 9 - It's Not a Gas >To: DB2-L@www.idugdb2-l.org > > because V is 5 and X is 10 it could be taken as DB2 > 15 (Or if you really remember your Roman numeracy > rules 5!) which is not what is intended!! > > Roy Boxwell > SOFTWARE ENGINEERING GMBH > -Product Development- > Robert-Stolz-Straße 5 > 40470 Düsseldorf/Germany > Tel. +49 (0)211 96149-675 > Fax [...] 4196 466 55_Antwort: Re: [DB2-L] AW: [DB2-L] DB2 9 - It's Not a Gas11_Roy Boxwell16_R.Boxwell@SEG.DE31_Wed, 10 Dec 2008 07:03:09 +0100438_iso-8859-1 as always the best things come from the UK!!! BSE, VX, Modern map of the wordl, World Chaos, etc. etc. however I don not think DB2 came from the UK or was there a UK developer on board "back then"...





Roy Boxwell SOFTWARE ENGINEERING GMBH -Product Development- Robert-Stolz-Straße 5 40470 Düsseldorf/Germany Tel. +49 (0)211 96149-675 Fax +49 (0)211 96149-32 Email: R.Boxwell@seg.de http://www.seg.de [...] 4663 123 21_Re: Read DB2 UDB logs13_Laila hosaini19_lshosaini@YAHOO.COM30_Tue, 9 Dec 2008 22:35:12 -0800361_iso-8859-1 HI IN unix ,i don't know ,but on z/os you can start trace on STATMENTS. START TRACE (PERFM) CLASS(3) IFCID(063) DEST(SMF)

although there is lots of information but you can extract statment by RECTRACE from gathered data.I gues on unix will be the same .

L.S.H

--- On Fri, 12/5/08, Deepak Goyal wrote: [...] 4787 97 62_Re: Help me to monitoring jdbc application from DB2 zOs angle.13_Laila hosaini19_lshosaini@YAHOO.COM30_Tue, 9 Dec 2008 22:47:48 -0800375_iso-8859-1 Hi Dear Renzo

As we have a jdbc application using type 4 driver ,I prefered to select a proper IFCID to monitor details on applications.If you review IFCID list you can find it. I can tell you more details if you need. Now i have a question ,Have you noticed the cursors remain open? what have you done about the open cursors? Thanks in advance. L.S.H [...] 4885 52 54_CF questions when recovering a db2 sysplex setup at DR0_17_ttaylor@CHUBB.COM31_Wed, 10 Dec 2008 08:41:22 -0500763_US-ASCII HI all

We are about to mirror just dasd from our prod site to our dr site... How is a DB2 recover handled when pages in the coupling facility are not at the dr site.

Tom





______________________________________________________________________

* IDUG 2009 Denver, CO, USA * May 11-15, 2009 * http://IDUG.ORG/lsNA * ______________________________________________________________________





The IDUG DB2-L Listserv is only part of your membership in IDUG. The DB2-L list archives, FAQ, and delivery preferences are at http://www.idug.org/lsidug under the Listserv tab. While at the site, you can also access the IDUG Online Learning Center, Tech Library and Code Place, see the latest IDUG [...] 4938 25 52_Re: Why is a REBIND necessary when changing LOCKSIZE12_Adam Baldwin22_adambaldwin@ES.IBM.COM31_Wed, 10 Dec 2008 14:17:05 +0000356_utf-8 ....I'm still waiting for a detailed explanation as to why the rebind is necessary when the granularity of locksize is increased.... Why does an alter in one direction require a rebind whilst in the other it's not necessary. I've continued looking elsewhere for an explanation but without success. Maybe someone has responded to Douwe off list? [...] 4964 106 52_Re: Why is a REBIND necessary when changing LOCKSIZE11_Mark Kimble25_mark.a.kimble@US.HSBC.COM31_Wed, 10 Dec 2008 08:24:35 -0600413_US-ASCII I haven't seen one either. The one thought I had was that standard lock escalation types of processes/code can be used for the row to page, page to table changes. There is no standard process to go the other direction.

Mark A Kimble Consultant - Database Management | HSBC Technology & Services 100 Mittel Drive, Wood Dale, IL 60191 __________________________________________________________ [...] 5071 64 52_Re: Why is a REBIND necessary when changing LOCKSIZE14_Grainger, Phil20_Phil.Grainger@CA.COM31_Wed, 10 Dec 2008 14:35:15 -0000283_utf-8 It looks like one of these "internal workings of DB2 are not really open for discussion" type of questions. And, to be fair, I can understand why IBM are reluctant to allow too much of the internals of DB2 to be discussed in public (otherwise there'd be a DB2 POP manual) [...] 5136 229 58_Re: CF questions when recovering a db2 sysplex setup at DR15_Douwe van Sluis21_d.b.van.sluis@HOME.NL31_Wed, 10 Dec 2008 15:49:52 +0100589_us-ascii Tom,

This IS handled by DB2 ;-). It would be to much to explain by mail how exactly everything works in detail. So take a look in the DB2 documentation about Datasharing Recovery. - SG24-7322 - DB2 Datasharing in a Nutshell - SG24-4791 - DB2 V4 Data Sharing Implementation (old, but most info still holds) - SG24-6370 - Disaster Recovery with DB2 for z/OS (Chapter 21) - DB2 Manual "Datasharing Planning and Administration. - And don't forget the DB2 Information Center http://publib.boulder.ibm.com/infocenter/dzichelp/v2r2/topic/com.ibm.db2 .doc.dshare/bjnzspsh.htm [...] 5366 270 24_Re: Including 14_Kannan Perumal21_db2l.kannan@GMAIL.COM31_Wed, 10 Dec 2008 11:15:51 -0500556_ISO-8859-1 Thank you all for your replies, I appreciate your time. We found out other options and okay to live with that for now. Thanks

On 12/5/08, Guido Verbraak wrote: > > > Hi > > Try James´option first. In the past however I received files that after > using FTP still had the problem you describe. There are two options to fix > the problem with these hidden characters. > The first option is to write a REXX program that fixes the problem. > The second option, the one I used, is using FTP to tranfser the files to [...] 5637 99 21_Re: Read DB2 UDB logs23_Jeffrey Benner, DB2 DBA22_consulting@EBENNER.COM31_Wed, 10 Dec 2008 10:27:36 -0600522_ISO-8859-1 I believe there is a db2audit option that tracks DML actions. Otherwise this is not routinely logged.

On Thu, Dec 4, 2008 at 15:53, Deepak Goyal wrote:

> Hi, > > I would like to know, how we can read the DB2 UDB logs means need to find > out the DML operations happened on a database during some timeframe. > > How can we do this ? Please do let me know the solution that I can use it > now > (means no tool solution, if tool is free then I would prefer that ). I have [...] 5737 30 79_Re: [z/OS] Any book available for exam 737 DB2 9 System Administrator for z/OS?15_Jorge Martelanz27_jorge_martelanz@HOTMAIL.COM31_Wed, 10 Dec 2008 16:41:49 +0000463_utf-8 Thanks Michael

I have found that the Study Guides are a good way to organize your actual knowledge of the subject focused on the exam structure and questions.

I learned that knowledge of the subject and experience alone is not enough to pass an exam, you have to study and practice on the questions. That part is very well covered in the Study Guides, but of course you can find the same base information in the DB2 manuals and redbooks. [...] 5768 24 79_Re: [z/OS] Any book available for exam 737 DB2 9 System Administrator for z/OS?15_Jorge Martelanz27_jorge_martelanz@HOTMAIL.COM31_Wed, 10 Dec 2008 16:43:36 +0000773_utf-8 Please give us a shout if you learn something new about the book.

Thanks

Jorge

______________________________________________________________________

* IDUG 2009 Denver, CO, USA * May 11-15, 2009 * http://IDUG.ORG/lsNA * ______________________________________________________________________





The IDUG DB2-L Listserv is only part of your membership in IDUG. The DB2-L list archives, FAQ, and delivery preferences are at http://www.idug.org/lsidug under the Listserv tab. While at the site, you can also access the IDUG Online Learning Center, Tech Library and Code Place, see the latest IDUG conference information and much more. If you have not yet signed up for Basic Membership in IDUG, available at no cost, [...] 5793 59 52_Re: Why is a REBIND necessary when changing LOCKSIZE14_Chad A. Walmer19_Cwalmer@RITEAID.COM31_Wed, 10 Dec 2008 11:49:27 -0500607_us-ascii Here is my understanding taken directly from an explanation provided by Bonnie Baker:

When you alter from a locksize of page to table/tablespace, the package is invalidated and will auto rebind (if you have that enabled) to pick up the new locksize. When you alter from a locksize of table/tablespace to page, the package is not invalidated and it continues to lock at the tablespace level. The reasoning behind this is the larger lock size ensures that there is not an integrity exposure (although you may have a concurrency problem.) But when you go from page to tablespace, you need [...] 5853 444 41_ARCHLOG Datasets that are Migrated by HSM9_Hale, Bob20_Bob.Hale@ROUNDYS.COM31_Wed, 10 Dec 2008 11:12:54 -0600306_us-ascii We send our Archlogs to dasd and then let HSM migrate them to tape. DB2 is setup to retain the logs for 15 days. HSM was setup to retain them for 14 days. Problem is that the migrated files never get deleted. Has anyone else run into this issue? Or does DB2 delete the logs by volser number? [...] 6298 32 52_Re: Why is a REBIND necessary when changing LOCKSIZE12_Adam Baldwin22_adambaldwin@ES.IBM.COM31_Wed, 10 Dec 2008 17:43:18 +0000570_utf-8 I know that this information is stored in the SPT and I understand Bonnie's reasoning but it still leaves me with a question. You go from page to tablespace and no binds / rebinds are necessary - I'm talking about explicit rather than auto. But when you go from tablespace to page, only those packages that are explicitly rebound use the new lock size. What I question is that this leaves a situation where LOCKRULE in SYSTABLESPACE can only be guaranteed to apply to those packages that have been rebound since the last alter, and, as not all alters require [...] 6331 31 34_DB2 for z/OS v9 migration question18_Sandra Lakenburger27_sandra.lakenburger@USDA.GOV31_Wed, 10 Dec 2008 18:44:13 +0000535_utf-8 I recently ran the premigration job (DSNTIJP9) on all of my v8 NFM subsystems and I have a questions concerning the plans identified as not having been bound since/before v7. Before I go to my application DBAs, what I'd like to know is if a plan was last bound prior to v7 and is strictly package bound (no dbrms bound directly to it), should it still be a candidate for rebinding as part of the migration plan or as long as the packages it contains are bound on a recent release, is it ok? I noticed I've got several plans [...] 6363 67 52_Re: Why is a REBIND necessary when changing LOCKSIZE15_Douwe van Sluis21_d.b.van.sluis@HOME.NL31_Wed, 10 Dec 2008 20:01:45 +0100321_US-ASCII Hi Adam,

No answer so far that explain why. The question that still puzzles me is "why is LOCKSIZE kind of locked into the bind process". I always try to explain things by the way DB2 is working, but this one leaves me puzzled.

If I get an answer off-list, I will post the answer on the list. [...] 6431 65 52_Re: Why is a REBIND necessary when changing LOCKSIZE14_Chad A. Walmer19_Cwalmer@RITEAID.COM31_Wed, 10 Dec 2008 15:29:00 -0500550_us-ascii Adam, I suspect you'll never get the reason that it was coded this way unless the developer speaks up (assuming they are even around or remember :)). I think you might be getting too caught up in the way it works. Once you understand that it is stored in the static package, then it doesn't really matter what is in the catalog (except for dynamic queries and I suspect those that are placed in the dynamic statement cache may work the same way as well.) So, if you ever change the locksize and you want to guarantee that everything is [...] 6497 26 35_DB2 LUW optimization level question12_Deepak Goyal20_goyal.mnit@GMAIL.COM31_Wed, 10 Dec 2008 22:19:42 +0000785_utf-8 Hi,

Do we have to specify any minimum query optimization level so that DB2 can consider using MQT ?

For example, DFT_QUERYOPT should be 5 or more to DB2 to consider MQT.

Thanks

______________________________________________________________________

* IDUG 2009 Denver, CO, USA * May 11-15, 2009 * http://IDUG.ORG/lsNA * ______________________________________________________________________





The IDUG DB2-L Listserv is only part of your membership in IDUG. The DB2-L list archives, FAQ, and delivery preferences are at http://www.idug.org/lsidug under the Listserv tab. While at the site, you can also access the IDUG Online Learning Center, Tech Library and Code Place, see the latest IDUG conference information and [...] 6524 140 52_Re: Why is a REBIND necessary when changing LOCKSIZE7_Ed Long19_rdhm99a@PRODIGY.NET31_Wed, 10 Dec 2008 15:04:30 -0800440_us-ascii To me, I don't see why locksize is stored in the package at all. Page size isn't and it has, arguably, more of an impact on performance than locksize.

The current design, assuming our understanding is correct, violates the basic design principal of transparency. This is true, in my opinion, because the catalog may not contain complete and accurate information about this rather important parameter yet it appears to. [...] 6665 43 38_Re: DB2 for z/OS v9 migration question12_Roger Miller19_millerrl@US.IBM.COM31_Thu, 11 Dec 2008 01:27:24 +0000281_utf-8 There are a few constructs for the plan header, and those change a little with almost every version. This one does not make much of a difference, but then it does not cost much to get the later format for the header, rather than do the work whenever the plan is loaded. [...] 6709 90 39_Re: DB2 LUW optimization level question23_Jeffrey Benner, DB2 DBA22_consulting@EBENNER.COM31_Wed, 10 Dec 2008 19:45:49 -0600592_ISO-8859-1 I use a MQT as an object of a direct query and as a result the optimization level doesn't matter. Otherwise I agree with your recitation of the documentation - that is what it says.

On Wed, Dec 10, 2008 at 16:19, Deepak Goyal wrote:

> Hi, > > Do we have to specify any minimum query optimization level so that DB2 can > consider using MQT ? > > For example, DFT_QUERYOPT should be 5 or more to DB2 to consider MQT. > > Thanks > > ______________________________________________________________________ > > * IDUG 2009 Denver, CO, USA * May [...] 6800 33 52_Re: Why is a REBIND necessary when changing LOCKSIZE12_Adam Baldwin22_adambaldwin@ES.IBM.COM31_Thu, 11 Dec 2008 07:43:48 +0000335_utf-8 Thanks for all of the various comments. I may be getting too caught up in this but then I like to know how and why things work as they do. One of our clients requested and alter of locksize from ANY to ROW and they wanted to know why the change implicated a REBIND..... and I wasn't satisfied with saying "Because it does". [...] 6834 35 35_Derek Moffett is out of the office.0_17_DMoffett@TSYS.COM31_Thu, 11 Dec 2008 04:01:30 -0500684_US-ASCII I will be out of the office starting 12/11/2008 and will not return until 01/05/2009.

I will respond to your message when I return.

----------------------------------------- The information contained in this communication (including any attachments hereto) is confidential and is intended solely for the personal and confidential use of the individual or entity to whom it is addressed. If the reader of this message is not the intended recipient or an agent responsible for delivering it to the intended recipient, you are hereby notified that you have received this communication in error and that any review, dissemination, copying, or unauthorized use [...] 6870 79 52_Re: Why is a REBIND necessary when changing LOCKSIZE19_Jardine, Lawrence J19_JardineLJ@AETNA.COM31_Thu, 11 Dec 2008 07:58:02 -0500391_US-ASCII I think you need to alter your view of the catalog. There are several values in the catalog that may or may not reflect reality depending external factors (like REORG or RUNSTATS). A few examples from SYSIBM.SYSTABLEPART:

Column When the value reflects reality PQTY after REORG SQTY true for the "next" extent; not past extents CARDF after runstats COMPRESS after REORG [...] 6950 134 47_DB2 z/OS DB2 9 CM - preformatted vs. used pages35_=?iso-8859-1?Q?Kurtz=2C_R=FCdiger?=28_Ruediger.Kurtz@HUK-COBURG.DE31_Thu, 11 Dec 2008 14:01:30 +0100874_iso-8859-1 Hi,

I'm posting this on behalf of a colleague.

When executing the copy-utility, will only the used pages be copied or the preformatted ones as well ? In other words, are preformatted pages considered to be "used" - even if they do not contain data ?

Thanks in advance

Ruediger

Abteilung Informatik Betrieb

HUK-COBURG Bahnhofsplatz 96444 Coburg Telefon 09561 96-44147 Telefax 09561 96-44104 E-Mail Ruediger.Kurtz@HUK-COBURG.de Internet www.HUK.de ============================================================= HUK-COBURG Haftpflicht-Unterstützungs-Kasse kraftfahrender Beamter Deutschlands a. G. in Coburg Reg.-Gericht Coburg HRB 100; St.-Nr. 9212/101/00021 Sitz der Gesellschaft: Bahnhofsplatz, 96444 Coburg Vorsitzender des Aufsichtsrats: Werner Strohmayr. Vorstand: Rolf-Peter Hoenen (Sprecher), Wolfgang Flaßhoff, [...] 7085 85 32_DB2 z/OS DB2 8 RECOVERY QUESTION14_Neff Stephen R22_Stephen.R.Neff@IRS.GOV31_Thu, 11 Dec 2008 10:00:26 -0500371_us-ascii Hello, We have deleted from the syscopy an image copy that we would now like to use to recover with. My question is if the image copy no longer exist in the syscopy can you use the IBM RECOVER utility? I saw where you can preallocate copy data set in the documentation but have not found an example of this. Even if you do so will the RECOVER utility work? [...] 7171 24 45_Re: ARCHLOG Datasets that are Migrated by HSM18_Sandra Lakenburger27_sandra.lakenburger@USDA.GOV31_Thu, 11 Dec 2008 15:11:07 +0000803_utf-8 If you're writing your archives to dasd, you'll need to scratch them manually once they pass their expiration date. It doesn't happen automatically the way it does with tape.

Sandra Lakenburger USDA-NITC

______________________________________________________________________

* IDUG 2009 Denver, CO, USA * May 11-15, 2009 * http://IDUG.ORG/lsNA * ______________________________________________________________________





The IDUG DB2-L Listserv is only part of your membership in IDUG. The DB2-L list archives, FAQ, and delivery preferences are at http://www.idug.org/lsidug under the Listserv tab. While at the site, you can also access the IDUG Online Learning Center, Tech Library and Code Place, see the latest IDUG conference information and much [...] 7196 22 38_Re: DB2 for z/OS v9 migration question18_Sandra Lakenburger27_sandra.lakenburger@USDA.GOV31_Thu, 11 Dec 2008 15:12:31 +0000785_utf-8 Thanks Roger. That's what I needed to know.

Sandra Lakenburger USDA-NITC

______________________________________________________________________

* IDUG 2009 Denver, CO, USA * May 11-15, 2009 * http://IDUG.ORG/lsNA * ______________________________________________________________________





The IDUG DB2-L Listserv is only part of your membership in IDUG. The DB2-L list archives, FAQ, and delivery preferences are at http://www.idug.org/lsidug under the Listserv tab. While at the site, you can also access the IDUG Online Learning Center, Tech Library and Code Place, see the latest IDUG conference information and much more. If you have not yet signed up for Basic Membership in IDUG, available at no cost, click on Member Services [...] 7219 66 52_Re: Why is a REBIND necessary when changing LOCKSIZE9_Mike Bell21_mbell11a1@VERIZON.NET31_Thu, 11 Dec 2008 09:33:25 -0600415_US-ASCII What DB2 runs from is the directory - execution does not require the catalog at all. I assume you have never had a DBD get out of sync with the catalog - it was caused by some MVS paging issues - I think and then we got to spend some quality time with IBM support. Because it was a test subsystem, it was easier to drop and rebuild but we had to use REPAIR to rebuild the DBD to even drop everything. [...] 7286 327 36_Re: DB2 z/OS DB2 8 RECOVERY QUESTION14_Sevetson, Phil22_PSevetson@FISA.NYC.GOV31_Thu, 11 Dec 2008 10:48:16 -0500578_us-ascii I take it that when you say "the image copy no longer exists in the SYSCOPY table", you mean that you have run MODIFY RECOVERY and eliminated history that includes your desired image copy.





If it's an image copy taken with SHRLEVEL NONE or SHRLEVEL REFERENCE, you can use it as the source for an UNLOAD, then LOAD the target data. Alternately, if it's a copy of a single dataset, you can use DSN1COPY to overlay the tablespace with the old copy, then do an immediate REORG (this last is risky if you've ALTERed the table since the image copy [...] 7614 61 45_Re: ARCHLOG Datasets that are Migrated by HSM14_Sevetson, Phil22_PSevetson@FISA.NYC.GOV31_Thu, 11 Dec 2008 10:49:25 -0500422_us-ascii If you're writing to SMS-managed volumes, you can set the archives up as a class which SMS will delete after a specified age.

--Phil S.

-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Sandra Lakenburger Sent: Thursday, December 11, 2008 10:11 AM To: DB2-L@www.idugdb2-l.org Subject: Re: [DB2-L] ARCHLOG Datasets that are Migrated by HSM [...] 7676 337 58_Re: CF questions when recovering a db2 sysplex setup at DR16_Robert Catterall21_rfcatterall@GMAIL.COM31_Thu, 11 Dec 2008 10:54:46 -0500579_ISO-8859-1 I'd say that there are dirty pages in the GBP, just as there are dirty pages in DB2 buffer pools in a non-data sharing environment (or in a data sharing environment for page sets that aren't group buffer pool dependent). What we mean by "dirty" is "updated but not yet written to the associated page set on disk." Having the corresponding data changes recorded in the log doesn't mean that the pages aren't dirty - it just means that DB2 can perform the necessary actions upon restart to get the database objects on disk (tablespaces and indexes) in a consistent [...] 8014 506 36_Re: DB2 z/OS DB2 8 RECOVERY QUESTION13_Bell, Raymond20_Raymond_Bell@BMC.COM31_Thu, 11 Dec 2008 16:10:30 -0000360_us-ascii Or, if you have the right tools, you could insert an entry into Syscopy with the details of the missing image copy and then IBM Recover would be happy. Or you could use a clever Recover utility to specify an input image copy dataset that's not registered in Syscopy. So, depending on what tools you've got, you might still be able to use the IC. [...] 8521 588 36_Re: DB2 z/OS DB2 8 RECOVERY QUESTION14_Grainger, Phil20_Phil.Grainger@CA.COM31_Thu, 11 Dec 2008 16:13:36 -0000407_us-ascii There are ways to reinsert accidentally deleted rows back into SYSCOPY (CA Merge/Modify being one, I think BMC also have a similar solution)





Without that though, you are stuck with either DSN1COPY or Unload - IBM Recover will not use ANY image copy that is not registered in SYSCOPY





Phil Grainger

CA

Senior Principal Product Manager [...] 9110 41 24_IBM Unload/Load Question9_Sam Orner21_sorner@AESSUCCESS.ORG31_Thu, 11 Dec 2008 11:05:20 -0500435_us-ascii I am using IBM's unload and load utilities under DB2 V8. The unload places spaces (x'40') for a decimal nullable field in the output unload file, as well as the leading null indicator (x'FF'). When I use the load without using field specifications, it fails when it reads the decimal field data. If I use field specifications in the load, I.E, for a decimal column....NULLIF a certain position = x'FF'....the load works. [...] 9152 133 52_Re: Why is a REBIND necessary when changing LOCKSIZE7_Ed Long19_rdhm99a@PRODIGY.NET31_Thu, 11 Dec 2008 08:26:56 -0800626_us-ascii Even if those are off, it doesn't create a help desk event. If Locksize is off, it can and will.

Edward Long



--- On Thu, 12/11/08, Jardine, Lawrence J wrote:

> From: Jardine, Lawrence J > Subject: Re: [DB2-L] Why is a REBIND necessary when changing LOCKSIZE > To: DB2-L@www.idugdb2-l.org > Date: Thursday, December 11, 2008, 7:58 AM > I think you need to alter your view of the catalog. There > are several > values in the catalog that may or may not reflect reality > depending > external factors (like REORG or RUNSTATS). A few examples [...] 9286 94 28_Re: IBM Unload/Load Question14_Grainger, Phil20_Phil.Grainger@CA.COM31_Thu, 11 Dec 2008 16:30:01 -0000497_us-ascii Before we answer that, what do you want to have loaded when you unload one of these null decimal columns?

Phil Grainger CA Senior Principal Product Manager Phone: +44 (0)1753 577 733 Mobile: +44 (0)7970 125 752 eMail: phil.grainger@ca.com

Ditton Park Riding Court Road Datchet Slough SL3 9LL

CA plc a company registered in England and Wales under company registration number 1282495 with its registered office at the address set out above. VAT number 697904179. [...] 9381 56 16_Parallel Sysplex24_SUBSCRIBE DB2-L Muthuraj24_muthuraj1983@YAHOO.CO.IN31_Thu, 11 Dec 2008 17:01:43 +0000421_UTF-8 Hi All,

In our shop, We have two MVS Parallel sysplexes each consisting of 2 LPARs (Total 4 LPARS). We have DB2 Datasharing Group consisting of 4 members in each MVS sysplex. DB2 Datasharing group name, Member names , Datasets names for Catalogs, BSDS, Active and Archive logs are same in both sysplexes. Application object names (viz. DB, Tables, Underlying Datasets ) are distinct in the two systems. [...] 9438 170 36_Re: DB2 z/OS DB2 8 RECOVERY QUESTION20_Bruzdzinski, Michael27_Michael.Bruzdzinski@SSA.GOV31_Thu, 11 Dec 2008 12:18:44 -0500743_us-ascii I think you best bet would be to use DSN1COPY at this point.



Michael Bruzdzinski

Social Security Administration

DCS/OESAE/DDBS/DBSB

410-966-7185







________________________________

From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Neff Stephen R Sent: Thursday, December 11, 2008 10:00 AM To: DB2-L@www.idugdb2-l.org Subject: [DB2-L] DB2 z/OS DB2 8 RECOVERY QUESTION



Hello, We have deleted from the syscopy an image copy that we would now like to use to recover with. My question is if the image copy no longer exist in the syscopy can you use the IBM RECOVER utility? I saw where you can preallocate copy data set [...] 9609 104 20_Re: Parallel Sysplex22_Leslie Pendlebury-Bowe21_db2dba@BTINTERNET.COM31_Thu, 11 Dec 2008 17:38:49 -0000448_US-ASCII Hi Merge is as it says Merge .. data from one to the other and your option 1 covers that.

Option 2 - yes you can do it.

You need of course to ensure that the vsam datasets are renamed, iefssn updated etc to reflect the new member names ... and so on and so forth. But as for a BSDS point of view you can actually modify the BSDS by way of a byte by byte update that will change the group name form one group to another. [...] 9714 35 20_Re: Parallel Sysplex24_SUBSCRIBE DB2-L Muthuraj24_muthuraj1983@YAHOO.CO.IN31_Thu, 11 Dec 2008 17:54:51 +0000570_UTF-8 Hi Leslie, I have read some redbook which talks about renaming the member and the std procedure to be followed. But I have had a question about renaming the datasharing group name. As far as I know, DSNJU003 doesnt provide any option to change the datasharing group name in BSDS. then i was thiknking of using IDCAMS, replace the old datasharing group name with new one, or create a new subsystem with new datasharing group name,use that BSDS to create the BSDS for the old datasharing group. in this case, i might loose the logs and all...after all these, I [...] 9750 35 35_Re: DB2 TS with Extended definition10_Nicola Nur17_nick_nur@YAHOO.CA31_Thu, 11 Dec 2008 11:49:16 -0800435_us-ascii Hi colleagues

Does anybody have experience or think there will be an issue for us when migration from DB2 z/OS v7 to v8 because of the following situation: 1) We have in production several compressed tablespaces that are defined as DSSIZE 64G. Our DFSMS level is 1.9.0.

Is there any issue that I should be aware of or watch for or test or worry about it during the v8 migration because of this definition ? [...] 9786 23 39_Re: DB2 LUW optimization level question12_Deepak Goyal20_goyal.mnit@GMAIL.COM31_Thu, 11 Dec 2008 20:22:16 +0000463_utf-8 I am just curious but why would you like to query MQT directly ? If I have to query MQT directly means changing my query but without changing my query the only way to use MQT is through query optimization only.

Thanks

______________________________________________________________________

* IDUG 2009 Denver, CO, USA * May 11-15, 2009 * http://IDUG.ORG/lsNA * ______________________________________________________________________ [...] 9810 87 39_Re: DB2 LUW optimization level question23_Jeffrey Benner, DB2 DBA22_consulting@EBENNER.COM31_Thu, 11 Dec 2008 15:13:36 -0600300_ISO-8859-1 I use a materialized view because a single query was run with very high frequency at very high performance cost. By changing a view to a MQT I heavily reduced the cost of running this query on the system by eliminating several expensive joins between several tables with static data. [...] 9898 151 24_DB2 Enhancement requests11_Suresh Sane21_data_arch@HOTMAIL.COM31_Thu, 11 Dec 2008 15:48:07 -0600472_Windows-1252

I am writing on behalf of the IDUG Requirements Committee.

As some of you may be aware, IDUG premier members can submit requirements and enhancement requests to IBM. They are first approved by a Requirements Committee, after that, they are opened for discussion in the IDUG requirements discussion forum and then opened for voting. They are then delivered to IBM for their evaluation. Finally, IBM may choose to implement the requirement. [...] 10050 25 31_MQT question for db2 v8 on z/OS12_Deepak Goyal20_goyal.mnit@GMAIL.COM31_Thu, 11 Dec 2008 22:16:16 +0000802_utf-8 Hi

Does anyone know about the MAINTYPE and REFSHAGE zparm ? We have a configuration where MAINTYPE=USER and REFSHAGE=ANY. What does this mean ? Will this configuration pick any MQT ( user/system) ?

Thanks

______________________________________________________________________

* IDUG 2009 Denver, CO, USA * May 11-15, 2009 * http://IDUG.ORG/lsNA * ______________________________________________________________________





The IDUG DB2-L Listserv is only part of your membership in IDUG. The DB2-L list archives, FAQ, and delivery preferences are at http://www.idug.org/lsidug under the Listserv tab. While at the site, you can also access the IDUG Online Learning Center, Tech Library and Code Place, see the latest IDUG conference information [...] 10076 39 45_Re: ARCHLOG Datasets that are Migrated by HSM14_James Campbell25_jacampbell@ACSLINK.NET.AU31_Thu, 11 Dec 2008 22:48:46 +0000321_utf-8 I suspect that when you write "DB2 is setup to retain the logs for 15 days" you mean you have ARCRETN=15 - or, in JCL terms, RETPD=15. These do not mean "keep for 15 days" - they actually mean "it expires on the 15'th day". ARCRETN/RETPD=1 means a dataset can be deleted today, not keep for the rest of today. [...] 10116 28 35_Re: MQT question for db2 v8 on z/OS24_SUBSCRIBE DB2-L Muthuraj24_muthuraj1983@YAHOO.CO.IN31_Fri, 12 Dec 2008 01:59:48 +0000508_UTF-8 Maintype and Refresh age will be used by AQR to decide MQT should be considered when deciding the accesspath. Both Zparms are related. Maintype can have System, user,all and none. Refresh age can have 0 or any. If refresh age is 0, no MQT will be considered. if refresh is any, then which MQTs should be considered will be decided by Maintype Zparm. Hope value itself is self explanatory. before all these, these values will be effective only for the MQTs defined with Enable optimization Clause. [...] 10145 28 52_Re: Why is a REBIND necessary when changing LOCKSIZE12_Adam Baldwin22_adambaldwin@ES.IBM.COM31_Fri, 12 Dec 2008 07:21:59 +0000588_utf-8 Ok, ok, to stop a deluge of comments re the catalog / directory I accept that my comment was a bit "wooly". Obviously stats etc are not necessarily accurate... My comment re locksize is that we're talking about an attribute defined with DDL. In the case of statistics we know that a plan/package bound prior to the current statistics wont be using them. In the case of a tablespace with LOCKSIZE altered to give more granularity, we don't know which plans/packages bound before the ALTEREDTS are using the current locksize. (Just by looking at the catalog). It's a moot point. [...] 10174 261 13_Perform Magic16_Bednarz, Michael24_michael.bednarz@CITI.COM31_Fri, 12 Dec 2008 09:40:25 -0000623_iso-8859-1 Hi folks and SQL experts,

Here is the magic SQL:

SET CURRENT SQLID = 'SYSDB'; DROP TABLE TTEST; (Only when exist) COMMIT; CREATE TABLE TTEST ( F1 CHAR(1)); GRANT SELECT ON TTEST TO PUBLIC; COMMIT; GRANT INSERT ON TTEST TO PUBLIC; COMMIT; GRANT UPDATE ON TTEST TO PUBLIC; COMMIT; SET CURRENT SQLID = 'HUGO'; CREATE VIEW VTEST AS SELECT * FROM SYSDB.TTEST; SET CURRENT SQLID = 'SYSDB'; (Maybe obsolet) (Didn't change the result) GRANT SELECT ON TTEST TO HUGO; (Maybe obsolet) (Didn't change the result) SELECT * FROM HUGO.VTEST; (SQLCODE 100) SET CURRENT SQLID = 'SYSDB'; REVOKE SELECT ON TTEST [...] 10436 492 17_Re: Perform Magic13_Bell, Raymond20_Raymond_Bell@BMC.COM31_Fri, 12 Dec 2008 09:53:30 -0000321_iso-8859-1 I imagine there's a few people that can, but you got me at a good time. The authority required to create a view is simply Select on the objects the view is on. If that Select privilege is revoked, so are your rights to have created that view, so it gets implicitly dropped. Can make for exciting times... [...] 10929 49 17_Re: Perform Magic16_Peter Adlersburg22_p.adlersburg@GMAIL.COM31_Fri, 12 Dec 2008 09:57:39 +0000576_utf-8 Hi,

As soon as you revoke the privilege from the group (in your case PUBLIC) under which a view has been created, the view is dropped implicitly. At the time of view-creation HUGO runs the CREATE as a member of the group PUBLIC. The following SQL-code works properly:

SET CURRENT SQLID = 'SYSDB'; DROP TABLE TTEST ; COMMIT ; CREATE TABLE TTEST ( F1 CHAR(1)); COMMIT ; GRANT SELECT ON TTEST TO HUGO; COMMIT ; GRANT SELECT ON TTEST TO PUBLIC; COMMIT; GRANT INSERT ON TTEST TO PUBLIC; COMMIT; GRANT UPDATE ON TTEST TO PUBLIC; COMMIT; SET CURRENT SQLID = [...] 10979 370 25_AW: [DB2-L] Perform Magic0_26_Walter.Janissen@ITERGO.COM31_Fri, 12 Dec 2008 11:25:07 +0100325_iso-8859-1 Hi Michael

Nice to hear from you

DB2 is working as usual. If a user has select-privilege on a table, he can create views which he owns. If the select privilege is revoked, the view will be dropped. This is working I think since the beginning of DB2

Mit freundlichen Grüßen Walter Janißen [...] 11350 141 17_Re: Perform Magic28_Balachandran Chandrasekaran119_balaccha@IN.IBM.COM31_Fri, 12 Dec 2008 16:16:48 +0530498_ISO-8859-1 This is an example of timestamp based privileges. Your Grant to public helped in the creation of HUGO. So, when you revoked the GRANT from public, the view got dropped though you had granted SELECT privilege for HUGO explicitly (Note: timestamp of this grant is latter than the Creation of the HUGO.TTEST view). I think this is the way DB2 has been working, checks the grantor and timestamp always. If your GRANT on HUGO had preceeded the CREATE VIEW, the view could have survived. [...] 11492 292 30_Antwort: [DB2-L] Perform Magic11_Roy Boxwell16_R.Boxwell@SEG.DE31_Fri, 12 Dec 2008 12:18:43 +0100484_iso-8859-1 no magic! Works as designed....here is the part of the SQL manual under REVOKE

SELECT Revokes the privilege to create a view or read data from the specified table or view. A view or a materialized query table is dropped when the SELECT privilege that was used to create it is revoked, unless the owner of the view or materialized query table was directly granted the SELECT privilege from another source before the view or materialized query table was created. [...] 11785 331 34_Antwort: Re: [DB2-L] Perform Magic11_Roy Boxwell16_R.Boxwell@SEG.DE31_Fri, 12 Dec 2008 12:19:30 +0100637_iso-8859-1 and I thought I was fast at typing!!! Raymond is superschnell!





Roy Boxwell SOFTWARE ENGINEERING GMBH -Product Development- Robert-Stolz-Straße 5 40470 Düsseldorf/Germany Tel. +49 (0)211 96149-675 Fax +49 (0)211 96149-32 Email: R.Boxwell@seg.de http://www.seg.de

Software Engineering GmbH Amtsgericht Düsseldorf, HRB 37894 Geschäftsführung: Siegfried Fürst, Gerhard Schubert









"Bell, Raymond" Gesendet von: DB2 Data Base Discussion List 12.12.2008 10:53 Bitte antworten an DB2 Database Discussion list at IDUG [...] 12117 621 38_Re: Antwort: Re: [DB2-L] Perform Magic13_Bell, Raymond20_Raymond_Bell@BMC.COM31_Fri, 12 Dec 2008 11:28:35 -0000531_iso-8859-1 I knew my Pitmans Elementary Typing Certificate would come in handy one day. For ridicule, mostly, but that's another story. Still, at the time there were 40 students and one bloke (yours truly) so, as a colleague of mine would say, do the math!









Raymond





From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Roy Boxwell Sent: 12 December 2008 11:20 To: DB2-L@www.idugdb2-l.org Subject: [DB2-L] Antwort: Re: [DB2-L] Perform Magic [...] 12739 148 25_Undertanding Access Paths16_Jair Montealegre18_jairmz@HOTMAIL.COM31_Fri, 12 Dec 2008 12:24:57 +0000 12888 69 43_AW: [DB2-L] DB2 TS with Extended definition0_26_Walter.Janissen@ITERGO.COM31_Fri, 12 Dec 2008 13:43:41 +0100601_iso-8859-1 Hi Nick

We only have two tablespace defined with DSSIZE 8G, but we haven't experienced any problems, when we migrated two years or so to V8 NFM.

Mit freundlichen Grüßen Walter Janißen

ITERGO Informationstechnologie GmbH Anwendungsentwicklung Laufzeitarchitektur Victoriaplatz 2 40198 Düsseldorf mailto:walter.janissen@itergo.com

Vorsitzender des Aufsichtsrats: Jürgen Vetter Geschäftsführung: Dr. Bettina Anders (Vorsitzende), Dr. Christian Nymphius, Dr. Michael Regauer, Wolfgang Schön. Sitz: Düsseldorf, Handelsregister: Amtsgericht Düsseldorf, HRB 37996 [...] 12958 223 42_Antwort: [DB2-L] Undertanding Access Paths11_Roy Boxwell16_R.Boxwell@SEG.DE31_Fri, 12 Dec 2008 15:18:55 +0100389_iso-8859-1 my gut reaction would be "size" - IO is also used in the access path and so accessing a "small" index and then getting data (which must also be done using the first index) is probably quicker (even if you add in a small in memory sort as well)

FEC_SALIDA has 35 distinct values and so with 123981 rows thats a lot of duplicates and possible a very small index io... [...] 13182 315 29_Re: Undertanding Access Paths22_Leslie Pendlebury-Bowe21_db2dba@BTINTERNET.COM31_Fri, 12 Dec 2008 15:21:58 -0000710_US-ASCII Hi

You might wish to restructure that email . see below.

Leslie





_____

From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Jair Montealegre Sent: 12 December 2008 12:25 To: DB2-L@www.idugdb2-l.org Subject: [DB2-L] Undertanding Access Paths





Hi guys, I am in an optimization group in my company and I have the following situation: There's a table that has 123,981 rows with the following indexes defined on it. P324.IWFMGPE1 COD_ESTADO 1 A FEC_SALIDA 2 A COD_MSWEMIB 3 A COD_VIAPAGO 4 A COD_DIISOALF 5 A IMP_VALOR 6 D P324.IWFMGPE2 COD_NUMBERRE 1 A COD_ESTADO 2 A P324.IWFMGPE3 FEC_SALIDA 1 A COD_ESTADO 2 [...] 13498 335 46_Re: Antwort: [DB2-L] Undertanding Access Paths16_Robert Catterall21_rfcatterall@GMAIL.COM31_Fri, 12 Dec 2008 10:14:59 -0500418_ISO-8859-1 I think that Roy is right on. Just because it seems to us that one access path would be "better" than another, more often than not (given rich and accurate catalog statistics) the path chosen by the DB2 optimizer is in fact the right one.

How many rows are in the result set? If it's not too many (e.g., a few hundred), the ORDER BY sort operation will complete very quickly and is no big deal. [...] 13834 40 35_Re: MQT question for db2 v8 on z/OS12_Deepak Goyal20_goyal.mnit@GMAIL.COM31_Fri, 12 Dec 2008 16:39:52 +0000519_utf-8 Thanks Muthu,

I have the following configuration on my dev system -

REFSHAGE = ANY MAINTYPE = USER

So as per manual, AQR should pick up any USER MQT but not SYSTEM MQT.

I did a simple test where I defined two MQTS (SYSTEM / USER) on the same table with enable query optimization. When I explained my sample SQL it picked up the SYSTEM MQT and when I dropped the SYSTEM MQT, it was not picking any MQT at all. So in brief it was not picking my USER MQT, that's surprised me a lot. [...] 13875 51 35_Re: MQT question for db2 v8 on z/OS13_David Simpson22_dsimpson@THEMISINC.COM31_Fri, 12 Dec 2008 12:41:57 -0500541_UTF-8 These zParms specify the default values for special registers CURRENT REFRESH AGE and CURRENT MAINTAINED TABLE TYPES FOR OPTIMIZATION (gotta love that one for being descriptive). It is the special registers for the session of the application that control the behavior of AQR (CURRENT REFRESH AGE determines if AQR is available, the other one determines which type of MQTs are eligible as described by Muthu below). The zParms control the default values for the special registers, but they may be changed at the application level. [...] 13927 53 54_Max storage for locks setting when upgrading to DB2 v813_Donna Domovic28_Donna.Domovic@EXELONCORP.COM31_Fri, 12 Dec 2008 17:52:23 +0000496_utf-8 Hello,

We're in the process of upgrading from DB2 v7 to DB2 v8 and I'm trying to figure out how to best estimate the value for the MLMT IRLM parameter (max storage for locks). I issued the modify command to display the current IRLM stats command for 2 of our systems with the following results but I don't know what to do from here to translate that in to the MLMT parameter value. I'd like to have an idea of if the default value of 2 will be large enough to handle our locks. [...] 13981 28 35_Re: MQT question for db2 v8 on z/OS12_Deepak Goyal20_goyal.mnit@GMAIL.COM31_Fri, 12 Dec 2008 18:13:10 +0000325_utf-8 Something I am not able to understand ...

We have a configuration where we have MAINTYPE = USER is defined on zparm. But when I queried the value for CURRENT MAINTAINED TABLE TYPES FOR OPTIMIZATION it returned me SYSTEM. I am not overwriting this value anywhere in my SQL. Then why it returned me SYSTEM ?? [...] 14010 29 24_DPSI question on z/os V812_Deepak Goyal20_goyal.mnit@GMAIL.COM31_Fri, 12 Dec 2008 21:05:22 +0000348_utf-8 Hi All,

We have a very huge 229 partitioned table (around 600 million rows) in our shop that have 5 NPIS defined on them. It's really very difficult to maintain the table especially reorg. Doing the reorg on NPI's takes lot of time so we are thinking of replacing the NPI's to DPSI. But would like to know the pros/cons for that. [...] 14040 89 28_Re: DPSI question on z/os V814_Grainger, Phil20_Phil.Grainger@CA.COM31_Fri, 12 Dec 2008 21:40:30 -0000521_iso-8859-1 Pros:

* DPSIs are partitioned so tend to be easier to manage (smaller datasets) * Under some circumstances you can get page range screening (if your SQL contains the partitioning key columns) * Partition level on-line reorgs do not have a BUILD" phase



Cons:

* DPSIs cannot be unique * Because DPSIs are partitioned they may compromise your SQL performance (keys are now not "in order of the DPSI key" but are "in order of the DPSI key WITHIN the order ofthe partitoning key") [...] 14130 166 28_Re: DPSI question on z/os V815_Kalena, Michael27_Michael.Kalena@JPMORGAN.COM31_Fri, 12 Dec 2008 18:55:32 -0500497_us-ascii One other thought that came up at the IBM Conference.

Consider if your queries will be able to limit the number of parts being referenced. Which in turn means it can limit the number of DPSI datasets to look at.

If the query is getting data from any/all partition and you have 229 DPSI datasets (229 partitions), the Optimizer would have to consider the cost of looking at them all. This may cause it to use another index or hurt performance vs. the NPI's performance. [...] 14297 61 35_Re: DB2 TS with Extended definition12_Roger Miller19_millerrl@US.IBM.COM31_Sat, 13 Dec 2008 01:49:43 +0000532_utf-8 That's not an area where we have seen any problems. If you think of the items where your shop is unique, those are great items to check. If you set zparms or have processes that the books indicate don't work. If you use unique products and have settings like no one else in the world, then only you can test them. If your DBAs and application programmers have chosen options no one else has thought or heard of, then that's what you need to test. If this is the most unique item on your list, then the shop tastes pretty [...] 14359 81 29_Re: Undertanding Access Paths12_Roger Miller19_millerrl@US.IBM.COM31_Sat, 13 Dec 2008 02:00:49 +0000300_utf-8 Even better would be using Visual Explain or OSC to gather all of the relevant information. You could check with the statistics advisor to see if information should be provided. If you think it's wrong, then you can use the service option to build an XML file with the needed information. [...] 14441 28 26_DB2 for z/OS books updated12_Roger Miller19_millerrl@US.IBM.COM31_Sat, 13 Dec 2008 02:21:50 +0000442_utf-8 Whether you like to RTFM or RTFW (read the friendly manual or read the friendly web), the books have just been updated. I counted about 20 refreshed books for DB2 9 and 4 key books for DB2 V8. I am updating my Information Center as I type this. http://publib.boulder.ibm.com/infocenter/imzic/

To pull individual books easily or get to the Information Center, try http://www.ibm.com/support/docview.wss?rs=64&uid=swg27011656 [...] 14470 116 58_Re: Max storage for locks setting when upgrading to DB2 v812_Roger Miller19_millerrl@US.IBM.COM31_Sat, 13 Dec 2008 02:45:32 +0000359_utf-8 2 gigabytes is more than you've had before. IRLM and DB2 can manage with applications that are not well behaved, but you do need the real memory, if applications are behaving very badly. If you do have the memory, then 3 would be adding 50% to what you had with PC=YES before. If you had PC=NO, then this provides a lot more than the prior MAXCSA. [...] 14587 28 29_Re: Undertanding Access Paths14_Peter Vanroose17_pvanroose@ABIS.BE31_Sat, 13 Dec 2008 12:36:03 +0000366_UTF-8 Roger Miller wrote:

> Even better would be using Visual Explain or OSC to gather all of > the relevant information. You could check with the statistics advisor > to see if information should be provided.

I thought that the Advisors were not part of the OSC, but only of the (paying) OE? Or did I miss something in OSC? [...] 14616 50 29_Re: Undertanding Access Paths13_Seibert, Dave26_Dave.Seibert@COMPUWARE.COM31_Sat, 13 Dec 2008 09:16:56 -0500330_us-ascii Hello Peter.

VE & Stats Advisor (for single query or workloads) are included in OSC; the other advisors are part of OE - query advisor, index advisor, access path advisor.

Our friend Pat Bossman has excellent presentations on the matter from IDUG NA & IOD this year. Get your hands on them if you can. [...] 14667 87 29_Re: Undertanding Access Paths14_Peter Backlund21_BacklundDB2@TELIA.COM31_Sat, 13 Dec 2008 16:24:56 +0100 14755 25 29_Re: Undertanding Access Paths24_SUBSCRIBE DB2-L Muthuraj24_muthuraj1983@YAHOO.CO.IN31_Sun, 14 Dec 2008 13:13:47 +0000797_utf-8 From my experience, DB2 Optimizer always tries to use the index in favor of some predicate which is restricting the more data retrieval tha Sorting.

I guess that applies here also.

Regards, Muthu

______________________________________________________________________

* IDUG 2009 Denver, CO, USA * May 11-15, 2009 * http://IDUG.ORG/lsNA * ______________________________________________________________________





The IDUG DB2-L Listserv is only part of your membership in IDUG. The DB2-L list archives, FAQ, and delivery preferences are at http://www.idug.org/lsidug under the Listserv tab. While at the site, you can also access the IDUG Online Learning Center, Tech Library and Code Place, see the latest IDUG conference information and [...] 14781 30 28_Re: DPSI question on z/os V824_SUBSCRIBE DB2-L Muthuraj24_muthuraj1983@YAHOO.CO.IN31_Sun, 14 Dec 2008 13:22:31 +0000497_utf-8 Deepak,

When converting from NPI to DPSI, there mighe be some restrictions like 'Not Unique'. But if you can convert it to DPSI, i would see there are many advantages from Data Maintenance perception. but there will be problems from performance point of view. I would recommend you to analyse the SQLs which are using the Indexes. Sometimes you might have to rewrite the SQLs to include the columns used in partitioning index to make sure that DPSI will be used in place of NPI. [...] 14812 19 35_Re: MQT question for db2 v8 on z/OS24_SUBSCRIBE DB2-L Muthuraj24_muthuraj1983@YAHOO.CO.IN31_Sun, 14 Dec 2008 13:29:59 +0000778_utf-8 May be ZPRAMS would have been changed dynamically.

______________________________________________________________________

* IDUG 2009 Denver, CO, USA * May 11-15, 2009 * http://IDUG.ORG/lsNA * ______________________________________________________________________





The IDUG DB2-L Listserv is only part of your membership in IDUG. The DB2-L list archives, FAQ, and delivery preferences are at http://www.idug.org/lsidug under the Listserv tab. While at the site, you can also access the IDUG Online Learning Center, Tech Library and Code Place, see the latest IDUG conference information and much more. If you have not yet signed up for Basic Membership in IDUG, available at no cost, click on Member Services at http://www.idug.org/lsms 14832 31 28_Re: DPSI question on z/os V814_Peter Vanroose17_pvanroose@ABIS.BE31_Sun, 14 Dec 2008 16:39:16 +0000410_UTF-8 On Fri, 12 Dec, Phil Grainger wrote:

> Pros: >* DPSIs are partitioned so tend to be easier to manage > (smaller datasets) > ...

Which implies an other (maybe less important) "Con": * DB2 will have to open *multiple* (smaller) datasets (up to 229 in your case) instead of one single dataset, especially for those queries that use the index but have no partition-constraining predicate. [...] 14864 86 28_Re: DPSI question on z/os V811_Nick Cianci19_ncianci@AU1.IBM.COM31_Mon, 15 Dec 2008 09:06:50 +1100345_ISO-8859-1 One word of advise ,and I'm sure it's been said, make sure that anything that uses the DPSI (static or Dynamic) specifies partitioning key (or at least a decent part of it) in the query. Otherwise you may end up with up to 229 (ie the number of partitions) index partitions being read, which will end up being a performance hit! [...]