1 WWW.IDUGDB2-L.ORG /home/listserv/home/db2-l November 2009, week 4 2 31 23_ADMIN:issues with DB2-l6_Robert29_robert.galambos@COMPUWARE.COM31_Sun, 22 Nov 2009 00:22:05 +0000325_UTF-8 some people, including myself, are still having issues with DB2-l.

f you are then you can access the posts, by going to the WEB interface at www.idug.org.

while this is not the best solution, it is a temporary one.

rest assured we are trying to track down what is causing the selective issue. [...] 34 25 42_Mike Hopkins/COR/AWG is out of the office.12_Mike Hopkins19_mhopkins@AWGINC.COM31_Sat, 21 Nov 2009 20:47:45 -0600669_US-ASCII I will be out of the office starting 11/20/2009 and will not return until 11/23/2009.

If you need immediate assistance, please e-mail edi@awginc.com.

_____________________________________________________________________

* IDUG North America * Tampa, Florida, * May 10-14 2010 * http://IDUG.ORG/NA * _____________________________________________________________________

http://www.idug.org/solutions-journal.html - home of the IDUG Solutions Journal Technical atricles from world famous authors in DB2's most prestigious, peer reviewed magazine now on-line! _____________________________________________________________________ [...] 60 39 27_Re: ADMIN:issues with DB2-l16_Brian K. Cocklin20_bcocklin@RITEAID.COM31_Sat, 21 Nov 2009 22:11:25 -0500417_utf-8 Thank you for the update.

-----Original Message----- From: IDUG DB2-L [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Robert Sent: Saturday, November 21, 2009 7:22 PM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] ADMIN:issues with DB2-l

some people, including myself, are still having issues with DB2-l.

f you are then you can access the posts, by going to the WEB interface at www.idug.org. [...] 100 46 35_Robert Badger is out of the office.13_Robert Badger24_robert.badger@AURORA.ORG31_Sat, 21 Nov 2009 22:12:59 -0600653_US-ASCII

I will be out of the office starting 11/20/2009 and will not return until 11/23/2009.

I will respond to your message when I return.

_____________________________________________________________________

* IDUG North America * Tampa, Florida, * May 10-14 2010 * http://IDUG.ORG/NA * _____________________________________________________________________

http://www.idug.org/solutions-journal.html - home of the IDUG Solutions Journal Technical atricles from world famous authors in DB2's most prestigious, peer reviewed magazine now on-line! _____________________________________________________________________ 147 384 60_Re: [z/OS V9] Default Behavior when no CLUSTER index exists?10_DB2sysprog21_bala.db2dba@GMAIL.COM31_Sun, 22 Nov 2009 15:20:07 +0530423_windows-1252 YES, the behaviour in discussion was implemented in v8. But, is this available in CM or only in NFM ? Anybody has experience taking advantage of this while in v8 CM itself ? we have got some hundreds of tables having implicit clustering index and reorgs are not doing the re-ordering. SQLs are suffering because of it. I don't have time to change those primary indexes into clustering while in DB2 v8 CM. [...] 532 193 29_Re: Listserv message slowdown11_Mike Turner19_mike.turner@GMX.COM31_Sun, 22 Nov 2009 10:51:51 -0000597_iso-8859-1 Listserv message slowdownThanks to all the Admins. I lost all messages for a week but this morning they are getting through OK.

Regards Mike Turner Email: mike.turner@gmx.com Tel: +44 (0)1565-873702 Web: www.michael-turner.ltd.uk Michael Turner Ltd. Registered in England & Wales No. 1433722. Registered Office: 131 King Street, Knutsford, Cheshire, WA16 6EJ. VAT No. GB 338 4159 44. ----- Original Message ----- From: Mark Labby Newsgroups: bit.listserv.db2-l To: DB2-L@WWW.IDUGDB2-L.ORG Sent: Saturday, November 21, 2009 5:41 PM Subject: [DB2-L] Listserv message slowdown [...] 726 489 60_Re: [z/OS V9] Default Behavior when no CLUSTER index exists?14_Peter Backlund21_backlunddb2@GMAIL.COM31_Sun, 22 Nov 2009 22:20:51 +0100616_windows-1252 I am a bit surprised about your comment "I don't have time to change those primary indexes into clustering while in DB2 v8 CM.

It is just an "Alter Index Creator.Indexname CLUSTER" and you are there...

Peter





On Sun, Nov 22, 2009 at 10:50 AM, DB2sysprog wrote:

> YES, the behaviour in discussion was implemented in v8. But, is this > available in CM or only in NFM ? Anybody has experience taking advantage of > this while in v8 CM itself ? we have got some hundreds of tables having > implicit clustering index and reorgs are not [...] 1216 542 60_Re: [z/OS V9] Default Behavior when no CLUSTER index exists?10_DB2sysprog21_bala.db2dba@GMAIL.COM31_Mon, 23 Nov 2009 12:10:02 +0530331_windows-1252 So, do I :-)

"It is just an "Alter Index Creator.Indexname CLUSTER" and you are there..."

Alter the Cluster is a v8 NFM feature. But, I found answer myself for my original question. The reorg seems to use implicit clustering index while in V8 CM itself. If that is wrong, someone please correct me. [...] 1759 143 17_Re: Reclaim Space4_Anil21_alisha_kale@YAHOO.COM31_Mon, 23 Nov 2009 07:17:58 +0000531_utf-8 Hi Al,

I can confirm to you that it doesn't have to do with what is specified as priqty. However, answering your question, the priqty is some cases was as low as 720. With the exception of a few, mostly all are MOD-9 DASDs. I did speak to a couple db2 folks from the labs, and they did say it is possible (the extents issue with reuse of previously used volume). Since I didn't get a conclusive and complete response to my question, I am wondering if it might help to contact folks from the z/os side. What say ? [...] 1903 56 54_General Question: Why abend in between DB2 utilities ?9_DB2DBAzOS21_bala.db2dba@GMAIL.COM31_Mon, 23 Nov 2009 12:49:24 +0530370_ISO-8859-1 Dear List,

I have looked at DB2 weekly/daily/monthly jobs that 'calls' the abort routine when the RC of the DB2 utility is unacceptable. Is there a special reason why this is being done ? On the scheduler acceptable code would have been set as 0 or <=4 and along with JCL COND parameter, the job could be forced to stop right after Utility error. [...] 1960 179 55_Presentations from Baltimore/Washington DB2 Users Group10_DB2usa !!!19_db2usa3@HOTMAIL.COM31_Mon, 23 Nov 2009 03:49:29 -0500467_iso-8859-1

Hi DB2 user,

Here is a non-profit Blog about DB2 for z/OS (IBM mainframes): http://db2usa.blogspot.com

Last update on Monday, November 23rd 2009



Here are several presentations from Baltimore/Washington DB2 Users Group, available on BWDB2UG website:





- Improving Performance with DB2 V9 for z/OS by David Beulke





- What's New in DB2 9 for Applications by Patrick Bossman [...] 2140 231 58_Re: General Question: Why abend in between DB2 utilities ?13_Michael Ebert18_mebert@AMADEUS.COM31_Mon, 23 Nov 2009 09:01:49 +0000533_ISO-8859-1 I'm not sure what "the abort routine" is. I had a step like this at the end of my SAS-generated maintenance JCL:

// IF RC>0 THEN // EXEC PGM=MYABEND // ENDIF

Previous job steps were e.g. a series of ICOPY steps. If any of these failed, I didn't want the whole job to fail, so I followed each image copy step with a cleanup step (TERM UTIL) that got executed only if the preceding RC was higher than 4/8, OR if it abended (which does NOT result in an RC). The cleanup step (a REXX) always had a high RC. [...] 2372 52 29_Re: v8.1 CM DBM1 Memory Creep13_Sheldon Davis21_sdavis@ISRACARD.CO.IL31_Mon, 23 Nov 2009 08:51:49 +0000741_utf-8 Hi Was this issue resolved ?

We have just migrated to DB2 V8 CM and are experiencing something similar.

When some DBM1's storage is paged out to aux storage the storage is never released.

From IFCID 225

Time Real storage Auxiliary storage 03:46:36 3576.5 859.8 03:51:36 3576.62 859.8 03:56:36 3576.56 859.8 04:01:36 3581.82 859.8 04:06:36 3577.42 859.8 04:11:36 2941.58 1366.91 04:16:36 2932.89 1376.19 04:21:36 2933.94 1376.19 04:26:36 2935.36 1376.19 04:31:36 2936.13 1376.19 04:36:39 2952.22 1376.19 04:41:39 3036.82 1376.07 04:46:39 3048.43 1375.95 04:51:39 3046.56 1375.95 04:56:39 3048.45 1375.93 05:01:39 3078.38 1375.87 05:06:39 3088.51 1375.87 05:11:45 3094.22 1375.69 05:16:45 3107.3 1375.69 [...] 2425 100 17_Re: Reclaim Space19_Sysdba_AHE/CORP/TPG34_Sysdba_AHE/CORP/TPG@GB.TNTPOST.COM31_Mon, 23 Nov 2009 09:31:02 +0000549_US-ASCII Hi Anil,

Unlike your other respondents, we have had this in the past.

Raising a z/OS question is a good idea. I think it's an MVS restriction, and probably applies equally to SMS.

AFAIK the part of a dataset on a particular volume has to consist of contiguously-numbered extents because of how VTOC entries chain volumes together. I think the volume list in the catalog (VVDS) also implies the sequence of volumes, and I don't believe you can repeat a volume serial number (apart from '*' for a candidate volume). [...] 2526 95 45_Changing limit-key on a Partition (revisited)11_Nick Cianci19_ncianci@AU1.IBM.COM31_Mon, 23 Nov 2009 23:24:16 +1100319_ISO-8859-1 Another sleepless night, and a pending repartitioning exercise has got me thinking.

The real hairy part of repartitioning is the ReOrg (beats me why it has to be SHR-NONE). If you can't shift the REORP flag you are "stuffed", once you get over that hill, a number of other possibilities open up. [...] 2622 44 20_DB2 for z/OS website20_Cuneyt Goksu (Gmail)22_cuneyt.goksu@GMAIL.COM31_Mon, 23 Nov 2009 15:00:22 +0200653_us-ascii This msg goes on behalf of Surekha Parekh from IBM.

Thanks, Cuneyt

---

Dear DB2 for z/OS Collegues & Friends

I am currently reviewing the DB2 for z/OS webpages - here is the URL

http://www-01.ibm.com/software/data/db2/zos/family/

I have created a simple survey - it should take you 5 mins please can you complete for me.

http://www.surveymonkey.com/s.aspx?sm=ix_2ba4DG56phkmh78AnoMNQ_3d_3d

Surekha Parekh DipM, MCIM. Chartered Marketer WW DB2 for z/OS Senior Market Manager/Strategist IBM Software Group Office +44 (0)1926 464080 Mobile +44 (0)7801 673018 Email:Surekha21@uk.ibm.com [...] 2667 431 13_ADMIN: UPDATE16_Galambos, Robert29_Robert.Galambos@COMPUWARE.COM31_Mon, 23 Nov 2009 09:34:36 -0500519_utf-8 Just an update.





Some people have been experiencing issues from receiving posts via email from the list, including myself.





Well it seems that I, for one, are once again, receiving emails.





Therefore, if anyone here is only able to see posts through the web interface, please let the support team know.





DB2-L-request@www.idugdb2-l.org





thanks





Robert Galambos CIPP/C CIPP/IT [...] 3099 110 49_Re: Changing limit-key on a Partition (revisited)13_Bell, Raymond20_Raymond_Bell@BMC.COM31_Mon, 23 Nov 2009 16:34:20 +0100563_iso-8859-1 Hey Nic,

You still in Canberra? Got access to any ISV Reorg products? To paraphrase Rolf Harris, 'Can you guess which one it is yet?'

It's possible at least one way to do a Shrlevel Change Reorg, altering the partitioning keys in the process. We (Oops. OK; I've been rumbled as to who the ISV is I have in mind) do eventually/briefly need exclusive access to the DBD to record the key range changes but that's not usually a problem. If it turns out to be, you need to cancel the task holding the lock (a batch job, long-running CICS [...] 3210 122 65_Re: ICF Catalog how to Split into Sub-Systems per IBM Recommended9_Ali, Omar25_Omar.Ali@CITYOFBOSTON.GOV31_Mon, 23 Nov 2009 10:39:28 -0500647_us-ascii Just an FYI concerning the ICF catalog, IBM is recommending that in the near future the BOOTSTRAPS, and LOGS be separated form the ICF Catalog moving to DB2 Next, there is a REDBOOK documenting the Processes, this is what my SYSTEM Programmers were asking us DBA's if anyone attempted this before? And how to go about separating this from the catalog I guest he is looking for suggestions before contacting IBM for advice Any suggestion Thanks in advance Omar -----Original Message----- From: IDUG DB2-L [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Ali, Omar Sent: Friday, November 20, 2009 8:23 AM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [...] 3333 95 22_RUNSTATS' Alternative?7_DB2 DBA21_the.db2.dba@GMAIL.COM31_Mon, 23 Nov 2009 13:36:07 -0500567_ISO-8859-1 Hello:

DB2 V7 z/OS 1.4

As we all are aware how resource consuming RUNSTATS is (particularly for large tables), is there any alternative for this with any of the vendors? Or is there a work around?

I am not only worried about the cost (don't blame me for this) but also the 'time'. The existing set up is to have RUNSTATS run on Sundays, when rest of the world 'sleeps'. THIS, usually takes around 4 hours. Now, it is being recommended that the set up be moved to the regular weekdays and try to 'fit it' in the maintenance window [...] 3429 13 63_Tomas Vasko/IEA/EAGA/EAG/AT ist au=?ISO-8859-1?Q?=DFer_?= Haus.11_Tomas Vasko23_tomas.vasko@GENERALI.AT31_Mon, 23 Nov 2009 19:45:45 +0100221_ISO-8859-1

Ich bin ab 23.11.2009 außer Haus und ab 26.11.2009 wieder im Büro. I'm out of the office from 23.11.2009 and will be back on 26.11.2009.

Ich werde Ihre Nachricht nach meiner Rückkehr beantworten. 3443 500 26_Re: RUNSTATS' Alternative?16_Rasmussen, Steen22_Steen.Rasmussen@CA.COM31_Mon, 23 Nov 2009 13:58:04 -0500330_us-ascii I'll leave out the vendor plug for now J

Why are you doing RUNSTATS every week ?

Are your tables really changing that much (cardinality etc.) - otherwise I would be more selective and only do RUNSTATS for those objects where you really had dramatic changes. You can use RTS to determine most of these. [...] 3944 35 34_DB2 Connect w/microsoft clustering10_John Lantz20_john.p.lantz@FRB.GOV31_Mon, 23 Nov 2009 19:10:13 +0000485_utf-8 We have DB2 Connect installed on a SQL Server machine so that it can use a linked server to retrieve data from DB2 on the mainframe. The system engineers want to implement Microsoft failover clustering. Is there anything special that needs to be done other then just having DB2 Connect installed on each machine? I've found stuff on InfoCenter that talks about how DB2 integrates with MS clustering by creating a resource type called DB2 Server. But I can't find much more. [...] 3980 548 26_Re: RUNSTATS' Alternative?14_Sevetson, Phil22_PSevetson@FISA.NYC.GOV31_Mon, 23 Nov 2009 14:12:50 -0500402_us-ascii Josh,

I frankly don't see why your firm finds it necessary to keep RUNSTATS current for all tables.

If you need to know what tables and indexes are becoming disorganized, Realtime Statistics will keep things up to date and you can build queries to find objects with too many index levels or large NEAROFFPOS and FAROFFPOS (I forget the exact names for the cognates in RTS). [...] 4529 48 49_Estimate EDM_SKELETON_POOL size for V9 migration?11_Gary Martin29_gary.martin@LIBERTYMUTUAL.COM31_Mon, 23 Nov 2009 19:36:05 +0000527_utf-8 We had a rude awakening after migrating a busy subsystem to V9 CM. Once the workload increased to prime-time levels, we saw performance degradation with these symptoms: • I/O delays on SPT01 • high # of PKG Loads We had relied on the install clist calculation for EDM_SKELETON_POOL of 10MB, but after a few painful hours of troubleshooting, increased it to 100MB. This made a significant improvement in reducing PKG Loads. We’ve subsequently increased it 120MB for another incremental performance improvement. [...] 4578 93 49_Re: Changing limit-key on a Partition (revisited)13_Mackey, Glenn20_GMackey@GUIDEONE.COM31_Mon, 23 Nov 2009 14:31:14 -0600297_iso-8859-1 Some other ideas..

1. If possible, redefine your own copy table and define to meet your needs. Then load the data "manually" and do a rename at the end of exercise. This makes assumptions on outage etc. But you never want to be left in a situation where you cannot recover. [...] 4672 35 50_Oracle Transparent Gateway for DB2 under z/OS 1.1113_Terry Banicki19_tbanicki@AMEREN.COM31_Mon, 23 Nov 2009 21:45:17 +0000396_utf-8 My first post and it's off topic. How embarrassing! Well at least it has DB2 in the name…

Is anyone still running the Oracle Transparent Gateway for DB2? Does anyone have it running under z/OS 1.11?

We are just starting a project to upgrade from z/OS 1.9 to 1.11 and we would like to know if there are any issues to be aware of. I have posted on the Oracle forum also. [...] 4708 130 58_Re: General Question: Why abend in between DB2 utilities ?9_DB2DBAzOS21_bala.db2dba@GMAIL.COM31_Tue, 24 Nov 2009 12:13:23 +0530560_ISO-8859-1 Hi Michael,

Thanks for the response with ideas. But, the 'abort' (just abend) I am talking about does nothing other than sending an abend code. So, I was thinking if there could be a reason, the return code or the abend code from DSNUTILB be unacceptable to stop the job.

On 11/23/09, Michael Ebert wrote: > I'm not sure what "the abort routine" is. I had a step like this at the > end of my SAS-generated maintenance JCL: > > // IF RC>0 THEN > // EXEC PGM=MYABEND > // ENDIF > > Previous job steps were e.g. [...] 4839 517 34_AW: [DB2-L] RUNSTATS' Alternative?35_Walter Jani=?WINDOWS-1252?Q?=DFen?=26_Walter.Janissen@ITERGO.COM31_Tue, 24 Nov 2009 08:04:04 +0100733_iso-8859-1 Very good answer, Phil.

Mit freundlichen Grüßen Walter Janißen

ITERGO Informationstechnologie GmbH Anwendungsentwicklung Laufzeitarchitektur Victoriaplatz 2 40477 Düsseldorf mailto:walter.janissen@itergo.com

Vorsitzender des Aufsichtsrates: Jürgen Vetter Geschäftsführung: Dr. Bettina Anders (Vorsitzende), Ina Kirchhof, Dr. Christian Nymphius, Dr. Michael Regauer, Wolfgang Schön. Sitz: Düsseldorf, Handelsregister: Amtsgericht Düsseldorf, HRB 37996





________________________________ Von: IDUG DB2-L [mailto:DB2-L@IDUGDB2-L.ORG] Im Auftrag von Sevetson, Phil Gesendet: Montag, 23. November 2009 20:13 An: DB2-L@WWW.IDUGDB2-L.ORG Betreff: Re: [DB2-L] RUNSTATS' Alternative? [...] 5357 42 62_AW: [DB2-L] Oracle Transparent Gateway for DB2 under z/OS 1.1112_PUSCH Othmar27_Othmar.Pusch@IT-AUSTRIA.COM31_Tue, 24 Nov 2009 09:32:26 +0100477_utf-8 Hi dear Terry !

o) Our shop is running DB2/V9 & z/OS-1.10 & OTGW (with WLM-ORACLE/OSDI-Subsystem 1.2.1.4) with no problems in two sysplex's with 4 db2_sharing-group's. We wanna to change to z/OS-1.11 too .. o) Oracle's Metalink was recently changed to 'my oracle support' --> right now I found no real docu-restriction's related to z/OS-1.11 & OTGW (OracleTransparentGateway). o) When ya get an answer from the oracle_forum ya'll please deploy it to me too ? [...] 5400 371 26_Re: RUNSTATS' Alternative?13_Bell, Raymond20_Raymond_Bell@BMC.COM31_Tue, 24 Nov 2009 09:41:42 +0100594_us-ascii Hi Josh,

If you're still on DB2 V7 and z/OS 1.4 I suspect you have bigger issues than the CPU consumption of Runstats. To hopefully accurately summarise some of what Mr. Sevetson quite rightly said, easiest thing might be simply to not run Runstats quite so often. After all, as has been used in a number of contexts, the fastest [insert action being discussed] is no [insert action being discussed]. For those remaining objects that do need Runstats, and for which reducing the Runstats job CPU/elapsed time is important, yes I believe there's at least one ISV that has a [...] 5772 45 26_Re: RUNSTATS' Alternative?12_Adam Baldwin22_adambaldwin@ES.IBM.COM31_Tue, 24 Nov 2009 10:09:21 +0000428_utf-8 Following on from what Phil and Raymond have said, you really need to ask the question "why do I want to run runstats?". As Phil stated there are specific scenarios when having current statistics is vital. He mentioned structure / business changes. Ideally you should have statistics as part of your standard change methodology in these cases - for example, following an alter to add columns with a reorg with stats. [...] 5818 98 26_Re: RUNSTATS' Alternative?10_Max Scarpa16_mscarpa@CESVE.IT31_Tue, 24 Nov 2009 11:59:34 +0100398_US-ASCII RTS is the first answer that came in my mind. But in a well-know DB2 environment (not always happens) every DBA knows if a table is growing fast or not or if it's stable in time.

But cardinality, after a table reach a 'knee' needs to increase a lot before, in general, providing a change in access path so running RUNSTATS could be useless. Running runstats for a million rows [...] 5917 58 54_[DB2 v8] Implicit Clustering index .. Still not done..9_DB2DBAzOS21_bala.db2dba@GMAIL.COM31_Tue, 24 Nov 2009 17:41:50 +0530450_ISO-8859-1 Hi List,

How does DB2 decide on implicit clustering ? Is it the OBID or ISOBID or the combination or the created_ts ? What I am seeing is that DB2 is picking the created_ts. But, there are two indexes that were created before V5 and hence the catalog has '0001-01-01-00.00.00.000000' . In such a case, how will the reorg know which one the implicit clustering index? Will it start looking at OBIDs next to created- timestamp ? [...] 5976 151 26_Re: RUNSTATS' Alternative?9_DB2DBAzOS21_bala.db2dba@GMAIL.COM31_Tue, 24 Nov 2009 17:45:42 +0530449_ISO-8859-1 Hi Josh,

Same situation here. Somebody (long time ago) wrote a small program to get ALL the tablespaces out of catalog and write Runstats control card for the monthly runstats job !

I am thinking of having RTS enabled for a week or month to see the amount of changes to cut down the analysis to half. Once done with trimming, I am planning to put the runstats into small jobs and run them in parallel on single Sunday. [...] 6128 312 55_DATATYPEID, TYPENAME and CREATEDTS in SYSIBM.SYSCOLUMNS12_Peter, Georg15_G.Peter@DZBW.DE31_Tue, 24 Nov 2009 14:40:32 +0100543_us-ascii Environment is DB2 for z/OS Version 8 CM,

-------------------------------------------------





Collegues,





given are two tables. In both tables is one column with the format VARCHAR FOR BIT DATA.

While comparing this two columns via SYSIBM.SYSCOLUMNS we see a difference in three cases:

The SQL was

select

substr(name,1, 12) as column, datatypeid, typename, createdts

from sysibm.syscolumns



where tbname in ( 'DEW61A' , 'DEW56A') [...] 6441 544 59_Re: DATATYPEID, TYPENAME and CREATEDTS in SYSIBM.SYSCOLUMNS13_Seibert, Dave26_Dave.Seibert@COMPUWARE.COM31_Tue, 24 Nov 2009 09:09:10 -0500530_utf-8 I'm not sure this answers all your questions, but the SQL Reference tells you that Datatypeid is zero if the table was created before V6. Similar info for typename and createdts.





This makes sense since Distinct Types were implemented first in V6.





Dave

From: IDUG DB2-L [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Peter, Georg Sent: Tuesday, November 24, 2009 8:41 AM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] DATATYPEID, TYPENAME and CREATEDTS in SYSIBM.SYSCOLUMNS [...] 6986 427 58_R: DATATYPEID, TYPENAME and CREATEDTS in SYSIBM.SYSCOLUMNS15_Moschelli Mauro34_mauro.moschelli@INTESASANPAOLO.COM31_Tue, 24 Nov 2009 15:11:50 +0100425_iso-8859-1 My guess is that the second column listed was created with an older DB2 Version where the three columns of SYSCOLUMNS didn't exist in the catalog (see default value in createdts).

HTH

Mauro Moschelli Intesa Sanpaolo Group Services

E-mail : mauro.moschelli@intesasanpaolo.com

IBM Certified Database Administrator - DB2 9 for z/OS IBM Certified System Administrator - DB2 9 for z/OS [...] 7414 411 59_AW: DATATYPEID, TYPENAME and CREATEDTS in SYSIBM.SYSCOLUMNS35_Walter Jani=?WINDOWS-1252?Q?=DFen?=26_Walter.Janissen@ITERGO.COM31_Tue, 24 Nov 2009 15:22:40 +0100611_iso-8859-1 Georg

My first guess is, these columns are fairly new and contain defaults for rows, which existed at the time these columns were appended.

Mit freundlichen Grüßen Walter Janißen

ITERGO Informationstechnologie GmbH Anwendungsentwicklung Laufzeitarchitektur Victoriaplatz 2 40477 Düsseldorf mailto:walter.janissen@itergo.com

Vorsitzender des Aufsichtsrates: Jürgen Vetter Geschäftsführung: Dr. Bettina Anders (Vorsitzende), Ina Kirchhof, Dr. Christian Nymphius, Dr. Michael Regauer, Wolfgang Schön. Sitz: Düsseldorf, Handelsregister: Amtsgericht Düsseldorf, HRB 37996 [...] 7826 486 59_Re: DATATYPEID, TYPENAME and CREATEDTS in SYSIBM.SYSCOLUMNS11_Mike Turner19_mike.turner@GMX.COM31_Tue, 24 Nov 2009 14:23:39 -0000733_iso-8859-1 NachrichtHi Georg

I suspect the table that contains column DEW61A_FNAME was created before Version 6. I found the following in the description of the SYSCOLUMNS column TYPENAME:

'TYPENAME is set only for columns created in Version 6 or later. The value for columns created earlier is not filled in'.

Regards Mike Turner Email: mike.turner@gmx.com Tel: +44 (0)1565-873702 Web: www.michael-turner.ltd.uk Michael Turner Ltd. Registered in England & Wales No. 1433722. Registered Office: 131 King Street, Knutsford, Cheshire, WA16 6EJ. VAT No. GB 338 4159 44. ----- Original Message ----- From: Peter, Georg Newsgroups: bit.listserv.db2-l To: DB2-L@WWW.IDUGDB2-L.ORG Sent: Tuesday, November 24, 2009 [...] 8313 293 18_Space issue in DB29_SrinivasG21_SRINIVASG@INFOSYS.COM31_Tue, 24 Nov 2009 20:32:55 +0530410_us-ascii Hi,

We are migrating an application that uses Oracle as Database to DB2. While the Data size in Oracle is 8 GB , when it is loaded on to DB2 , it uses up almost 28 3390-9 volumes. The Tablespaces are created using an automated script.

How do I know if the VSAM Datasets Created for the Table Spaces are fully utilized? If the Dataset is over allocated , how do I reclaim the space? [...] 8607 66 53_Re: Estimate EDM_SKELETON_POOL size for V9 migration?12_McLaren Phil32_Phil.McLaren@AXAWINTERTHUR.CO.UK31_Tue, 24 Nov 2009 15:09:22 +0000452_utf-8 Gary,

I can't say I relied on the install clist, but then I had the flexibility to go high on the value for the new EDM-SKEL_POOL value and tune afterwards. I ended up with an approximate split, halving the original EDMPOOL and giving the other half to the new SKEL_POOL. If I remember correctly, much of the new pool was not consumed until we rebound. Final results we ended up with were as below (for the main production subsystem) [...] 8674 294 22_Re: Space issue in DB214_Steurs Patrick21_Patrick.Steurs@NBB.BE31_Tue, 24 Nov 2009 16:27:41 +0100538_UTF-8 Hi,





Don't use space-parameters in db2. Let SMS do the job and allocate your vsam with primary & secondary quantity equal to -1





greetings,





Patrick





From: IDUG DB2-L [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of SrinivasG Sent: dinsdag 24 november 2009 16:03 To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] Space issue in DB2





Hi,





We are migrating an application that uses Oracle as Database to DB2. [...] 8969 490 67_AW: [DB2-L] DATATYPEID, TYPENAME and CREATEDTS in SYSIBM.SYSCOLUMNS12_Peter, Georg15_G.Peter@DZBW.DE31_Tue, 24 Nov 2009 16:29:17 +0100708_iso-8859-1 Thanks a lot to Dave, Walter, Mauro and Mike.

Now it is clear why we see this differences.......

Sometimes it should be good to have the manuals nearby... ;-))

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 ---------------------------------------------------------------------- P Think before you print. [...] 9460 866 71_Re: AW: [DB2-L] DATATYPEID, TYPENAME and CREATEDTS in SYSIBM.SYSCOLUMNS13_Seibert, Dave26_Dave.Seibert@COMPUWARE.COM31_Tue, 24 Nov 2009 11:03:47 -0500589_utf-8 The info center is your friend.





http://www-01.ibm.com/support/docview.wss?rs=64&uid=swg27011656





Dave

From: IDUG DB2-L [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Peter, Georg Sent: Tuesday, November 24, 2009 10:29 AM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] AW: [DB2-L] DATATYPEID, TYPENAME and CREATEDTS in SYSIBM.SYSCOLUMNS





Thanks a lot to Dave, Walter, Mauro and Mike.

Now it is clear why we see this differences.......

Sometimes it should be good to have the manuals nearby... ;-)) [...] 10327 339 22_Re: Space issue in DB29_SrinivasG21_SRINIVASG@INFOSYS.COM31_Tue, 24 Nov 2009 21:43:50 +0530342_utf-8 Thanks Patrick. I will try that.

Actually , I create SMS Routines and place Datasets on Certain storage groups based on VCAT qualifiers.

The scripts are run as follows : using VCAT and create STOGROUP stogroup-name Volume(‘*’) VCAT vcatname.

Is this fine? So now I give primary and secondary equal to -1 ? [...] 10667 403 22_Re: Space issue in DB214_Steurs Patrick21_Patrick.Steurs@NBB.BE31_Tue, 24 Nov 2009 17:32:34 +0100561_UTF-8 ok, but don't forget to adapt the zparm MGEXTSZ to the value "YES" + compile the DSNTIJUZ-module.





greetings,





Patrick Steurs





From: IDUG DB2-L [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of SrinivasG Sent: dinsdag 24 november 2009 17:14 To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] Space issue in DB2





Thanks Patrick. I will try that.





Actually , I create SMS Routines and place Datasets on Certain storage groups based on VCAT qualifiers. [...] 11071 407 22_Re: Space issue in DB216_Rasmussen, Steen22_Steen.Rasmussen@CA.COM31_Tue, 24 Nov 2009 11:37:21 -0500367_UTF-8 Using negative one for your allocation will make sure you don’t run out of extents until you hit the limit. You might get to a point where the secondary extent is so big that you waste some space (hence the name Sliding Scheme).

Since you seem to use a lot more DASD compared to Oracle – did you compress the objects (COMPRESS YES) in the DDL ? [...] 11479 250 58_Re: [DB2 v8] Implicit Clustering index .. Still not done..12_Hepp Shery C21_Shery.Hepp@SRPNET.COM31_Tue, 24 Nov 2009 09:41:41 -0700693_us-ascii Why would you want to use implicit clustering instead of specifying cluster on an index?





From: IDUG DB2-L [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of DB2DBAzOS Sent: Tuesday, November 24, 2009 5:12 AM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] [DB2 v8] Implicit Clustering index .. Still not done..





Hi List,





How does DB2 decide on implicit clustering ? Is it the OBID or ISOBID or the combination or the created_ts ? What I am seeing is that DB2 is picking the created_ts. But, there are two indexes that were created before V5 and hence the catalog has '0001-01-01-00.00.00.000000' . In such a case, how will [...] 11730 424 22_Re: Space issue in DB29_SrinivasG21_SRINIVASG@INFOSYS.COM31_Tue, 24 Nov 2009 22:15:38 +0530357_utf-8 Thanks a lot.

I will try this solution. But it will take time to implement since we have to make changes to all the scripts. We have 5+ subsystems.

From: IDUG DB2-L [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Steurs Patrick Sent: Tuesday, November 24, 2009 10:03 PM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] Space issue in DB2 [...] 12155 81 58_Re: [DB2 v8] Implicit Clustering index .. Still not done..12_Martin Hubel17_Martin@MHUBEL.COM31_Tue, 24 Nov 2009 16:50:52 -0500 12237 75 57_Free Webinar ! DB2 for z/OS optimizing INSERT performance14_Surekha Parekh20_surekha21@UK.IBM.COM31_Tue, 24 Nov 2009 23:39:17 +0000461_utf-8 Hi

Just a quick reminder that John Campbell - IBM DE will be presenting on 1st Dec 2009. Here are event details & URL to register

http://www-01.ibm.com/software/os/systemz/telecon/dec1/index.html? S_TACT=109AY53M&S_CMP=SurekhaIDUG

Broadcast date: December 1, 2009, 11 a.m., Eastern Standard Time, 4:00 UTC Developed for: IT managers, application architects, DBAs, system administrators, systems programmers and system architects [...] 12313 224 22_Re: Space issue in DB216_Gregory Palgrave28_gpalgrave@STAFF.IINET.NET.AU31_Wed, 25 Nov 2009 09:16:40 +0800307_utf-8 Once upon a time I would have known how big a 3390-9 was, but I have to admit I needed to look it up this time – roughly 8.5 GB.

That’s a huge difference in space usage from the Oracle db, and it makes me wonder if possibly the Oracle db is using row/table compression and DB2 is not? [...] 12538 25 42_Mike Hopkins/COR/AWG is out of the office.12_Mike Hopkins19_mhopkins@AWGINC.COM31_Tue, 24 Nov 2009 19:55:40 -0600678_US-ASCII I will be out of the office starting 11/24/2009 and will not return until 11/30/2009.

If you need immediate assistance, please e-mail edi@awginc.com.

_____________________________________________________________________

* IDUG North America * Tampa, Florida, * May 10-14 2010 * http://IDUG.ORG/NA * _____________________________________________________________________

http://www.idug.org/db2-videos.html has hundreds of video presentations! Did you miss out on attending an IDUG conference? Many of the presentations were recorded and are available on our website! _____________________________________________________________________ [...] 12564 25 41_Doug Holliday/Alcan is out of the office.13_Doug Holliday23_doug.holliday@ALCAN.COM31_Tue, 24 Nov 2009 21:06:02 -0500660_US-ASCII I will be out of the office starting 11/24/2009 and will not return until 11/25/2009.

I will respond to your message when I return.

_____________________________________________________________________

* IDUG North America * Tampa, Florida, * May 10-14 2010 * http://IDUG.ORG/NA * _____________________________________________________________________

http://www.idug.org/db2-videos.html has hundreds of video presentations! Did you miss out on attending an IDUG conference? Many of the presentations were recorded and are available on our website! _____________________________________________________________________ [...] 12590 248 58_Re: [DB2 v8] Implicit Clustering index .. Still not done..9_DB2DBAzOS21_bala.db2dba@GMAIL.COM31_Wed, 25 Nov 2009 10:39:46 +0530311_ISO-8859-1 Why would I care ?

I have a huge list of tablespaces that do not have explicit clustering index. Looks to me that only partitioning index (of v7 and back) have had Cluster and no other index has. I am trying to define (alter!) explicit clustering index (primary key index in most cases). [...] 12839 218 26_Re: RUNSTATS' Alternative?11_Roy Boxwell16_R.Boxwell@SEG.DE31_Wed, 25 Nov 2009 07:25:58 +0100621_ISO-8859-1 dont run RUNSTATS...very simple...and then only run one when you need to run one





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







DB2 DBA Gesendet von: IDUG DB2-L 23.11.2009 19:36 Bitte antworten an IDUG DB2-L [...] 13058 154 61_SV: [DB2-L] Estimate EDM_SKELETON_POOL size for V9 migration?13_Olle Brostrom25_olle.brostrom@SWEDBANK.SE31_Wed, 25 Nov 2009 08:12:14 +0100695_iso-8859-1 Gary and Phil, We also went through the same experience.

V8: EDMPOOL=176000,

V9 NFM when migrated and using DSNTINST calc values, bad performance in EDM_SKELETON: EDMPOOL=176000, EDM_SKELETON_POOL=34514,

After tuning we ended up with the following: EDMPOOL=32000, EDM_SKELETON_POOL=88000,

The EDMPOOL could be reduced from 176000K in V8 down to 32000K without any performance degradation. The problem we got was to monitor the EDMPOOL because our monitor (Mainview for DB2) do not signal any performance problems when reducing the EDMPOOL. Due to the high Free Pages in EDMPOOL we beleived we could change the EDMPOOL further down, but when we did we [...] 13213 49 26_Re: RUNSTATS' Alternative?14_Peter Vanroose17_pvanroose@ABIS.BE31_Wed, 25 Nov 2009 08:15:15 +0000606_UTF-8 A RUNSTATS frequency of once per week, taking 4 hours, is probably too frequent. This of course depends on the "stability" of your data (statistically speaking), but having a systematic (full) RUNSTATS once in three months is probably more than enough. And indeed, as others pointed out, just selectively do a RUNSTATS on those tables (and their indexes) that changed drastically (like e.g. a DDL change, or a serious data move, or an accumulatively large amount of INSERTs or UPDATEs). But in most of those cases, you most likely planned a REORG for those tables, which allows an (almost free, [...] 13263 477 22_Re: Space issue in DB214_Steurs Patrick21_Patrick.Steurs@NBB.BE31_Wed, 25 Nov 2009 09:23:11 +0100464_UTF-8 Hi,





Compression is also a nice feature. I use it only for tablespaces larger than 1 Gigabyte and when the space gained > 30% . ( see compression report )





Greetings,





Patrick Steurs





From: IDUG DB2-L [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Rasmussen, Steen Sent: dinsdag 24 november 2009 17:37 To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] Space issue in DB2 [...] 13741 751 58_Re: [DB2 v8] Implicit Clustering index .. Still not done..13_Phil Grainger30_philg@PHILGRAINGER.DEMON.CO.UK31_Wed, 25 Nov 2009 09:09:25 -0000533_US-ASCII Let's ask another question





How are you ACCESSING the data?





A clustering index is only relevant if it defines or supports some processing sequence





If you are only ever (or most of the time) accessing the data randomly then who cares what sequence it's in - DB2 certainly doesn't





However, if you have many (or perhaps a few time-critical) processes accessing the data in a given sequence, then that should be your clustering key [...] 14493 513 58_Re: [DB2 v8] Implicit Clustering index .. Still not done..9_DB2DBAzOS21_bala.db2dba@GMAIL.COM31_Wed, 25 Nov 2009 14:53:11 +0530531_windows-1252 Ok, I quit. I guess everybody is in the "asking mode".

On Wed, Nov 25, 2009 at 2:39 PM, Phil Grainger < philg@philgrainger.demon.co.uk> wrote:

> Let’s ask another question > > > > How are you ACCESSING the data? > > > > A clustering index is only relevant if it defines or supports some > processing sequence > > > > If you are only ever (or most of the time) accessing the data randomly then > who cares what sequence it’s in – DB2 certainly doesn’t > > > > However, if you have many (or perhaps a [...] 15007 706 58_Re: [DB2 v8] Implicit Clustering index .. Still not done..11_Mike Turner19_mike.turner@GMX.COM31_Wed, 25 Nov 2009 10:18:07 -0000476_Windows-1252 OK, I will try to answer your original question! I seem to remember (and at my age that doesn't guarantee accuracy) that the selection of the implicit clustering index is not based on anything stored in the Catalog. It is dependent on how the object definitions are chained together within the DBD in the Directory. This will initially point DB2 to the first index defined but, after drops and creates of indexes, the exact order of chaining is not certain. [...] 15714 1090 58_Re: [DB2 v8] Implicit Clustering index .. Still not done..13_Phil Grainger30_philg@PHILGRAINGER.DEMON.CO.UK31_Wed, 25 Nov 2009 10:35:06 -0000459_us-ascii Please don't be offended by the questions





We're just trying to make sure you get the RIGHT answer





And the right answer isn't (necessarily) to have a clustering index on every table





You may well be fine the way things are





(and be grateful no-one has yet said "it depends" :-) )





Phil Grainger

Grainger Database Solutions Ltd [...] 16805 655 58_Re: [DB2 v8] Implicit Clustering index .. Still not done..9_DB2DBAzOS21_bala.db2dba@GMAIL.COM31_Wed, 25 Nov 2009 16:40:47 +0530344_windows-1252 Thanks, Mike. I observed that there is no simple logic on Catalogs that DBAs can understand in following the implicit clustering. Because, over the last two reorgs, one of the table changed totally because one of the index was dropped and recreated (thanks to V8 reorg feature that picks implicit clustering while reloading). [...] 17461 723 58_Re: [DB2 v8] Implicit Clustering index .. Still not done..9_DB2DBAzOS21_bala.db2dba@GMAIL.COM31_Wed, 25 Nov 2009 16:45:16 +0530580_windows-1252 Hi Phil,

I wasn't offended because, I do understand it would be monotonous to provide just answers all the time :-)

And about 'it depends' answers, well, it depends ...







On Wed, Nov 25, 2009 at 4:05 PM, Phil Grainger < philg@philgrainger.demon.co.uk> wrote:

> Please don’t be offended by the questions > > > > We’re just trying to make sure you get the RIGHT answer > > > > And the right answer isn’t (necessarily) to have a clustering index on > every table > > > > You may well be fine the way things are > > [...] 18185 43 30_Commit upon exiting subprogram10_Mike Baker23_goldofophir@LIVE.COM.AU31_Wed, 25 Nov 2009 11:14:41 +0000438_utf-8 Hi,

We have Easytrieve Plus, but we do not possess the "Easytrieve Plus / DB2"; we only have the plain (vanilla) version and also the DLI version.

We would like to be able to access DB2 from plain (vanilla) Easytrieve. (PS: or via the DLI version, if this might be better.)

One way of doing this would be for the Easytrieve program to call an Assembler/COBOL subprogram, which would handle all of the SQL. [...] 18229 92 34_Re: Commit upon exiting subprogram14_Fazio, Richard21_RFAZIO@TRANSUNION.COM31_Wed, 25 Nov 2009 06:35:59 -0600445_us-ascii There are techniques to created a called module that will open a connection, leave it open and allow re-entry to perform other work.

Call OPEN Call Function1 Call Function2 Call CLOSE

The issue may be linkage in the program (you may have to statically link the called modules into your compiled Easytrieve program). But you can try this approach simply by writing a tiny test routine that does these three functions. [...] 18322 29 34_Re: Commit upon exiting subprogram11_Ted MacNEIL18_eamacneil@YAHOO.CA31_Wed, 25 Nov 2009 12:50:35 +0000374_- >However, its my understanding that upon exiting the subprogram, that all of the DB2 work will be committed. And... this could be a potential problem.

>Is there some way around this?? >Please advise.

Why don't you want the commit? It's there to 'harden' any changes and guarantee the integrity of any updated tables. - Too busy driving to stop for gas! [...] 18352 250 22_Re: Space issue in DB29_SrinivasG21_SRINIVASG@INFOSYS.COM31_Wed, 25 Nov 2009 18:50:57 +0530430_utf-8 I have it confirmed. Compression was disabled in Oracle at all levels. I am clueless..

From: IDUG DB2-L [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Gregory Palgrave Sent: Wednesday, November 25, 2009 6:47 AM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] Space issue in DB2

Once upon a time I would have known how big a 3390-9 was, but I have to admit I needed to look it up this time – roughly 8.5 GB. [...] 18603 317 22_Re: Space issue in DB214_Steurs Patrick21_Patrick.Steurs@NBB.BE31_Wed, 25 Nov 2009 14:42:54 +0100491_UTF-8 Hi,





What is your default encoding scheme in db2 ? Verify this in your DSNHDECP-job.





greetings,





Patrick





From: IDUG DB2-L [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of SrinivasG Sent: woensdag 25 november 2009 14:21 To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] Space issue in DB2





I have it confirmed. Compression was disabled in Oracle at all levels. I am clueless.. [...] 18921 250 33_Re: Tape on DASD for DB2 backups?13_Johnny Wilder28_Johnny.Wilder@MOTION-IND.COM31_Wed, 25 Nov 2009 07:52:10 -0600571_us-ascii We use Luminex http://www.luminex.com/ for our virtual tape system. Both DB2 archive logs and image copies are now virtual tape (disk) for us. Image copy clock time was cut in half after we switched. I was a little nervous using this for DB2 archive logs, but there has never been a problem. We also use Data Domain's deduplication http://www.datadomain.com/ to send all of our virtual tape backups to our DR site. For DB2 we use dual logging and deduplicate both to our DR site at archive log time. We have an in house written DR process for DB2 that runs [...] 19172 80 22_Re: Space issue in DB213_Mick P Graley16_mgraley2@CSC.COM31_Wed, 25 Nov 2009 14:12:17 +0000471_UTF-8 Hi Srinivas,

To see if your VSAM clusters are over-allocated do a listcat and check that the HI-A-RBA and HI-U-RBA are similar. If HI-A-RBA is much larger than HI-U-RBA then your PRIQTY is probably way too big.

Also as I remember (it's a long time ago) in Oracle all CHARs are stored as VARCHARs internally. This could make a big difference if all the columns are defined as CHAR in Oracle and DB2 if they have a lot of blank padding at the end. [...] 19253 36 34_Re: Commit upon exiting subprogram10_Mike Baker23_goldofophir@LIVE.COM.AU31_Wed, 25 Nov 2009 19:02:01 +0000497_utf-8 Richard,

Cheers, thanks for this. We'll try out this technique, or rather someone else will... Not quite sure what they'll do with the Easytrieve yet, but we just wanted further clarification on this issue (for the moment). My arrival at the party is late.

After second thoughts, our guys might actually be coding DLI Easytrieve, rather than vanilla, so it could potentially involve both IMS & DB2 updates. This would be the reason behind the concern with commit control. [...] 19290 148 26_Re: RUNSTATS' Alternative?7_DB2 DBA21_the.db2.dba@GMAIL.COM31_Wed, 25 Nov 2009 14:21:53 -0500374_ISO-8859-1 Peter, Roy, Bala, Max, Adam, Bell, Steen AND ofcourse Phil Sevetson... THANK YOU ALL. I agree, it isn't necessary to run RUNSTATS for anything AND everything under the sun PLUS as frequently as a week.

Although, I've been thinking about RTS for some time, it just didn't happen. I never knew that it's available from V7, until I referred the manuals. [...] 19439 79 34_Re: Commit upon exiting subprogram14_Fazio, Richard21_RFAZIO@TRANSUNION.COM31_Wed, 25 Nov 2009 13:39:20 -0600411_iso-8859-1 Ah, I was wondering why the reference to "DLI" specifically.

Again, not sure the love of Easytrieve. It only uses dynamic SQL (can't code a singleton select)...at least under the releases of Easytrieve+ I've used. I can't spell IMS so, I'm can't help there...but there's this new language called "Cobol". It works great with both and from what my idol Grace Hopper says, it's the bomb :) [...] 19519 60 34_Re: Commit upon exiting subprogram9_Stan Hoey22_shoey@CIRCLE-GROUP.COM31_Wed, 25 Nov 2009 19:47:34 -0000455_us-ascii I'm not sure if I missed something here, but if you are using DLI Easytrieve running under the IMS-DB2 attachment code, you won't be able to issue a COMMIT in your sub-program. The IMS-DB2 attachment disallows it.

-----Original Message----- From: IDUG DB2-L [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Mike Baker Sent: Wednesday, November 25, 2009 7:02 PM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] Commit upon exiting subprogram [...] 19580 95 57_DB2 z/OS V8 NFM and Oracle Transparent Gateway 10.2.0.3.013_David S Waugh16_dsw-dba@JUNO.COM29_Wed, 25 Nov 2009 22:16:43 GMT400_windows-1252 Greetings all:

My client recently made an attempt to go from DB2 V8 CM to DB2 V8 ENFM, then to NFM on their Test subsystem.

Before doing this, they upgraded their Oracle Transparent Gateway for this subsystem to Oracle HS Agent, TG4DB2 Version 10.2.0.3.0 (which is the one that supposedly supports DB2 V8 and V9). The Oracle Servers they're using vary from 9i to 10g. [...] 19676 314 22_Re: Space issue in DB215_Campbell, Wayne28_Wayne.Campbell@COURTS.WA.GOV31_Wed, 25 Nov 2009 14:28:24 -0800547_utf-8 A couple of things I would look at besides VARCHARs being packed with spaces are in the TableSpace definition the FREEPAGE, PCTFREE, and MAXROWS. MAXROWS controls how many rows will be put on a page. I’ve recently received some scripts from IBM that had MAXROWS set to 1. It came from the Server side of the house and not the Mainframe. I assumed that they were worried about lock contention, so a page lock would only lock one row. It would be a lot better to switch to row level locking for the table. The row length was around 100 [...] 19991 674 22_Re: Space issue in DB214_Sevetson, Phil22_PSevetson@FISA.NYC.GOV31_Wed, 25 Nov 2009 17:37:38 -0500673_us-ascii Wayne,

Good catch. I wouldn't have thought of that until late in the process.

--Phil Sevetson

________________________________ From: IDUG DB2-L [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Campbell, Wayne Sent: Wednesday, November 25, 2009 5:28 PM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] Space issue in DB2

A couple of things I would look at besides VARCHARs being packed with spaces are in the TableSpace definition the FREEPAGE, PCTFREE, and MAXROWS. MAXROWS controls how many rows will be put on a page. I've recently received some scripts from IBM that had MAXROWS set to 1. It came from the Server side of the house and [...] 20666 487 22_Re: Space issue in DB29_SrinivasG21_SRINIVASG@INFOSYS.COM31_Thu, 26 Nov 2009 10:02:30 +0530693_utf-8 Hi Patrick,

The Default encoding is UNICODE.

AGCCSID=65534, SCCSID=1047, MCCSID=65534, GCCSID=65534, USCCSID=367, UMCCSID=1208, UGCCSID=1200, ENSCHEME=UNICODE, APPENSCH=UNICODE,



And if I specify

PRIQTY -1 SECQTY -1

Will that be ok? Since TSQTY is 7200 and IXQTY is also 7200 and MGEXTSZ=YES is in effect currently.

I am also planning to give COMPRESS YES and DEFINE NO.

In fact the current Create Tablespace statement reads like the one below. I was created by the programmers. I could not check it before since the setup here is a bit complicated. But now they have come back to me since they are having performance issues. [...] 21154 107 22_Re: Space issue in DB212_Isaac Yassin20_yassini@BEZEQINT.NET31_Thu, 26 Nov 2009 06:44:51 +0200582_UTF-8 Hi

Your BP32k may be too small and this may cause performance problems

Isaac Yassin

---- Original message ---- >Date:Â Â Thu, 26 Nov 2009 10:02:30 +0530 >From:Â Â SrinivasG >Subject:Â Â Re: [DB2-L] Space issue in DB2 >To:Â Â DB2-L@WWW.IDUGDB2-L.ORG > > Hi Patrick, > > > > The Default encoding is UNICODE. > > > > AGCCSID=65534, > > SCCSID=1047, > > MCCSID=65534, > > GCCSID=65534, > > USCCSID=367, > > UMCCSID=1208, > > UGCCSID=1200, > > ENSCHEME=UNICODE, > > APPENSCH=UNICODE, > > > > > > And if I specify > > > > PRIQTY -1 [...] 21262 123 69_AW: [DB2-L] DB2 z/OS V8 NFM and Oracle Transparent Gateway 10.2.0.3.012_PUSCH Othmar27_Othmar.Pusch@IT-AUSTRIA.COM31_Thu, 26 Nov 2009 09:40:47 +0100709_iso-8859-1 Hi David !

o) We've had code-page_problems in this area too ... similar to the prob's in your eMAIL-info. o) We tried with ORACLE & IBM a lot (z/OS-1.10, DB2/V9 and OTGW/10.2.0.3.0 at problem-time) ... trace-level's (4,16, ..., 255), dump's etc. ... o) Solution for us is/was ORACLE's Patch-Set B6682743 for OTGW-Mainframe ... since all our (historical) ccsid's of the different oracle_client's are correct supported. With this Mainframe-B6682743 ya bring up OTGW/Oracle-Part to 10.2.0.3.92 and the dyn. z/OS-subsystem OTGW/OSDI-Part to 1.2.1.4. Don't forget to code in DD-ORA$ENV-Parmlib of the Gateway-Part of OTGW a (new) FDS-Class_Statement ---> (FDS_CLASS_VERSION = '10.2.0.3.92') [...] 21386 51 42_DB2 Connect in Windows Cluster environment16_Andreopoulou Zoi24_z.andreopoulou@TTBANK.GR31_Thu, 26 Nov 2009 10:46:58 +0200530_us-ascii We have a SQL Server 2005 Enterprise Edition database installed on a Windows Cluster (MSCS) running MS Windows 2003 Server Enterprise Edition 32-bit. We need to connect the SQL Server database to a number of DB2 V8 for z/OS subsystems. We have available the "DB2 Connect Unlimited Edition for zSeries V9.1 for Linux, UNIX and Windows". Should we install the DB2 Connect on both servers? Do we need to configure the Cluster Administrator in some way in order to make this possible? Any patches that we need to apply? [...] 21438 685 22_Re: Space issue in DB214_Steurs Patrick21_Patrick.Steurs@NBB.BE31_Thu, 26 Nov 2009 10:01:54 +0100709_UTF-8 Hi,





We are not using UNICODE as default.





DSNHDECM CHARSET=ALPHANUM,

ASCCSID=1252,

AMCCSID=65534,

AGCCSID=65534,

SCCSID=500,

MCCSID=65534,

GCCSID=65534,

USCCSID=367,

UMCCSID=1208,

UGCCSID=1200,

ENSCHEME=EBCDIC,

APPENSCH=EBCDIC,





If you use UNICODE, you are using more diskspace than with EBCDIC. But, if your source data comes from Oracle ( & uses UNICODE data ) then you don't have a choice.





The system zparms that have an impact on allocation are TSQTY, IXQTY and MGEXTSZ. The descriptions below are from the Installation Guide. [...] 22124 369 33_Re: Tape on DASD for DB2 backups?0_17_jtonchick@AOL.COM31_Thu, 26 Nov 2009 08:42:07 -0500402_utf-8

You still might consider copying your archived logs from the virtual tapes at the DR site to disk to allow for parallel recover jobs to use the same archlog at the same time. I mention this because while your V-tape device uses internal disk with its I/O speed advantages, it is still genned to the operating system as a physical tape device that cannot be shared by multiple tasks. [...] 22494 13 63_Tomas Vasko/IEA/EAGA/EAG/AT ist au=?ISO-8859-1?Q?=DFer_?= Haus.11_Tomas Vasko23_tomas.vasko@GENERALI.AT31_Thu, 26 Nov 2009 16:02:17 +0100221_ISO-8859-1

Ich bin ab 23.11.2009 außer Haus und ab 30.11.2009 wieder im Büro. I'm out of the office from 23.11.2009 and will be back on 30.11.2009.

Ich werde Ihre Nachricht nach meiner Rückkehr beantworten. 22508 133 27_Oracle to DB2 9.7 migration10_Raul Baron20_jbaron@CALCULO-SA.ES31_Thu, 26 Nov 2009 18:40:55 +0000388_- Hi list, we are currently facing an Oracle to DB2 migration project using DB2 9.7 but we are finding certain issues that could make this impossible.

Here are some of them:

TYPE DECLARATION ================ TYPE declaration is not supported in Functions, Procedures, Triggers or anonymous blocks. Does anybody know if this is going to be solved short or medium term ? [...] 22642 111 42_Allowed to have V9 Trigger using XMLPARSE?13_Steve Mallett21_malutjuta@HOTMAIL.COM31_Fri, 27 Nov 2009 05:44:18 +0000359_utf-8 Hey, my first V9 question

Our developers are trying to create a trigger similar to below in DB2V9 for Z/OS 1.9 and getting -873 SQLCODE 53090

Can't find anything specifically forbidding this but mention is made to declare the XML doc as a :HV in other situations (which can't be done in trigger of course) . Is this syntax allowable? [...] 22754 55 49_DB2 for z/OS Version 8 - lost DSNTIDxx parameters12_Jim McAlpine22_jim.mcalpine@GMAIL.COM31_Fri, 27 Nov 2009 15:19:30 +0000556_ISO-8859-1 We have recently moved from a FLEX-ES system to the Remote Development Program at Dallas and I have unfortunately forgotten to copy across our DSNTIDxx parameter list from the previous system which is no longer available. Is there any way to be able to recreate our DSNTIDxx parameters.

Jim McAlpine

_____________________________________________________________________

* IDUG North America * Tampa, Florida, * May 10-14 2010 * http://IDUG.ORG/NA * _____________________________________________________________________ [...] 22810 75 53_Re: DB2 for z/OS Version 8 - lost DSNTIDxx parameters12_Jim McAlpine22_jim.mcalpine@GMAIL.COM31_Fri, 27 Nov 2009 16:09:14 +0000553_ISO-8859-1 Panic over, I've found them. I knew I had copied them somewhere :-) And NO they were'nt in DSNXXX.SDSNSAMP.

Jim McAlpine

On Fri, Nov 27, 2009 at 3:19 PM, Jim McAlpine wrote:

> We have recently moved from a FLEX-ES system to the Remote Development > Program at Dallas and I have unfortunately forgotten to copy across our > DSNTIDxx parameter list from the previous system which is no longer > available. Is there any way to be able to recreate our DSNTIDxx parameters. > > Jim McAlpine > > > [...]