1 WWW.IDUGDB2-L.ORG /home/listserv/home/db2-l August 2000, week 3 2 12 54_Greg Fernholz/Deluxe Corporation is out of the office.13_Greg Fernholz24_Greg_Fernholz@EFUNDS.COM31_Tue, 15 Aug 2000 01:02:12 -0500312_us-ascii I will be out of the office from 08/15/2000 until 08/16/2000.

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 15 98 49_Re: Problems inserting LOB data greater than 16MB15_Keith Phimister31_keith.phimister@VIRGINNET.CO.UK31_Tue, 15 Aug 2000 08:30:34 +0100420_iso-8859-1 Thanks again James, That seems a good and normal solution. I was wondering if the other solution might be to have UDB for NT installed, do the insert into that using file locators (which you cannot use for OS390 yet) and then set up a trigger on that NT table to insert the LOB row up to os390 via DB2 Connect. A delete could then follow so that, in effect, I can use file locators on the OS390 DB2 UDB. [...] 114 33 8_DSn1copy25_Srirengan Venkatesh Kumar42_SRIRENGAN_VENKATESH_KUMAR/IRAS@IRAS.GOV.SG31_Tue, 15 Aug 2000 15:46:08 +0800433_us-ascii Hi,

I have two questions

1. One of our tablespaces which is segmented and has only one table defined in it has gone to multiple files i.e a A001 and A002. Our DR statergy is to DSN1COPY our backups (not registered in syscopy) which are sent offsite. I would like to know whether at DR site I would still be able to DSN1COPY as the files has split because the sysut2 dataset of the dsn1copy refer to a001. [...] 148 339 49_Re: Problems inserting LOB data greater than 16MB11_KUHN,Ulrich37_ulrich.kuhn@EMPLOYMENTNATIONAL.COM.AU31_Tue, 15 Aug 2000 17:46:50 +1000619_- Keith,

as far as I am aware, triggers in DB2 NT etc cannot reference tables in other databases, therefore the insert to OS/390 would be a problem.

I could be wrong, of course...

Cheers, Ulrich

> -----Original Message----- > From: Keith Phimister [SMTP:keith.phimister@virginnet.co.uk] > Sent: Tuesday, August 15, 2000 5:31 PM > To: DB2-L@RYCI.COM > Subject: Re: Problems inserting LOB data greater than 16MB > > Thanks again James, > That seems a good and normal solution. I was wondering if the other > solution > might be to have UDB for NT installed, do the insert into that using [...] 488 51 12_Re: DSn1copy10_Pete Coles28_peter.coles@SOMERFIELD.CO.UK31_Tue, 15 Aug 2000 09:53:09 +0100461_iso-8859-1 dsn1copy will work ok. you will need to define the a002 dataset yourself as when you create the tablespace db2 will only define the a001 one.

i'm not aware of the second point but it sounds very dangerous mucking about with syscopy, be careful

-----Original Message----- From: Srirengan Venkatesh Kumar [mailto:SRIRENGAN_VENKATESH_KUMAR/IRAS@IRAS.GOV.SG] Sent: Tuesday, August 15, 2000 7:46 AM To: DB2-L@RYCI.COM Subject: DSn1copy [...] 540 36 24_V5 to V6 migration error0_24_James_Hughes@MANDG.CO.UK31_Tue, 15 Aug 2000 09:19:28 +0000470_us-ascii David, Seems that you may have this problem already covered. My penny for what it is worth. As we are in the processes of upgrade from v4 to v6 (via v5) it has highlighted passwords as being incompatible with version 6. This is not the only one that maybe highlighted. Suggest running DSNTIJPM in your sample library, which will highlight and build what you need. If you have this sorted, or have already done this, just ignore my ramblings. Regards James [...] 577 127 28_AW: V5 to V6 migration error12_Peter, Georg15_G.Peter@DZBW.DE31_Tue, 15 Aug 2000 11:01:29 +0200370_iso-8859-1 Hi James,

I fully agree with your statement "This is not the only one that maybe highlighted." Version 6 is - as far as I know - the first version of DB2 where IBM has removed some features.......

Several things will go away (not longer supported) when you migrate to Version 6 and it is essential to know what will go through the roof ;-) [...] 705 23 27_Reason code 00C20200 ??????0_19_Sven.Heidorn@CSN.SE31_Tue, 15 Aug 2000 11:31:24 +0200495_ISO-8859-1 Hi all,

we get a -904, resource unavailable on a data-set with the reason code 00C20200. I can't find the reason code in the manuals or on the web. Does anyone have any clue what the reason code stands for?

TIA

Sven Heidorn

===============================================To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 729 53 31_Re: Reason code 00C20200 ??????38_Toine Michielse/Santa Teresa/Contr/IBM18_vndobtm@US.IBM.COM31_Tue, 15 Aug 2000 11:49:43 +0100483_us-ascii Hello Sven,

DB2 was unable to open an index dataset because the index was formatted incorrectly. Formatted as type 1 while the catalog has it registered as typ2 for instance.

Run recover/rebuild INDEX.

Regards,

Toine Michielse





Sven.Heidorn@CSN.SE@RYCI.COM> on 08/15/2000 10:31:24 AM

Please respond to DB2 Data Base Discussion List

Sent by: DB2 Data Base Discussion List [...] 783 97 31_Re: Reason code 00C20200 ??????9_Andy Hunt33_Andy.Hunt@SCOTTISH-SOUTHERN.CO.UK31_Tue, 15 Aug 2000 10:54:06 +0100496_us-ascii APAR Identifier ...... PQ40454 Last Changed ........ 00/08/03 RC00C20200 TRYING TO OPEN DIRECTORY INDEXES IN THE DSNDB01 DATABASE

Symptom ...... IN INCORROUT Status ........... OPEN Severity ................... 1 Date Closed ......... Component .......... 5740XYR00 Duplicate of ........ Reported Release ......... 610 Fixed Release ............ Component Name 5740 IBM DATABA Special Notice Current Target Date ..00/09/22 Flags SCP ................... Platform ............ [...] 881 86 28_Re: V5 to V6 migration error16_Skrivanek, David28_David.Skrivanek@DOWJONES.COM31_Tue, 15 Aug 2000 08:20:02 -0400541_iso-8859-1 Linda,

You were correct, I was able to restart in V5 without recovering the catalog, remove the datasets passwords, and re-migrate to V6. I did, however, run into another problem. When the catmaint failed the first time, catalog indexes DSNTAX02, DSNAUH01 and DSNDXX02 were left in rebuild pending status (which is consistent with what is documented below). If you attempt to go back to V6 without recovering the indexes, the catmaint will receive a -601 on create index sysibm.dsnoxx01. This is described in PQ26982. [...] 968 201 44_Re: Shutting Down Instances on AIX at Reboot11_Emery, Matt19_matt.emery@JNLI.COM31_Tue, 15 Aug 2000 08:28:11 -0400612_iso-8859-1 This is our script..

db2 list application show detail

db2 force application all

sleep 5

db2stop force



Then we move all db2dump files (eg. db2diag.log, *.000, ARCHIVE, USEREXIT, etc.) to a save area and timestamp the file names.

We've never had any problems restarting an instance.

Matthew J. Emery Jackson National Life Insurance Company Technology Department - Database Administrator voice: 517-367-3715 fax: 517-394-0986 pager: 517-232-1617 email: Matt.Emery@jnli.com e-page: 5172321617@epage.arch.com [...] 1170 90 0_14_Dean Montevago18_DMonteva@VNSNY.ORG31_Tue, 15 Aug 2000 09:06:24 -0400535_- Thank you.

> -----Original Message----- > From: IBM1 [SMTP:IBM1@EGEBANK.COM.TR] > Sent: Monday, August 14, 2000 12:13 PM > To: DB2-L@RYCI.COM > Subject: > > dean, > > try and find out if your "local friendly" IBM have set up course CA81 (DB2 > OS/390 for DBAs Advanced). > > some IBM educations provide this 4-day course - and it focuses mainly on > recovery and backup, with all the new stuff for v 6 included!! > > I have done it some time ago and it's good, if a bit.... rare to find!!! > > ciao! > > Aurora Dell'Anno [...] 1261 53 0_14_Dean Montevago18_DMonteva@VNSNY.ORG31_Tue, 15 Aug 2000 09:06:51 -0400597_- Thanks

> -----Original Message----- > From: David Oberstadt/Santa Teresa/IBM [SMTP:davido@US.IBM.COM] > Sent: Monday, August 14, 2000 12:01 PM > To: DB2-L@RYCI.COM > Subject: > > Dean, > > You can find the DB2 for OS/390 education page at this address: > > www.ibm.com/software/db2os390/education.html > > I know only of one DB2 for OS/390 recovery class--DB2 for OS/390 System > Recovery Workshop, CF910. This class is linked from the list in "DB2 for > OS/390 course catalog". I have not taken the class, so all I know is what > the page says. Good luck! > > David Oberstadt > DB2 [...] 1315 63 12_Re: DSn1copy19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Tue, 15 Aug 2000 08:13:29 -0500637_ISO-8859-1 Hi Srirengan, Why not use RECOVERYDDN for the DR site tapes and consider creating two tapes as well.

Regards, Rick Davis

"This e-mail and any files transmitted with it are the property of SBC, are confidential, and are intended solely for the use of the individual or entity to whom this e-mail is addressed. If you are not one of the named recipient(s) or otherwise have reason to believe that you have received this message in error, please notify the sender at 314-235-6854 and delete this message immediately from your computer. Any other use, retention, dissemination, forwarding, printing, or copying [...] 1379 196 12_Re: DSn1copy10_Bill Disch23_bill.disch@GENMILLS.COM31_Tue, 15 Aug 2000 08:25:02 -0500523_iso-8859-1 Something to look out for: We used IBM and BMC to COPY the individual datasets A001, A002, etc. and sent the tapes offsite. At the DR site we used DSN1COPY to restore the datasets and found that they must be restored in ascending sequence. If A002 was restored first, then when A001 was restored, it wrote (formatted the next segment?) at the beginning of A002. This was discovered with DB2 V3.1 and our DR procedures were adjusted accordingly. I have not investigated if this still occurs at V4.1 or V5.1. [...] 1576 22 12_Re: DSn1copy17_Venkat Srinivasan28_venkatsrinivasan@HOTMAIL.COM31_Tue, 15 Aug 2000 08:31:08 -0500571_- As DSN1COPY operates outside of DB2, you don't need any special considerations within the catalog while trying to do DR, except the usual LOG RBA reset and translations. If your dataset has gone into multiple datasets then your recovery procedure must start with the first dataset and then in seriatim. To make sure this works,you can simulate this in your local site itself. In otherwords, recover A0001 first and then 2,3....and so on. Regarding catalog updates, I don't think the list discussed any hidden ZPARMs, that will open up the catalog. Regards, Venkat [...] 1599 18 28_FW: automatic key generation14_John Lawniczak26_jlawniczak@RTSWIRELESS.COM31_Tue, 15 Aug 2000 10:07:40 -0400407_iso-8859-1 HI list, Does anyone know if db2/udb v7.1 has the capability to automatically generate a key (ID data type)? Thank you.

John Lawniczak RTS Wireless

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 1618 37 12_Re: DSn1copy20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM31_Tue, 15 Aug 2000 10:18:24 -0400685_us-ascii The list did discuss about ZPARM changes to update catalog but definitely it is too risky. Besides SYSCOPY structure has changed between V5 & V6.

-----Original Message----- From: Venkat Srinivasan [SMTP:venkatsrinivasan@HOTMAIL.COM] Sent: Tuesday, August 15, 2000 9:31 AM To: DB2-L@RYCI.COM Subject: Re: DSn1copy

As DSN1COPY operates outside of DB2, you don't need any special considerations within the catalog while trying to do DR, except the usual LOG RBA reset and translations. If your dataset has gone into multiple datasets then your recovery procedure must start with the first dataset and then in seriatim. To make sure this works,you can simulate [...] 1656 35 28_Re: automatic key generation20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM31_Tue, 15 Aug 2000 10:22:15 -0400452_us-ascii There is an identity type column, which is introduced in DB2 V7 but also comes as APAR in DB2 V6. This provides capability to automatically generate a key. This needs two more new catalog tables.

Check APARs PQ30652, PQ30684, PQ36328 & PQ36452.

-----Original Message----- From: John Lawniczak [SMTP:jlawniczak@RTSWIRELESS.COM] Sent: Tuesday, August 15, 2000 10:08 AM To: DB2-L@RYCI.COM Subject: FW: automatic key generation [...] 1692 38 21_Datajoiner & Runstats18_Whittaker, Stephen26_stephen.whittaker@CPLC.COM31_Tue, 15 Aug 2000 10:30:12 -0400544_iso-8859-1 If anyone out there is using Datajoiner (V2) with OS/390 DB2 V5 I would like to know how you are doing 'Runstats' against the Datajoiner nicknames. We found out that if I use the runstats command that is generated using the Datajoiner Control Center I can bring down the NT box that datajoiner is running on. To much info for the box to handle. I've tried using 'script files' to run these same SQL runstat stmts but again at certain times they will bring the NT box to its knees. I've come up with a script where I just update [...] 1731 25 45_significant increase in DSN4kxx usage with V612_Myron Miller22_myronwmiller@YAHOO.COM31_Tue, 15 Aug 2000 08:06:55 -0700373_us-ascii We upgraded to V6 from V5 of DB2 in production this weekend and ran batch for the first time last night. Our usage of DSN4kxx files went from about 50 cylinders to failing at over 2300 cylinders. No changes to SQL happened.

Has anyone else noticed this type of increase in DSN4k usage? Can anyone explain why the dramatic growth in space needs for V6? [...] 1757 39 49_Re: significant increase in DSN4kxx usage with V620_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM31_Tue, 15 Aug 2000 11:10:06 -0400520_iso-8859-1 Did you check any access path changes where index usage converted to SORT ?

-----Original Message----- From: Myron Miller [SMTP:myronwmiller@YAHOO.COM] Sent: Tuesday, August 15, 2000 11:07 AM To: DB2-L@RYCI.COM Subject: significant increase in DSN4kxx usage with V6

We upgraded to V6 from V5 of DB2 in production this weekend and ran batch for the first time last night. Our usage of DSN4kxx files went from about 50 cylinders to failing at over 2300 cylinders. No changes to SQL happened. [...] 1797 48 49_Re: significant increase in DSN4kxx usage with V60_19_mike.holmans@BT.COM31_Tue, 15 Aug 2000 16:15:09 +0100336_iso-8859-1 Did anything else change during the V6 upgrade? For instance, did the size of SORTPOOL reduce because the installation did a default calculation and you didn't specify a much higher value? Are the VDWQT and DWQT thresholds set as high as they were before?

Mike Holmans BT ISE Technical Design mike.holmans@bt.com [...] 1846 25 28_Attention all UK subscribers14_Grainger, Phil20_Phil.Grainger@CA.COM31_Tue, 15 Aug 2000 16:24:13 +0100451_iso-8859-1 If you are not resident in the UK, then please delete NOW (and please accept my apologies for disturbing you).

Sorry to do this, but I need some 'volunteers'!!

The UK/Eire DB2 GUIDE meeting is looming and I am looking for volunteer user speakers FROM THE UK!!

If you would like this opportunity to shine in front of your peers, then please drop me a line at phil.grainger@ca.com and NOT by replying to this eMail. [...] 1872 198 27_Re: Problem with Image Copy13_Carlton Enuda24_carltonenuda@HOTMAIL.COM29_Tue, 15 Aug 2000 11:26:57 EDT346_- Hi List, Thanks to all for your help in getting me started in creating a backup procedure for our shop. Here's a related question dealing with Tablespace name and the associated GDG dataset thus:

1.) How do you create a model DSCB needed to create new GDGs? I have no clue what it does, but it seems like it's a required parameter. [...] 2071 51 49_Identity Columns >> WAS: automatic key generation13_Seibert, Dave26_Dave_Seibert@COMPUWARE.COM31_Tue, 15 Aug 2000 11:38:36 -0400423_iso-8859-1 Yes, there is the Identity feature added via PTF UQ38405.

You need to investigate very carefully its usage.

We are discovering some questionable results with its implementation.

You can probably use it very happily if you never intend to 1) use it in an RI relationship, 2) rely on the values to have permanence, 3) ever drop & recreate the table or 4) rely on the values to be unique. [...] 2123 17 49_Re: significant increase in DSN4kxx usage with V614_Philip Gunning18_pgunning@PHEAA.ORG31_Tue, 15 Aug 2000 11:44:36 -0400403_us-ascii Transition tables associated with triggers used DSNDB07, and global temporary tables. But since you just converted you are probably not using triggers extensively. There is a notice in the V6 What's New manual indicating that you may need to increase DSNDB07 space because of the use of triggers. But nothing else is mentioned. Did you enable star join? You may be hitting a bug. HTH Phil [...] 2141 37 49_Re: significant increase in DSN4kxx usage with V612_Myron Miller22_myronwmiller@YAHOO.COM31_Tue, 15 Aug 2000 08:47:21 -0700607_us-ascii We're not using triggers and global temp table usage has not changed. --- Philip Gunning wrote: > Transition tables associated with triggers used > DSNDB07, and global > temporary tables. But since you just converted you > are probably not using > triggers extensively. There is a notice in the V6 > What's New manual > indicating that you may need to increase DSNDB07 > space because of the use > of triggers. But nothing else is mentioned. Did > you enable star join? > You may be hitting a bug. HTH Phil > > ================================================ > To change [...] 2179 249 27_Re: Problem with Image Copy15_Billings, Linda30_linda.billings@DOA.STATE.WI.US31_Tue, 15 Aug 2000 10:47:33 -0500451_iso-8859-1 Hi, Carlton,

1) A model DSCB provides DCB information for the GDG. It is not necessary in an SMS enviroment as you can use a storage class to define the DCB information. If I remember correctly, a model DSCB is no more than a dummy data set that can be defined using IEFBR14 and providing DCB info that matches what you want for your image copies. I have a book at home that provides this information but don't have it with me. [...] 2429 76 49_Re: significant increase in DSN4kxx usage with V612_Myron Miller22_myronwmiller@YAHOO.COM31_Tue, 15 Aug 2000 08:49:11 -0700570_us-ascii VDWQT and DWQT didn't change. Size of SORTPOOL was set to same as before, thru zparms. --- mike.holmans@BT.COM wrote: > Did anything else change during the V6 upgrade? For > instance, did the size > of SORTPOOL reduce because the installation did a > default calculation and > you didn't specify a much higher value? Are the > VDWQT and DWQT thresholds > set as high as they were before? > > Mike Holmans > BT ISE Technical Design > mike.holmans@bt.com > > This post represents the views of the author and > does not necessarily > accurately represent the [...] 2506 263 27_Re: Problem with Image Copy15_Billings, Linda30_linda.billings@DOA.STATE.WI.US31_Tue, 15 Aug 2000 10:51:15 -0500631_iso-8859-1 Ooooops! For number 2, you would define a DATACLAS, not a STORCLAS, in an SMS environment to provide DCB information.

Sorry,

Linda

-----Original Message----- From: Billings, Linda Sent: Tuesday, August 15, 2000 10:48 AM To: 'DB2 Data Base Discussion List' Subject: RE: Problem with Image Copy



Hi, Carlton,

1) A model DSCB provides DCB information for the GDG. It is not necessary in an SMS enviroment as you can use a storage class to define the DCB information. If I remember correctly, a model DSCB is no more than a dummy data set that can be defined using IEFBR14 and [...] 2770 175 49_Re: Problems inserting LOB data greater than 16MB13_OConner, John26_John.OConner@COURTS.WA.GOV31_Tue, 15 Aug 2000 08:52:19 -0700314_iso-8859-1 James,

We haven't started using CLOBs here yet, so I'm just asking for a little clarification. The DB2 documentation (I can't put my finger on the specific reference) seems to indicate that you cannot reassign a value to a locator. Rather, you must first do a FREE. For example, rather than [...] 2946 43 53_Re: Identity Columns >> WAS: automatic key generation12_MARTIN WOLFF31_martin_wolff@GLOBALCROSSING.COM31_Tue, 15 Aug 2000 12:01:34 -0400522_us-ascii

> > > The key problem (ok, call it an opportunity) is drop/recreate of the table. > Say you have 10,000 rows in a table. Assume your startnum =1 and your > increment=1. > You have values 1-10,000 in the column. > > Say you drop and recreate the table, reload the table, and your application > resumes insertion of rows into the table. You'd probably expect (or at > least want) the next number to be 10,000 or greater. > Guess what. > > It starts over at 1. > > This is the way its working for [...] 2990 275 27_Re: Problem with Image Copy15_Billings, Linda30_linda.billings@DOA.STATE.WI.US31_Tue, 15 Aug 2000 11:24:26 -0500407_iso-8859-1 Ooooops on the oooops! That is pertaining to item 1. More caffeine, please!

Linda

-----Original Message----- From: Billings, Linda Sent: Tuesday, August 15, 2000 10:51 AM To: 'DB2 Data Base Discussion List' Subject: RE: Problem with Image Copy



Ooooops! For number 2, you would define a DATACLAS, not a STORCLAS, in an SMS environment to provide DCB information. [...] 3266 19 53_Re: Identity Columns >> WAS: automatic key generation13_Seibert, Dave26_Dave_Seibert@COMPUWARE.COM31_Tue, 15 Aug 2000 12:24:31 -0400570_iso-8859-1 >Are you saying that the 'START WITH nnnn' option doesn't work?

"Start with ..." works great if you're willing to edit your DDL and change it after having found out what value your new "start with ..." should use.

David Seibert Compuware Corporation File-AID Product Architect Dave.Seibert@Compuware.com

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 3286 51 51_Data Sharing: Non-stop services during DB2 Upgrade?29_Chris Munson/Santa Teresa/IBM18_cmunson@US.IBM.COM31_Tue, 15 Aug 2000 10:29:40 -0600611_us-ascii Anita, The basic design of data sharing is to provide higher levels of availability than a stand-alone DB2 subsystem. This includes application of DB2 maintenance, coupling facility maintenance and system migration (V5 to V6 as in your example). If you have 2 V5 members you can migrate one to V6 while the other V5 system is up and running. This includes during CATMAINT execution. There is, of course, a period of time when the catalog will be unavailable during CATMAINT. How long depends on various factors including size and (dis)organization of your catalog. The obvious rule of thumb is to [...] 3338 81 25_Re: Datajoiner & Runstats11_David Ayers24_david.ayers@HIGHMARK.COM31_Tue, 15 Aug 2000 12:38:09 -0400480_us-ascii Steve, we are running datajoiner on an AIX box(data sources DB2 v5 OS 390 / Oracle ) and are able to use the runstats command. We do have problems when trying to execute the runstats on large tables (basically we get a memory and the task abends). Remember when datajoiner builds the nicknames he goes to the data source and extracts basic statistics from the catalog that is used to populate the datajoiner global catalog. For our large tables we do the following: [...] 3420 99 53_Re: Identity Columns >> WAS: automatic key generation14_Richard Yevich21_ryevich@ATTGLOBAL.NET31_Tue, 15 Aug 2000 13:12:47 -0400518_iso-8859-1 Dave,

I want to add a more critical concern, but would also like some more info from you on some of your issues.

Regarding your issues: (1). What are the issues you found here, and not the drop and recreate because there is a workaround on that. As I understand it, the drop removes the controls in the catalog tables that were in support of the table that now (DROP) no longer exists. In the redefine, use START WITH... (2). Again, if other that DROP/RECREATE, please explain (4). Same. [...] 3520 18 23_OS/390 DB2 DBA POSITION7_Richard21_rick.olle@TALBOTS.COM31_Tue, 15 Aug 2000 14:09:07 -0400466_us-ascii We are currently looking for a DB2 DBA with a minimum of 2 years experience. The position is located in Tampa Florida and is a full time permanent position.



Please send resume to rick.olle@talbots.com

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 3539 82 49_Re: significant increase in DSN4kxx usage with V612_Myron Miller22_myronwmiller@YAHOO.COM31_Tue, 15 Aug 2000 11:38:20 -0700592_us-ascii We had a few of these, but nothing significant.

We think we did find the problem. One major program was bound CS and degree(any). Although no tables were partitioned and all of the tables are very small (less than 1 million rows), several cursors are opened and I think the cursor processing was being performed in parallel. When we turned parallelism off (where have we heard this before), the dramatic growth in size of the workfiles stopped. I was able to watch the DSN4kxx files literally grow by 1000s of tracks in a few minutes time before turning off parallelism. [...] 3622 35 27_Re: OS/390 DB2 DBA POSITION22_Ditchkus, Vincent, Jr.18_VDitchkus@FDIC.GOV31_Tue, 15 Aug 2000 14:28:04 -0400456_iso-8859-1 Vince Ditchkus FDIC/DB2 Database Administration Phone: (703)845-3160 Pager: (703)705-9550



-----Original Message----- From: Richard [mailto:rick.olle@TALBOTS.COM] Sent: Tuesday, August 15, 2000 2:09 PM To: DB2-L@RYCI.COM Subject: OS/390 DB2 DBA POSITION



We are currently looking for a DB2 DBA with a minimum of 2 years experience. The position is located in Tampa Florida and is a full time permanent position. [...] 3658 64 53_Re: Identity Columns >> WAS: automatic key generation17_Kornblum, Kenneth24_Kenneth_Kornblum@BMC.COM31_Tue, 15 Aug 2000 14:24:12 -0500425_iso-8859-1 Dave,

Regarding your example... I haven't played with this myself yet, but wouldn't the LOAD Utility cause the current high-value (MAXASSIGNEDVAL) of the identity column to be set back up to 10,000? Thus the first INSERT (after the LOAD) would, in fact, get value 10,001.

Or, is this the bug/opportunity you referenced in your message? If so, it would seem to be an issue with the LOAD utility. [...] 3723 20 53_Re: Identity Columns >> WAS: automatic key generation13_Seibert, Dave26_Dave_Seibert@COMPUWARE.COM31_Tue, 15 Aug 2000 15:43:46 -0400539_iso-8859-1 Hi Ken, No. The load util does not reset this. I was told this is documented, but have not seen it myself.

I can't find it in the Technical Update (SG24-6108) Redbook or in the APAR.

David Seibert Compuware Corporation File-AID Product Architect Dave.Seibert@Compuware.com

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 3744 33 53_Re: Identity Columns >> WAS: automatic key generation13_Seibert, Dave26_Dave_Seibert@COMPUWARE.COM31_Tue, 15 Aug 2000 15:47:48 -0400631_iso-8859-1 Hi Richard,

I have a longer list from our development organization that I'm trying to recreate and document myself, so I can't give you an exhaustive list yet. They do mostly pertain to drop/recreate scenarios and do include the reorg pending exposure.

Here's one that makes RI an unattractive choice:

If you use the >Generated Always< clause, drop/recreate does not allow you to reload your Identity column values and as the clause implies new values are generated. If you had gaps in your sequence numbers(from deleted rows or from rollbacks or ??), you now have different values for some of [...] 3778 17 55_Re: Data Sharing: Non-stop services during DB2 Upgrade?12_Roger Miller19_millerrl@US.IBM.COM31_Tue, 15 Aug 2000 14:57:19 -0500440_- There is no simple characterization of the other members during the CATMAINT. If the applications are running, but not accessing the catalog and directory, they are not likely to be affected. The time for the CATMAINT for V5 to V6 is often in the range of 3 to 5 minutes, so many customers choose to keep the interruption short and make the process simple. In any case, the suggestion is to practice the process with your situation. [...] 3796 33 26_Re: OS/390 certification??12_Roger Miller19_millerrl@US.IBM.COM31_Tue, 15 Aug 2000 15:08:30 -0500393_- Announcing DB2 V7 Family Certification

With the product announcement of DB2 V7 for Unix, Windows and OS/2, we are updating our certification exams, targeted for July 2000 availability to cover - -DB2 UDB V7 Family Fundamentals Exam- Test 512 -DB2 UDB V7 Database Administration on Unix, Windows and OS/2 Exam - Test 513 -DB2 UDB V7 Application Development Family Exam - Test 514 [...] 3830 66 53_Re: Identity Columns >> WAS: automatic key generation14_Richard Yevich21_ryevich@ATTGLOBAL.NET31_Tue, 15 Aug 2000 15:35:21 -0400735_iso-8859-1 Dave,

Beginning to look like IDENTITY columns only have a limited scope of use, like maybe a nice choice for DECLARED TEMP TABLES, etc. Hum...

Hope this helps, Richard Yevich +======+======+======+ Richard_Yevich@attglobal.net http://www.YLAssoc.com -- DB2 Performance Journal http://www.ryci.com -- RYC Knowledge Database http://www.DB2-Portal.com



> -----Original Message----- > From: DB2 Data Base Discussion List [mailto:DB2-L@RYCI.COM]On Behalf Of > Seibert, Dave > Sent: Tuesday, August 15, 2000 3:48 PM > To: DB2-L@RYCI.COM > Subject: Re: Identity Columns >> WAS: automatic key generation > > > Hi Richard, > > I have a longer list from our development organization that > I'm trying [...] 3897 23 53_Re: Identity Columns >> WAS: automatic key generation13_Seibert, Dave26_Dave_Seibert@COMPUWARE.COM31_Tue, 15 Aug 2000 16:46:48 -0400560_iso-8859-1 >Beginning to look like IDENTITY columns only have a limited scope of use...

Naw!

This is such an important feature and IBM are so smart, I'm sure they'll fix it.

(I AM NOT being sarcastic)

David Seibert Compuware Corporation File-AID Product Architect Dave.Seibert@Compuware.com

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 3921 26 27_Re: To commit or not commit12_Roger Miller19_millerrl@US.IBM.COM31_Tue, 15 Aug 2000 15:43:16 -0500575_- Perhaps it's my turn to rant. You have a case where you want your cake and you want to eat it too. Database consistency requires a commit to provide a point of consistency. Without the identified point, then your data integrity doesn't meet the ACID properties. If applications need to share the data, they need to have some rules. We are trying to resolve a problem in an area with poorly-behaved applications where the other vendor does not have the base function. We have the usage guidelines on our home page, performance discussions on the red books web page and [...] 3948 18 32_Re: AW: V5 to V6 migration error12_Roger Miller19_millerrl@US.IBM.COM31_Tue, 15 Aug 2000 15:52:09 -0500404_- I'd like to twist just a bit on what you said.

We have removed some minor items in almost every release and tried to highlight them in the section on incompatibilities. With V6, some of the old structures needed more rework. This list is just the major ones, and so running the premigration job DSNTIJPM and getting people to read the list of incompatibilities is more important than ever. [...] 3967 57 53_Re: Identity Columns >> WAS: automatic key generation14_Richard Yevich21_ryevich@ATTGLOBAL.NET31_Tue, 15 Aug 2000 16:09:57 -0400760_iso-8859-1 Dave,

I agree with you completely. I am sure that it will follow precedence, although I am betting that it is not fixes but -- infrastructure 1.0, enhancements 2.0, sophisticated enhancements 3.0, etc...

Hope this helps, Richard Yevich +======+======+======+ Richard_Yevich@attglobal.net http://www.YLAssoc.com -- DB2 Performance Journal http://www.ryci.com -- RYC Knowledge Database http://www.DB2-Portal.com



> -----Original Message----- > From: DB2 Data Base Discussion List [mailto:DB2-L@RYCI.COM]On Behalf Of > Seibert, Dave > Sent: Tuesday, August 15, 2000 4:47 PM > To: DB2-L@RYCI.COM > Subject: Re: Identity Columns >> WAS: automatic key generation > > > >Beginning to look like IDENTITY columns only have a [...] 4025 14 32_Re: FW: automatic key generation12_Roger Miller19_millerrl@US.IBM.COM31_Tue, 15 Aug 2000 15:54:56 -0500418_- The identity column seems to match your request. It's in DB2 for UNIX, Windows, OS/2 V7.1 and shipped as an APAR on DB2 for OS/390 V6 about the same time.

Roger Miller

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 4040 349 27_Re: Problem with Image Copy14_Jack O'Connell26_jack.oconnell@MCMASTER.COM31_Tue, 15 Aug 2000 19:00:34 -0500431_iso-8859-1 Within SMS , unit types and other dataset characteristics can be specified in filter lists :

FILTLIST DB2_VALID4_HLQ INCLUDE('DB2X','DB2XB') FILTLIST TAPE INCLUDE('3480','3490','9840', AFF*)

along with code to direct the dsns to approved unit types

WHEN ((&HLQ = &DB2_VALID4_HLQ) && (&UNIT ¬= &TAPE)) DO SET &STORCLAS = 'SCXXXXXX'



it sounds like 'tape' is not in the filter list [...] 4390 20 28_Commit & Close Cursor w/CICS0_14_DPetro@AOL.COM29_Tue, 15 Aug 2000 20:19:21 EDT363_US-ASCII What are peoples standards/experiences regarding the use of explicit Commits and Close Cursors in CICS programs? Are there any advantages/disadvantages to having them or not having them? Are there any problems if they're not there? One of my co-workers seems to remember some potential problems with threads if Close Cursor was not explicitly done. [...] 4411 23 40_SQL 1336N Remote Host not found problem?13_Carlton Enuda24_carltonenuda@HOTMAIL.COM29_Tue, 15 Aug 2000 20:19:56 EDT363_- Hi list, I installed a DB2 Connect following steps outlined in the Quick beginings for DB2 Connect. I logged on to the control center and tried to do a connect to the host when I got the SQL1336N Host not found message. It's obvious I did not do something right and what it is I can't tell. Could someone enlighten me on how to resolve my coonect problem? [...] 4435 67 20_db2 stored procedure12_Rakesh Singh14_rasun@JUNO.COM31_Tue, 15 Aug 2000 21:17:09 -0700273_- Dear list, I am trying to call a db2 stored procedure (DB2 ver 5.1 os/390) from a java program. The stored procedure has a fixed character input of 150 bytes and a VARCHAR output of 3000 bytes. When i execute the client program written in java i get following error [...] 4503 14 42_iI will return your note after I get back.15_Mark Hardenburg30_mark.hardenburg@PRUDENTIAL.COM31_Wed, 16 Aug 2000 01:04:20 -0400379_us-ascii I will be out of the office from 08/15/2000 until 08/17/2000.

For DB2 help while I am out, contact Frank Yao at 716-8644.

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 4518 65 44_Re: SQL 1336N Remote Host not found problem?23_Pendlebury-Bowe, Leslie34_Leslie.Pendlebury-Bowe@EU.SONY.COM31_Wed, 16 Aug 2000 06:16:38 +0100521_iso-8859-1 ...where do I begin.

Some questions I think you may be asked by other list peeps :

1. Are you accessing DB2 OS390 using Connect? 2. Assuming you are .. do you have DDF up and running on the host? 3. Are you happy that you have correctly defined the host in your hosts file? (IP address etc - i.e. where do you wish to talk to) 4. Have you defined the correct Services entries on your PC to match the ports on the host? (i.e. what door you going to talk through) 5. If you have not setup DDF [...] 4584 14 49_Peter Faller/HRD/Prudential is out of the office.12_Peter Faller27_peter.faller@PRUDENTIAL.COM31_Wed, 16 Aug 2000 01:32:13 -0400365_us-ascii I will be out of the office from 08/12/2000 until 08/22/2000.

I will respond to your message when I return.

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 4599 69 26_Re: OS/390 certification??14_Julian Stuhler27_julian.stuhler@TRITON.CO.UK31_Tue, 15 Aug 2000 22:20:08 -0700553_iso-8859-1 Hi Eric.

I can confirm that IBM will definitely be offering the new DB2 for OS/390 certification tests at this year's European IDUG conference in October - the first time the new tests will have been offered at any major event.

As usual, the certification tests will be offered for free, on a first-come, first-served basis. As we expect demand to even higher than usual this year, I would encourage anyone planning to take the tests at the conference to book theirs as soon as possible when they arrive at the conference. [...] 4669 201 49_Re: Problems inserting LOB data greater than 16MB14_James Campbell29_James.Campbell@HANCORP.COM.AU31_Wed, 16 Aug 2000 15:57:54 +1000518_iso-8859-1 John,

Nope, we do use EXEC SQL SET :HV-LOC = :HV-LOC || :HV-DATA

The recommendation to free a lob locator only applies when one has finished with it.

/* standard disclaimer */ James Campbell DBA Hansen Corporation, Doncaster +61 3 9843 8442 James.Campbell@Hancorp.com.au -----Original Message----- From: OConner, John [mailto:John.OConner@COURTS.WA.GOV] Sent: Wednesday, August 16, 2000 1:52 AM To: DB2-L@RYCI.COM Subject: Re: [DB2-L] Problems inserting LOB data greater than 16MB [...] 4871 77 44_Triggers - Where can I get more information?0_34_L-IS.Kirkpatrick@MUTUALOFOMAHA.COM31_Tue, 15 Aug 2000 16:47:34 -0500499_us-ascii I have one general question and two specific questions regarding documentation on triggers (for DB2 version 6 for OS/390).

My general question is: Where do I go to get some more information on how to implement triggers? I have set up some triggers in a test environment and the documentation that I have used (The "SQL reference" and the "Application programming and SQL guide") seem to be sketchy at best. Is there somewhere else to go for additional detail and MORE EXAMPLES? [...] 4949 105 53_Re: Identity Columns >> WAS: automatic key generation14_James Campbell29_James.Campbell@HANCORP.COM.AU31_Wed, 16 Aug 2000 16:06:19 +1000426_iso-8859-1 I might throw in some comments at this point (ie earlier than some later discussion):

- it is interesting that the catalog tables have SYSSEQUENCES holding the current position of the sequences and SYSSEQUENCESDEP, as a child, holding the details of the identity column. To me this implies some pre-positioning to allow use of sequences without an identity column. At some time in the future. Possibly? [...] 5055 50 32_Re: Commit & Close Cursor w/CICS7_Sanjeev21_Sanjeev@INTERRAIT.COM31_Wed, 16 Aug 2000 11:30:35 +0530588_iso-8859-1 In my training course of DB2 V5 from IBM ,i just heard a sentence from the person who was conducting the class that "If you have RETURN TRANSID in your application the DB2 performs Implicit Commit and there is some extra bit of code which DB2 needs to execute when you are issuing the COMMIT explicitly". This is the reason i think that for read only DB2 transaction there is no need to issue COMMIT explicitly(presuming relatively small amount of data handled) but for the transactions doing update,insert and delete it is required to have explicit COMMIT for having the [...] 5106 84 32_Re: Commit & Close Cursor w/CICS15_Keith Phimister31_keith.phimister@VIRGINNET.CO.UK31_Wed, 16 Aug 2000 08:29:55 +0100517_iso-8859-1 From my knowledge..

A DB2 commit is not possible from within a CICS program and shouldn't be used anyway as the other CICS resources should be kept in line. For this reason, a CICS SYNCPOINT would be used.

Sanjeev is correct in that when an EXEC CICS RETURN is issued (with any other parameters, e.g. TRANSID, COMMAREA, IMMEDIATE), then this is equivalent to a COBOL GOBACK. This ends the program and so does an implicit commit as happens with all programs when they terminate normally. [...] 5191 186 64_Repeat Message--Bonnie Baker at SIRDUG (Charlotte, NC) 8/24-8/2511_Paul Turpin22_pault704@EARTHLINK.NET31_Wed, 16 Aug 2000 02:18:32 -0500609_ISO-8859-1 I sending this again, for those in SIRDUG area that might have missed the original announcement (August 2000 Week 1).

SIRDUG Southeastern Inter-Relational Database Users' Group P.O. BOX 34641 Charlotte, NC 28234

Website: http://users.vnet.net/sirdug Email: SIRDUG@VNET.NET Information/Message Line: (704) 556-6621



Meeting Announcement ========================================================================

WHEN? Thursday, August 24th & Friday, August 25th

WHERE? Charlotte Marriott Executive Park 5700 Westpark Dr. Charlotte NC 28217 704) 527-9650 [...] 5378 17 12_3 Part query13_Viswanathan N33_Viswanathan.N@POWAIMAIL.LTITL.COM31_Wed, 16 Aug 2000 13:45:23 +0530472_us-ascii Hi !! Does any one know how to create a JOIN between two tables that are on different databases. I am using DB2 Ver 5.2 on NT. Can we use 3 part qualifiers to differentiate databases in the same DB2 instance.

Vishy

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 5396 33 36_Question about shrlevel change reorg19_Briggs, N. - Neil -25_N.Briggs@CANON-EUROPA.COM31_Wed, 16 Aug 2000 10:31:27 +0200535_iso-8859-1 Dear All

OS390 2.7 DB2 version 5.1

When performing a sharelevel change reorg, a shadow dataset is created. This shadow dataset is then switched to be the real dataset. Can somebody please advise me. Is it possible to define this dataset on a volume that is not in the storage group of the database that is being reorged. The problem I have is that I have large tablespace (not partitioned) and do not have the space in the storage group to contain both the tablespace dataset and the shadow copy dataset. [...] 5430 69 16_Re: 3 Part query0_29_Greg.Palgrave@BANKWEST.COM.AU31_Wed, 16 Aug 2000 16:34:26 +0800542_us-ascii I don't think you can do it unless you have a product like Datajoiner or similar other 3rd party software.

Cheers, Greg Palgrave Database Administration IT Services Bank of Western Australia eMail : greg.palgrave@bankwest.com.au



---------------------------------------- Message History ----------------------------------------



From: Viswanathan N on 16/08/2000 16:15





Please respond to DB2 Data Base Discussion List [...] 5500 27 40_Re: Question about shrlevel change reorg14_Massimo Scarpa23_Massimo_Scarpa@DEBIS.IT31_Wed, 16 Aug 2000 10:43:27 +0100368_us-ascii You can work with your SMS & DASD manager to try to resolve this problem (the names of tablespace and shadow copy

have minimal differences) . But often SMS managers don't like to use restrictive rules to allow this game.

And after some times ALL your tablespaces will reside in the other DASD(s), so you must move this tablespace from this [...] 5528 122 48_Re: Triggers - Where can I get more information?14_James Campbell29_James.Campbell@HANCORP.COM.AU31_Wed, 16 Aug 2000 18:51:38 +1000310_iso-8859-1 Examples?? Hope springs eternal! A lack of appropriate examples seems a feature of V6 - just look at the example UDFs - all written in C. (But maybe that's not inappropriate - maybe you're supposed to only use C for UDF's - certainly the parameter format is designed that way .) [...] 5651 64 40_Re: Question about shrlevel change reorg19_Briggs, N. - Neil -25_N.Briggs@CANON-EUROPA.COM31_Wed, 16 Aug 2000 10:52:57 +0200405_iso-8859-1 Max

Thanks for your reply, but I must apologise I haven't a clue what you are talking about?? :-)

For your information at this site we do not have a DASD manager or a SMS manager.

Please explain "And after some times ALL your tablespaces will reside in the other DASD(s), so you must move this tablespace from this 'new' pool to the original pool (almost empty....)." [...] 5716 22 40_Re: Question about shrlevel change reorg14_Massimo Scarpa23_Massimo_Scarpa@DEBIS.IT31_Wed, 16 Aug 2000 10:59:38 +0100524_us-ascii Don't you use SMS for DB2 tablespace? It sound strange to me.....but not so strange if I consider some installation's

requirements.

I'm afraid that without SMS you cannot resolve ur problem, as I experimented in previous shop.

Regards

Max Scarpa

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 5739 92 40_Re: Question about shrlevel change reorg23_Pendlebury-Bowe, Leslie34_Leslie.Pendlebury-Bowe@EU.SONY.COM31_Wed, 16 Aug 2000 09:59:55 +0100464_iso-8859-1 are we talking about the same DB2s here ..

Max is talking about OS390 ... what are you talking about?

Leslie

-----Original Message----- From: Briggs, N. - Neil - [mailto:N.Briggs@CANON-EUROPA.COM] Sent: Wednesday, August 16, 2000 9:53 AM To: DB2-L@RYCI.COM Subject: Re: Question about shrlevel change reorg



Max

Thanks for your reply, but I must apologise I haven't a clue what you are talking about?? :-) [...] 5832 62 40_Re: Question about shrlevel change reorg19_Briggs, N. - Neil -25_N.Briggs@CANON-EUROPA.COM31_Wed, 16 Aug 2000 11:22:23 +0200521_iso-8859-1 Max,



I have reread you previous email and it falls into place. When a sharelevel reorg is preformed DB2 creates the shadow dataset in the storage group of the tablespace that is being reorged. I do not have the space in the storage group to contain 2 large tablespaces. The shadow tablespace is then renamed. Hence if a temporary volume is added to the stogroup the shrlevel reorg will end up moving the tablespace being reorged to the temporary volume. This is what I was trying to avoid. [...] 5895 91 40_Re: Question about shrlevel change reorg23_Pendlebury-Bowe, Leslie34_Leslie.Pendlebury-Bowe@EU.SONY.COM31_Wed, 16 Aug 2000 10:28:59 +0100698_iso-8859-1 Neil

...interesting solution to the issue of the storage group now having enough dasd to support both datasets ....

Leslie

-----Original Message----- From: Briggs, N. - Neil - [mailto:N.Briggs@CANON-EUROPA.COM] Sent: Wednesday, August 16, 2000 10:22 AM To: DB2-L@RYCI.COM Subject: Re: Question about shrlevel change reorg



Max,



I have reread you previous email and it falls into place. When a sharelevel reorg is preformed DB2 creates the shadow dataset in the storage group of the tablespace that is being reorged. I do not have the space in the storage group to contain 2 large tablespaces. The shadow tablespace is then renamed. [...] 5987 19 42_Re - Question about shrlevel change reorg-14_Massimo Scarpa23_Massimo_Scarpa@DEBIS.IT31_Wed, 16 Aug 2000 11:33:18 +0100478_us-ascii You can use DFSMSDSS (ADRDSSU) to move the tablespace to be reorged in the 'temp' DASD(s), it's faster and

you can move many tablespaces with wildcards in one job. Just a suggestion :-)))

Regards

Max Scarpa

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 6007 123 40_Re: Question about shrlevel change reorg19_Briggs, N. - Neil -25_N.Briggs@CANON-EUROPA.COM31_Wed, 16 Aug 2000 11:47:00 +0200571_iso-8859-1 Leslie,

Thanks for offering help, I work at a small shop. 1 DBA/SYSPROG (me) 1 systems prog. 650 users 5 databases 798 tables. Mainframe is slowly being wound down. So no cash to buy nice software tools and DASD. So I have to do these little frigs.

Though I would not recommend it because it really goes against the grain (philosophy) of the shrlevel reorg. But for us 24*7 availabilty is not really an issue. I have a couple of hours in the early moring to stop and start tablespaces if required. I use shrlevel reference because I find it [...] 6131 34 40_Re: Question about shrlevel change reorg13_Hanne Lyssand10_han@VPS.NO31_Wed, 16 Aug 2000 11:34:46 +0200378_us-ascii Dear Neil! You can solve your problem by using some dummy datasets. Add a new volume to your storage group. Allocate a large file on that volume at least the size of your tabespace When you do the reorg delete this dataset as the first step. As the last step in your reorg, allocate a new dummy dataset to the volume where your tablespace origanaly was allocatet. [...] 6166 57 40_Re: Question about shrlevel change reorg19_Briggs, N. - Neil -25_N.Briggs@CANON-EUROPA.COM31_Wed, 16 Aug 2000 12:07:07 +0200336_iso-8859-1 Dear Hanne,

Good solution, but unfortunately I have not got a spare volume to leave in the stogroup. It would be difficult to justify, to have a volume in the stogroup that would have a 30,000 track dummy dataset on it. If I had more volumes I would use your solution. It has been saved in my solutions database. [...] 6224 14 19_DB2 V5 to V7 Direct17_Harvey J. Puckett23_hpuckett@SPRINGMAIL.COM31_Wed, 16 Aug 2000 06:12:24 -0400471_- Any experience base out there on going from DB2 V51 to V71 w/o going to V61? We are at O/S390 V26 (or will be very soon now).

Good or bad results that you can share would be appreciated...Lessons learned, etc.

TIA

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 6239 36 0_20_Jahangir Nematollahi28_J_NEMATOLLAHI@ISC.IRANET.NET31_Mon, 14 Aug 2000 15:21:10 +043015_iso-8859-2 mail 6276 32 23_Re: DB2 V5 to V7 Direct16_Rasmussen, Steen22_Steen.Rasmussen@CA.COM31_Wed, 16 Aug 2000 11:45:46 +0100580_- remember V7 of DB2 (UDB server for OS/390 V7) is not available yet. Latest I have heard is GA december 2000.

Best regards Steen Rasmussen Computer Associates Denmark

> -----Original Message----- > From: Harvey J. Puckett [SMTP:hpuckett@SPRINGMAIL.COM] > Sent: 16 August 2000 12:12 > To: DB2-L@RYCI.COM > Subject: DB2 V5 to V7 Direct > > Any experience base out there on going from DB2 V51 to V71 w/o going to V61? We are at O/S390 V26 (or will be very soon now). > > Good or bad results that you can share would be appreciated...Lessons learned, etc. > > TIA [...] 6309 36 0_20_Jahangir Nematollahi28_J_NEMATOLLAHI@ISC.IRANET.NET31_Mon, 14 Aug 2000 15:35:15 +043025_iso-8859-2 set db2-l mail 6346 43 32_Datajoiner as an extraction tool13_Jaydeep Ghosh29_Jaydeep_Ghosh@PROGRESSIVE.COM31_Wed, 16 Aug 2000 07:45:42 -0400612_us-ascii Hi , I had posted a query pertaining to ETL tools which could be used for populating DB2 for OS390 tables , still waiting for responses . A solution which came up in our shop was Datajoiner , because some other dept. is using it , although not as an extraction tool . Could datajoiner be the suitable tool for the scenario described below ? Any suggestions , comments , etc. would be greatly appreciated . --------------------------------------- ( repost of my earlier mail ) ------------------------------------------------ Hi , I would appreciate suggestions for an ETL , or just an extract tool [...] 6390 79 40_Re: Question about shrlevel change reorg24_Peter_Schwarcz (Bigpond)26_Peter_Schwarcz@BIGPOND.COM31_Wed, 16 Aug 2000 22:23:48 +1000580_iso-8859-1 Neil,

The following process might achieve your requirement.

1. ALTER the current STOGROUP defined TABLESPACE to be VCAT defined. 2. IDCAMS allocate the Shadow data set on an appropriate volume. 3. REORG SHARELEVEL CHANGE to the shadow data set. Because we are VCAT defined DB2 will not do VSAM DELETES so after the REORG you have got both datasets on volumes of your choice. 4. If you want the dataset to go back to its original spot in the SMS pool, all I have to do is run another REORG SHARELEVEL CHANGE 5. DELETE the shadow dataset. 6. ALTER the [...] 6470 81 36_Re: Datajoiner as an extraction tool13_Umair Hussain25_umair_hussain@HOTMAIL.COM29_Wed, 16 Aug 2000 08:05:19 CDT556_- Yes, datajoiner would be a good tool for extracting data out of your different flavors of databases. Or you could intall db2udb v7 (aix,nt ...etc) which provides basic datajoiner functionality that I think would fulfill your needs. That way you wont have to pay separate for DJ.

regards Umair



Umair Hussain******************************** DBA Consultant * DB2 UDB, Oracle, Sybase, DataJoiner * IBM Certified Solution Expert - DB2 UDB DBA * IBM Certified AIX System Administrator * ********|All Disclaimers Apply|************** [...] 6552 33 22_VarChar as a Data type15_Morris, Deborah20_DMorris@UNCH.UNC.EDU31_Wed, 16 Aug 2000 09:06:31 -0400367_iso-8859-1 I have a question about using VarChar as a data type in a DB2 (Version 5 or 6) table.

Does the general rule still apply that you should not use it on a column if it's maximum length is less than 30?

If this is true, why is that the case, and what is the lower limit on the maximum length that is most efficient with a VarChar data type? [...] 6586 174 48_Re: Triggers - Where can I get more information?0_18_damcon2@US.IBM.COM31_Wed, 16 Aug 2000 09:40:52 -0400499_us-ascii Larry,

I would think you'd be able to setup a trigger similar to the following:

create trigger... before insert... begin atomic set new.col1 = (select col1 from parent_table where primary_key = new.primary_key and any other criteria that would determine which value from col1 to choose)

But, maybe the logic necessary to determine col1 is too much for one sql statement and needs to be in a stored procedure or udf. Then again maybe I'm totally missing something. [...] 6761 98 36_Re: Datajoiner as an extraction tool18_Krawetzky, Peter J21_KrawetzkyPJ@AETNA.COM31_Wed, 16 Aug 2000 09:46:33 -0400438_- We are using datajoiner to extract data from Oracle v7.3.4 to UDB V6.2 on solaris. One word of warning. You may have to add server mappings types or reverse mapping types to allow data to be compatiable between Informix and DB2. I hit problems with dates and need to add mapping types. The Datajoiner documenation has a pretty good explaination of this. The actual UDB views are SYSCAT.REVTYPEMAPPINGS and SYSCAT.SERVER_DATATYPES. [...] 6860 72 26_Re: VarChar as a Data type19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Wed, 16 Aug 2000 08:59:10 -0500621_ISO-8859-1 Deborah, See V6 Admin Guide, 2.4.3.4.1 String data types. An extract:

"Choosing CHAR or VARCHAR: VARCHAR saves DASD space, but costs a 2-byte overhead for each value and the additional processing required for varying-length records. Thus, CHAR is preferable to VARCHAR, unless the space saved by the use of VARCHAR is significant. The savings are not significant if the maximum length is small or the lengths of the values do not have a significant variation. In general, do not define a column as VARCHAR(n) unless n is at least 18. (Consider, also, using data compression if your main concern is [...] 6933 109 40_Re: Question about shrlevel change reorg15_Tonmoy Dasgupta32_tonmoy.dasgupta@MAIL.STATE.AR.US31_Wed, 16 Aug 2000 09:16:49 -0500652_iso-8859-1 Neil,

Why are you considering "shrlevel change" reorg when 1) you are contemplating using DSN1COPY to move your tablespace around and 2) you are severly short of DASD ?

If you can afford the luxury of DSN1COPYing around your tablespace chances are that you do not need "shrlevel change reorgs". You have to make the tablespace un-available to DB2 when you use DSN1COPY...whereas the other name of "shrlevel change reorg" is "online read-write reorg". You carry out these reorgs when availibility is critical. The reorg can run concurrently with quick-committing transactions (typically OLTP). At the tail end of the reorg [...] 7043 173 48_Re: Triggers - Where can I get more information?12_Steve Grimes30_Steve_Grimes@AISMAIL.WUSTL.EDU31_Wed, 16 Aug 2000 09:17:09 -0500623_us-ascii Two additional sources that may prove helpful are:

DB2 UDB Server fo OS/390 Version 6 Technical Update DB2 Mag., Qrtr 1, 2000: Spring.

As for a trigger on the child table which will use contents the parent table column, I think your out of luck without the UDF. I'm forwarding the response I received from IBM about this. (I encourage everyone to press for removal of this limitation. Minimally, the "Set with Subselect" should be enabled for "before" trigger transitional variables. Additionally, a stored procedure should be able to manipulate and return transitional variable as parameters.) [...] 7217 115 26_Re: VarChar as a Data type15_Slot, JP (Jaap)23_J.P.Slot@RF.RABOBANK.NL31_Wed, 16 Aug 2000 16:23:42 +0200450_iso-8859-1 Deborah,

Before compressing consider your already existing Varchars. The already existing Varchars will be treated as normal CHAR including the 2 byte overhead. Just adding my 2 cents worth of thought.

Jaap Slot



Phone 0031 (0)30 215 2220 Mail



Reserve your AGENDA for next DB2 Users Group Conference IDUG Geneva - 23/26 October 2000 - [...] 7333 42 34_Recovery from Stacked Image Copies12_Hall, Robert25_Robert.Hall@OCC.TREAS.GOV31_Wed, 16 Aug 2000 10:34:37 -0400569_iso-8859-1 Hello,

I had to recover 92 tablespaces today (in panic mode of course!) for the first time since I began "stacking" image copies on a single tape. Each stacked file on the tape is cataloged, so I figured I could just run the recover commands (without specifying 92 copy dd statements) and all would be fine. OS390 wanted to mount the tape for EVERY recovery statement! Needless so say the operators now hate me and the job took so long I eventually had to cancel it and specify each of the dd names with the vol=(,retain) statement. Once I coded [...] 7376 109 42_Chicago and Chicago suburbs, Please help!!12_Steve Fuller18_vantech@RADIKS.NET31_Wed, 16 Aug 2000 09:40:24 -0500577_iso-8859-1 Respected List,

I have a young lady, a DB2/IMS/SYBASE DBA, who just contacted me. Her husband just accepted a position in Chicago, and they have bought a home in Lisle, IL. They are moving from New York, and her husband's company is picking up the relocation.

My question is are any of you aware of a permanent (not contracting or consulting) position available? She is willing to ride the train downtown, and/or, drive to a suburb site. However, probably anything farther North or Northeast of Chicago O'Hare Airport, would be to long a commute. [...] 7486 206 48_Re: Triggers - Where can I get more information?0_18_damcon2@US.IBM.COM31_Wed, 16 Aug 2000 10:43:20 -0400487_us-ascii Larry,

Just saw Steve's reply. Didn't realize the subselect wasn't allowed on OS/390. I've only used triggers in the UNIX world where this does work.

Cheers, Jay



Jay Reavill damcon2@us.ibm.com IBM Global Services Tampa, Florida Tel: (813) 801-7303, Tie Line 8-427-7303 ----------------------------------------------------- Happiness is not around the corner. Happiness is the corner. - BMW ----------------------------------------------------- [...] 7693 241 48_Re: Triggers - Where can I get more information?14_Richard Yevich17_ryevich@YAHOO.COM31_Wed, 16 Aug 2000 07:47:08 -0700589_us-ascii Larry,

Triggers do not support subqueries as a method to update transition variables. Can only be done using a UDF in the SET statement:

SET A.B = UDF(parms);

Yes, subqueries can be used in the SET of a normal update, but not in the SET in a trigger.

--- damcon2@US.IBM.COM wrote: > Larry, > > I would think you'd be able to setup a trigger similar to the > following: > > create trigger... > before insert... > begin atomic > set new.col1 = (select col1 from parent_table where primary_key = > new.primary_key and any other criteria that would > [...] 7935 88 44_Re: SQL 1336N Remote Host not found problem?13_Carlton Enuda24_carltonenuda@HOTMAIL.COM29_Wed, 16 Aug 2000 10:51:07 EDT620_- Goodday Leslie and to All, No, leslie I don't think you are off track in anything...I will get answers to these Qs and take it from there... Thanks Carlton



>From: "Pendlebury-Bowe, Leslie" >Reply-To: DB2 Data Base Discussion List >To: DB2-L@RYCI.COM >Subject: Re: SQL 1336N Remote Host not found problem? >Date: Wed, 16 Aug 2000 06:16:38 +0100 > >...where do I begin. > >Some questions I think you may be asked by other list peeps : > >1. Are you accessing DB2 OS390 using Connect? >2. Assuming you are .. do you have DDF up and running on the [...] 8024 28 38_Re: Recovery from Stacked Image Copies0_23_pjackson@SUMMITBANK.COM31_Wed, 16 Aug 2000 10:47:25 -0400481_us-ascii Try using the MOUNT operator's command to assign the tape to a specific drive.

From a QuickRef manual:

MOUNT

4.22 MOUNT Command

Use the MOUNT command to allow allocation of an I/O device to all job steps that require a particular volume without intervening demountings and remountings of the volume. Because the system must schedule MOUNT commands, there is a short delay between when you issue the command and when the volume is mounted. [...] 8053 71 38_Re: Recovery from Stacked Image Copies23_Pendlebury-Bowe, Leslie34_Leslie.Pendlebury-Bowe@EU.SONY.COM31_Wed, 16 Aug 2000 15:50:21 +0100399_iso-8859-1 Hi Robert, I do beleive we chewed the fat on this last year .. you have to do the work. I have rexx that does it for us and in addition a 3rd party product. Sorry Leslie

-----Original Message----- From: Hall, Robert [mailto:Robert.Hall@OCC.TREAS.GOV] Sent: Wednesday, August 16, 2000 3:35 PM To: DB2-L@RYCI.COM Subject: Recovery from Stacked Image Copies Sensitivity: Private [...] 8125 101 38_Re: Recovery from Stacked Image Copies12_Hall, Robert25_Robert.Hall@OCC.TREAS.GOV31_Wed, 16 Aug 2000 11:05:41 -0400408_iso-8859-1 Hi Leslie,

Thanks for the info....I thought this had been discussed b4, but my search of the archives was unsucessful.... Also, the Utilities manual talks about how to prevent multiple mounts with JCL when the datasets are uncataloged, but doesn't specifically address cataloged datasets. Now I know why :)

At least now I know what I am up against for the next disaster drill!! [...] 8227 55 38_Re: Recovery from Stacked Image Copies23_Pendlebury-Bowe, Leslie34_Leslie.Pendlebury-Bowe@EU.SONY.COM31_Wed, 16 Aug 2000 16:03:36 +0100444_iso-8859-1 ahh .. but it does keep re-winding it doesn't it .... ?????? I think it does. which means you will be there all day long ... Les

-----Original Message----- From: pjackson@SUMMITBANK.COM [mailto:pjackson@SUMMITBANK.COM] Sent: Wednesday, August 16, 2000 3:47 PM To: DB2-L@RYCI.COM Subject: Re: Recovery from Stacked Image Copies



Try using the MOUNT operator's command to assign the tape to a specific drive. [...] 8283 68 40_Re: Question about shrlevel change reorg10_Bill Brown25_bill_brown@SPRINGMAIL.COM31_Wed, 16 Aug 2000 11:14:41 -0400602_- Using this method of DSN1COPY means that you will have to take an outage. If this is the case why use online reorg. Just take the outage and reorg in place.



On Wed, 16 Aug 2000 11:22:23 +0200 "Briggs, N. - Neil -" wrote:

Max,



I have reread you previous email and it falls into place. When a sharelevel reorg is preformed DB2 creates the shadow dataset in the storage group of the tablespace that is being reorged. I do not have the space in the storage group to contain 2 large tablespaces. The shadow tablespace is then renamed. [...] 8352 17 60_Re: DB2 for OS/390 Performance Monitor - which one? - thanks11_Shigang Mai16_smai@METLIFE.COM31_Wed, 16 Aug 2000 10:03:40 -0500487_- I find the discussion is very interesting. I just happen to be in the process of evaluating MAINVIEW FOR DB2 and INSIGHT FOR DB2. Is there anyone have good understanding of both and may share you finding? Thanks lot.





Sam

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 8370 85 40_Re: Question about shrlevel change reorg0_20_bjnigh@HOUSEHOLD.COM31_Wed, 16 Aug 2000 09:13:57 -0700747_us-ascii How did you get around SMS with this solution?









"Briggs, N. - Neil -" on 08/16/2000 02:22:23 AM

Please respond to DB2 Data Base Discussion List





To: DB2-L@RYCI.COM cc: (bcc: Barbara J. Nigh/Household International) Subject: Re: Question about shrlevel change reorg





Max,



I have reread you previous email and it falls into place. When a sharelevel reorg is preformed DB2 creates the shadow dataset in the storage group of the tablespace that is being reorged. I do not have the space in the storage group to contain 2 large tablespaces. The shadow tablespace is then renamed. Hence if [...] 8456 20 19_DB2 Tech Conference14_Wayne Driscoll24_Wayne.Driscoll@QUEST.COM31_Wed, 16 Aug 2000 09:24:15 -0700544_iso-8859-1 Is there a web site with the agenda of the upcoming DB2 Tech Conference? I searched IBM's site and found nothing, but I probably didn't use the right keywords. Thanks in advance. Wayne Driscoll Product Developer Quest Software Inc. wdriscoll@quest.com Note: All opinions are strictly my own.

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 8477 36 23_Re: DB2 Tech Conference13_Terry Shockey30_Terry.Shockey@OCSE.STATE.AR.US31_Wed, 16 Aug 2000 11:30:55 -0500584_iso-8859-1 HERE'S THE WEB SITE

http://www-3.ibm.com/services/learning/conf/index.html

-----Original Message----- From: Wayne Driscoll [mailto:Wayne.Driscoll@QUEST.COM] Sent: Wednesday, August 16, 2000 11:24 AM To: DB2-L@RYCI.COM Subject: DB2 Tech Conference



Is there a web site with the agenda of the upcoming DB2 Tech Conference? I searched IBM's site and found nothing, but I probably didn't use the right keywords. Thanks in advance. Wayne Driscoll Product Developer Quest Software Inc. wdriscoll@quest.com Note: All opinions are strictly my own. [...] 8514 40 23_Re: DB2 Tech Conference16_Rasmussen, Steen22_Steen.Rasmussen@CA.COM31_Wed, 16 Aug 2000 17:35:56 +0100727_- Hi Wayne;

Use this link and ther point at AGENDA

http://www-3.ibm.com/services/learning/conf/db2/index.html

Best rgds Steen Rasmussen Computer Associates Denmark

> -----Original Message----- > From: Wayne Driscoll [SMTP:Wayne.Driscoll@QUEST.COM] > Sent: 16 August 2000 18:24 > To: DB2-L@RYCI.COM > Subject: DB2 Tech Conference > > Is there a web site with the agenda of the upcoming DB2 Tech Conference? I > searched IBM's site and found nothing, but I probably didn't use the right > keywords. > Thanks in advance. > Wayne Driscoll > Product Developer > Quest Software Inc. > wdriscoll@quest.com > Note: All opinions are strictly my own. > > ================================================ [...] 8555 55 23_Re: DB2 Tech Conference9_Mike Mays26_MikeM@EGGERSCONSULTING.COM31_Wed, 16 Aug 2000 11:36:47 -0500690_- Does anyone know of a site similar to this for AS400? thanks in advance

Mike Mays Eggers Consulting



> -----Original Message----- > From: Terry Shockey [SMTP:Terry.Shockey@OCSE.STATE.AR.US] > Sent: Wednesday, August 16, 2000 11:31 AM > To: DB2-L@RYCI.COM > Subject: Re: DB2 Tech Conference > > HERE'S THE WEB SITE > > http://www-3.ibm.com/services/learning/conf/index.html > > -----Original Message----- > From: Wayne Driscoll [mailto:Wayne.Driscoll@QUEST.COM] > Sent: Wednesday, August 16, 2000 11:24 AM > To: DB2-L@RYCI.COM > Subject: DB2 Tech Conference > > > Is there a web site with the agenda of the upcoming DB2 Tech Conference? > I > searched IBM's site [...] 8611 44 23_Re: DB2 Tech Conference36_Enarson, Mike (CC-Database Services)24_Mike.Enarson@CONAGRA.COM31_Wed, 16 Aug 2000 11:36:19 -0500748_iso-8859-1 Wayne, the IBM web pages generally have a site for all their tech conferences. Selecting DB2 from that page takes me to:

www-3.ibm.com/services/learning/conf/db2/index.html





Regards Mike Enarson ConAgra, Inc.



-----Original Message----- From: Wayne Driscoll [mailto:Wayne.Driscoll@QUEST.COM] Sent: Wednesday, August 16, 2000 11:24 AM To: DB2-L@RYCI.COM Subject: DB2 Tech Conference



Is there a web site with the agenda of the upcoming DB2 Tech Conference? I searched IBM's site and found nothing, but I probably didn't use the right keywords. Thanks in advance. Wayne Driscoll Product Developer Quest Software Inc. wdriscoll@quest.com Note: All opinions are strictly my own. [...] 8656 73 23_Re: DB2 Tech Conference16_Rasmussen, Steen22_Steen.Rasmussen@CA.COM31_Wed, 16 Aug 2000 17:54:20 +0100672_- Hi Mike - use this link and you have ALL IBM's conferences

http://www-3.ibm.com/services/learning/conf/

best regards Steen Rasmussen Computer Associates Denmark

> -----Original Message----- > From: Mike Mays [SMTP:MikeM@EGGERSCONSULTING.COM] > Sent: 16 August 2000 18:37 > To: DB2-L@RYCI.COM > Subject: Re: DB2 Tech Conference > > Does anyone know of a site similar to this for AS400? thanks in advance > > Mike Mays > Eggers Consulting > > > > -----Original Message----- > > From: Terry Shockey [SMTP:Terry.Shockey@OCSE.STATE.AR.US] > > Sent: Wednesday, August 16, 2000 11:31 AM > > To: DB2-L@RYCI.COM > > Subject: Re: DB2 Tech Conference > > [...] 8730 78 23_Re: DB2 Tech Conference36_Enarson, Mike (CC-Database Services)24_Mike.Enarson@CONAGRA.COM31_Wed, 16 Aug 2000 11:39:46 -0500529_iso-8859-1 Mike, select www-3.ibm.com/services/learning/conf/index.html and then select the AS400 technical conference. That web address contains links to all of IBM's technical conferences.



Regards Mike Enarson ConAgra, Inc.





-----Original Message----- From: Mike Mays [mailto:MikeM@EGGERSCONSULTING.COM] Sent: Wednesday, August 16, 2000 11:37 AM To: DB2-L@RYCI.COM Subject: Re: DB2 Tech Conference



Does anyone know of a site similar to this for AS400? thanks in advance [...] 8809 35 60_Re: DB2 for OS/390 Performance Monitor - which one? - thanks20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM31_Wed, 16 Aug 2000 13:23:08 -0400560_us-ascii Will CA continue to support CA-INSIGHT in future? Or are they offering Platinum detector as a replacement?



-----Original Message----- From: Shigang Mai [SMTP:smai@METLIFE.COM] Sent: Wednesday, August 16, 2000 11:04 AM To: DB2-L@RYCI.COM Subject: Re: DB2 for OS/390 Performance Monitor - which one? - thanks

I find the discussion is very interesting. I just happen to be in the process of evaluating MAINVIEW FOR DB2 and INSIGHT FOR DB2. Is there anyone have good understanding of both and may share you finding? Thanks lot. [...] 8845 45 60_Re: DB2 for OS/390 Performance Monitor - which one? - thanks13_Thomas, Janis19_Janis.Thomas@CA.COM31_Wed, 16 Aug 2000 13:29:22 -0400449_iso-8859-1 There are absolutely no plans to replace CA-Insight. It will continue to be supported & enhanced.

Janis Thomas DB2 Product Management Computer Associates International, Inc. janis.thomas@ca.com 630-481-1830





-----Original Message----- From: Pillay, Venkat (PCA) [mailto:venkat_pillay@ML.COM] Sent: Wednesday, August 16, 2000 12:23 PM Subject: Re: DB2 for OS/390 Performance Monitor - which one? - thanks [...] 8891 38 23_Re: DB2 Tech Conference18_Teegarden, Stephen22_Teegarden@AEGONUSA.COM31_Wed, 16 Aug 2000 12:30:41 -0500699_- http://www-3.ibm.com/services/learning/us/

Steve Teegarden, DBA AEGON USA 319-398-8066

> -----Original Message----- > From: Wayne Driscoll [SMTP:Wayne.Driscoll@QUEST.COM] > Sent: Wednesday, August 16, 2000 11:24 AM > To: DB2-L@RYCI.COM > Subject: DB2 Tech Conference > > Is there a web site with the agenda of the upcoming DB2 Tech Conference? > I > searched IBM's site and found nothing, but I probably didn't use the right > keywords. > Thanks in advance. > Wayne Driscoll > Product Developer > Quest Software Inc. > wdriscoll@quest.com > Note: All opinions are strictly my own. > > ================================================ > To change your subscription options or to [...] 8930 87 38_Re: Recovery from Stacked Image Copies19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Wed, 16 Aug 2000 12:29:43 -0500562_ISO-8859-1 Hi Les, pjackson, We were sneaky at a site I use to work at -- we Image Copied locally in the order we recovered at the DR site to stacked tapes and used the MOUNT command.

HTH, Rick Davis

"This e-mail and any files transmitted with it are the property of SBC, are confidential, and are intended solely for the use of the individual or entity to whom this e-mail is addressed. If you are not one of the named recipient(s) or otherwise have reason to believe that you have received this message in error, please notify the sender at [...] 9018 25 56_Any way to load another database using Full Image copies16_Thirumal Valavan31_thirumal.valavan@PRUDENTIAL.COM31_Wed, 16 Aug 2000 13:35:59 -0400391_us-ascii Hi,

I have got a small doubt. Is there any way to load another database in the same subsystem using full image copies of a database. I used platinum FAST UNLOAD .But while loading data in to another database in the same subsystem It is encountering problems as the length does n't match. Shall I know the reason?. Is anything I am missing or anything I am doing wrongly [...] 9044 34 32_Re: AW: V5 to V6 migration error11_Hayden, Lee18_Lee_Hayden@CSX.COM31_Wed, 16 Aug 2000 13:42:57 -0400403_iso-8859-1 Roger,

I this DSNTIJPM job available on the net somewhere if we have not yet ordered/installed V6 ? I would like to be as pro-active as I can. ;-)



-----Original Message----- From: Roger Miller [mailto:millerrl@US.IBM.COM] Sent: Tuesday, August 15, 2000 4:52 PM Subject: Re: AW: V5 to V6 migration error



I'd like to twist just a bit on what you said. [...] 9079 125 32_Re: Commit & Close Cursor w/CICS19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Wed, 16 Aug 2000 12:45:34 -0500536_ISO-8859-1 Hi All, Please considered the following CICS/DB2 situation/environment:

Several of your 'bread-and-butter' transactions tend to hit the same DB2 DASD hot spots (gee, what's new in that, hehe). These transactions have very high arrival rates. They always show up together. Several transactions change 'stuff' others simply read.

Thus: maximum concurrency is required. In this situation, if any read-only transaction contained two or more LUWs, adding a SYNCPOINT after each LUW will allow more concurrency. [...] 9205 23 28_FW:migrating from 5.2 to 7.114_John Lawniczak26_jlawniczak@RTSWIRELESS.COM31_Wed, 16 Aug 2000 13:50:03 -0400503_iso-8859-1 Hi list, Has anyone migrated DB2 UDB on NT from 5.2 to 7.1 ? Did it matter how much data you had or is it a "system" only migration? Thanks,

John Lawniczak DBA Database Operations Phone: (516) 918-4571 email: johnl@rtswireless.com RTS Wireless

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 9229 47 60_Re: Any way to load another database using Full Image copies13_Shapiro, Dave26_Shapiro.Dave@PRINCIPAL.COM31_Wed, 16 Aug 2000 12:53:35 -0500513_iso-8859-1 Thirumal,

You might try using DSN1COPY. Just follow the instructions in the Utility Guide for converting the DBID, PSID, OBID etc. The structure of the source table and the target table must match, but image copies are a valid source for this facility.

Dave

-----Original Message----- From: Thirumal Valavan [mailto:thirumal.valavan@prudential.com] Sent: Wednesday, August 16, 2000 12:36 PM To: DB2-L@RYCI.COM Subject: Any way to load another database using Full Image copies [...] 9277 19 20_Linux DB2 Case Study30_Jeffrey I Condon/Cupertino/IBM19_jicondon@US.IBM.COM31_Wed, 16 Aug 2000 14:06:08 -0400360_us-ascii I was at the Linux World conference and I ran across a non-IBM girls who was yelling, "I love DB2". Yesterday IBM developerWorks Open Source Zone posted a Case Study on the early developments of IBM DB2 for Linux. In this piece members of the IBM dB2 team tell their story and offer insights. I just though some of you DB2 lovers may enjoy this. [...] 9297 23 15_DDF Performance11_Kevin Davis18_kevin.davis@DB.COM31_Wed, 16 Aug 2000 13:23:20 -0600483_us-ascii Hi, Everyone

I have a question maybe someone can help me with. We've moved our Dev subsystem on a different machine from our Prod. When using DDF to access the Dev or Prod, through SPUFI were finding it to be extremely slow, but the response when performing the same SQL in QMF is acceptable. Ex. retrieving 900 rows is taking 5 to 6 minutes in SPUFI and just a few seconds in QMF. We've looked at the plans and zparms. Can anyone shed some light on this issue? [...] 9321 75 60_Re: Any way to load another database using Full Image copies19_Rajeev P. Dhanawade22_rdhanawade@STATE.DE.US29_Wed, 16 Aug 2000 14:18:56 EDT375_us-ascii Dave,

I think since he is getting a length mismatch with the Platinum Load, he will also encounter a problem with DSN1COPY.

Thirumal,

Before you do a DSN1COPY, drop the table definition from the target database and redefine the table using the DDL from the source database. It should be easy to migrate the definition using Platinum tools. [...] 9397 47 19_Re: DDF Performance13_Fletcher, Jim21_jim.fletcher@BSIS.COM31_Wed, 16 Aug 2000 14:24:26 -0400685_iso-8859-1 I believe QMF automatically appends FOR FETCH ONLY to queries, causing network blocking. Try adding FOR FETCH ONLY to the SPUFI query and see what happens.

jim.fletcher@bsis.com

-----Original Message----- From: Kevin Davis [mailto:kevin.davis@DB.COM] Sent: Wednesday, August 16, 2000 3:23 PM To: DB2-L@RYCI.COM Subject: DDF Performance



Hi, Everyone

I have a question maybe someone can help me with. We've moved our Dev subsystem on a different machine from our Prod. When using DDF to access the Dev or Prod, through SPUFI were finding it to be extremely slow, but the response when performing the same SQL in QMF is acceptable. Ex. [...] 9445 47 19_Re: DDF Performance9_Mike Mays26_MikeM@EGGERSCONSULTING.COM31_Wed, 16 Aug 2000 13:35:50 -0500591_- Forgive my question again: Can anyone tell me where to find a discussion users list Just like this one but about AS400?? not the IBM site either. Thanks in advance

Mike Mays Eggers Consulting

> -----Original Message----- > From: Kevin Davis [SMTP:kevin.davis@DB.COM] > Sent: Wednesday, August 16, 2000 2:23 PM > To: DB2-L@RYCI.COM > Subject: DDF Performance > > Hi, Everyone > > I have a question maybe someone can help me with. We've moved our Dev > subsystem on a different machine from our Prod. When using DDF to access > the Dev or Prod, through SPUFI were finding [...] 9493 70 60_Re: Any way to load another database using Full Image copies12_Keister, Jim16_JKeister@QRS.COM31_Wed, 16 Aug 2000 11:38:40 -0700518_iso-8859-1 Thirumal,

If you also own Platinum's Fast Recover utility, you can use an image copy input from one tablespace to recover to another tablespace. Check out the OBIDXLAT option of Platinum Fast Recover. This assumes that you are not doing RESUME-type load processing.



-----Original Message----- From: Shapiro, Dave [mailto:Shapiro.Dave@PRINCIPAL.COM] Sent: Wednesday, August 16, 2000 10:54 AM To: DB2-L@RYCI.COM Subject: Re: Any way to load another database using Full Image copies [...] 9564 188 19_Re: DDF Performance11_Pedro Cunha20_pplfilho@SUPER11.NET31_Wed, 16 Aug 2000 15:37:48 -0300404_iso-8859-1 Hello Kevin,

This can be due to other parameters in your Operational System Environment. Check with your MVS/OS390 systems programmers to see what could be causing the delay. Could be TSO AS Dispatching priority too low or something to do with WLM if your shop uses WLM to manage your environment. Is QMF running under CICS. So there are quite a few questions to be answered here. [...] 9753 42 32_Re: FW:migrating from 5.2 to 7.114_Marilyn Stokes28_mstokes@OPPENHEIMERFUNDS.COM31_Wed, 16 Aug 2000 12:58:15 -0600330_us-ascii I've been told by IBM that you cannot skip 6.1, on this platform. The OS/390 you could skip 6.1. m

>>> "John Lawniczak" 08/16 11:50 AM >>> Hi list, Has anyone migrated DB2 UDB on NT from 5.2 to 7.1 ? Did it matter how much data you had or is it a "system" only migration? Thanks, [...] 9796 19 9_Rollbacks9_Sally Mir22_Sally.Mir@WACHOVIA.COM31_Wed, 16 Aug 2000 15:15:00 -0400508_us-ascii Can you please verify that the only negative SQL code that will rollback changes is the -911, or is there any other condition (other than an explicit ROLLBACK call) that will rollback work within a unit of work?

Thanks,

Sally Mir Wachovia Bank

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 9816 74 20_ACCESS PATH QUESTION16_Jeff A L'Italien25_Jeff.A.L'Italien@AEXP.COM31_Wed, 16 Aug 2000 12:38:48 -0700320_- Esteemed list,

I'd like to know if anyone would be able to offer some advice specific to access path selection. The problem in question is trying to get the optimizer to choose one index over another index. The SQL is as follows:

SELECT * FROM TT3.TBC_PLST WHERE NBR_CARDH_ACCT = ? AND NBR_PLST = ? [...] 9891 103 24_Re: ACCESS PATH QUESTION9_Mark Ruhe19_Mark.Ruhe@QUEST.COM31_Wed, 16 Aug 2000 12:53:18 -0700459_iso-8859-1 Jeff,

I am assuming your using DB2 OS/390. If your version 6.0 you can use hints to get the result you wish. If not give this a shot;

SELECT * FROM TT3.TBC_PLST WHERE NBR_CARDH_ACCT = ? AND (NBR_PLST = ? or 0 = 1)

Adding the 0=1 predicate should eliminate consideration of column NBR_PLST. You could also add additional NBR_CARDH_ACCT = ? predicates to increase the filter factor for this column making it more selective. [...] 9995 101 24_Re: ACCESS PATH QUESTION20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM31_Wed, 16 Aug 2000 16:15:01 -0400567_iso-8859-1 Jeff

The SQL you have shown us has "=" predicate. This means that you will always get one row as result and if you are not executing this SQL in a loop then prefetch is not of any use.

If you are executing the same SQL in a loop than make sure that your input data should come in ascending order of NBR_CARDH_ACCT to kick in dynamic prefetch. In this situation the second index could give a good performance due to very effective "index look-aside" (because input data is in the order). To encourage use of second index catalog tweaking [...] 10097 68 34_Permanent Delete of DB, TS, and TB16_Brookman, Gerald32_gerald.brookman@US.ORIGIN-IT.COM31_Wed, 16 Aug 2000 16:06:03 -0400491_iso-8859-1 Environment: DB2 for OS/390 V3.1 (not a typo!)

We have identified over 100 tables where the underlying VSAM clusters have been deleted (not migrated) sometime in the last 2-3 years. RUNSTATS on the DB.TS abends (of course) with the VSAM data set as an "unavailable resource". We wish to delete these permanently from the DB2 catalogs. DB2 Developer's Guide (4th Ed.) suggests DROP ..., then MODIFY RECOVERY ... DELETE AGE(*). Any other ideas? REPAIR DBD DROP...? TIA. [...] 10166 38 38_Re: Permanent Delete of DB, TS, and TB15_Billings, Linda30_linda.billings@DOA.STATE.WI.US31_Wed, 16 Aug 2000 15:24:22 -0500493_iso-8859-1 Hi, Gerald,

First, our sympathies for being on V3 yet....All you have to do is drop the object and everything goes away. There is no need to perform the modify. The drop gets rid of all image copies, and entries in the catalog about the object.

Linda Billings

-----Original Message----- From: Brookman, Gerald [mailto:gerald.brookman@US.ORIGIN-IT.COM] Sent: Wednesday, August 16, 2000 3:06 PM To: DB2-L@RYCI.COM Subject: Permanent Delete of DB, TS, and TB [...] 10205 65 32_Re: FW:migrating from 5.2 to 7.118_Gert van der Kooij15_geko@WANADOO.NL31_Wed, 16 Aug 2000 22:37:38 +0200522_iso-8859-1 Hi, I migrated directly from NT 5.2 to 7.1 (Workgroup Edition) without any problem but it's not a system only migration. You have to migrate your databases as well. To be sure they can be migrated run the db2chkmg tool before installing 7.1 from the 7.1 CD.

Hope this helps

----- Original Message ----- From: "Marilyn Stokes" Newsgroups: bit.listserv.db2-l To: Sent: Wednesday, August 16, 2000 8:58 PM Subject: Re: FW:migrating from 5.2 to 7.1 [...] 10271 38 13_Re: Rollbacks14_Grainger, Phil20_Phil.Grainger@CA.COM31_Wed, 16 Aug 2000 22:15:55 +0100351_iso-8859-1 Sally,

An error in a stored procedure or UDF will normally cause a rollback and will give a different SQL code.

For example, calling a Stored Procedure that performs a database update from a before trigger (where database updates are not allowed) gives you a -817 and a rollback.

Phil Grainger Computer Associates [...] 10310 37 36_IBM Online Reorg - Words of Wisdom??13_Frost, Pamela27_Pamela.Frost@PACIFICORP.COM31_Wed, 16 Aug 2000 14:15:36 -0700408_iso-8859-1 We are planning on going to the use of "online" reorgs in my shop. We are currently on DB2 V5 and it is not clear at this time when we might upgrade to DB2 V6 or V7. * Have you found any significant incidents of application outages when using online reorg? (I know, "significant" is a relative term - depends on your client!) What steps did you take to help minimize any application outages. [...] 10348 31 109_Sept 26th-27th - Part I/Part II of Stored Procedures - Linda Claussen - Central PA DB2 Users Group - $150 PP.10_Cathy Peck23_cathy.peck@HIGHMARK.COM31_Wed, 16 Aug 2000 17:52:32 -0400384_us-ascii Hi everyone! Just a reminder that Linda Claussen is coming to Central PA to teach 2 separate 1 day Stored Procedure Classes. Part 1 basically covers coding the Stored Procedure. Part 2 basically covers the connectivity to the Stored Procedures. If you are interested in Linda's bio & the course abstracts, see the Calendar of Events on our UG website: www.db2parug.org. [...] 10380 32 18_Identity Revisited27_MORRISON, WILLIAM F (SBCSI)21_wm8728@TXMAIL.SBC.COM31_Wed, 16 Aug 2000 16:45:04 -0500382_ISO-8859-1 Team DB2,

I have been following this Identity column thread closely, and would like to add to the discussion (and confusion??).

(1) Is it true that if your Identity col is set up a Generated Always, you cannot use the Load utility? If so, would anyone other than myself want the Load utility modified to allow this? We are also pursuing this with IBM. [...] 10413 58 22_Re: Identity Revisited13_Seibert, Dave26_Dave_Seibert@COMPUWARE.COM31_Wed, 16 Aug 2000 20:15:38 -0400584_iso-8859-1 Hold the phone! Bil.

I didn't see anyone say and I sure hope I didn't say that you can't use the Load utility with Identity columns specified as generated always. Perhaps the multiple messages and comments on messages misled you to that conclusion.

Here are observations I've shared: If 1) you have Generated Always Identity columns, and 2) you want to use the generated ID number for some visible key or as part of an RI relationship, then 3) you need to be aware that the generated number is not guaranteed to be preserved in the case of unload, drop, [...] 10472 31 49_Help with Increasing Archive Log space allocation13_Carlton Enuda24_carltonenuda@HOTMAIL.COM29_Wed, 16 Aug 2000 21:16:59 EDT633_- Hi list, I started noticing the event today evry time start the development DB2 for OS/390 subsystem: I want to increase the space allocation for the archive logs: What is the best/standard way to accomplish. The error message Iget each time is listed below:





IEC030I B37-04,IFG0554A,DSN1MSTR,DSN1MSTR,SYS00008,0306,DSK306, DSN610.ARCHLOG1.D00229.T2109140.A0000944 IEA989I SLIP TRAP ID=XB37 MATCHED. JOBNAME=DSN1MSTR, ASID=00FF. DSNJ128I - LOG OFFLOAD TASK FAILED FOR ACTIVE LOG DSN610.LOGCOPY1.DS0 IEC030I B37-04,IFG0554A,DSN1MSTR,DSN1MSTR,SYS00010,0306,DSK306, DSN610.ARCHLOG2.D00229.T2109140.A0000944 [...] 10504 162 24_Re: ACCESS PATH QUESTION14_Richard Yevich21_ryevich@ATTGLOBAL.NET31_Wed, 16 Aug 2000 20:23:58 -0400556_iso-8859-1 Jeff/Venkat,

The following should be the order of "fixing" a single statement, based on the statements in this thread, assuming there is no other way of writing the statement.

1 - Use :hv CONCAT '' or :hv + 0 to remove the column from the index selection process. This expression is removed and the predicate will still be available for stage 1 processing. Stop using the older methods of 0=1 or 0 <> 0. 2 - Use hints in V6 only if no other method can achieve the result (unlikely in this case) 3 - Catalog updates - but this [...] 10667 69 53_Re: Help with Increasing Archive Log space allocation14_James Campbell29_James.Campbell@HANCORP.COM.AU31_Thu, 17 Aug 2000 12:35:00 +1000527_iso-8859-1 Carlton

Re-run the install CLIST and modify the space allocation. Generally, the space should be the save as the active logs, and blksize 24k (for 3390)

/* standard disclaimer */ James Campbell DBA Hansen Corporation, Doncaster +61 3 9843 8442 James.Campbell@Hancorp.com.au



-----Original Message----- From: Carlton Enuda [mailto:carltonenuda@HOTMAIL.COM] Sent: Thursday, August 17, 2000 11:17 AM To: DB2-L@RYCI.COM Subject: [DB2-L] Help with Increasing Archive Log space allocation [...] 10737 107 22_Re: Identity Revisited14_James Campbell29_James.Campbell@HANCORP.COM.AU31_Thu, 17 Aug 2000 12:46:27 +1000453_iso-8859-1 Bil

1) I think rather than modifying the load utility, it would be better if one could alter the table to change between generated always and by default and to change the generated numbers. (And, Dave, I think Bil is referring to the circumstances that produce a DSNU269I message.)

2) What problem do you foresee in using CACHE n with RI or uniqueness? (Dave had better be carefull, he might end up with the Janitor's id!) [...] 10845 74 38_Re: Permanent Delete of DB, TS, and TB14_James Campbell29_James.Campbell@HANCORP.COM.AU31_Thu, 17 Aug 2000 12:51:29 +1000459_iso-8859-1 Linda,

Remember DROP doesn't cleanup SYSLGRNG? Which is why Gerald should do his MODIFY RECOVERY first.

/* standard disclaimer */ James Campbell DBA Hansen Corporation, Doncaster +61 3 9843 8442 James.Campbell@Hancorp.com.au -----Original Message----- From: Billings, Linda [mailto:linda.billings@DOA.STATE.WI.US] Sent: Thursday, August 17, 2000 6:24 AM To: DB2-L@RYCI.COM Subject: Re: [DB2-L] Permanent Delete of DB, TS, and TB [...] 10920 20 50_Mark Anzmann/MCIG/USA/Zurich is out of the office.12_Mark Anzmann25_mark.anzmann@ZURICHUS.COM31_Thu, 17 Aug 2000 01:01:18 -0400475_us-ascii I will be out of the office from 08/10/2000 until 08/24/2000.

If you need DBA support, please contact John Lantz, Rob Seay, or John Gray.

I will respond to your message when I return.

Thanks!

Mark

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 10941 99 24_Re: ACCESS PATH QUESTION7_Sanjeev21_Sanjeev@INTERRAIT.COM31_Thu, 17 Aug 2000 11:02:42 +0530474_iso-8859-1 Hi Jeff, I believe this is not the only sql running on the table.So, it is better to do modifications in the query rather than changing the catalog statistics. I think (0=1) technique is the best possible solution for this.

Hope this helps

Regards

Sanjeev

-----Original Message----- From: Jeff A L'Italien [mailto:Jeff.A.L'Italien@AEXP.COM] Sent: Thursday, August 17, 2000 1:09 AM To: DB2-L@RYCI.COM Subject: ACCESS PATH QUESTION [...] 11041 113 38_Re: Recovery from Stacked Image Copies23_Pendlebury-Bowe, Leslie34_Leslie.Pendlebury-Bowe@EU.SONY.COM31_Thu, 17 Aug 2000 06:49:34 +0100485_iso-8859-1 Rick I still not see how this will help .. i.e. do a mount private .. so the tape is on the drive for good .. then you issue multiple, in order, recoveries(same step ... but it will keep rewinding the tape - ????

I need to check the mount command .. maybe it can prevent the rewind from occuring. I did look at this in detail last year and I came to the conclusion that there was no way of stopping the rewind unless you coded RETAINS. your thoughts Rick et al. [...] 11155 42 30_Sync. and Async. writes(Again)7_Sanjeev21_Sanjeev@INTERRAIT.COM31_Thu, 17 Aug 2000 11:37:06 +0530383_iso-8859-1 Hi All, I was going through the new presentation on bufferpool tuning by Joel Goldstien in the resposive system website and got some new things(confusion) about the synch. writes from BP to DASD. DB2 writes asynchronously when DWQT,VDWQT reaches, checkpoint takes place. DB2 writes synchronously when IWTH,DMTH and SPTH reaches,DSMAX reaches and two more conditions. [...] 11198 44 34_Re: Sync. and Async. writes(Again)0_35_paul.a.redhead@TRANSPORT.QLD.GOV.AU31_Thu, 17 Aug 2000 17:37:51 +1000604_us-ascii Sanjeev, You'll have to forgive me if I'm still confused about the question. However, I'll attempt to answer what I think you asked by saying that my understanding is that DB2 will NOT externalize any page that has not been committed...period. This would cause the recovery problems I believe your alluding to. The two checkpoint stuff refers, I believe, to "hot" pages that are constantly the subject of update. If two consecutive checkpoints pass (and any number of commits) where the page is not externalized then at the time of the next commit it will be synchronously written to disk. [...] 11243 70 34_Re: Sync. and Async. writes(Again)7_Sanjeev21_Sanjeev@INTERRAIT.COM31_Thu, 17 Aug 2000 14:07:51 +0530609_iso-8859-1 Hi Paul, Thanks for the reply and you are correct with the understanding of my question.However, your words DB2 will not externalized the page unless until it is commited needs some more discussion from my point of view. Suppose there is a scenario like this :

-----A------B------C

Let's say Page-1 is an updated page scheduled to be written at checkpoint A, just before DB2 encounters the checkpoint at A, page-1 is taken by other process for update.So cann't be written at A.Again at point B that process issues a Commit. 1) Will DB2 externalize that page ? The other scenario [...] 11314 148 22_Re: Identity Revisited0_18_mebert@AMADEUS.NET31_Thu, 17 Aug 2000 10:48:49 +0200624_us-ascii We don't use Identity columns (being V5/OS390), and I hope nobody is planning to use them. The Unload/Drop/Create/Load technique is a standard process here (probably elsewhere as well) for changing column datatypes, dropping columns, changing partitioning indexes and so on. I would also think that in many if not most cases an Identity column will be in a primary key and/or involved in RI (DB2 or app). If the above process (specifically, the LOAD utility) scrambles a (existing, visible) data value, it's far worse than having it not work at all, in my opinion. Somebody, sometime will certainly badly mess [...] 11463 53 40_Re: IBM Online Reorg - Words of Wisdom??14_Grainger, Phil20_Phil.Grainger@CA.COM31_Thu, 17 Aug 2000 09:55:03 +0100394_iso-8859-1 I think most of the major changes in OnLine Reorg between V5 and V6 concern the processing during the switch phase. This includes such things as parallel dataset renames (as well as parallel dataset create/delete during initialisation and termination). When I did my benchmarks (admittedly over a year ago) there was no appreciable difference in the Reorg time other than this. [...] 11517 140 40_Re: IBM Online Reorg - Words of Wisdom??0_18_mebert@AMADEUS.NET31_Thu, 17 Aug 2000 11:19:09 +0200596_us-ascii We have implemented OLR in our OLTP system about one year ago, and it is working fine. Sometimes there are problems (B37s etc.) but recovering is easy - just TERM the utility. But it's not "always better" than standard REORG:

1. The main problem is with partitioned Tablespaces with NPIs: either you reorg by partition, in which case you have a long BUILD2 phase, making the NPIs (the logical partition) unavailable, or you reorg the full TS, which will require a looong SWITCH phase, lots of disk for the shadow datasets, may do much unnecessary work (if not all partitions [...] 11658 114 34_Re: Sync. and Async. writes(Again)14_James Campbell29_James.Campbell@HANCORP.COM.AU31_Thu, 17 Aug 2000 19:21:10 +1000321_iso-8859-1 If DB2 never externalised a page unless it was committed, then it would never have to read the UNDO records in the log (because the pre-update data is still in the source cluster). But DB2 _does_ read the UNDO records, therefore DB2 must think that it is possible for uncommitted data to be externalised. [...] 11773 135 34_Re: Sync. and Async. writes(Again)15_Lankester, Andy22_Andy_Lankester@BMC.COM31_Thu, 17 Aug 2000 04:44:40 -0500495_iso-8859-1 As I understand it the rule is 'DB2 never externalises _unlogged_ data'. This is so that externalised uncommitted data can be backed out (using, as James says, the UNDO records) in the event of a rollback, system restart etc.

Andy Lankester BMC Software Ltd mailto:andy_lankester@bmc.com

-----Original Message----- From: James Campbell [mailto:James.Campbell@HANCORP.COM.AU] Sent: 17 August 2000 10:21 To: DB2-L@RYCI.COM Subject: Re: Sync. and Async. writes(Again) [...] 11909 41 13_DDF CPU Usage14_David Williams25_David-O.Williams@UBSW.COM31_Thu, 17 Aug 2000 10:30:33 +0100328_US-ASCII We are experiencing a strange problem when using DB2 Connect. We have an application running on a Sun Solaris box which accesses DB2 OS390 stored procedures via DB2 Connect. We are getting a large amount of CPU usage in the SPAS address space, as expected, but between 5 - 10 times this amount in the DDF enclave! [...] 11951 25 0_9_James Tse18_jamestse@YOUPY.COM31_Thu, 17 Aug 2000 09:55:41 +0000336_us-ascii I install dbi for Linux DB2 and get error message as "DB2_HOME" value must set to DB2 install directory.

Some expert of perl dbi teach me that "DB2_HOME must be set to /usr/IBMdb2/v6.3" not to set "usr/IBMdb2/

Does anyone know how to set DB2_HOME? using command or edit the parameter file or something else? [...] 11977 103 24_Re: ACCESS PATH QUESTION10_John Hardy27_john_j_hardy@CONSULTANT.COM31_Thu, 17 Aug 2000 05:05:55 -0500623_- Jeff,

I was just wondering about the rationale behind your always 'priming' the stats prior to a bind.

You will obviously always have to 'fix' this particular query, as you are aware of some information that the optimiser cannot possibly know, i.e. that you are issuing the query iteratively in the order of the second index, and therefore wish to exploit dynamic runtime options (namely sequential detection and index lookaside). However, in the general case, are you sure that the optimiser would not usually choose the best possible access path if it had accurate Statistics from a recent Runstats? [...] 12081 14 13_Timestampdiff18_Krawetzky, Peter J21_KrawetzkyPJ@AETNA.COM31_Thu, 17 Aug 2000 07:03:31 -0400424_- Does anyone have the layout and an example of the timestampdiff function? No matter what combinations I try, it won't work. Thanks!

Peter J. Krawetzky DataBase Administrator

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 12096 105 16_Re: 3 Part query14_William Gannon41_wgannon@INET-SERVICES.PALMBEACH.K12.FL.US31_Thu, 17 Aug 2000 07:29:03 -0400304_us-ascii I don't know if this applies to DB2 Ver 5.2 on NT - but Oracle supports two things called Database links and Synonyms. Using a Database link you define another instance, db and signon and defining a synonym on that definition you could make the remote table appear local to the application [...] 12202 103 38_Re: Permanent Delete of DB, TS, and TB15_Billings, Linda30_linda.billings@DOA.STATE.WI.US31_Thu, 17 Aug 2000 07:09:55 -0500536_iso-8859-1 James,

In our situation, if an underlying VSAM is deleted it is because it hasn't been accessed for over 2 years by anyone. SMS rolls it to ML1 (slow DASD) and then ML2 (tape) and then it is deleted at the end of two years. In that time I think it is safe to assume that no entries have been made in SYSLGRNG for the object. We execute MODIFY RECOVERY on a regular basis and only keep three weeks of image copies. My assumption is that nothing is left in SYSLGRNG to delete by the time SMS deletes the underlying [...] 12306 48 27_Re: OS/390 DB2 DBA POSITION11_rick creech18_ykcirc@HOTMAIL.COM29_Thu, 17 Aug 2000 12:26:38 GMT549_- Richard, How are things going lately? I saw a bunch of BMC people at IDUG. Hennie is still very unhappy. They put him in a consulting group that was supposed to specialize in going into companies and doing things for them only in the area of BMC tools, but could not get enough business; and now they are slowly transferring the people into other groups. We were thinking that such a consulting group should make themselves available to work on anything in DB2 that a customer needs and not just limit it to BMC tools tasks. It looks like my [...] 12355 88 53_Re: Help with Increasing Archive Log space allocation27_Hilton, Tina, BmS - NMI -PM21_Tina.Hilton@BMSUS.COM31_Thu, 17 Aug 2000 07:35:07 -0500625_- Or just change dsnzparm and reassemble & link it. Rerunning the clist to generate dsnzparm is dangerous if any changes have been made to zparm outside of the clist. You will lose those changes. The keywords are priqty and secqty in the dsn6arvp macro.

Tina

> -----Original Message----- > From: James Campbell [SMTP:James.Campbell@HANCORP.COM.AU] > Sent: August 16, 2000 9:35 PM > To: DB2-L@RYCI.COM > Subject: Re: Help with Increasing Archive Log space allocation > > Carlton > > Re-run the install CLIST and modify the space allocation. Generally, the > space should be the save as the active logs, and [...] 12444 121 38_Re: Permanent Delete of DB, TS, and TB15_Billings, Linda30_linda.billings@DOA.STATE.WI.US31_Thu, 17 Aug 2000 07:39:02 -0500457_iso-8859-1 I am glad that you brought this question up. On further thought we should issue the modify recovery command ourselves before dropping these objects. Our modify recovery procedures do keep the most current image copy no matter how old. By dropping the objects we do keep a few entries in SYSLGRNG that are "orphaned" by the dropping of the object. I have made a suggestion to my group to try performing the modify before dropping the object. [...] 12566 126 53_Re: Help with Increasing Archive Log space allocation23_Pendlebury-Bowe, Leslie34_Leslie.Pendlebury-Bowe@EU.SONY.COM31_Thu, 17 Aug 2000 13:38:02 +0100435_iso-8859-1 Tina et al.

you have raised a question I have wished to ask everyone ..

I was always led to believe ths clist was the way to go when modifying the zparm .. but in my experience I have found it is easier to manage it away from the clist and then spend a couple hours making sure of your TIDs when doing an upgrade .. and obviously making double sure by checking the ZPARM out of the clist to the old one . [...] 12693 46 34_Re: Sync. and Async. writes(Again)14_Joel Goldstein29_joel_goldstein@COMPUSERVE.COM31_Thu, 17 Aug 2000 08:57:22 -0400576_ISO-8859-1 See comments inline flagged by ++++

Regards, Joel

Message text written by DB2 Data Base Discussion List > I was going through the new presentation on bufferpool tuning by Joel Goldstien in the resposive system website and got some new things(confusion) about the synch. writes from BP to DASD. DB2 writes asynchronously when DWQT,VDWQT reaches, checkpoint takes place. DB2 writes synchronously when IWTH,DMTH and SPTH reaches,DSMAX reaches and two more conditions. +++++++ Not true for SPTH and DSMAX.... these have no relevance for synch writes [...] 12740 34 18_Isolation level UR18_Whittaker, Stephen26_stephen.whittaker@CPLC.COM31_Thu, 17 Aug 2000 09:00:22 -0400509_iso-8859-1 Can you issue a 'update or insert or delete' stmt using a package this is bound with an Isolation level of UR (dirty read). I was under the impression that UR could only be used with 'Select only'. We have a client that is coming through Datajoiner and his desktop is configured so that the packages use UR's. He says that he can issue 'updates' when using the UR package. I'm going through the DB2 V5 manual but can't seem to find anything really clear-cut on the use of UR. Can anyone HELP? [...] 12775 86 34_Re: Sync. and Async. writes(Again)7_Sanjeev21_Sanjeev@INTERRAIT.COM31_Thu, 17 Aug 2000 18:43:26 +0530604_iso-8859-1 Thanks for your reply Joel, I would like to cut and paste the portion of the article (previous version) which confuses me : "Another indicator that is not totally understood by many people is Synchronous Writes. These are triggered if the IWTH is reached; however, there are two other instances that will cause this counter to be incremented that are not of the same overall performance impact. The other two situations are: closing a tablespace when updated buffers are still in the pool, and a second checkpoint occurring while updated pages that have not been written are still in the [...] 12862 36 24_Re: ACCESS PATH QUESTION20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM31_Thu, 17 Aug 2000 09:27:12 -0400370_us-ascii Hi Richard

I wouldn't disagree with anything that you've mentioned. I myself wouldn't prefer to update the catalog as a first choice but given a choice that program changes are difficult to make or difficult to persuade, I have resorted to careful catalog updates many times, still keeping a very tight control over runstats and catalog statistics. [...] 12899 69 22_Re: Isolation level UR38_Toine Michielse/Santa Teresa/Contr/IBM18_vndobtm@US.IBM.COM31_Thu, 17 Aug 2000 15:27:40 +0100398_us-ascii Hello Steve,

Yes you can issue all other SQL. They just do not use UR of course since this is restricted to read-only SQL. So the updates will take the appropriate locks...

Regards,

Toine Michielse DB2 for OS/390 Technical Specialist Silicon Valley Laboratory Mobile: +31 6 537 23 256 Email: vndobtm@us.ibm.com Lotus: Toine Michielse/Santa Teresa/Contr/IBM@IBMUS [...] 12969 42 46_Passing data in stored procedures via cursors.12_MARTIN WOLFF31_martin_wolff@GLOBALCROSSING.COM31_Thu, 17 Aug 2000 09:50:57 -0400410_us-ascii Dear list,

We have a potential situation where we need to access a large amount of non-DB2 data from an intranet caller. The actual data cannot be accessed in the stored procedure but can from a CICS transaction. The obvious solution is have the SP do a EXCI call to

CICS and rertieve the data via the returned COMMAREA. The SP then, via a temp table can pass it back via a cursor. [...] 13012 21 16_Time Zone Change11_Mike Fatula17_mfatula@PHEAA.ORG31_Thu, 17 Aug 2000 10:14:34 -0400500_us-ascii Hi:

We are running DB2 5.2 AIX. We have two instances defined on a single node. Can anyone tell me if I can have one instance use eastern standard time and the other instance use Pacific time. And how to do this.





Thanks

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 13034 59 17_Re: Timestampdiff0_18_damcon2@US.IBM.COM31_Thu, 17 Aug 2000 10:16:43 -0400605_us-ascii Peter,

Something like the following should work...

select timestampdiff(8,char(timestamp1 - timestamp2)) from table;

The 8 represents hours to be calculated.

Cheers, Jay



Jay Reavill damcon2@us.ibm.com IBM Global Services Tampa, Florida Tel: (813) 801-7303, Tie Line 8-427-7303 ----------------------------------------------------- Happiness is not around the corner. Happiness is the corner. - BMW -----------------------------------------------------



"Krawetzky, Peter J" @RYCI.COM> on 08/17/2000 07:03:31 AM [...] 13094 150 38_Re: Permanent Delete of DB, TS, and TB15_Lankester, Andy22_Andy_Lankester@BMC.COM31_Thu, 17 Aug 2000 09:30:57 -0500516_iso-8859-1 FYI from the APAR that rejected the RQ to cleanup SYSLGRNG on DROP:

COMMENTS: THIS APAR HAS BEEN CLOSED SUG BECAUSE IT WOULD REQUIRE A DESIGN CHANGE TO DELETE THE RELATED SYSLGRNG ROWS WHEN A TABLESPACE IS DROPPED. CURRENTLY, THE SYSLGRNG ROWS IN DSNDB01.SYSLGRNG FOR A DROPPED TABLESPACE ARE DELETED WHEN THE DBID (DATABASE ID) AND PSID (PAGESET ID) ARE REUSED ON THE CREATE OF A NEW TABLESPACE. IT WAS DESIGNED THIS WAY BECAUSE THE DROP TABLESPACE COMMAND MAY BE ROLLBACKED FOR SOME REASON. [...] 13245 71 17_Re: Timestampdiff18_Krawetzky, Peter J21_KrawetzkyPJ@AETNA.COM31_Thu, 17 Aug 2000 10:31:17 -0400717_- Excellent, thank you Jay!

-----Original Message----- From: damcon2@US.IBM.COM [mailto:damcon2@US.IBM.COM] Sent: Thursday, August 17, 2000 10:17 AM To: DB2-L@RYCI.COM Subject: Re: Timestampdiff



Peter,

Something like the following should work...

select timestampdiff(8,char(timestamp1 - timestamp2)) from table;

The 8 represents hours to be calculated.

Cheers, Jay



Jay Reavill damcon2@us.ibm.com IBM Global Services Tampa, Florida Tel: (813) 801-7303, Tie Line 8-427-7303 ----------------------------------------------------- Happiness is not around the corner. Happiness is the corner. - BMW ----------------------------------------------------- [...] 13317 133 34_Re: Sync. and Async. writes(Again)20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM31_Thu, 17 Aug 2000 10:38:54 -0400377_iso-8859-1 DB2 does externalize uncommitted pages. I think it could be during two checkpoints have passed. A simple experiment using SPUFI demonstrates this.

Update data using SPUFI with auto-commit no. DSN1PRNT shows old page. Wait for considerable time, in my case I waited 3 minutes and did DSN1PRNT again, the updated uncommitted page was externalized to DASD. [...] 13451 54 22_Re: Isolation level UR14_Manas Dasgupta25_manas.dasgupta@PFSFHQ.COM31_Thu, 17 Aug 2000 10:49:06 -0400589_iso-8859-1 Yes you can...INSERT,UPDATE,DELETE would use CS.

> -----Original Message----- > From: DB2 Data Base Discussion List [mailto:DB2-L@RYCI.COM]On Behalf Of > Whittaker, Stephen > Sent: Thursday, August 17, 2000 9:00 AM > To: DB2-L@RYCI.COM > Subject: Isolation level UR > > > Can you issue a 'update or insert or delete' stmt using a package this is > bound with > an Isolation level of UR (dirty read). I was under the impression that UR > could only be > used with 'Select only'. > We have a client that is coming through Datajoiner and his desktop is > configured so [...] 13506 77 22_Re: Isolation level UR18_Whittaker, Stephen26_stephen.whittaker@CPLC.COM31_Thu, 17 Aug 2000 11:00:53 -0400476_iso-8859-1 Thanks all for the help. I verified the CS isolation when an update was issued using our TMON for DB2 monitor.

--Steve....

Steve Whittaker DB2/DBA Carolina Power & Light Mail stop: CPB 17A4 - Raleigh Email: stephen.whittaker@cplc.com Phone: (919) 546-7267



-----Original Message----- From: Manas Dasgupta [mailto:manas.dasgupta@PFSFHQ.COM] Sent: Thursday, August 17, 2000 10:49 AM To: DB2-L@RYCI.COM Subject: Re: Isolation level UR [...] 13584 166 38_Re: Permanent Delete of DB, TS, and TB15_Billings, Linda30_linda.billings@DOA.STATE.WI.US31_Thu, 17 Aug 2000 10:05:46 -0500676_iso-8859-1 Thanks for the update! I will share it with my team members.

Linda

-----Original Message----- From: Lankester, Andy [mailto:Andy_Lankester@BMC.COM] Sent: Thursday, August 17, 2000 9:31 AM To: DB2-L@RYCI.COM Subject: Re: Permanent Delete of DB, TS, and TB



FYI from the APAR that rejected the RQ to cleanup SYSLGRNG on DROP:

COMMENTS: THIS APAR HAS BEEN CLOSED SUG BECAUSE IT WOULD REQUIRE A DESIGN CHANGE TO DELETE THE RELATED SYSLGRNG ROWS WHEN A TABLESPACE IS DROPPED. CURRENTLY, THE SYSLGRNG ROWS IN DSNDB01.SYSLGRNG FOR A DROPPED TABLESPACE ARE DELETED WHEN THE DBID (DATABASE ID) AND PSID (PAGESET ID) ARE REUSED ON THE [...] 13751 23 48_please remove my personal email sent by accident11_rick creech18_ykcirc@HOTMAIL.COM29_Thu, 17 Aug 2000 15:25:36 GMT393_- Dear list-serve, Please remove a personal note I sent this morning. I recently sent a personal email to the list by accident. I am very embarrassed and do apologize greatly. I have a very high regard for BMC and hope that I did not cause them any negative feed-back for this. Anything expressed in the email was only my personal opinion about something that was not my business anyway. [...] 13775 207 34_Re: Sync. and Async. writes(Again)13_Mohan, Ashish21_Ashish.Mohan@NIKE.COM31_Thu, 17 Aug 2000 08:47:58 -0700502_iso-8859-1 Can't help adding...

Andy is absolutely correct that DB2 never externalizes unlogged data. Imagine the kind of I/O in the bufferpool if DB2 were to synchronously write pages back at every commit. Purely for performance reasons, the logs are written out to stable storage at every commit but not the data from buffer pools . Just so that there is no 'unlogged data' , any change that is made to DB2 is first written to log and only then is the actual change made to the DB2 data. [...] 13983 25 23_Re: DB2 V5 to V7 Direct12_Roger Miller19_millerrl@US.IBM.COM31_Thu, 17 Aug 2000 10:35:36 -0500509_- First you wait for general availability. The early customers are moving from V6, so they won't have much information along these lines. Then you should wait about the same time as you generally migrate from release to release - a few months for a few customers, a few years for others. My net message for migration from V5 to V7 direct is that the process is for customers who are still on V4 or V3 or need to avoid performing a migration for at least a year from now. Otherwise V6 is a better target. [...] 14009 231 34_Re: Sync. and Async. writes(Again)15_Lankester, Andy22_Andy_Lankester@BMC.COM31_Thu, 17 Aug 2000 10:59:20 -0500450_iso-8859-1 Ashish wrote:

Imagine the kind of I/O in the bufferpool if DB2 were to synchronously write pages back at every commit.

But that's just how IMS works (except FastPath)! Maybe that's why they need WADS.

Andy Lankester



-----Original Message----- From: Mohan, Ashish [mailto:Ashish.Mohan@NIKE.COM] Sent: 17 August 2000 16:48 To: DB2-L@RYCI.COM Subject: Re: Sync. and Async. writes(Again) [...] 14241 25 7_cursors15_Kuashal Trivedi17_c_triveka@BAM.COM31_Thu, 17 Aug 2000 11:59:14 -0400525_iso-8859-1 Hi All, One of the program is driven by a main cursor and for each fetch done, it opens one another cursor and this inturn does one more, currently I am looking to change the program by opening all the cursors at the beginning and doing a serial compare , as there is a criteria that runs common among all the cursors, My question is about the volume down the line, the volume would increase in the near future on these tables, and I would like to know if opening 3 such big cursors would be a good idea, Any [...] 14267 58 32_Re: AW: V5 to V6 migration error12_Roger Miller19_millerrl@US.IBM.COM31_Thu, 17 Aug 2000 10:44:59 -0500423_- I responded to your email yesterday. You would want the job after editing is done, then there are problems with service, documentation and the rest of the install process. Try these queries. No guarantees, no service, ... Get the book from the web to see other situations. Also check the red book, SG24-6108 and the Info APARs II11442 and II12343. These queries have been in the presentations for a couple of years. [...] 14326 157 53_Re: Help with Increasing Archive Log space allocation13_OConner, John26_John.OConner@COURTS.WA.GOV31_Thu, 17 Aug 2000 09:23:18 -0700464_iso-8859-1 Leslie,

When we migrate, we screen print the clist panels and save a hard copy. We also print the zparms. When we make a zparm change, we mark it on the hard copy and put a comment in the zparm jcl member. At the next migration we check the hard copy and the current zparm member (we check both hoping to catch any accidental omissions) for the changes we had made since the last migration. Not very elegant, but for our size shop it works. [...] 14484 43 11_Re: cursors15_Tonmoy Dasgupta32_tonmoy.dasgupta@MAIL.STATE.AR.US31_Thu, 17 Aug 2000 11:29:24 -0500633_iso-8859-1 Hi KT,

Am I correct in assuming that each of the three cursors will have "order by" clause to enable the serial compare?



-----Original Message----- From: Kuashal Trivedi [mailto:c_triveka@BAM.COM] Sent: Thursday, August 17, 2000 10:59 AM To: DB2-L@RYCI.COM Subject: cursors



Hi All, One of the program is driven by a main cursor and for each fetch done, it opens one another cursor and this inturn does one more, currently I am looking to change the program by opening all the cursors at the beginning and doing a serial compare , as there is a criteria that runs common among all [...] 14528 68 11_Re: cursors11_rick creech18_ykcirc@HOTMAIL.COM29_Thu, 17 Aug 2000 16:35:23 GMT672_- One of the advantages of having three cursors at the top is that the possibility of doing sequential prefetch on each table increases. Random i/o being done when two cursors are opened for each fetch can significantly slow down a batch process.

HTH, Rick Creech



>From: Tonmoy Dasgupta >Reply-To: DB2 Data Base Discussion List >To: DB2-L@RYCI.COM >Subject: Re: cursors >Date: Thu, 17 Aug 2000 11:29:24 -0500 > >Hi KT, > >Am I correct in assuming that each of the three cursors will have "order >by" >clause to enable the serial compare? > > >-----Original Message----- >From: Kuashal Trivedi [...] 14597 59 11_Re: cursors15_Kuashal Trivedi17_c_triveka@BAM.COM31_Thu, 17 Aug 2000 12:33:40 -0400544_iso-8859-1 yup they will.

Kaushal Trivedi

-----Original Message----- From: Tonmoy Dasgupta [mailto:tonmoy.dasgupta@MAIL.STATE.AR.US] Sent: Thursday, August 17, 2000 12:29 PM To: DB2-L@RYCI.COM Subject: Re: cursors



Hi KT,

Am I correct in assuming that each of the three cursors will have "order by" clause to enable the serial compare?



-----Original Message----- From: Kuashal Trivedi [mailto:c_triveka@BAM.COM] Sent: Thursday, August 17, 2000 10:59 AM To: DB2-L@RYCI.COM Subject: cursors [...] 14657 44 34_Re: Sync. and Async. writes(Again)11_Mike Turner29_Michael_Turner@COMPUSERVE.COM31_Thu, 17 Aug 2000 12:41:44 -0400699_ISO-8859-1 Paul

Sorry, but you are wrong. DB2 will write pages that have uncommitted updates. The Log Write Ahead protocol ensures that it write the log records needed to rollback the update before it writes the page from the buffer pool.

Regards Mike Turner Email: michael_turner@compuserve.com Home Page: http://ourworld.compuserve.com/homepages/michael_turner

Message text written by DB2 Data Base Discussion List >Sanjeev, You'll have to forgive me if I'm still confused about the question. However, I'll attempt to answer what I think you asked by saying that my understanding is that DB2 will NOT externalize any page that has not been committed...period. This would [...] 14702 125 40_Re: Question about shrlevel change reorg12_Isaac Yassin23_yassin@NETVISION.NET.IL31_Thu, 17 Aug 2000 19:51:22 +0200613_x-user-defined Hi,

I would suggest reorg shrlevel reference with "reuse".

Isaac Yassin

Tonmoy Dasgupta wrote: > > Neil, > > Why are you considering "shrlevel change" reorg when 1) you are > contemplating using DSN1COPY to move your tablespace around and 2) you are > severly short of DASD ? > > If you can afford the luxury of DSN1COPYing around your tablespace chances > are that you do not need "shrlevel change reorgs". You have to make the > tablespace un-available to DB2 when you use DSN1COPY...whereas the other > name of "shrlevel change reorg" is "online read-write reorg". You [...] 14828 27 41_Few questions about DB2 store procedure .15_Philip, Sibimon20_SPhilip@CSXLINES.COM31_Thu, 17 Aug 2000 12:48:59 -0400448_iso-8859-1 We are using DB2 version 5 on OS/390 version 2.6. Planning to start using DB2 stored procedures. Even though I checked the Archive, I have few questions.

1. What are the testing tools usually companies use to test Stored procedure? We have Xpeditor/TSO, does it support DB2 stored procedure. I saw a message in archive from compuware saying the future version support this. But what is that feature version, if anybody know. [...] 14856 93 11_Re: cursors15_Tonmoy Dasgupta32_tonmoy.dasgupta@MAIL.STATE.AR.US31_Thu, 17 Aug 2000 12:24:26 -0500632_iso-8859-1 Kaushal,

Go for it...your program should finish much faster.

Your processing would involve single passes over the data possibly using prefetch vs using excessive index access by the repeatedly opening cursors. The only time-consuming area might be the three "order by".

This may not be applicable to your site but I have found that external DFSORT works faster than internal DB2 sort (could be the way we have our different pools are set up...I wish someone would give me some pointers on this) and in one extreme case found excellent results by 1)putting the tables in RO status 2) downloading [...] 14950 24 24_Re: ACCESS PATH QUESTION12_Roger Miller19_millerrl@US.IBM.COM31_Thu, 17 Aug 2000 12:06:52 -0500414_- Let me be sure I understand the facts. It appears that you are choosing to avoid most of the advantages of optimization, as noted in several of the other appends.

When you said that you prime the values, does that mean you don't run RUNSTATS and put in the correct information? If so, then you are not providing the real statistics to the optimizer - but only using constants for all of the indexes. [...] 14975 24 22_Re: Identity Revisited12_Roger Miller19_millerrl@US.IBM.COM31_Thu, 17 Aug 2000 12:16:33 -0500565_- The only items I'd tend to add would be

6) The updated pdf books SQL Reference, Utilities Guide and Reference, ... have the changes in them. 7) send me money and time

I try to be hard on mistrakes, since I neverrrr make any myself. My personal observation is that this work has several different sets of user comments that seem to have groupings: a) people who want to use it for migration from Sybase b) people who want to use it for migration from Oracle, and want sequences instead c) people who want a more general capability, designed for [...] 15000 24 16_Data Dictonaries18_Gerrit Groenendaal22_groenen@WORLDONLINE.NL31_Thu, 17 Aug 2000 19:52:45 +0200299_us-ascii Hi list,

I have two questions, we are using Predict from Software A.G. as a data dictionary for our DB2 definitions. 1 Are there other shops that also using Predict and what are there experiences? 2 Are there other data dictionaries in the market with support of db2 v6 and v7? [...] 15025 124 36_Re: Datajoiner as an extraction tool22_Piontkowski Michael ML35_michael.piontkowski@ASTRAZENECA.COM31_Thu, 17 Aug 2000 13:51:37 -0400819_iso-8859-1 Maybe Data Propagator (DPropR) will do what you need.

http://www-4.ibm.com/software/data/dbtools/datarepl.html



Mike Piontkowski Voice/Fax: 302.886.4612 mailto:michael.piontkowski@astrazeneca.com





-----Original Message----- From: Krawetzky, Peter J [mailto:KrawetzkyPJ@AETNA.COM] Sent: Wednesday, August 16, 2000 9:47 To: DB2-L@RYCI.COM Subject: Re: [DB2-L] Datajoiner as an extraction tool



We are using datajoiner to extract data from Oracle v7.3.4 to UDB V6.2 on solaris. One word of warning. You may have to add server mappings types or reverse mapping types to allow data to be compatiable between Informix and DB2. I hit problems with dates and need to add mapping types. The Datajoiner documenation has a pretty good explaination of this. [...] 15150 30 17_Job Opportunities12_Diane Barela28_Info@PALLADIANCONSULTING.COM31_Thu, 17 Aug 2000 11:09:05 -0700659_iso-8859-1 Dear Colleagues,

Palladian Consulting specializes in providing RDBMS consulting and technical support. Our clients in the Healthcare, Government, Banking, Marketing, Education, and Commercial industries are always seeking candidates with a wide variety of technical knowledge in RDBMSs including DB2. We have clients currently in need of Programmers and Database Administrators with senior level skills. These positions are located throughout the United States. US citizenship is necessary. If you are interested in pursuing opportunities please feel free to contact me directly or log onto our web site at www.PalladianConsulting.com. [...] 15181 31 20_CICS RCT/PPT Entries0_20_bjnigh@HOUSEHOLD.COM31_Thu, 17 Aug 2000 11:54:54 -0700407_us-ascii We have a non-DB2 CICS program (program 1) calling/linking to another non-DB2 CICS program (program 2) which links/calls a DB2 program (program 3).

Program 1 has an entry in the RCT/RDO, programs 2 & 3 do not as they are called/linked to. We are finding program 3 goes to DB2 and never comes back. We found the thread with a status of N (thread is in either IDENTIFY or SIGNON status). [...] 15213 55 24_Re: CICS RCT/PPT Entries16_Thirumal Valavan31_thirumal.valavan@PRUDENTIAL.COM31_Thu, 17 Aug 2000 15:22:11 -0400318_us-ascii I think it depends on the nature of linking. My assumption is program 2 is linked statically from program1 where as program 3 is linked dyanmically. Please ensure this, Then that will solve your problem. If it is dynamically called , then you needs RCT and PPT entry seperately for this program Thirumal [...] 15269 185 38_Re: Recovery from Stacked Image Copies14_Jack O'Connell26_jack.oconnell@MCMASTER.COM31_Thu, 17 Aug 2000 14:31:07 -0500543_iso-8859-1 If you are using (ca)Platinum and recovering a set to a given RBA, and you are recovering in the same order they were copied onto the tape, you don't need to specify any DD statements to begin with. You will need to specify the RETAIN YES parm, and will need to run these within the same step execution. (ca)Platinum also has limits to the number of recover statements in a step ( they recommend 100 or less ) but we have successfully used over 500 recover statements, so it depends on the size of your system. I think you get [...] 15455 23 20_Ad Hoc Report Writer11_Dale Miller23_dmille@TRADESERVICE.COM31_Thu, 17 Aug 2000 15:54:24 -0400360_us-ascii We are looking for suggestions for a good ad hoc report writer. We are currently in the process of porting to DB2 on AIX and must replace our current report writer. Any advice or experience would be appreciated. Also, does anyone have experience using DecisionBase Reporter from CA? It looks like it might fit the bill. Thank you, Dale R. Miller [...] 15479 31 17_Stored Procedures0_22_wjmartin@HOUSEHOLD.COM31_Thu, 17 Aug 2000 16:04:14 -0400519_us-ascii We've started using stored procedures on our development system, accessed from a Powerbuilder front end. Now, we're preparing to move them to production. Our problem is that we have both the development and QA databases on the same subsystem. This means that to use the stored procedures on one or the other database, we have to change SYSIBM.SYSPROCEDURES for the collection id, and bind the stored procedures. This works fine, but doesn't allow developers to work on the same things that QA is testing. [...] 15511 70 50_Re: Passing data in stored procedures via cursors.11_John Wallin24_john.wallin@ZURICHUS.COM31_Thu, 17 Aug 2000 10:58:20 -0500419_us-ascii This is more of a CICS question than DB2 . You might try using a temp storage area or Trans Data Que to store the incomming data . That will allow transactions to share the data without a commarea being a limiting force.









MARTIN WOLFF on 08/17/2000 08:50:57 AM

Please respond to DB2 Data Base Discussion List [...] 15582 254 36_Re: DB2 Trigger calling UDF in COBOL23_Jain, Sanjay (Exchange)19_sanjayjain@BEAR.COM31_Thu, 17 Aug 2000 16:33:58 -0400400_- Here is a working sample.

1. Trigger definition

CREATE TRIGGER DB2DBA.TSTUDF NO CASCADE BEFORE INSERT ON DSN8610.PROJACT REFERENCING NEW AS NROW FOR EACH ROW MODE DB2SQL WHEN (NROW.ACSTAFF = 0) BEGIN ATOMIC SET NROW.ACSTAFF = UDFPGM01(0); END#

2. UDF definition - some of the parameters below, are default for Create function statement, but I chose to include them anyway [...] 15837 105 50_Re: Passing data in stored procedures via cursors.14_Laredo, Andrew19_Andy_Laredo@BMC.COM31_Thu, 17 Aug 2000 15:42:37 -0500465_iso-8859-1 If there is NO db2 commit going on in the processing of the cics transaction (I'm not sure about EXCI calls) then the temp table contents should be preserved back to the caller....

Of course you probably would want to run a test bed...

-----Original Message----- From: John Wallin [mailto:john.wallin@ZURICHUS.COM] Sent: Thursday, August 17, 2000 11:58 AM To: DB2-L@RYCI.COM Subject: Re: Passing data in stored procedures via cursors. [...] 15943 39 25_Urgent Position Available19_Patrick Lewis - 28319_plewis@ACT1CORP.COM31_Thu, 17 Aug 2000 13:46:56 -0700425_- List Members,

I apologize for this posting, however I just received a very urgent need and wanted to find out if anyone is interested.

My client is in need of an experienced DB2 professional to be part of a systems integration team to design a large scale data warehouse using DB2 UDB EEE on an AIX platform. This person will assist with the physical data base design and implementation of the database. [...] 15983 21 9_sync sink15_Kuashal Trivedi17_c_triveka@BAM.COM31_Thu, 17 Aug 2000 17:12:25 -0400350_iso-8859-1 Hi All, Like Dynamic prefetch gets activated when the criteria for a certain number of pages within a group of pages are accessed, is there anything that could trigger the opposite, one of my explains shows a Sequestial prefetch ,and looking in the Tmon shows a ton of sync I?O being done, what factors could effect this, any ideas . [...] 16005 25 25_Upgrade to CAE version 6.0_34_L-IS.Kirkpatrick@MUTUALOFOMAHA.COM31_Thu, 17 Aug 2000 16:15:44 -0500283_us-ascii We are currently pursuing a desktop upgrade from CAE version 5.x to CAE version 6. Does any of you have any experience with this upgrade in your shop? If so, were there any problems with the upgrade or did it go without a hitch? What problems do I need to be aware of? [...] 16031 37 24_Re: Ad Hoc Report Writer12_Reid, Louise25_ReidL@CI.FORT-WORTH.TX.US31_Thu, 17 Aug 2000 16:20:50 -0500606_iso-8859-1 We use Qmf for Windows and the users like it a lot!!!!

-----Original Message----- From: Dale Miller [mailto:dmille@TRADESERVICE.COM] Sent: Thursday, August 17, 2000 2:54 PM To: DB2-L@RYCI.COM Subject: Ad Hoc Report Writer



We are looking for suggestions for a good ad hoc report writer. We are currently in the process of porting to DB2 on AIX and must replace our current report writer. Any advice or experience would be appreciated. Also, does anyone have experience using DecisionBase Reporter from CA? It looks like it might fit the bill. Thank you, Dale R. Miller [...] 16069 15 36_DB2 Connect EE V7 and DB2 CAE V5.2 ?13_Kim Bundgaard10_kib@DDF.DK31_Thu, 17 Aug 2000 16:06:54 -0500317_- Anyone know if there are problem using DB2 CAE V5.2 as a client, when accessing DB2 OS/390 V6 through a DB2 Connect V7 gateway.

We are planning to upgrade directly from DB2 Connect EE V5.2 to DB2 Connect EE V7, but we are not able to change all our client (from DB2 CAE V5.2 to DB2 Runtime V7) at once. [...] 16085 19 13_Re: sync sink0_26_truman.g.brown@VERIZON.COM31_Thu, 17 Aug 2000 17:39:25 -0400416_us-ascii Sequential prefetch can be reduced or disabled when there aren't enough pages in the VBP or read engines to support it ; this is usually a concurrency issue for pages or a DASD contention issue for read engines. Once it's turned off for a thread it will not restart, and sync reads occur. When enough pages become available again the thread may revert to dynamic prefetch through sequential detection. [...] 16105 77 40_Re: DB2 Connect EE V7 and DB2 CAE V5.2 ?13_Kurt Sahlberg26_Kurt.Sahlberg@EXPERIAN.COM31_Thu, 17 Aug 2000 16:43:18 -0500467_US-ASCII You can only skip 1 release. V5.X will work with V6 but 5.X will NOT work with V7. HTH Kurt



>>> kib@DDF.DK 08/17/00 04:06PM >>> Anyone know if there are problem using DB2 CAE V5.2 as a client, when accessing DB2 OS/390 V6 through a DB2 Connect V7 gateway.

We are planning to upgrade directly from DB2 Connect EE V5.2 to DB2 Connect EE V7, but we are not able to change all our client (from DB2 CAE V5.2 to DB2 Runtime V7) at once. [...] 16183 111 11_Re: cursors15_Keith Phimister31_keith.phimister@VIRGINNET.CO.UK31_Thu, 17 Aug 2000 23:05:10 +0100378_iso-8859-1 Why not do a nice big left join and get it all in one go? It makes the coding a little more difficult but depending upon the data, it might be a solution. Keith ----- Original Message ----- From: "Tonmoy Dasgupta" Newsgroups: bit.listserv.db2-l To: Sent: Thursday, August 17, 2000 6:24 PM Subject: Re: cursors [...] 16295 134 24_S0C4 Abends in DB2/OS39012_Hall, Robert25_Robert.Hall@OCC.TREAS.GOV31_Thu, 17 Aug 2000 18:14:45 -0400550_iso-8859-1 Hi,

We are at V5 DB2/OS390 (9805 maint. level), under OS390 V1.3. We have been experiencing sporadic S0C4 abends in our DB2 development subsystem for over a year. We opened an APAR with IBM and sent them 11 dumps, but since the 0C4 error occurs in different sections of DB2 code, IBM has closed the APAR and told us the problem is a "programming error" on our part. I personally think this is a bogus response from IBM in that we ship a query to DB2 (via DDF or DSNTEP2), how could we have a programming error? Either the DBMS [...] 16430 143 50_Re: Passing data in stored procedures via cursors.11_Suresh Sane21_data_arch@HOTMAIL.COM29_Thu, 17 Aug 2000 17:40:03 CDT579_- No DB2 commit but a CICS RETURN which will issue a DB2 commit! - I am assuming your CICS transctions are not conversatonal.

In this case, your only choices seem to be TS or TS queues (as John suggests) or multiple executions. We have implemented an architecture that requires multiple executions and overall it works well. The situations where a large answer set is built/sorted is a problem but this is a training issue as well as a question of writing tight code that limits such occurences. The fact that CICS is required implies you are accessing VSAM or other [...] 16574 163 28_Re: S0C4 Abends in DB2/OS39011_Robert Jans21_Jans.Robert@AMSTR.COM31_Thu, 17 Aug 2000 17:05:00 -0600350_- First question: Do you have DB2 parallelism enabled? Second question: Are you data sharing? There are problems with DB2 parallelism (aggravated by data sharing) in V5. There are APARS to fix them and the fixes work. DB2 V6 also has a problem with data sharing and parallelism. We tripped over it and haven't got a resolution on it yet........ [...] 16738 53 24_Re: CICS RCT/PPT Entries11_Tom Gleason21_tagleas@ATTGLOBAL.NET31_Thu, 17 Aug 2000 19:23:54 -0400618_us-ascii I suggest you check a few items and then if you still have a problem list the actual entries you have for DB2 resources and Transaction Definition. The Program Resource (PPT) has no effect on the DB2 Attach facility since the connection of CICS to DB2 is enabled by TRANSACTION/PLAN association. The TRANID(CICS) should be identified in a DB2TRAN RDO CSD entry for example DEF DB2TRAN(TXXA) TRAN(TXXA) DB2E(TXXA) - the association to the plan is made thru the DB2ENTRY RDO CSD entry ...(partial) : DEF DB2ENTRY(TXXA) ......PLAN(TXXA)..... .. You should be able to trace the flow of the transaction using [...] 16792 65 34_Re: Sync. and Async. writes(Again)0_35_paul.a.redhead@TRANSPORT.QLD.GOV.AU31_Fri, 18 Aug 2000 09:46:04 +1000577_us-ascii Sanjeev, I think your getting lost in the little picture and losing sight of the big picture...the number one goal (I believe) is to protect your data, I think this has a higher priority than performance. Again, I think both your questions can be answered by accepting the fact that uncommitted pages aren't externalized. 1) If a page has been queued for write at checkpoint time and is then part of an update so that it's removed from the queued checkpoint writes, then when that unit of work is committed the page is written synchronously. 2) DB2 won't write a [...] 16858 202 28_Re: S0C4 Abends in DB2/OS39013_OConner, John26_John.OConner@COURTS.WA.GOV31_Thu, 17 Aug 2000 16:55:49 -0700446_iso-8859-1 Robert,

We had a process (for reloading tables on a weekly basis) that was corrupting some of our indexes. We didn't see a problem until after the application was running for a while. Then we would get peculiar error messages that didn't say anything about indexes. A CHECK INDEX would NOT detect the initial index corruption. Once we started getting errors, a CHECK INDEX (or even a RECOVER TABLESPACE!!) could CRASH DB2. [...] 17061 58 29_Re: Upgrade to CAE version 6.8_LAM, Leo17_l.lam@IOOF.COM.AU31_Fri, 18 Aug 2000 10:15:52 +1000600_iso-8859-1 Larry,

I'm interested to know the compatibility problems you have. Regards

Leo

-----Original Message----- From: L-IS.Kirkpatrick@MUTUALOFOMAHA.COM [mailto:L-IS.Kirkpatrick@MUTUALOFOMAHA.COM] Sent: Friday, 18 August 2000 7:16 To: DB2-L@RYCI.COM Subject: Upgrade to CAE version 6.



We are currently pursuing a desktop upgrade from CAE version 5.x to CAE version 6. Does any of you have any experience with this upgrade in your shop? If so, were there any problems with the upgrade or did it go without a hitch? What problems do I need to be aware of? [...] 17120 35 34_Re: Sync. and Async. writes(Again)0_35_paul.a.redhead@TRANSPORT.QLD.GOV.AU31_Fri, 18 Aug 2000 10:21:06 +1000442_us-ascii Mike, Thanks for that. I wish I'd read this post before responding again incorrectly to Sanjeev's subsequent question...guess I should read ALL my mail first.

>>> Mike Turner wrote: Paul

Sorry, but you are wrong. DB2 will write pages that have uncommitted updates. The Log Write Ahead protocol ensures that it write the log records needed to rollback the update before it writes the page from the buffer pool. >>> [...] 17156 78 24_Re: CICS RCT/PPT Entries14_James Campbell29_James.Campbell@HANCORP.COM.AU31_Fri, 18 Aug 2000 11:18:29 +1000340_iso-8859-1 Which version of CICS are you running?

CICS post V?.? uses RDO for DB2 interface information, and you should follow Tom Gleason's advice. However, I doubt that you are since the RDO definitions are based on the transaction name - and, if I follow you situation correctly, all the programs are in the one transaction. [...] 17235 196 28_Re: S0C4 Abends in DB2/OS3900_29_Greg.Palgrave@BANKWEST.COM.AU31_Fri, 18 Aug 2000 09:24:01 +0800416_us-ascii Robert,

We were getting occasional mystery 0C4 abends in DB2 jobs (Binds, Utilities (LOAD, REORG etc) mostly) back in July/August 98. Jobs would usually rerun OK.

These went away after we applied maintenance. As you're at 9805 maint level, I'd guess your situation is similar. You seem to be a long way behind on maint at the moment....I'd keep pushing for the maintenance to be updated. [...] 17432 198 53_Re: Help with Increasing Archive Log space allocation14_James Campbell29_James.Campbell@HANCORP.COM.AU31_Fri, 18 Aug 2000 11:53:38 +1000442_iso-8859-1 What I do is:

- All changes made by running the install clist and a new NEWXn.SDSNSAMP library/ DSNTIDXn/ DSNZPAXn member created.

- the install job is compared (with the X option) against the previous job. All differences are noted, dissected etc. If necessary, the old DSNTIDXn member will be modified and the process rerun, so that the new TIDXn member and install job are consistant and include all changes. [...] 17631 58 34_Re: Sync. and Async. writes(Again)14_Joel Goldstein29_joel_goldstein@COMPUSERVE.COM31_Thu, 17 Aug 2000 21:56:00 -0400356_ISO-8859-1 Paul, I'm not sure I follow your statements here. Uncommitted pages can and are written (externalized), if they could never be written, then every system out there would be hitting a 100% pool full.... That's why there is Backout (and sometimes really big ones) - to remove the updates from pages that were written before a commit occured. [...] 17690 90 50_Re: Passing data in stored procedures via cursors.14_James Campbell29_James.Campbell@HANCORP.COM.AU31_Fri, 18 Aug 2000 12:19:54 +1000401_iso-8859-1 Martin

In addition the other options, you could have a permanent table for temporary data. If there is too much data for COMMAREA, the CICS program populates this table, with a unique key (eg CICS task number) and sequence number (for an ORDER BY clause), it then passes the unique key back to the calling SP. The SP can then manipulate the table to return whatever's required. [...] 17781 31 60_Restarting an RC/Migrator stragegy on another LPAR/subsystem12_Raymond Bell20_raymondb@NZ1.IBM.COM31_Fri, 18 Aug 2000 15:55:30 +1200357_us-ascii Colleagues,

Not strictly a DB2 question but certainly related. I want to use RC/Migrator to migrate a database from one DB2 subsystem to another. Easy, you say. Me too, except when the source and target subsystems are on different LPARs. I know how to do it in the BMC equivalent tool, Change Manager, but can't find how to do it here. [...] 17813 31 34_Out of list question..............25_Sinha, Ranjay Kumar (CTS)24_SRanjay@CAL.CTS-CORP.COM31_Fri, 18 Aug 2000 11:20:08 +0530656_iso-8859-1 Hi list, We have a non-DB2 cics/cobol program source code. There is no "exec cics ............... end-exec " cics command. Instead I found some "exec boss initliaze end-exec " command. Can any one there in the list shed some light in this "BOSS" area. Thanks in advance.

Regards Ranjay





This e-mail and any files transmitted with it are for the sole use of the intended recipient(s) and may contain confidential and privileged information. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. Any unauthorised review, use, disclosure, [...] 17845 54 64_Re: Restarting an RC/Migrator stragegy on another LPAR/subsystem12_Robert Brock28_brock@KRYPTON-CONSULTING.COM31_Fri, 18 Aug 2000 02:14:58 -0500599_iso-8859-1 Hi Raymond,

You can use the Unload Mode options on the Batch Processor Interface screen to do what you want. Assuming you have shared DASD between the two subsystems, run the migration analysis output on the source subsystem using Unload Mode "U" to do the data unloads. After that finishes, run the same strategy on the target system using Unload Mode "N". That will do the creates and loads. If you don't have shared DASD you will have to manually move the migration analysis output and the unload datasets to the target system before the second execution. Hope that helps. [...] 17900 209 40_Re: Question about shrlevel change reorg0_18_mebert@AMADEUS.NET31_Fri, 18 Aug 2000 09:31:47 +0200505_us-ascii You can't use REUSE with SHRLEVEL REFERENCE. I guess you mean SHRLEVEL NONE with REUSE, which is what I would recommend for a space-constrained REORG (along with SORTKEYS, probably NOSYSREC and possibly SORTDATA; also consider Inline Copy) when maximum availability is not an issue. With NOSYSREC, you can't restart an abended REORG: so make sure you've got a valid image copy for the TS (no need to take one immediately before every REORG, just make sure you _can_ RECOVER if you have to). [...] 18110 74 38_Re: Out of list question..............14_James Campbell29_James.Campbell@HANCORP.COM.AU31_Fri, 18 Aug 2000 18:00:24 +1000316_iso-8859-1 Ranjay

What process is used to compile the program? Both CICS and DB2 have preprocessors that take 'EXEC CICS' or 'EXEC SQL' code and turn it into pure COBOL (or PLI or C or ... as the case might be). You will probably find that the code is passed through a similar processor. Investigate it. [...] 18185 62 60_Re: Any way to load another database using Full Image copies10_John Hardy27_john_j_hardy@CONSULTANT.COM31_Fri, 18 Aug 2000 02:59:34 -0500357_- In addition, if you are sure that your source and target tables are structured identically, then you will encounter problems if you have altered the source table (to add a new column), and have not REORGed it. If you use Platinum products to REORG your tablespaces as well as copy them, then I don't know if a Platinum Reorg will 'fix' this problem. [...] 18248 116 80_Re: DB2 v5.1 Reorg shrlevel change on a single partition of a 10 partition table0_18_mebert@AMADEUS.NET31_Fri, 18 Aug 2000 11:11:49 +0200567_us-ascii Maybe it's a little bit late but I can't remember a response to this one (I just recently returned from vacation)... So: if the problem is in the RELOAD phase, then a probable cause is that the source data is quite unclustered (check clusterratio). In that case, the mapping index will be built in a random manner, requiring high CPU and large elapsed time during the RELOAD phase. You may have to reorg more often (and if you're using LOAD to populate the TS, make sure the input is sorted correctly). With partitioned TS, if you've got NPIs, then the [...] 18365 38 17_Websphere problem0_19_david.a.owen@BT.COM31_Fri, 18 Aug 2000 11:11:43 +0100339_- Can anyone help please ?

We have just put live a set of applications that use Enterprise Java Beans on Websphere on an AIX that has access to DB2 OS/390.

This worked fine during test and training where we hit max of 30 users.

However since going live we have now 200 users accessing the Websphere applications. [...] 18404 63 21_Re: Websphere problem20_Beighton-Dykes, Hugh27_hugh.beighton-dykes@EDS.COM31_Fri, 18 Aug 2000 12:06:29 +0100478_iso-8859-1 There has been discussion of this problem on ibm.software.websphere recently.

If you go to www.deja.com/usenet and put write_inet_server_socket in the search box you will see 17 matches.

Hope this helps.

Did we meet at IBM South Bank a few months ago?

Hugh Beighton-Dykes

-----Original Message----- From: david.a.owen@BT.COM [mailto:david.a.owen@BT.COM] Sent: 18 August 2000 11:12 To: DB2-L@RYCI.COM Subject: Websphere problem [...] 18468 185 50_SV: Passing data in stored procedures via cursors.0_19_Sven.Heidorn@CSN.SE31_Fri, 18 Aug 2000 13:49:34 +0200513_ISO-8859-1 If you are at V6.1 on DB2, have a look at DECLARED GLOBAL TEMPORARY TABLE, it's my understanding of the manual that if used with the ON COMMIT PRESERVE ROWS clause, the rows and the table should still be available to the application process after a commit operation. You should be able to do the declare in the SP or the CICS program, do the EXCI call and have the CICS program insert the rows, come back to the SP and open a cursor over the declared temp. I would of course test this thoroughly. [...] 18654 38 50_Re: Passing data in stored procedures via cursors.12_MARTIN WOLFF31_martin_wolff@GLOBALCROSSING.COM31_Fri, 18 Aug 2000 07:59:15 -0400291_us-ascii I guess I hoped that CICS Transaction Server would leave it up to WLM to issue the commit.

Yes, this is actually CA-Datacom data. While we are able to combine DB2 and Datacom in CICS and batch programs, we have not yet found a way to combine them in a stored procedure. [...] 18693 97 50_Re: Passing data in stored procedures via cursors.12_MARTIN WOLFF31_martin_wolff@GLOBALCROSSING.COM31_Fri, 18 Aug 2000 08:00:00 -0400567_us-ascii Yes, don't we forget the really obvious ones!

Martin.

James Campbell wrote:

> Martin > > In addition the other options, you could have a permanent table for > temporary data. If there is too much data for COMMAREA, the CICS program > populates this table, with a unique key (eg CICS task number) and sequence > number (for an ORDER BY clause), it then passes the unique key back to the > calling SP. The SP can then manipulate the table to return whatever's > required. > > /* standard disclaimer */ > James Campbell > DBA > Hansen [...] 18791 31 36_Comparing Teradata to DB2 and Oracle15_Michael McGuire23_mmcguire@NET2000.COM.AU31_Fri, 18 Aug 2000 21:50:34 +1000384_iso-8859-1 I am beginning a process of evaluating which DBMS to use for a new data warehouse - only 250 GB or raw data, but adhoc queries are prepared via a query tool. A key component of the costs is how much DBA support is required for each type of DBMS. What I have heard so far is that the ratio of DBA support is 5 Oracle DBAs compared to 3 DB2 AIX DBAs and 1 Teradata DBA. [...] 18823 66 21_Re: Stored Procedures14_Richard Yevich21_ryevich@ATTGLOBAL.NET31_Fri, 18 Aug 2000 07:01:57 -0400657_iso-8859-1 You can use AUTHID and LUNAME to have multiple rows in SYSPROCEDURES to allow multiple load modules, collections, etc., for a same name SP. There is a defined search order that allows the specific entires to be matched first.

Hope this helps, Richard Yevich

> -----Original Message----- > From: DB2 Data Base Discussion List [mailto:DB2-L@RYCI.COM]On Behalf Of > wjmartin@HOUSEHOLD.COM > Sent: Thursday, August 17, 2000 4:04 PM > To: DB2-L@RYCI.COM > Subject: Stored Procedures > > > We've started using stored procedures on our development > system, accessed from > a Powerbuilder front end. Now, we're preparing to move them [...] 18890 138 11_Re: cursors19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Fri, 18 Aug 2000 07:32:38 -0500630_ISO-8859-1 Hi Kaushal, You didn't mention it but I assume you are dealing with a batch program! If possible, design the program to ensure each cursor is opened only once and closed as soon as possible. When they are opened isn't important provided the following is possible. Make sure the ORDER BY clause for each cursor supports your 'serial compare'. To avoid full materialization of the result sets of your cursors at open (addresses your volume concerns), for each cursor ensure the ORDER BY and SELECT columns are supported by the appropriate index(es). Use RELEASE(DEALLOCATE) to exploit sequential detection and index [...] 19029 550 21_Re: Help asap please.9_Harish P.21_emailharish@YAHOO.COM31_Fri, 18 Aug 2000 07:23:19 -0500607_- A quick update for those who replied and helped.

Well, the good news is the DBAs have been convinced for multiple copies of the database to ease testing and an elixir to all other problems.

Again, thanks a ton.

Harish.





On Tue, 8 Aug 2000 21:54:04 +0200, Isaac Yassin wrote:

>Hi, >I had to deal many times with people who worked on ORACLE before and used to do nearly everything by >themselves. >Giving the separate DB is the best way to lead them to continue working in DB2 as they really hate >to wait for DBA to do >some [...] 19580 71 64_Re: Restarting an RC/Migrator stragegy on another LPAR/subsystem12_craig patton21_prgpatton@HOTMAIL.COM29_Fri, 18 Aug 2000 12:42:32 GMT573_- Raymond,

I have done this in the past by splitting the strategy manually. (The following works best if using shared DASD.) Take the MEMBER created, make a copy, delete from the UNLOAD LPAR member, all statements AFTER the UNLOADs. Take MEMBER '2' Delete all of the UNLOAD statements. You should now have 2 PDS members, 1 that unloads from SSID 'A' and 1 that takes the data and LOADS it to 'B' (can even have DDL). At this point, run the unload strategy through the Batch PRocessor. When Done, run the LOAd strategy through the Batch Processor on the 'other' [...] 19652 142 53_Re: Help with Increasing Archive Log space allocation11_Hayden, Lee18_Lee_Hayden@CSX.COM31_Fri, 18 Aug 2000 08:45:07 -0400389_iso-8859-1 This is basically what I have done also. Many shops do not use the install clist to make simple changes. In any case it is a good idea to know how to update the 'TID' to match current ZPARM/DSNDECP output. I have not always had the luxury of documented history of ZPARM changes when it came time to migrate. This can also help if your last good 'TID' is lost or corrupted. [...] 19795 27 28_Autoreply when not in office19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Fri, 18 Aug 2000 07:48:08 -0500550_ISO-8859-1 Hi All, Please don't Autoreply to DB2-L. I hate to sound like a complainer . . . Just for grins I checked my in basket and discovered that I receive about 3 -- 5 autoreplys for each post to DB2-L . . .

Regards, Rick Davis "This e-mail and any files transmitted with it are the property of SBC, are confidential, and are intended solely for the use of the individual or entity to whom this e-mail is addressed. If you are not one of the named recipient(s) or otherwise have reason to believe that you have received this message [...] 19823 193 54_Re: SV: Passing data in stored procedures via cursors.12_MARTIN WOLFF31_martin_wolff@GLOBALCROSSING.COM31_Fri, 18 Aug 2000 08:52:34 -0400597_iso-8859-1 Yes, there is some info in the V6 Technical Update but not much. The DB2 V6 manuals I have don't have anything on DECLARED GLOBAL TEMPORARY TABLEs.

Can any one point me to any more info on the 'ON COMMIT PRESERVE ROWS' option beyond the Technical Update?

Thanks, Martin Wolff.

Sven.Heidorn@CSN.SE wrote:

> If you are at V6.1 on DB2, have a look at DECLARED GLOBAL TEMPORARY TABLE, > it's my understanding of the manual that if used with the ON COMMIT PRESERVE > ROWS clause, > the rows and the table should still be available to the application process [...] 20017 31 0_10_Pete Coles28_peter.coles@SOMERFIELD.CO.UK31_Fri, 18 Aug 2000 15:05:05 +0100486_iso-8859-1 Can anyone actually tell me what steps are involved in migrating from db2 version 5 direct to version 7 on os/390. Is it the case you have to migrate to v6 and then to v7 in two operations or can you just do one migration?

















> ------------------------------------------------------------------ > Pete Coles - Systems Integration > > Email: peter.coles@somerfield.co.uk > > Tel: +44 117 935 6726 > > > [...] 20049 19 28_Re: S0C4 Abends in DB2/OS3909_Colin Fay13_cfay2@CSC.COM31_Fri, 18 Aug 2000 09:08:47 -0400445_us-ascii Hi,

Do you have parallel processing enabled?

This is one thing that has widespread impact , perhaps like you are seeing.

If you do , try disabling it for awhile and monitor.

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 20069 31 32_Re: Autoreply when not in office15_Lankester, Andy22_Andy_Lankester@BMC.COM31_Fri, 18 Aug 2000 08:26:49 -0500404_iso-8859-1 Unfortunately those of us saddled with MS Outlook for their sins to not have the ability selectively to autoreply based on the sender of the incoming messages. DB2-L will however only get one message for the whole 'out-of-office' period, not one for each post.

The good news is that you can filter out the out-off-office messages so that you don't see them, so win some/lose some. [...] 20101 15 54_Greg Fernholz/Deluxe Corporation is out of the office.13_Greg Fernholz24_Greg_Fernholz@EFUNDS.COM31_Fri, 18 Aug 2000 08:16:02 -0500437_us-ascii I will be out of the office from 08/18/2000 until 08/25/2000.

I will be attending a Class in Bloomington. MN Please leave me a voice mail at 651 483-7627 and I will call you back.

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 20117 44 52_SQLSTATE 42707 in DB2 for OS/390 Version 5 Release 112_Peter, Georg15_G.Peter@DZBW.DE31_Fri, 18 Aug 2000 15:26:49 +0200383_iso-8859-1 Hello List.

One of our programmers came back from a class and he was happy. He has seen that Version 5.1 allows columns in the ORDER BY clause that are not in the SELECT clause (?!).

We tried this in one of our DB2 develeopment systems and ran into a SQLSTATE 42707 "THE ORDER BY CLAUSE IS INVALID BECAUSE COLUMN xxxxx IS NOT PART OF THE RESULT TABLE". [...] 20162 267 54_SV: SV: Passing data in stored procedures via cursors.0_19_Sven.Heidorn@CSN.SE31_Fri, 18 Aug 2000 15:32:33 +0200729_ISO-8859-1 Martin,

you can find the latest manuals at:

http://www-4.ibm.com/software/data/db2/os390/library.html

There has been a refresh of the SQL REFERENCE and has to be the CS26-9014-01 version to contain the proper information.

Hope this helps

Sven Heidorn



> -----Ursprungligt meddelande----- > Från: martin_wolff@GLOBALCROSSING.COM > [SMTP:martin_wolff@GLOBALCROSSING.COM] > Skickat: den 18 augusti 2000 14:59 > Till: DB2-L@RYCI.COM > Ämne: Re: SV: Passing data in stored procedures via cursors. > > Yes, there is some info in the V6 Technical Update but not much. The DB2 > V6 manuals I > have don't have anything on DECLARED GLOBAL TEMPORARY TABLEs. > > Can any one [...] 20430 60 32_Re: Autoreply when not in office0_22_The-Thanh.Luong@AGF.BE31_Fri, 18 Aug 2000 15:31:50 +0200395_- Hi,

What I do in case I'm out is 1) send a message SET DB2-L NOMAIL to listserv@ryci.com 2) set MS out of office autoreply as needed When I return, I send a SET DB2-L MAIL to listserv@ryci.com

I am then sure that I don't annoy anybody on the list with the autoreply, and if I need to read something posted during the period of my absence, I can still search the archives... [...] 20491 57 56_Re: SQLSTATE 42707 in DB2 for OS/390 Version 5 Release 111_rick creech18_ykcirc@HOTMAIL.COM29_Fri, 18 Aug 2000 13:48:07 GMT614_- It UQ31706.

Regards, Rick Creech



>From: "Peter, Georg" >Reply-To: DB2 Data Base Discussion List >To: DB2-L@RYCI.COM >Subject: SQLSTATE 42707 in DB2 for OS/390 Version 5 Release 1 >Date: Fri, 18 Aug 2000 15:26:49 +0200 > >Hello List. > >One of our programmers came back from a class and he was happy. He has seen >that Version 5.1 allows columns in the ORDER BY clause that are not in the >SELECT clause (?!). > >We tried this in one of our DB2 develeopment systems and ran into a >SQLSTATE >42707 "THE ORDER BY CLAUSE IS INVALID BECAUSE COLUMN xxxxx [...] 20549 100 32_Re: Autoreply when not in office4_IBM119_IBM1@EGEBANK.COM.TR31_Fri, 18 Aug 2000 17:00:49 +0300325_- that's excellent practice.

however, let's not get too hard on autoreplies - I might have sent a few myself by mistake in the past and I was written off the list!!! not joking - after a few autoreplies I came back from the hols to find a message which said "buzz off" due to replies received from your inbox :-) [...] 20650 69 40_Re: Comparing Teradata to DB2 and Oracle19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Fri, 18 Aug 2000 09:10:25 -0500626_ISO-8859-1 Hi Michael, I can't address the ratios. However, please consider the following: 1. Are any of the DBMSs currently in-house and how many DBAs would be available for this new requirement. 2. The key components I would be considered before DBA support costs are: a) can the platform/DBMS handle the load (be sure to include latent demand), b) are tools available for the DBMS that will meet the user's requirements for adhoc queries c) what is the criticality of the data housed in the chosen DBMS and it's requirements for availability, rapid restore/recovery times, d) consider maintenance requirements and how [...] 20720 22 46_DB2 OS390 V5 READ ACCESS TO SQLSERVER DATABASE13_Terry Shockey30_Terry.Shockey@OCSE.STATE.AR.US31_Fri, 18 Aug 2000 09:21:41 -0500403_iso-8859-1 Good morning list

I have been asked to come up with a process for accessing tables in an sqlserver database from CICS transactions , is there anyone out there that

is doing that sort of thing and if so what product(s) are you using to accomplish this . The access to the sqlserver database tables would be read only and the sqlserver is windows NT soon to be windows 2000. [...] 20743 33 25_DB2 Connect on Peoplesoft14_Gary Bernhardt18_GBernhardt@CSI.COM31_Fri, 18 Aug 2000 09:42:46 -0500400_ISO-8859-1 We recently converted from Centura to DB2 Connect for our Peoplesoft Financials app and have experienced a 15% increase in CPU.

We are running 5.2 of DB2 Connect and V5 of DB2 OS/390 and 7.57 of PS Tools. We have WLM implemented in Goal Mode and used the suggested DDF classifications given to us from Peoplesoft. We also had to tweak them somewhat to work in our environment. [...] 20777 80 30_Stored procedures LE refresh??15_Stovall, Martin16_MStovall@USG.COM31_Fri, 18 Aug 2000 09:51:47 -0500460_- We are just starting to work with stored procedures; and the V5 application and programming (6.2.3.2) guide states:

Refresh Language Environment.

Do this when someone has modified a load module for a stored procedure, and that load module is cached in a stored procedures address space. When you refresh Language Environment, the cached load module is purged. On the next invocation of the stored procedure, the new load module is loaded. [...] 20858 71 29_Re: DB2 Connect on Peoplesoft14_David Williams25_David-O.Williams@UBSW.COM31_Fri, 18 Aug 2000 16:05:51 +0100281_ISO-8859-1 Can you determine where the extra CPU is being consumed? We have noticed a huge amount of CPU being attributed to DDF enclave but as yet are unable to determine the cause. It would be ineterseting to hear if you have a similar problem. We are using DB2 connect V6. [...] 20930 26 9_DEFINE NO0_29_paul.packham@POSTOFFICE.CO.UK31_Fri, 18 Aug 2000 16:15:22 +0000384_us-ascii Hi,

This comment is really for our IBM colleagues on the list to reply to. The DEFINE NO option which was introduced in V6 to me seems somewhat useless. My reason for this is that objects which were created with DEFINE NO can only have a limited set of utilities run against them. At the moment Copy, Recover, Quiesce etc. issues message DSNU185I and terminates. [...] 20957 126 34_Re: Stored procedures LE refresh??12_Steve Grimes30_Steve_Grimes@AISMAIL.WUSTL.EDU31_Fri, 18 Aug 2000 10:15:58 -0500387_us-ascii



Hello,

We are using LE Cobol Stored Procedures under V6. We've have added the following step to the bottom of our Stored Procedure compile/bind JCL.. This accomplishes a refresh of the specific SP being recompiled. We only refresh the entire SPAS when we want to clean out the SYSOUT Display messages. (There is also a "QUEUE" option for the ACTION) [...] 21084 120 64_Re: Restarting an RC/Migrator stragegy on another LPAR/subsystem9_Rob Crane22_racrane@CONCENTRIC.NET31_Fri, 18 Aug 2000 09:45:55 -0600475_us-ascii Bob Brock's suggestion about using the UNLOAD MODE is the best way to go. Craig's suggestion will work but you are then editing what the product creates and I personally favor to not edit what the product creates if at all possible. You are probably executing the analysis output by placing an S next to it within Migrator. Once you do this you will get the Batch Processor Screen with your strategy analysis output hard coded into Batch Processor Input Field. [...] 21205 115 29_Re: DB2 Connect on Peoplesoft12_craig patton21_prgpatton@HOTMAIL.COM29_Fri, 18 Aug 2000 15:31:14 GMT562_- DAve, The main reason for the high usage by DDF enclave is the nature of the PEopleSoft beast!.. All work is passed through DDF. Watch for running of NVISION reports as the method of setting the reports up can greatly effect DDF usage. If choosing to NOT do JOINS or SUBQUERIES, NVISION will run a bunch of repetative SQLs, passing data back and forth to NViSION continuously. This has effected our CPU as well. BUT in most cases, not doing the joins and subqueries has improved report timings over 90% in some cases, but has (very rare) also slowed some [...] 21321 34 17_RID pool failures16_Alison Pelletier26_alison_pelletier@YAHOO.COM31_Fri, 18 Aug 2000 08:46:46 -0700297_us-ascii Lost my old email, so have to start a new thread here... Ganesh Seetharaman wrote in an email...

"only one RID block(4096 RIDS) is allowed per SQL statement I mean that 16 KB...Even if there is a shortage the SQL can process 4096 RIDS before it turns of the Tablespace scan." [...] 21356 181 64_Re: Restarting an RC/Migrator stragegy on another LPAR/subsystem12_craig patton21_prgpatton@HOTMAIL.COM29_Fri, 18 Aug 2000 15:59:24 GMT586_- Rob,

I agree with using UNLOAD MODE as a better solution (I am old school BP and have been editing strategies for over 12 years now)

BUT

I don't agree with running the analysis of strategies multiple times. Analysis is a VERY heavy RESOURCE user and there is a COST to running multiple analysis'. This should be weighed against the cost of running the UNLOAD MODE option or doing a simple 5 second manual update to the PDS Member. When working in a PeopleSoft type of environment and perhaps loading all production data to a Test environment, there are over [...] 21538 216 40_Re: Question about shrlevel change reorg12_Isaac Yassin23_yassin@NETVISION.NET.IL31_Fri, 18 Aug 2000 19:13:21 +0200613_x-user-defined Hi, My mistake. Of course shrlevel none reuse.

Isaac Yassin

mebert@AMADEUS.NET wrote: > > You can't use REUSE with SHRLEVEL REFERENCE. I guess you mean SHRLEVEL NONE with > REUSE, which is what I would recommend for a space-constrained REORG (along with > SORTKEYS, probably NOSYSREC and possibly SORTDATA; also consider Inline Copy) > when maximum availability is not an issue. > With NOSYSREC, you can't restart an abended REORG: so make sure you've got a > valid image copy for the TS (no need to take one immediately before every REORG, > just make sure you _can_ RECOVER if [...] 21755 62 64_Re: Restarting an RC/Migrator stragegy on another LPAR/subsystem0_20_bjnigh@HOUSEHOLD.COM31_Fri, 18 Aug 2000 09:37:12 -0700357_us-ascii Ray,

You can run the strategy in batch mode using processing option "U" (indicates unload only) on the first lpar. You then resubmit the stategy in batch mode using processing option "N" (indicates that unloads are bypassed) on the second lpar. We do this all the time at our shop. Please email me directly if you have other questions. [...] 21818 47 17_Quoting Scripture27_MORRISON, WILLIAM F (SBCSI)21_wm8728@TXMAIL.SBC.COM31_Fri, 18 Aug 2000 12:23:05 -0500522_ISO-8859-1 Team DB2,

I want to apologize for the way I tried to quote someone on the list as a means of starting a discussion. It would have been better if I had just popped in an asked the question instead of beating around the bush. This topic (Identity) has been weighing in on my mind for over a month now, and the sudden burst of activity regarding this topic got me very excited. It is true that I misconstrued some discussion about the Identity column from various posts to this forum. Some background: [...] 21866 91 21_Re: Stored Procedures11_Robert Jans21_Jans.Robert@AMSTR.COM31_Fri, 18 Aug 2000 11:52:00 -0600412_- Sorry, Richard. We followed this path when we developed our V5 stored procs and it worked fine there. Now we're in the process of migrating to V6 and we're having a ton of trouble migrating the stored procs to V6. Here's the catch, according to the IBM lab the V6 cat maint will NOT migrate any stored proc that uses an AUTHID to V6. Also, AUTHID is not supported. The correct V6 route is to use schemas. [...] 21958 73 42_Calling a stored proc from a batch program15_Stovall, Martin16_MStovall@USG.COM31_Fri, 18 Aug 2000 13:12:19 -0500267_iso-8859-1 I'm calling a stored proc from a TSO batch program and getting the following abend -

DSNE949E COMMIT PROCESSING ABNORMAL TERMINATION, REASON CODE X'00E30104' DSN END

Has anyone encountered this problem??

Any help is appreciated. Thanks 22032 56 13_Re: DEFINE NO15_Murley, Michael22_Michael_Murley@BMC.COM31_Fri, 18 Aug 2000 13:30:19 -0500447_iso-8859-1 Paul,

I know you asked for an IBM response, but I just wanted to note that BMC COPY PLUS for DB2* Version 6.1.00 and RECOVER PLUS for DB2* Version 3.3.00 both work OK with DEFINE NO spaces. There are some cases when COPY PLUS calls the DB2 Quiesce Utility; in those cases we fail because the Quiesce is not successful if a table space or one of its related indexes is DEFINE NO. But we try to work in all cases when we can. [...] 22089 25 11_Re: Quoting13_Seibert, Dave26_Dave_Seibert@COMPUWARE.COM31_Fri, 18 Aug 2000 14:59:52 -0400373_iso-8859-1 Hi Bil, I don't think anyone wants you to apologize. I sure hope you don't think I expected that.

I hope that my light-hearted response would have come through as intended. I merely wanted to clarify what I had tried to say. I missed it if there was any pitting one against another going on (unless you're referring to my joking reference to Roger). [...] 22115 32 46_Re: Calling a stored proc from a batch program12_Ragam, Sai J18_Sai.Ragam@BNSF.COM31_Fri, 18 Aug 2000 14:19:08 -0500419_iso-8859-1 Are you issuing a commit within the stored procedure? This is not supported in db2 v5 or v6 for os/390.

-----Original Message----- From: Stovall, Martin [mailto:MStovall@USG.COM] Sent: Friday, August 18, 2000 1:12 PM To: DB2-L@RYCI.COM Subject: Calling a stored proc from a batch program





I'm calling a stored proc from a TSO batch program and getting the following abend - [...] 22148 54 13_Re: DEFINE NO19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Fri, 18 Aug 2000 14:43:56 -0500400_ISO-8859-1 Hi Paul, SAP/PeopleSoft needs to be able to create only the tables they need by purchasable product or piece. See the archives for more on this topic. I think this is really a problem created by SAP/PeopleSoft and IBM is trying to help their customers overcome it. And I'm sure, over time, the process will improve. SAP/PeopleSoft, its after 10 PM, do you know where your tables are? [...] 22203 29 26_How is QUERYNO determined?13_Brian Charles20_bcharles@RITEAID.COM31_Fri, 18 Aug 2000 15:51:45 -0400338_iso-8859-1 We have a process that alerts us to access path changes resulting from rebinds. It is based upon the premise that the QUERYNOs don't change after being rebound. This concept is obviously invalid when a program's source code has changed because once the QUERYNOs change, the last 2 access plans cannot be equally compared. [...] 22233 58 21_Re: RID pool failures14_Richard Yevich21_ryevich@ATTGLOBAL.NET31_Fri, 18 Aug 2000 15:08:10 -0400570_iso-8859-1 1 - 16 millions rids 2 - 25% of rids from an index selected 3 - rid pool short on storage

Hope this helps, Richard Yevich

> -----Original Message----- > From: DB2 Data Base Discussion List [mailto:DB2-L@RYCI.COM]On Behalf Of > Alison Pelletier > Sent: Friday, August 18, 2000 11:47 AM > To: DB2-L@RYCI.COM > Subject: RID pool failures > > > Lost my old email, so have to start a new thread > here... Ganesh Seetharaman wrote in an email... > > "only one RID block(4096 RIDS) is allowed per SQL > statement I mean that 16 KB...Even if there [...] 22292 119 21_Re: Stored Procedures14_Richard Yevich21_ryevich@ATTGLOBAL.NET31_Fri, 18 Aug 2000 15:13:03 -0400590_iso-8859-1 Yes, but this person is still V5 and was made aware privately of the V6 conversion issues.

Hope this helps, Richard Yevich

> -----Original Message----- > From: DB2 Data Base Discussion List [mailto:DB2-L@RYCI.COM]On Behalf Of > Robert Jans > Sent: Friday, August 18, 2000 1:52 PM > To: DB2-L@RYCI.COM > Subject: Re: Stored Procedures > > > Sorry, Richard. We followed this path when we developed our V5 > stored procs > and it worked fine there. Now we're in the process of migrating to V6 and > we're having a ton of trouble migrating the stored procs to V6. [...] 22412 19 50_Can a change in HPSEQT increase buffer pool paging13_Mohammad Khan20_mkkhan88@HOTMAIL.COM31_Fri, 18 Aug 2000 15:52:47 -0500507_- Well the title says it all. We decreased HPSEQT value in order to give more room to randomly accessed pages. Later I noticed that there is substantial amount of paging being shown in our monitor ( CA-NSIGHT ), it might have been there before the change but I don't know. I notified the DBA who called IBM support and they blamed it on the most recent change - the HPSEQT. AFAIK bp paging is caused by not enough memory being available to back the buffer pool allocation, unless I'm out of my depth ! [...] 22432 30 54_Re: Can a change in HPSEQT increase buffer pool paging0_26_truman.g.brown@VERIZON.COM31_Fri, 18 Aug 2000 17:27:24 -0400342_us-ascii All subsystems experience paging; normally paging is to ex- panded storage and is asynchronous. If you're paging to ex- panded a lot, or to aux storage (DASD), then the problem be- comes serious.

Take a look at the paging averaged over the period of time the subsytem has been up to get a better idea of how heavy it is. [...] 22463 18 16_Plans-2-Packages19_Freeling, Marv, ITD25_mfreelin@CO.ALAMEDA.CA.US31_Fri, 18 Aug 2000 15:05:28 -0700334_- Hi,

We have on our plate a plan to packages conversion. We have not yet assembled a plan to tackle the conversion, but I understand from the last IDUG that there was a redbook published on that topic many years ago, and it's currently out of print. We've been trying to locate a copy. Does anyone, perhaps, have a copy? [...] 22482 39 20_Re: Plans-2-Packages9_Don Alden25_Don.Alden@I-STRUCTURE.COM31_Fri, 18 Aug 2000 16:02:19 -0600655_iso-8859-1 Howdy Mary,

The easiest and quickest way to convert Plans to Packages is to buy a product form CA known as "Package It". It does most of the work for you.

Good Luck, Don

-----Original Message----- From: Freeling, Marv, ITD [mailto:mfreelin@CO.ALAMEDA.CA.US] Sent: Friday, August 18, 2000 4:05 PM To: DB2-L@RYCI.COM Subject: Plans-2-Packages



Hi,

We have on our plate a plan to packages conversion. We have not yet assembled a plan to tackle the conversion, but I understand from the last IDUG that there was a redbook published on that topic many years ago, and it's currently out of print. We've [...] 22522 53 30_Re: How is QUERYNO determined?11_Suresh Sane21_data_arch@HOTMAIL.COM29_Fri, 18 Aug 2000 18:05:47 CDT684_- Brian:

Could one of the INLCLUDEs be different(e.g any DCLGEN) - even when the source did not change? This would change the relative position (line #) causing a different queryno?

Suresh



>From: Brian Charles >Reply-To: bcharles@riteaid.com >To: DB2-L@RYCI.COM >Subject: How is QUERYNO determined? >Date: Fri, 18 Aug 2000 15:51:45 -0400 > >We have a process that alerts us to access path changes resulting from >rebinds. It is based >upon the premise that the QUERYNOs don't change after being rebound. This >concept is obviously >invalid when a program's source code has changed because once the QUERYNOs >change, the last 2 [...] 22576 44 54_Re: Can a change in HPSEQT increase buffer pool paging14_Joel Goldstein29_joel_goldstein@COMPUSERVE.COM31_Fri, 18 Aug 2000 21:32:36 -0400269_ISO-8859-1 I can't see any way this can affect paging. If your workload is mostly random in the pool you changed, you did the right thing. I often lower HPSEQT to 20%, and sometimes even 5% for large HPs. The paging reported by insight is at the VP level, not HP. [...] 22621 14 55_Daryl G Spletzer/Pr/Consumers/CMS is out of the office.16_Daryl G Spletzer24_dgspletzer@CMSENERGY.COM31_Sat, 19 Aug 2000 02:38:25 -0400365_us-ascii I will be out of the office from 08/18/2000 until 08/29/2000.

I will respond to your message when I return.

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 22636 18 44_Pamela Harris/IT/SEARS is out of the office.0_17_pharri1@SEARS.COM31_Sat, 19 Aug 2000 01:57:28 -0500537_us-ascii I will be out of the office from 08/18/2000 until 08/21/2000.

If you need assistance with DB2, please use the primary DB2 pager, for informix use the primary informix pager, for network issues please call Bill Pugh, PIN # 7270063.

DB2 pin # 7270526, Informix pin # 7270525

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 22655 225 54_Re: SV: Passing data in stored procedures via cursors.14_James Campbell29_James.Campbell@HANCORP.COM.AU31_Sat, 19 Aug 2000 17:00:12 +1000521_iso-8859-1 One question I have about this technique is: is the CICS transaction generated through an EXCI call the same DB2 transaction as the caller? Anyone checked it out?

/* standard disclaimer */ James Campbell DBA Hansen Corporation, Doncaster +61 3 9843 8442 James.Campbell@Hancorp.com.au

-----Original Message----- From: Sven.Heidorn@CSN.SE [mailto:Sven.Heidorn@CSN.SE] Sent: Friday, August 18, 2000 9:50 PM To: DB2-L@RYCI.COM Subject: [DB2-L] SV: Passing data in stored procedures via cursors. [...] 22881 254 54_Re: SV: Passing data in stored procedures via cursors.14_Richard Yevich21_ryevich@ATTGLOBAL.NET31_Sat, 19 Aug 2000 03:21:08 -0400698_iso-8859-1 From the caller thru the Stored Procedure, through WLM, RRSAF to EXCI to CICS --> same DB2 thread.

Hope this helps, Richard Yevich +======+======+======+ Richard_Yevich@attglobal.net http://www.YLAssoc.com -- DB2 Performance Journal http://www.ryci.com -- RYC Knowledge Database

> -----Original Message----- > From: DB2 Data Base Discussion List [mailto:DB2-L@RYCI.COM]On Behalf Of > James Campbell > Sent: Saturday, August 19, 2000 3:00 AM > To: DB2-L@RYCI.COM > Subject: Re: SV: Passing data in stored procedures via cursors. > > > One question I have about this technique is: is the CICS transaction > generated through an EXCI call the same DB2 transaction as the [...] 23136 26 19_How to set DB2_HOME9_James Tse18_jamestse@EMAIL.COM31_Sat, 19 Aug 2000 08:17:20 -0400336_us-ascii I install dbi for Linux DB2 and get error message as "DB2_HOME" value must set to DB2 install directory.

Some expert of perl dbi teach me that "DB2_HOME must be set to /usr/IBMdb2/v6.3" not to set "usr/IBMdb2/

Does anyone know how to set DB2_HOME? using command or edit the parameter file or something else? [...] 23163 74 13_Re: DEFINE NO12_Isaac Yassin23_yassin@NETVISION.NET.IL31_Sat, 19 Aug 2000 17:41:37 +0200568_x-user-defined Hi, Someone told they're on the disks somewhere ...;-))

Isaac Yassin

"DAVIS, RICK (SBCSI)" wrote: > > Hi Paul, > SAP/PeopleSoft needs to be able to create only the tables they need > by purchasable product or piece. See the archives for more on this topic. > I think this is really a problem created by SAP/PeopleSoft and IBM > is trying to help their customers overcome it. And I'm sure, over time, the > process will improve. > SAP/PeopleSoft, its after 10 PM, do you know where your tables are? > > Regards, > Rick Davis > > "This [...] 23238 73 21_Re: Quoting Scripture12_Isaac Yassin23_yassin@NETVISION.NET.IL31_Sat, 19 Aug 2000 17:44:36 +0200573_x-user-defined Hi, .

It's not the lurk , it's the chimes.

Isaac Yassin





"MORRISON, WILLIAM F (SBCSI)" wrote: > > Team DB2, > > I want to apologize for the way I tried to quote someone on the list as a > means of starting a discussion. It would have been better if I had just > popped in an asked the question instead of beating around the bush. This > topic (Identity) has been weighing in on my mind for over a month now, and > the sudden burst of activity regarding this topic got me very excited. [...] 23312 13 29_Test Message - Please Ignore.13_NoreenLaFleur26_Noreen.LaFleur@PRODIGY.NET31_Sat, 19 Aug 2000 14:34:38 -0400282_iso-8859-1 Test Message - Please Ignore.

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 23326 19 28_Re: S0C4 Abends in DB2/OS39012_Raymond Bell20_raymondb@NZ1.IBM.COM31_Mon, 21 Aug 2000 10:59:35 +1200501_us-ascii Robert,

Sounds like you're having fun. Can't vouch for the other codes, but 00C90101 is fairly common. Usually seen in the company of DSN1COPY jobs that didn't do what they were intended to do. Good luck with the others.



Raymond

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 23346 34 64_Re: Restarting an RC/Migrator stragegy on another LPAR/subsystem12_Raymond Bell20_raymondb@NZ1.IBM.COM31_Mon, 21 Aug 2000 11:41:45 +1200621_us-ascii Robert/Steen/Ken/Craig/Rob/anyone else I've missed off,

Thanks for the feedback. I saw the option to unload only, exclude unloading, etc. on the Batch Processor panel but didn't twig you could invoke it without having a related strategy. I haven't tried it but I'm sure it'll work.

I've 'faked' it before by manually editing the generated worklist, er, I mean strategy (oops, BMC history coming out there) as you suggest, Craig, but I really wanted to avoid all that editing rubbish if I could. Looks like invoking the Batch Processor outside a migration strategy is the way to go. And yes, I [...] 23381 14 33_John Ordman is out of the office.0_29_John_Ordman@READERSDIGEST.COM31_Mon, 21 Aug 2000 01:05:23 -0400365_us-ascii I will be out of the office from 08/18/2000 until 08/28/2000.

I will respond to your message when I return.

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 23396 21 37_Advantage of RAW DEVICE (and SMS/DMS)13_Kim Bundgaard10_kib@DDF.DK31_Mon, 21 Aug 2000 02:09:24 -0500370_- What are the advantage/disadvantage of using raw devices with DB2 UDB tablespaces ?

Some of the disadvantage must be managing, where copying the tablespace is not possible if the tablespace are not part of the UNIX file system (except with DB2 backup of course).

And what if I choose not to use raw devices, should I use the SMS or DMS tablespace ? [...] 23418 89 56_AW: SQLSTATE 42707 in DB2 for OS/390 Version 5 Release 112_Peter, Georg15_G.Peter@DZBW.DE31_Mon, 21 Aug 2000 09:27:27 +0200372_iso-8859-1 Thank you, Rick.

With kind regards - mit freundlichen Grüssen, G e o r g H . P e t e r

-----Ursprüngliche Nachricht----- Von: rick creech [mailto:ykcirc@HOTMAIL.COM] Gesendet am: Freitag, 18. August 2000 15:48 An: DB2-L@RYCI.COM Betreff: Re: SQLSTATE 42707 in DB2 for OS/390 Version 5 Release 1

It UQ31706.

Regards, Rick Creech [...] 23508 87 79_Re: DB2 Education - for dean montevago - FW: Welcome to Issue XII I of DB2Today4_IBM119_IBM1@EGEBANK.COM.TR31_Mon, 21 Aug 2000 11:05:14 +0300512_- dean,

on saturday my eye fell on the following update to the IBM US course calendar!! and that's the course I mentioned in my latest - somebody must've been reading my mind at our eduservices!!

ciao!

Aurora Dell'Anno Systems Engineer IBM Software Support e-mail: ibm1@egebank.com.tr aurora.dellanno@sistinf.it _____________________________________________________ No trees were killed in the sending of this message. However, a large number of electrons were terribly inconvenienced. [...] 23596 20 16_SQL maintenance.9_Harish P.21_emailharish@YAHOO.COM31_Mon, 21 Aug 2000 05:02:18 -0500507_- Is there any way all SQLs relating to a table be kept in one source file (or more specifically a COPYBOOK) and then this would copied in the COBOL program by say INCLUDE TAB1SQL,SQLNUM=5. This would help keeping all SQLs clean and optimisation easier.

Now my problem is, this works fine with assembler programs, but with COBOL I dont think this is directly achievable. The only method I can think is write an exit routine. But was thinking of other 'industry-standard' solutions. Any ideas??? [...] 23617 89 17_AW: Timestampdiff12_Peter, Georg15_G.Peter@DZBW.DE31_Mon, 21 Aug 2000 12:39:00 +0200399_iso-8859-1 Jay,

which DB2 and what version is used when you're dealing with TIMESTAMPDIFF ?

With kind regards - mit freundlichen Grüssen, G e o r g H . P e t e r DB2 Data Base Administrator c/o

-----Ursprüngliche Nachricht----- Von: damcon2@US.IBM.COM [mailto:damcon2@US.IBM.COM] Gesendet am: Donnerstag, 17. August 2000 16:17 An: DB2-L@RYCI.COM Betreff: Re: Timestampdiff [...] 23707 32 70_Frank Ingrasia Courses - High Performance Optimization for DB2 ar gggg23_Pendlebury-Bowe, Leslie34_Leslie.Pendlebury-Bowe@EU.SONY.COM31_Mon, 21 Aug 2000 12:02:46 +0100732_iso-8859-1 I have been trying to get in touch with IMSI who run his courses ... but nobody responds to my requests.

If they fail at this hurdle should I even bother to go on their course?(High Performance Optimization for DB2) are there better courses out there?

regards

Leslie



********************************************************************** This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. Sony cannot accept liability for statements made which are clearly the sender's own and not made on behalf of Sony. (01) ********************************************************************** [...] 23740 56 75_Re: Frank Ingrasia Courses - High Performance Optimizatio n for DB2 ar gggg0_19_mike.holmans@BT.COM31_Mon, 21 Aug 2000 12:10:29 +0100293_- I would doubt whether there are any better courses that Frank's. I've been twice (once when V2.3 was the new one, and last year). You've probably missed him for this year, at least in the UK: the UK agents who organise the courses here are GSoft, and their phone no is +44 1494 680459. [...] 23797 67 74_Re: Frank Ingrasia Courses - High Performance Optimization for DB2 ar gggg13_Adrian Savory24_adrian.savory@ZURICH.COM31_Mon, 21 Aug 2000 12:16:53 +0100296_us-ascii Hi Les,

Gsoft are the people who administer this course in the UK - it used to be handled by Geoff Petherwick (Geoff are you out there?). The number I've got for them is 01494 680459. Of course they'd have a web site as well.

It's a good course - go on it if you can. [...] 23865 77 76_Re: Frank Ingrasia Courses - High Performance Optimizatio n for D B2 ar gggg23_Pendlebury-Bowe, Leslie34_Leslie.Pendlebury-Bowe@EU.SONY.COM31_Mon, 21 Aug 2000 12:19:52 +0100419_iso-8859-1 Mike If you think they are good and so does Adrian (S) then I might aswell stick with them. will call the people you say deal with the courses in the uk

thanks Les

-----Original Message----- From: mike.holmans@BT.COM [mailto:mike.holmans@BT.COM] Sent: Monday, August 21, 2000 12:10 PM To: DB2-L@RYCI.COM Subject: Re: Frank Ingrasia Courses - High Performance Optimizatio n for DB2 ar gggg [...] 23943 88 75_Re: Frank Ingrasia Courses - High Performance Optimization for DB 2 ar gggg23_Pendlebury-Bowe, Leslie34_Leslie.Pendlebury-Bowe@EU.SONY.COM31_Mon, 21 Aug 2000 12:28:21 +0100607_iso-8859-1 Ade just called them .. they are waiting for info from the printers and the details will be sent out.

Les

-----Original Message----- From: Adrian Savory [mailto:adrian.savory@ZURICH.COM] Sent: Monday, August 21, 2000 12:17 PM To: DB2-L@RYCI.COM Subject: Re: Frank Ingrasia Courses - High Performance Optimization for DB2 ar gggg



Hi Les,

Gsoft are the people who administer this course in the UK - it used to be handled by Geoff Petherwick (Geoff are you out there?). The number I've got for them is 01494 680459. Of course they'd have a web site as well. [...] 24032 33 46_cant connect to DB2 OS/390 v6 from DB2 Aix 6.121_Karthikeyan Vinayagam18_vkarthik@YAHOO.COM31_Mon, 21 Aug 2000 06:15:09 -0500403_- Hi all,

when I try to connect to Db2 os/390 v6 from aix db2 6.1 EE, I am getting

"SQL5048N The release level of the database client is not supported by the release level of the database server"

I am getting rc 00d3101c on os/390, which says

"DB2 has detected an error while processing a connection request from DDCS. A DB2RA connection request was received from DDCS" [...] 24066 41 50_Re: DB2 OS390 V5 READ ACCESS TO SQLSERVER DATABASE10_Mike Evans28_EVANSMP@CO.MECKLENBURG.NC.US31_Mon, 21 Aug 2000 08:20:10 -0400368_US-ASCII Terry, we have a similar need. I talked with IBM and they suggested that we use Datajoiner to replicate the SQL Server data to a DB2 table and read the DB2 table from CICS. CICS works in 2 phase commit mode which prohibits a direct read to SQL Server. Another option we are exploring is using MQSeries. If you come up with anything better, please share. [...] 24108 24 29_Refresh Language Environment.23_Mary Petras/Norwalk/IBM19_marypetr@US.IBM.COM31_Mon, 21 Aug 2000 06:48:59 -0600373_us-ascii For DB2-managed stored procedures, issue -STOP followed by -START to tell DB2 that the stored procedure needs to be reloaded.

For WLM-managed stored procedures, load the new stored procedure by issuing a V WLM,APPLENV=DB2TSP,REFRESH, which has the effect of recycling the WLM stored procedure address space as well as loading the new stored procedure. [...] 24133 66 54_Re: Can a change in HPSEQT increase buffer pool paging23_Pendlebury-Bowe, Leslie34_Leslie.Pendlebury-Bowe@EU.SONY.COM31_Mon, 21 Aug 2000 14:04:30 +0100551_iso-8859-1 Joel can I have your web site address? thanks Les

-----Original Message----- From: Joel Goldstein [mailto:joel_goldstein@COMPUSERVE.COM] Sent: Saturday, August 19, 2000 2:33 AM To: DB2-L@RYCI.COM Subject: Re: Can a change in HPSEQT increase buffer pool paging



I can't see any way this can affect paging. If your workload is mostly random in the pool you changed, you did the right thing. I often lower HPSEQT to 20%, and sometimes even 5% for large HPs. The paging reported by insight is at the VP level, not HP. [...] 24200 106 54_AW: Can a change in HPSEQT increase buffer pool paging12_Peter, Georg15_G.Peter@DZBW.DE31_Mon, 21 Aug 2000 15:23:09 +0200530_iso-8859-1 Leslie,

take a look at http://www.responsivesystems.com/ That is the url you are looking for.

With kind regards - mit freundlichen Grüssen, G e o r g H . P e t e r DB2 Data Base Administrator c/o ------------------------------------------------------------------------ Datenzentrale Baden-Wuerttemberg Department Product Support Krailenshaldenstrasse 44, 70469 Stuttgart, Germany Phone 0049-711-8108-271 PC-Fax 004971189696071 ------------------------------------------------------------------------ [...] 24307 137 21_Re: AW: Timestampdiff0_18_damcon2@US.IBM.COM31_Mon, 21 Aug 2000 09:36:45 -0400534_iso-8859-1 Georg,

I'm using this with DB2 UDB V6.1 for AIX. It's in the UDB for AIX V5.2 manual, but I don't know if it existed prior to that. Don't believe it's valid on OS/390. Is it not working for you?

Cheers, Jay



Jay Reavill damcon2@us.ibm.com IBM Global Services Tampa, Florida Tel: (813) 801-7303, Tie Line 8-427-7303 ----------------------------------------------------- Happiness is not around the corner. Happiness is the corner. - BMW ----------------------------------------------------- [...] 24445 56 26_LOAD LOG NO - Log activity14_William Gannon41_wgannon@INET-SERVICES.PALMBEACH.K12.FL.US31_Mon, 21 Aug 2000 10:14:52 -0400568_us-ascii Ok - I am perplexed !

I created a subsystem that is intended to be a query only environment. The only activity in the environment will be client ODBC access via Crystal Reports and Weekly refresh Loads jobs of the data (using Load Log NO). I initially allocated the active logs very small thinking there would only be a small amount of data being recorded (no inserts, updates or deletes). Boy was I wrong. Over the weekend I loaded the 48 tables (48 million records) and to my astonishment, and the operators misery, I spun over 50 archive logs. [...] 24502 45 30_Re: LOAD LOG NO - Log activity20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM31_Mon, 21 Aug 2000 10:19:48 -0400670_us-ascii Bill

If you have NPI on the table, then the structure changes are logged despite LOG NO.

Regards, Venkat Pillay

-----Original Message----- From: William Gannon [SMTP:wgannon@INET-SERVICES.PALMBEACH.K12.FL.US] Sent: Monday, August 21, 2000 10:15 AM To: DB2-L@RYCI.COM Subject: LOAD LOG NO - Log activity

Ok - I am perplexed !

I created a subsystem that is intended to be a query only environment. The only activity in the environment will be client ODBC access via Crystal Reports and Weekly refresh Loads jobs of the data (using Load Log NO). I initially allocated the active logs very small thinking there would only be [...] 24548 111 29_Re: Upgrade to CAE version 6.0_34_L-IS.Kirkpatrick@MUTUALOFOMAHA.COM31_Mon, 21 Aug 2000 09:17:34 -0500582_us-ascii Leo - Since I am not a Visual Basic programmer, I have to relate this incompatibility from my second hand experience.

When we originally upgraded to DB2 version 6 on one of our test platforms, it caused our Visual Basic programs that called stored procedures to quit working. These programs were coded using the "RDO" API. It turned out that the cause of our problems was that our CAE version (CAE version 5.2 fixpack ??) was not aware of the DB2 version 6 stored procedure changes (of which there are many). We retested using the latest fixpack of CAE version [...] 24660 69 30_Re: How is QUERYNO determined?14_Grainger, Phil20_Phil.Grainger@CA.COM31_Mon, 21 Aug 2000 15:25:49 +0100427_iso-8859-1 And even IBM catches up with the problem (eventually) :)

DB2 V7 has an extra clause on all embedded SQL statements "QUERYNO nnnnn" so you can hard-code those query numbers in your apps!

Phil Grainger Computer Associates

-----Original Message----- From: Suresh Sane [mailto:data_arch@HOTMAIL.COM] Sent: 19 August 2000 00:06 To: DB2-L@RYCI.COM Subject: Re: [DB2-L] How is QUERYNO determined? [...] 24730 158 21_AW: AW: Timestampdiff12_Peter, Georg15_G.Peter@DZBW.DE31_Mon, 21 Aug 2000 16:28:20 +0200300_iso-8859-1 Jay,

Thanks for your answer. You are right. It does not work in our DB2 for OS/390 Version 5. If I tried this the SQLSTATE 42601 came up. May be (and I hope) in a newer release this will work........

With kind regards - mit freundlichen Grüssen, G e o r g H . P e t e r [...] 24889 12 4_Test14_Philip Gunning18_pgunning@PHEAA.ORG31_Mon, 21 Aug 2000 11:02:12 -0400266_us-ascii This is a test.

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 24902 13 13_Re: DEFINE NO12_Dan Sullivan28_daniel.sullivan@ZURICHUS.COM31_Mon, 21 Aug 2000 10:02:25 -0500339_us-ascii I thought the 6.1.0 version of Copy Plus for DB2 would not be avaliable until September?

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 24916 27 8_Re: Test12_Mike O'Neill29_mike.oneill@CUSTOMS.TREAS.GOV31_Mon, 21 Aug 2000 11:08:57 -0400537_US-ASCII could you do your TESTing via some other medium?



______________________________ Reply Separator _________________________________ Subject: Test Author: "DB2 Data Base Discussion List" at smtplink Date: 8/21/00 11:02 AM



This is a test.

================================================ To change your subscription options or to cancel your subscription visit the DB2-Lwebpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. [...] 24944 50 35_C++ Include files for UDB vs. OS39011_Jeff Brokaw24_rufus_mcdoofus@YAHOO.COM31_Mon, 21 Aug 2000 08:13:17 -0700530_us-ascii Trying to port code from NT, Linux, Solaris, and AIX that uses DB2/UDB V6 APIs and CLI calls, and this code has various #includes, in particular these are giving me trouble right now when compiling on OS390 :

#include #include #include

These files don't exist in my DB2 / OS390 V6 include library (DSN610.SDSNC.H). That lib has the following members : DSNCA,DSNCLI,DSNCLI1,DSNEXT,DSNSQL,DSNSTATE,DSNSYSTM,DSNUDF SQL,SQLCA,SQLCLI,SQLCLI1,SQLEXT,SQLSTATE,SQLSYSTM,SQLUDF [...] 24995 21 8_Re: Test13_Mohammad Khan20_mkkhan88@HOTMAIL.COM31_Mon, 21 Aug 2000 10:26:29 -0500414_- On Mon, 21 Aug 2000 11:08:57 -0400, Mike O'Neill wrote:

> could you do your TESTing via some other medium? >

Probably this could be another feature for people writing listserv software to consider, pretty much like the "loopback" interface that TCP/IP guys have so that some people can have the warm feeling of being connected without irritating other people ! [...] 25017 27 54_Re: Can a change in HPSEQT increase buffer pool paging14_Joel Goldstein29_joel_goldstein@COMPUSERVE.COM31_Mon, 21 Aug 2000 11:45:45 -0400534_ISO-8859-1 Les, I saw Georg Peter sent it to you (thanks Georg, but I'll repeat anyway. Are you coming to IDUG/Geneva? Regards, Joel

www.responsivesystems.com





Message text written by DB2 Data Base Discussion List >Joel can I have your web site address? thanks Les<

===============================================To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 25045 28 30_Re: How is QUERYNO determined?11_Mike Turner29_Michael_Turner@COMPUSERVE.COM31_Mon, 21 Aug 2000 11:46:59 -0400451_ISO-8859-1 QUERYNO nnnnn is in Version 6 of DB2 for OS/390.

Regards Mike Turner Email: michael_turner@compuserve.com Home Page: http://ourworld.compuserve.com/homepages/michael_turner

Message text written by DB2 Data Base Discussion List >And even IBM catches up with the problem (eventually) :)

DB2 V7 has an extra clause on all embedded SQL statements "QUERYNO nnnnn" so you can hard-code those query numbers in your apps! [...] 25074 36 48_Loops in REBUILD INDEX (ALL) after DSN1COPY used12_Myron Miller22_myronwmiller@YAHOO.COM31_Mon, 21 Aug 2000 08:52:04 -0700566_us-ascii I've found that REBUILD INDEX (ALL) will in some cases create a permanent CPU loop if DSN1COPY is used to copy an image copy of a table with different segment sizes.If you use REBUILD INDEX (specific) for each of the indexes, you won't get a loop, you'll get a quick 00C90101 abend for each index.IBMs official response to this when I reported this situation to them was a) don't have the problem (obviously); b) "The looping is our method with existing code that says a problem exists.We could not allow the REBUILD to bypass our code and as a result [...] 25111 128 30_Re: LOAD LOG NO - Log activity0_18_mebert@AMADEUS.NET31_Mon, 21 Aug 2000 18:02:16 +0200311_us-ascii Hi Venkat,

sorry but I don't understand your answer - what do you mean with "structure changes"?

Another thing that I would think is going to be logged by LOG NO utilities is the updates to SYSCOPY, SYSUTILX and SYSLGRNX (but probably not 50 logs worth even if they are very small). [...] 25240 22 19_Host variable value15_Kuashal Trivedi17_c_triveka@BAM.COM31_Mon, 21 Aug 2000 12:02:06 -0400408_iso-8859-1 Hi All, One of the program , has a cursor wherein a host variable if used it takes about a couple of hours to run , and when hard coded the value it runs in about 5 minutes, One of the things I was told is when the host variable was used with REPOT(VARS) still did not work, I wanted to know of any such experiences and if changing this to a dynamic SQL might work, Any input is appreciated. [...] 25263 186 30_Re: LOAD LOG NO - Log activity20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM31_Mon, 21 Aug 2000 12:09:11 -0400392_iso-8859-1 Hello Dr.

There is a little description in UTILITY GUIDE page 127, chapter 2-9.

" If the tablespace has a non-partitioning index, some logging might occur in the build phase if DB2 is accessing another partition for an SQL statement or utility. Also DB2 logs index structure changes to allow recoverability of the nonpartitioning index in case an abend occurs." [...] 25450 105 23_Re: Host variable value0_18_mebert@AMADEUS.NET31_Mon, 21 Aug 2000 18:11:27 +0200352_us-ascii Hi Kaushal (or is it Kuashal?),

probably the datatype of the host variable and the DB2 column are different. In that case, the host variable won't be used in an indexed access. Different length is enough to cause this.

Hth

Dr. Michael Ebert DB2 Database Administrator aMaDEUS Data Processing Erding / Munich, Germany [...] 25556 42 25_DB2 Connect on Peoplesoft0_39_brian_holroyd.consultant@PEOPLESOFT.COM31_Mon, 21 Aug 2000 09:21:37 -0700437_us-ascii Craig,

Please refrain from broadcasting disparaging remarks about PeopleSoft's DB2 expertise to the whole Listserv, and making comments about PeopleSoft's testing strategy, when you obviously don't have a clue about either.

All PeopleSoft major and minor releases are fully platform tested on DB2 for OS/390 using the same test scripts as are used by every other database platform they support. End of story. [...] 25599 40 23_Re: Host variable value15_Tonmoy Dasgupta32_tonmoy.dasgupta@MAIL.STATE.AR.US31_Mon, 21 Aug 2000 11:23:29 -0500433_iso-8859-1 Did you check if the host-variable definition is exactly right for the column? For example you have a col col_x as char(5) and define host variable hv-col-x as PIC x(06)...you will not use an index on col_x if there exists one...

Tonmoy Dasgupta. -----Original Message----- From: Kuashal Trivedi [mailto:c_triveka@BAM.COM] Sent: Monday, August 21, 2000 11:02 AM To: DB2-L@RYCI.COM Subject: Host variable value [...] 25640 26 40_Re: Comparing Teradata to DB2 and Oracle12_Roger Miller19_millerrl@US.IBM.COM31_Mon, 21 Aug 2000 11:04:41 -0500441_- When I look at DBA tasks, there is so much variation in what they do to make the simple figures close to meaningless. There are not many DB2 DBAs who only work on the query side. There are some situations where the DB2 world is shared, and that means working with other people to deliver a lower total cost. The warehouses tend to have more data than the operational side, and fewer complex relationships. But everything is changing. [...] 25667 27 17_PeopleSoft Tables19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Mon, 21 Aug 2000 11:34:59 -0500611_iso-8859-1 Brian Holroyd, Would you care to comment on when PeopleSoft will be able to install only the required tables needed to support a saleable piece or function within PeopleSoft?

Rick Davis "This e-mail and any files transmitted with it are the property of SBC, are confidential, and are intended solely for the use of the individual or entity to whom this e-mail is addressed. If you are not one of the named recipient(s) or otherwise have reason to believe that you have received this message in error, please notify the sender at 314-235-6854 and delete this message immediately from your [...] 25695 23 29_Re: DB2 Connect on Peoplesoft12_Roger Miller19_millerrl@US.IBM.COM31_Mon, 21 Aug 2000 11:25:05 -0500618_- Are you fairly current on DB2 service? We work very closely with our friends just up the road at PeopleSoft, like Brian Holroyd. We have some IBMers with offices in both locations, since we're just 45 minutes apart when the traffic is not bad.

Dynamic statement caching is probably the biggest opportunity for savings, and caching also provides an easier technique for finding the long-running statements. There were many solid sessions on PeopleSoft tuning at the last IDUG. If you have a colleague who attended, perhaps you can browse the foils and notes. There were sessions with general tuning, like [...] 25719 29 21_Re: PeopleSoft Tables0_39_brian_holroyd.consultant@PEOPLESOFT.COM31_Mon, 21 Aug 2000 10:00:33 -0700353_us-ascii Rick,

PS delivered tables by product up to PS 7 However, because many tables are shared by multiple products, it was a nightmare to deliver additional product to customers because some customers would already have the tables, and some customers wouldn't. Starting in PS 7.5, we deliver everything, and it's likely to stay that way. [...] 25749 32 40_Administrative Tools for UDB v6.1 on AIX0_22_BILL_GALLAGHER@PHL.COM31_Mon, 21 Aug 2000 13:00:45 -0400484_us-ascii Hi,

Just a quick poll looking for some advice/suggestions for administrative tools for UDB v6.1 running on AIX. We're just about getting to the point where our workload of changes in this environment is getting a little too complex for using the native tools that are shipped with UDB, and are looking for a set of third-party admin tools to help support the environment. Implementing database structure changes, change management, etc., is what I'm looking for. [...] 25782 80 44_Re: Administrative Tools for UDB v6.1 on AIX21_Sean M O'Keefe/AC/VCU15_sokeefe@VCU.EDU31_Mon, 21 Aug 2000 13:09:38 -0400364_us-ascii Bill, At VCU we are an OS/390 & AIX DB2 shop. We use BMC for OS/390, and DataBaseGuys for AIX. (http://www.database-guys.com). The AIX toolkit is more monitoring/tuning that Admin-type functions, but check them out anyway. It's an excellent suite of tools, fairly priced, that provide a lot of insights into what happening in DB2 wolrd for AIX. HTH, [...] 25863 32 30_Re: How is QUERYNO determined?12_Roger Miller19_millerrl@US.IBM.COM31_Mon, 21 Aug 2000 11:56:21 -0500348_- There are several complexities that make simple comparisons fail. There are differences from release to release and possibly even in a PTF. This is one type of problem we try to warn you about - using or depending upon something that is not an interface. Here is the text from the V6 Application Programming and SQL Guide, with an overview. [...] 25896 31 13_Re: DEFINE NO15_Di Carlo, Donna23_Donna_Bermender@BMC.COM31_Mon, 21 Aug 2000 12:52:02 -0500389_iso-8859-1 V6.1.00 of COPY PLUS for DB2 has been GA since July 31, 2000.

Donna Di Carlo BMC Software

-----Original Message----- From: Dan Sullivan [mailto:daniel.sullivan@ZURICHUS.COM] Sent: Monday, August 21, 2000 10:02 AM To: DB2-L@RYCI.COM Subject: Re: DEFINE NO



I thought the 6.1.0 version of Copy Plus for DB2 would not be avaliable until September? [...] 25928 13 44_Re: Administrative Tools for UDB v6.1 on AIX12_Dan Sullivan28_daniel.sullivan@ZURICHUS.COM31_Mon, 21 Aug 2000 12:58:31 -0500372_us-ascii You might try Embarcadero.com. They make DBArtisan. I have used this for Sybase and Oracle and I'm sure they support DB2.

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 25942 18 17_Replication tools15_Jardim, Luciano18_LJardim@BANDAG.COM31_Mon, 21 Aug 2000 13:12:18 -0500460_- We would like to buy replication tools that could read from/write to DB2/OS390 and MS-SQL Server. Any suggestions or comments?

Thanks in advance,

Luciano C. Jardim Bandag, Inc. - Database Administration

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 25961 64 21_Re: PeopleSoft Tables19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Mon, 21 Aug 2000 13:44:41 -0500511_iso-8859-1 Brian, I think there may be a simple way to escape the "grief". To avoid trying to install tables that already exist, simply scan the DB2 Catalog. If you get a hit don't run the CREATE but check for an ALTER that might be needed for that product/release for that table. If table not found, run CREATE, don't run the ALTER. Now, DEFINE NO isn't needed and you can avoid all the "grief" related to utilities. You solve PS's problem and install only those tables needed by product. Happy Ending :) [...] 26026 23 21_COBOL Copybook to DDL13_Keith Gilbert17_kgilbert@CSA1.COM31_Mon, 21 Aug 2000 13:57:44 -0500504_iso-8859-1 Hello List,

Does anyone know of a utility, program, or magic potion that will take a COBOL copybook and create the DDL for a table. Sort of like a reverse DCLGEN.

Thanks

Keith Gilbert DBA Central Service Association Tupelo, MS USA

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 26050 22 23_"DCLGEN" on DB2 on AIX?15_Tonmoy Dasgupta32_tonmoy.dasgupta@MAIL.STATE.AR.US31_Mon, 21 Aug 2000 14:05:09 -0500550_iso-8859-1 How does one generate a program host-variable copybook for a table in UDB V5.1 for AIX? (mainframe equivalent is DCLGEN)...and I am interested in a COBOL copybook...

TIA.

Tonmoy Dasgupta State of Arkansas, Database Administration Department of Information Systems. Ph: (501)-682-5109

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. 26073 48 27_Re: "DCLGEN" on DB2 on AIX?10_Leo Pedron19_lpedron@BOSCOVS.COM31_Mon, 21 Aug 2000 15:24:47 -0400492_iso-8859-1 You could try the db2dclgn command. This will generate a DCLGEN for you to use with a cobol program. I have been using in under version 6.1 for a while. We have found a few problems with the dclgen's created not being exactly to the way our programmers would want them, but nothing serious. We are actually FTPing the dclgens created to our mainframe for cobol programs that run there and do remote calls to NT/AIX databases. See the Command Reference for a full description. [...] 26122 56 25_Re: COBOL Copybook to DDL20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM31_Mon, 21 Aug 2000 15:40:53 -0400327_us-ascii Once we had this big project going, to convert thousands of copybooks to DDL and import it to ERWIN. I was given the task for writing REXX-ISPF EDIT MACRO. I finally figured out the rules had many loop holes also I was nearly writing a COBOL parser.

With much fuss, we found there we had two more choices, [...] 26179 777 42_Repost: Welcome to Issue XIII of DB2 Today13_Morrill, John12_JohnM@VP.NET31_Mon, 21 Aug 2000 13:43:38 -0600567_iso-8859-1 Dear DB2 Interested Parties,

Below is the current issue of IBM's online DB2 newsletter, "DB2 Today", as well as information on how to subscribe. DB2 Today is published monthly and summarizes key IBM data management related news since the last issue. As you can see from Issue XIII below, news items are very short with links back to details on the IBM Data Management Internet site. An edition is produced for each geography and subscribers can define a "My DB2 Today" profile so that the summary is tailored to your individual interest areas. [...] 26957 90 101_Repost: IBM DB2 Everyplace Version 7.1 -- A Relational Database M anagement System for Mobile Devices13_Morrill, John12_JohnM@VP.NET31_Mon, 21 Aug 2000 13:44:15 -0600383_iso-8859-1 Dear DB2 Interested Parties,

DB2 Everyplace Version 7.1 is now generally available, see below for a summary bulletin. (Note: DB2 Everyplace is the renamed, new version the DB2 for mobile and pervasive computers formerly known as DB2 Everywhere.) For more information and a trial copy of DB2 Everyplace, go to http://www.ibm.com/software/data/db2/everyplace/ . [...] 27048 100 97_Repost: CA to Grant International Awards for Innovation in DB2 fo r OS/390 Information Management13_Morrill, John12_JohnM@VP.NET31_Mon, 21 Aug 2000 13:44:42 -0600519_iso-8859-1 -----Original Message----- From: Michael Swift/Santa Teresa/IBM [mailto:dbswift@us.ibm.com] Sent: Wednesday, August 16, 2000 10:23 AM To: DB2 Interested Parties Subject: RESEND (corrected URL): CA to Grant International Awards for Innovation in DB2 for OS/390 Information Management



Dear DB2 Interested Parties,

I am resending the bulleting below with the corrected URL for CA's Awards for Innovation in DB2 for OS/390 Information Management, http://www.ca.com/products/db2/award/ [...] 27149 280 88_Repost: SAP AG and IBM Announce General Availability of the S/390 SAP Application Server13_Morrill, John12_JohnM@VP.NET31_Mon, 21 Aug 2000 13:45:26 -0600401_iso-8859-1 Dear DB2 Interested Parties,

A significant new capability for running the SAP R/3 Application Server on the S/390 is now available (previously only the SAP R/3 Database Server could be run on S/390, and the SAP R/3 Application Server on UNIX or NT). Below is the SAP AG press release, as well as 2 articles about the announcement that appeared in InformationWeek and InfoWorld. [...] 27430 80 105_Repost: Redbook Announcement: SG24-5641-00 Business Intelligence Architecture on S/390 Presentation Guide13_Morrill, John12_JohnM@VP.NET31_Mon, 21 Aug 2000 13:46:00 -0600416_iso-8859-1 Dear DB2 Interested Parties,

The new "Business Intelligence Architecture on S/390" redbook (pdf file) is available on the web at http://www.redbooks.ibm.com/abstracts/sg245641.html

In presentation format, the book gives a comprehensive overview of BI architectures and user tools in the S/390 environment. The announcement bulletin is appended below.

Regards, Mike + + + + + + + [...] 27511 105 51_FW: DB2 UDB Fastpath Training for Experienced DBA's13_Morrill, John12_JohnM@VP.NET31_Mon, 21 Aug 2000 13:46:42 -0600383_iso-8859-1 Dear DB2 Interested Parties,

See below for a new DB2 UDB for UNIX, Windows, OS/2 training offering targeted for DBA's with experience administering other relational DBMSs. It is labeled "Fastpath" because it is designed to be completed in only 1 day (8 hours). This is offered via "distance learning" (interactive Web delivery, or orderable in CD ROM format). [...] 27617 59 54_Repost: New Book: DB2 High Performance Design & Tuning13_Morrill, John12_JohnM@VP.NET31_Mon, 21 Aug 2000 13:47:27 -0600417_iso-8859-1 Dear DB2 Interested Parties,

A new book addressing DB2 UDB for OS/390 performance tuning is now available from Richard Yevich and Susan Lawson. Below is a bulletin. Check it out!

Regards, Mike + + + + + + +

DB2 High Performance Design & Tuning ------------------------------------

by Richard Yevich and Susan Lawson

Foreword by Roger Miller, IBM Lead DB2 Strategist [...] 27677 28 28_History of Max Dataset Sizes11_Burch, Todd18_Todd_Burch@BMC.COM31_Mon, 21 Aug 2000 15:08:30 -0500436_iso-8859-1 Can someone summarize for me, by year (preferably) or DB2 (for S/390) release, the maximum dataset sizes allowed for partitioned and non-partitioned tablespaces?

For instance, in 1985, DB2 V1R1, max non-partitioned size allowed was 2gb, and since you could have a maximum of 64 partitions in a partitioned tablespace, was the limit 128gb, or something less? (that seems high - can't recall... memory goes first). [...] 27706 48 32_Re: History of Max Dataset Sizes22_Killen, Martin W - CNF21_Killen.Martin@CNF.COM31_Mon, 21 Aug 2000 13:19:31 -0700478_iso-8859-1 4 gig limit for 1-16 partitons 2 gig limit for 17-32 parts 1 gig limit for 33-64

V5.. HTH

Marty

-----Original Message----- From: Burch, Todd [mailto:Todd_Burch@BMC.COM] Sent: Monday, August 21, 2000 1:09 PM To: DB2-L@RYCI.COM Subject: History of Max Dataset Sizes



Can someone summarize for me, by year (preferably) or DB2 (for S/390) release, the maximum dataset sizes allowed for partitioned and non-partitioned tablespaces? [...] 27755 64 32_Re: History of Max Dataset Sizes22_Killen, Martin W - CNF21_Killen.Martin@CNF.COM31_Mon, 21 Aug 2000 13:23:33 -0700331_iso-8859-1 Here's what comes out of the SQL Reference guide for partitoned..

64 for table spaces that are not defined with LARGE or a DSSIZE greater than 2G 254 for table spaces that are defined with LARGE or a DSSIZE greater than 2G

For table spaces that are not defined with LARGE or a DSSIZE greater than 2G: [...] 27820 63 32_Re: History of Max Dataset Sizes12_craig patton21_prgpatton@HOTMAIL.COM29_Mon, 21 Aug 2000 20:25:45 GMT517_- Todd,

I started working with V1.3 in 1989 and partitioned TAblespaces did not yet exist. the limit was still 64g as a simple TS allowed for 32 secondary dsns (still does as far as I know). V2.1 introduced Partitioning in late 89 or 3 early '90 and the limit was STILL 64G. V2.2 and V2.3 in the '90 as well. V5 introduced the Large Tablespace structure and that was in '95 or '96 (I think) and this is where you can get to 4T of data (I don't believe it is 4T), in a single TS (255 4G partitions) Simple [...] 27884 24 29_Re: DB2 Connect on Peoplesoft17_Venkat Srinivasan28_venkatsrinivasan@HOTMAIL.COM31_Mon, 21 Aug 2000 15:42:09 -0500601_- Craig, I would not use this listserv to comment about the expertise Peoplesoft has in DB2 and about Peoplesoft's testing strategy and methodology. Seriously, What makes you think that you are qualified enough to pass a comment?. It only exposes your ignorance and 'in-depth-knowledge' that you have in trying to understand Peoplesoft platform on DB2. I would first suggest you sign up for a tuning class for Peoplesoft on DB2 and then get a competency level to even recommend to others on what they should listen and what not. Please refrain from using the LISTSERV to satisfy your grudges and [...] 27909 82 30_DB2 OS390 Stored Proc Question10_Missy Case28_Missy.Case@FIRSTDATACORP.COM31_Mon, 21 Aug 2000 16:19:40 -0500341_us-ascii All, I'm coming to the list to gather opinions and any ideas our group may have overlooked.

We have Account Tables that a batch process inputs into nightly. This table is 30 partitions & is reloaded 'fresh' weekly. It has about 500 million rows, an average of 2 million transactions / day use the query described below. [...] 27992 127 44_Re: Administrative Tools for UDB v6.1 on AIX10_Deb Jenson20_Deb.Jenson@QUEST.COM31_Mon, 21 Aug 2000 14:24:38 -0700659_iso-8859-1 Bill, I saw your post regarding tools for DB2 6.1 on AIX and would like to respond with information about Quest Software's solution. We are very excited about bringing a comprehensive database management suite to DB2 UDB users. We've worked hard to create a lightweight and easy-to-use product that provides database administration, performance monitoring, space management and SQL tuning capabilities. Quest Central is currently in beta testing and general availability is anticipated early October. The first release supports DB2 UDB 6.1 and higher for not only AIX, but also Windows, all flavors of UNIX, and OS/2. For more information see [...] 28120 63 0_0_22_Rohn.Solecki@MTS.MB.CA31_Mon, 21 Aug 2000 17:00:58 -0500437_us-ascii Take a look at the V7 What's new doc, http://www-4.ibm.com/software/data/db2/os390/pdf/whatv7.pdf. The doc says that you can do a direct 1 step conversion from V5 to V7. But you have to be aware of incompatiblities introduced by V7 (and V6), that have to be resolved before doing the actual conversion ie: - Type 1 index no longer supported - password protection removed - host variable now MUST have a preceding colon etc [...] 28184 40 25_Re: COBOL Copybook to DDL12_Tal Lifshitz21_Tal.Lifshitz@MSDW.COM31_Mon, 21 Aug 2000 18:02:40 -0400410_us-ascii Keith,

I wrote a REXX script that can do that. The script is executing the COBOL compiler in the background to compile the copybook in a small dummy cobol program that just COPY the copybook. Then I am analyzing the compiler output data types section to determine the column names, types and lengths and create the CREATE TABLE statement. If you are interested, I can send you the script. [...] 28225 38 25_DB2 LOAD Utility with SAP12_Higgins John25_HigginsJohn@JOHNDEERE.COM31_Mon, 21 Aug 2000 17:14:55 -0500521_us-ascii Fellow DB2 list members:

We are faced with a performance problem as we try to load additional factories into a live SAP system.

In outline, I want to do this: Load the next factory into a "staging" SAP system. This "staging" SAP system would be configured as a clone of the live system. This load process would verify all the referential integrity, etc. Use DB2 utilities to extract the loaded rows from the "staging" system. Use DB2 utilities to load the extracted rows into the live system. [...] 28264 15 27_Accessing UDB EEE from CICS7_Db2 Dba24_data_analyst@HOTMAIL.COM31_Mon, 21 Aug 2000 19:24:29 -0500382_- Hi,

Here is a simple question. What do I need to have set up in order to allow CICS get to a UDB EEE on AIX database?

Thanks

================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.