1 WWW.IDUGDB2-L.ORG /home/listserv/home/db2-l July 2000, week 1 2 128 24_Re: PROBUILD replacement12_Isaac Yassin23_yassin@NETVISION.NET.IL30_Sat, 1 Jul 2000 13:23:30 +0200501_x-user-defined Hi, We have our own "home written" REXX-DB2. Sorry, but I can't put it as a tool because it's proprietary. We use a main PL/1 dynamic sql program with assembler routines to access REXX. We manipulate memory areas to make it run faster. We used it for may years now and it beats any product that we've checked in terms of processing speed, resources consumed and ease of writing, moreover it is maintenance free (3 OS/390 upgrades and 3 DB2 versions). Can't ask for more then that. [...] 131 105 57_Re: User Defined Functions: infix notations and qualified14_James Campbell29_James.Campbell@HANCORP.COM.AU30_Sun, 2 Jul 2000 19:40:55 +1000561_iso-8859-1 Bill,

I think you are forgetting that the syntax parser doesn't know what any of the things are that will eventually be used to process the SQL - all it can recognise are syntactical structures. At the syntax level, * is not the same as "*": * is a known operator, "*" is an sql identifier. When the syntax processor sees consequtive identifiers in a select clause (either SELECT KTO SCHEMA1 ... or, I think - I haven't tested it, SELECT KTO "*" ... it treats the second identifier as an implied aliased column rename - which can only be [...] 237 123 26_Log Analyzer & Datacapture14_Walter Schrott21_walter.schrott@VOL.AT30_Sun, 2 Jul 2000 17:11:34 +0200502_iso-8859-1 Dear Friends,

in our production-system we have at least 200 GB of DB2-data and a lot of data manipulation activity. Most of our tablespaces are compressed.

Some month ago our internal revision decided to buy Platinum Log Analyzer (LA) to analyze the log for "critical manipulations". Now we have 2 problems:

1. in case of an update, DB2 logs only the part of the record from the updated column to the end of the record. So LA can only show this part of the record. [...] 361 84 30_Re: Log Analyzer & Datacapture13_McDonald, Ken20_Ken_McDonald@BMC.COM30_Sun, 2 Jul 2000 10:59:17 -0500590_iso-8859-1 Hi Walter,

I must say I am very surprised that Log Analyzer does not complete and decompress partially logged update rows. There may be something procedurely which can be done to allow for some form of completion/decompression. As for the two questions... With DCC, the data should be logged in compressed format. With DATA CAPTURE NONE, for variable length data, DB2 will log from the first changed byte to the end of the row. With DATA CAPTURE CHANGES, DB2 will ensure that either the undo or redo image will be completely logged, but, it should look as it does on [...] 446 19 30_APPC from DB2 Stored procedure22_Ramesh Balasubramanyan15_RAMBAL@SAIF.COM30_Sun, 2 Jul 2000 11:33:39 -0700499_- Hello Folks, Has anybody attempted to use APPC(Advanced program to program communication) from DB2 stored procedure to CICS applications?. If so can you please get in touch with me. I am desperate for some help (and sample code). Thanks !!! regards Ramesh

================================================ 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. 466 56 30_Trigger+StoredProc Performance15_Benjamin Reches24_Benjamin.Reches@MSDW.COM30_Sun, 2 Jul 2000 16:48:57 -0400520_us-ascii Hi All!

Environment:: DB2 OS/390 6.1

I am running some benchmarks to establish what our DB2 throughput capacity is.

I created a C program which reads a flat file and INSERTs each record into a DB2 table with minimal reformatting. There exists an AFTER trigger which calls a STORPROC, which populates a set of DB2 tables for each insert on the 'staging' table. There could be anywhere from 1 to 5 INSERTs performed by the storproc per 'staging' table insert. On the average, about 2-3. [...] 523 24 20_Re: FW: SQLCODE -31015_Matthias Pohlig27_matthias.pohlig@T-ONLINE.DE30_Mon, 3 Jul 2000 04:58:11 -0500395_- Oops ..sorry then... Have only been doing standard "C" , not too much "C++" and don't by far know all them nice classes that are avail.

I'm sure you also know that the -310 gives you an additional hint which variable content DB2 does not like ... in your case, it says the "hundredth" (thats the 100 in the SQLERRMC) position of SQLDA is incorrect. Which leads me to the question: [...] 548 84 33_00C90101 in index and data pages.11_Adam Foster22_Adam.Foster@BULL.CO.UK30_Mon, 3 Jul 2000 13:37:59 +0100449_us-ascii Dear List,

Wonder if anyone on the list has seen this before and knows how to fix.

We're running DB2 v4 on os/390.

When running a select, we get the following error message

The tablespace is partitioned with 34 parts.



an Abend 04E000 hex occurred processing command.

The master address space contains the following

13.28.55 STC22745 DSNI014I @ DSNIRNXT DATA IN USE DURING ABEND [...] 633 182 51_Re: DB2I defaults disappearing when migrating to V615_Robert Lawrence21_rlawrence@BOSCOVS.COM30_Mon, 3 Jul 2000 08:43:48 -0400439_us-ascii our defaults are set by SDSNCLST dataset member DSNEMC01 for each of the DB2I options and set ISPF variables which are used by the various underlying clist/REXX execs HTH

Bob Lawrence DBA Boscov's Dept Stores Inc



-----Original Message----- From: tim malamphy [SMTP:timalamphy@YAHOO.COM] Sent: Friday, June 30, 2000 5:44 PM To: DB2-L@RYCI.COM Subject: Re: DB2I defaults disappearing when migrating to V6 [...] 816 220 60_DB2 UDB V6 on Solaris & IBM's Secure Directory - LDAP abend.11_Louw, Jan J23_JANLouw@MAIL.SBIC.CO.ZA30_Mon, 3 Jul 2000 14:37:58 +0200588_iso-8859-1 hi I am experiencing problems with DB2 UDB and the IBM SecureWay Directory - LDAP products. It would seem like when a load is applied to LDAP for UPDATE , DB2's "LOCK MANAGER" abend, thus forcing us to restart LDAP services. Anybody experiencing the same. What I see is when the -902 (FFFFF51D) happened, LDAP still have threads connected to DB2, but LDAP then start complaining about "Connection failures" when it tries to issue commits. The abend also causes db2 to take a dump of the abend. I've included entries from the db2diag.log around the time of the abend. thx. [...] 1037 160 41_Re: FW: 00C90101 in index and data pages.11_Adam Foster22_Adam.Foster@BULL.CO.UK30_Mon, 3 Jul 2000 15:53:16 +0100324_us-ascii Thanks for the help Kathy.

We don't have Platinum at our shop so all IBM utilities, not sure if I mentioned but the IBM unload failed with the same error message. One ray of light, when we reorg'd some of the individual partitions the reorg worked ok, so we should be able to unload with reorg 'pause'. [...] 1198 65 76_Nested SQL procedures, parameter handling differs between UDB 390 and UDB NT0_19_Sven.Heidorn@CSN.SE30_Mon, 3 Jul 2000 17:38:28 +0200497_ISO-8859-1 Hi all,

we have some SQL procedures which are created and tested using the SP builder. The client application calls one procedure that in turn calls the other procedures. In the call statement I use the parameters declared in the create procedure statement.

## calling procedure ##

CREATE PROCEDURE X090DBT.K191Z ( INOUT XXX_NUMMER INT, IN YYY_NUMMER SMALLINT, IN ARGRPBET CHAR(6), IN INNEVDATUM DATE, OUT REGISTRERINGS_DATU DATE, OUT NAMN CHAR(40), ......... [...] 1264 40 15_SHRLEVEL CHANGE10_David Swan29_david_x_swan@STANDARDLIFE.COM30_Mon, 3 Jul 2000 16:27:55 +0100540_iso-8859-1 The Standard Life Assurance Company, Standard Life House, 30 Lothian Road, Edinburgh EH1 2DH, is registered in Scotland (No. SZ4) and regulated by the Personal Investment Authority. Tel: 0131 225 2552 - calls may be recorded or monitored. This confidential e-mail is for the addressee only. If received in error, do not retain/copy/disclose it without our consent and please return it to us. We virus scan all e-mails but are not responsible for any damage caused by a virus or alteration by a third party after it is sent. [...] 1305 70 19_Re: SHRLEVEL CHANGE23_Pendlebury-Bowe, Leslie34_Leslie.Pendlebury-Bowe@EU.SONY.COM30_Mon, 3 Jul 2000 16:40:57 +0100721_iso-8859-1 David

We do this all the time and does not cause us one single problem.

regards

Leslie Pendlebury-Bowe DB2 DBA

-----Original Message----- From: David Swan [mailto:david_x_swan@STANDARDLIFE.COM] Sent: Monday, July 03, 2000 4:28 PM To: DB2-L@RYCI.COM Subject: SHRLEVEL CHANGE



The Standard Life Assurance Company, Standard Life House, 30 Lothian Road, Edinburgh EH1 2DH, is registered in Scotland (No. SZ4) and regulated by the Personal Investment Authority. Tel: 0131 225 2552 - calls may be recorded or monitored. This confidential e-mail is for the addressee only. If received in error, do not retain/copy/disclose it without our consent and please return it to [...] 1376 27 54_How to alter Bufferpool from 4k to 32k of a tablespace20_Baskaran Subramaniam27_subramaniam.baskaran@DB.COM30_Mon, 3 Jul 2000 20:07:22 +0530343_us-ascii Hello Listusers

We have a tablespace of Bufferpool 4K, we want to alter it to 32K, but when we tried , we are getting the SQLCODE = -671, which states that the

SQLCODE = -671, ERROR: THE BUFFERPOOL ATTRIBUTE OF THE TABLE SPACE CANNOT BE ALTERED AS SPECIFIED BECAUSE IT WOULD CHANGE THE PAGE SIZE OF THE TABLE SPACE [...] 1404 57 58_Re: How to alter Bufferpool from 4k to 32k of a tablespace14_John P Shipley24_jpshiple@DUKE-ENERGY.COM30_Mon, 3 Jul 2000 12:27:31 -0400496_us-ascii Drop and recreate the object. You must change the page size and this requires drop and recreate of the physical object. Good Luck



Patrick..











Baskaran Subramaniam on 07/03/2000 10:37:22 AM

Please respond to DB2 Data Base Discussion List

To: DB2-L@RYCI.COM cc: (bcc: John P Shipley/Corp/DukePower) Subject: How to alter Bufferpool from 4k to 32k of a tablespace [...] 1462 62 58_Re: How to alter Bufferpool from 4k to 32k of a tablespace15_Billings, Linda30_linda.billings@DOA.STATE.WI.US30_Mon, 3 Jul 2000 11:28:39 -0500345_iso-8859-1 Hi, Baskar,

It seems to me that DB2 is not letting you do this for a reason. I have never seen 4K tablespaces paired with 32K bufferpools. Are you changing the size of the row so that it is greater than 4K? If so, I don't see an alternative to changing the size of the pages in the tablespace. Why do you want to do this? [...] 1525 69 58_Re: How to alter Bufferpool from 4k to 32k of a tablespace12_HEPP SHERY C17_schepp@SRPNET.COM30_Mon, 3 Jul 2000 09:40:41 -0700410_ISO-8859-1 You don't specify what version of DB2 you are currently on- but in the SQL Reference guide for V5 under alter tablespace it specifies that the "bpname must identify an activated buffer pool with the same page size as the tablespace". So I think the only way to accomplish this you will have to recreate as larger or create a new tablespace as 32k and move your table to the new tablespace. HTH [...] 1595 105 19_Re: SHRLEVEL CHANGE13_Randy Krueger22_Kruegerr@SCHNEIDER.COM30_Mon, 3 Jul 2000 11:35:21 -0500509_us-ascii Our catalog backups run nightly in the off hours and the only catalog table we found actively update was SYSUTILX & SYSCOPY, for those we use Shrlevel Change.

The others tables are updated via prime time activties and have run as Shrlevel reference in the off hours with no conflicts.







"Pendlebury-Bowe, Leslie" cc: Sent by: DB2 Data Base Subject: Re: SHRLEVEL CHANGE Discussion List [...] 1701 67 20_It's broke again....12_Steve Grimes30_Steve_Grimes@AISMAIL.WUSTL.EDU30_Mon, 3 Jul 2000 12:24:52 -0500580_us-ascii Hello.

Once upon a time there was batch program which executed a DB2 V6 Stored Procedure which called another procedure which had a tight code loop (no SQL). The loop had a display so it filled the spool. My name was taken in vain repeatedly....

Here's the problem though. I cancel the thread, then attempted to cancel the job. The thread is gone, but the job remains. And no one can talk to the Stored Procedure Address space. A -DIS PROCEDURE hangs my TSO session such that I can't cancel it either!. STOP DB2 MODE(FORCE) did nothing on Friday night [...] 1769 18 16_DB2 presentation13_Luis Martinez24_luis_martinez@ES.IBM.COM30_Mon, 3 Jul 2000 12:18:05 -0500340_- I'm looking for a general (and up-to-date) DB2 for OS/390 presentation. I'm able to access a lot of information about DB2, but I can't find a presentation that covers topics like the following: architecture overview, buffering and locking mechanisms, DB2 utilities, ... I could prepare it but the customer is urgently asking for it. [...] 1788 80 24_Re: It's broke again....12_MARTIN WOLFF31_martin_wolff@GLOBALCROSSING.COM30_Mon, 3 Jul 2000 13:43:30 -0400297_us-ascii This won't help your current problem but DB2 V5 has a ASUTIME that limits the CPU each stored procedure can use. I assume V6 has something similar. Your definitions may be set to zero (no limit) which is not a good idea for testing.

At least that may stop it happening again. [...] 1869 14 20_Re: DB2 presentation8_Bob Abad16_BobRabad@AOL.COM28_Mon, 3 Jul 2000 13:55:49 EDT347_US-ASCII Luis, look up Roger Miller at the IBM intranet and I am quite sure he can assist you..Take care!

================================================ 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. 1884 165 41_Re: FW: 00C90101 in index and data pages.12_Isaac Yassin23_yassin@NETVISION.NET.IL30_Mon, 3 Jul 2000 21:20:24 +0200533_x-user-defined Hi, Try to see if moving it to another disk helps. I'm guessing at disk i/o error on that area.

Isaac Yassin

Adam Foster wrote: > > Thanks for the help Kathy. > > We don't have Platinum at our shop so all IBM utilities, not sure if I > mentioned but the IBM unload failed with the same error message. One ray of > light, when we reorg'd some of the individual partitions the reorg worked > ok, so we should be able to unload with reorg 'pause'. > > Anyway, cheers for the help...see if IBM can come [...] 2050 53 20_Re: DB2 presentation16_Richard A Yevich21_ryevich@BELLSOUTH.NET30_Mon, 3 Jul 2000 13:24:59 -0400741_iso-8859-1 Go to www.software.ibm.com/data/dba/os390 and there are all the manuals there, including whats new,etc as well as announcement brochures, etc.

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





> -----Original Message----- > From: DB2 Data Base Discussion List [mailto:DB2-L@RYCI.COM]On Behalf Of > Luis Martinez > Sent: Monday, July 03, 2000 1:18 PM > To: DB2-L@RYCI.COM > Subject: DB2 presentation > > > I'm looking for a general (and up-to-date) DB2 for OS/390 > presentation. I'm > able to access a lot of information about DB2, but I can't find a > [...] 2104 23 34_Re: Trigger+StoredProc Performance12_Roger Miller19_millerrl@US.IBM.COM30_Mon, 3 Jul 2000 13:32:23 -0500447_- Check the red book SG24-6108 Version 6 Technical Update on the web for some other parameters in trigger and stored procedure performance. Start at www.ibm.com/redbooks Put 6108 in the search and click Go. I usually download the pdf file and this is about 3.2 MB, so, maybe not on the 28Kb line. The section is chapter 2, section 2.3 for trigger performance. Invoking triggers and stored procedures can easily take more time than an insert. [...] 2128 45 19_Re: SHRLEVEL CHANGE12_BOB JEANDRON21_BOB.JEANDRON@USDA.GOV30_Mon, 3 Jul 2000 14:08:00 -0400602_- We use Change on SYSUTILX and SYSCOPY only. We also do a log archive when the backup is done. We have had no problems so far. >>> DB2-L@RYCI.COM@inter2 07/03/00 10:25AM >>> The Standard Life Assurance Company, Standard Life House, 30 Lothian Road, Edinburgh EH1 2DH, is registered in Scotland (No. SZ4) and regulated by the Personal Investment Authority. Tel: 0131 225 2552 - calls may be recorded or monitored. This confidential e-mail is for the addressee only. If received in error, do not retain/copy/disclose it without our consent and please return it to us. We virus scan all e-mails but [...] 2174 24 24_Re: It's broke again....12_Steve Grimes30_Steve_Grimes@AISMAIL.WUSTL.EDU30_Mon, 3 Jul 2000 16:39:50 -0500632_us-ascii Thank you Martin.

It turns out it was more of a "hang" than a loop. Anyway, per IBM we "killed it dead" by canceling the IRLMPROC. Thanks again.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Steve Grimes Washington University Information Systems OS/390 V2R6 DB2 UDB 6.1 W: 314-935-4376 H: 636-928-5005 Isa.64:6 Steve_Grimes@aismail.wustl.edu ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

================================================ 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. 2199 45 49_PeopleSoft: Programmer access to production data.13_Mackey, Glenn21_GMackey@GUIDEMAIL.COM30_Mon, 3 Jul 2000 17:02:46 -0500346_iso-8859-1 Hi,

NB: I have posted this on two list server lists (DB2-l and PeopleSoft)

We have a security requirement to remove access from production data from programmers in a PeopleSoft environment. We have no problem with doing this, but I have been asked to see if this is common practice for PeopleSoft type environments. [...] 2245 552 41_Re: FW: 00C90101 in index and data pages.15_Richard Simpson25_rsimpson@MACQUARIE.COM.AU30_Tue, 4 Jul 2000 09:33:27 +1000606_iso-8859-1 Well, every single time we have got this message we have either restored an Image Copy from before an "alter add columns" (see APAR PQ08211) or have done something dumb with DSN1COPY.

Thankfully we have only every had this problem in Test.

We have found that sometimes 'reorging' has fixed the problem (when we had a stuffed up with DSN1COPY and copied between 2 copies of a tablespace with different segsizes - obviously not your situation as you are partitioned). Another time we recovered to a point before the problem was created. Another way we have got around this sort [...] 2798 64 53_Re: PeopleSoft: Programmer access to production data.21_Smith, Ronald M - CNF20_smith.ronald@CNF.COM30_Mon, 3 Jul 2000 16:58:06 -0700586_- Brief answers then :-) a. No b. Yes (controlled Secondary Authids) c. Yes d. Yes

> -----Original Message----- > From: Mackey, Glenn [SMTP:GMackey@GUIDEMAIL.COM] > Sent: Monday, July 03, 2000 3:03 PM > To: DB2-L@RYCI.COM > Subject: PeopleSoft: Programmer access to production data. > > Hi, > > NB: I have posted this on two list server lists (DB2-l and PeopleSoft) > > We have a security requirement to remove access from production data from > programmers in a PeopleSoft environment. We have no problem with doing > this, > but I have been asked to see if this is common [...] 2863 137 37_Re: 00C90101 in index and data pages.0_29_Greg.Palgrave@BANKWEST.COM.AU30_Tue, 4 Jul 2000 08:32:21 +0800337_us-ascii Adam,

You may want to try a REPAIR DBD REBUILD DATABASE as a last resort - we have had a similar instance recently as a result (we think) of datasets in a training environment getting out of synch due to non-DB2 backup/restores. The DBD rebuild resolved the errors for us - you may want to do a test run beforehand. [...] 3001 141 47_Re: Needed query help - Exists Check variation.14_Michael Hannan24_mhannan@C031.AONE.NET.AU30_Tue, 4 Jul 2000 11:09:48 +1000564_us-ascii Kaushal,

Existance Checks.

SELECT 1 is usually better than SELECT COUNT(*), for non-materializing access paths, due to stopping the query scan after finding 2nd row rather than scanning all of them. If there are only 1 or 2 it would not matter much.

If there is a case where finds 1st row quickly, but has to scan a lot before reaching the end and finding NO 2nd row, then can be expensive. In this case, and in cases where OPTIMIZE FOR 1 ROW needed to prevent expensive materialization, Fetch 1 row from a Cursor can be better. [...] 3143 98 30_Trigger+StoredProc Performance14_Michael Hannan24_mhannan@C031.AONE.NET.AU30_Tue, 4 Jul 2000 11:11:19 +1000414_us-ascii Benjamin,

The cost of executing your INSERTs may be quite low if they are not doing random access I/O.

While the time inside the Stored Procedure may be less than < 0.001 secs, the overhead of calling a stored procedure is quite high, perhaps equivalent to executing a dozen simple SQLs or more. I don't have exact numbers. However your performance numbers looked higher than expected. [...] 3242 112 37_Re: 00C90101 in index and data pages.15_Michael McGuire32_Michael.McGuire@COLESMYER.COM.AU30_Tue, 4 Jul 2000 14:11:27 +1000452_iso-8859-1 Was a DSN1COPY run?

I've had this error after a supposedly successful DSN1COPY and the most common cause is that the "to" tablespace does not match the structure (tables/indexes) of the "from" tablespace.

Michael ____________________________________________________________________ Michael McGuire IMS/DB2 DBA Team Leader ITD, Coles Myer Ltd Phone: 9829 4125, mobile: 0417 550 604 email : Michael.McGuire@colesmyer.com.au [...] 3355 69 24_DB2 Maintenance Strategy0_18_mebert@AMADEUS.NET30_Tue, 4 Jul 2000 12:25:52 +0100458_us-ascii Hello List,

a couple of weeks ago I asked whether anyone else had a problem with LOAD DATA REPLACE REUSE INTO TABLE PART nn not working right (the file is not reused, but delete-defined); I got only one response saying "It works here". I did not find any problem report on Engine. I concluded that the problem is most likely fixed by a PTF meant for some different problem, which caused me to enquire about our DB2 maintenance strategy. [...] 3425 135 28_Re: DB2 Maintenance Strategy0_19_mike.holmans@BT.COM30_Tue, 4 Jul 2000 13:14:42 +0100592_iso-8859-1 The standard advice is "be as current as possible and apply everything". There's little doubt that this is the ideal.

We don't follow that standard advice.

The main reason we don't is that we don't have the resources to manage it.

I don't know how many subsystems we have off the top of my head, but it's about 30 production subsystems and 50 or so development/test subsystems. We can't afford to have each subsystem on a different maintenance level: we have to have a standard base level so that when a problem occurs, we don't spend half the day trying [...] 3561 33 28_Re: DB2 Maintenance Strategy14_Massimo Scarpa23_Massimo_Scarpa@DEBIS.IT30_Tue, 4 Jul 2000 14:27:53 +0100423_us-ascii Hi Mike & Michael...

My personal standard some time ago was 'Let the others sysprogs apply the most recent PTFs and let them cry for first'. It

was....

Now with 43 DB2 subystems (+ data sharing) I MUST agree with you, Mike, you MUST have a common base to udpate

all running subsystems in a fast and secure way. Otherways there's the risk to have some system updated and others not [...] 3595 17 16_db2text extender16_Tyapuhin, Andrew42_Tyapuhin@SCIENCESOFT.INVENTION-MACHINE.COM30_Tue, 4 Jul 2000 18:21:01 +0300501_koi8-r Hello all I have a problem with start db2 text extender (db2 v 6.01,Windows nt sp6) when I put command "enable server for db2text" I get error "DES9996N An internal Text Extender error occurred.Reason code :708". Can anybody help me solve this problem?

================================================ 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. 3613 116 28_Re: DB2 Maintenance Strategy9_Andy Hunt33_Andy.Hunt@SCOTTISH-SOUTHERN.CO.UK30_Tue, 4 Jul 2000 16:30:34 +0100361_us-ascii As with Massimo and Michael we have a common base which is rolled out to all DB2 sub-systems.

We apply a CBPDO every 4-6 months. This level would be to the RSU recommended in the CBPDO. During this rollout I would also check the PSP to see if there are any other PTFs which we believe are relevant and could be rolled out at the same time. [...] 3730 46 27_Re: DB2 for LINUX on OS/3900_15_leon@CA.IBM.COM30_Tue, 4 Jul 2000 13:33:20 -0400477_us-ascii Linux/390 is brand new and since it is real Linux there are no books that I know of written about the operating system itself. DB2 for Linux/390 is not available yet even in beta so there are certainly no books on that.

If you want to learn Linux there are lots of books on the subject. DB2 for Linux/390 will be a straight port of DB2 for Linux for Intel so you might as well get your hands on the DB2 for Linux and run that on a PC to get your feet wet. [...] 3777 17 14_Bind parameter12_Ranjay Sinha24_rsinha@LOT.TATASTEEL.COM28_Tue, 4 Jul 2000 18:24:36 GMT447_us-ascii Hello ! Can any one tell me what are the various parameter in bind step . The different value that can be given to each parameter and the s significance of each parameter.

Thanks in advance

================================================ 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. 3795 39 18_Re: Bind parameter0_22_Rohn.Solecki@MTS.MB.CA30_Tue, 4 Jul 2000 15:13:58 -0500578_us-ascii Take a look in the "Command Reference" bookreader manual for what ever OS and DB2 version you are using. On web you can try looking at: http://www.multimania.com/db2usa/ebookmanager.htm http://www.s390.ibm.com:80/bookmgr-cgi/bookmgr.cmd/Shelves/DSNSH0G2 OS/390 DB/2 V6 bookshelf http://www.s390.ibm.com/bookmgr-cgi/bookmgr.cmd/Shelves the ENTIRE OS/390 library of manual (s)

HTH









Ranjay Sinha @RYCI.COM> on 2000/07/04 01:24:36 PM

Sent by: DB2 Data Base Discussion List [...] 3835 170 28_Re: DB2 Maintenance Strategy15_Toine Michielse18_vndobtm@US.IBM.COM30_Wed, 5 Jul 2000 01:07:04 +0100522_us-ascii Hello Andy,

Thank you very much for that last paragraph.........

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



Andy Hunt @RYCI.COM> on 07/04/2000 04:30:34 PM

Please respond to DB2 Data Base Discussion List

Sent by: DB2 Data Base Discussion List [...] 4006 27 18_Existance Checking12_Raymond Bell20_raymondb@NZ1.IBM.COM30_Wed, 5 Jul 2000 11:24:12 +1200552_us-ascii Mike,

Saw your response to Kaushal about minimising existence checking. Basically agree with what you're saying - boy, I bet you're relieved... :o) - but I have a question for you. You say it's better to select a literal i.e. 'select 1' rather than 'select count(*)', which it is, but that the fetching will stop after finding the 2nd row. I always thought that the point of saying 'where exists' was that it stops looking for rows once it finds the first qualifying row. If it doesn't work this way I'd appreciate a confirmation. [...] 4034 49 18_Existance Checking14_Michael Hannan24_mhannan@C031.AONE.NET.AU30_Wed, 5 Jul 2000 10:24:29 +1000463_us-ascii Raymond,

Sorry if there was any confusion. 'SELECT 1' over 'SELECT COUNT(*)' referred to a technique without any EXISTS subquery. Having no subquery saves some overhead but causes singleton SELECT to look for a 2nd row even if there isn't one. Later in my answer I mentioned another technique where query was moved into an EXISTS subquery with outer SELECT changing to a select from dummy one row table. Now EXISTS looks for single row only. [...] 4084 45 22_Re: Existance Checking12_Raymond Bell20_raymondb@NZ1.IBM.COM30_Wed, 5 Jul 2000 14:09:36 +1200352_us-ascii Cheers, Mike. Gotcha now. In a direct select, DB2 has to find out whether or not there's another row to return so it knows whether to pass back SQL code +100, 0 or -811. When specifying EXISTS in a subquery DB2 can terminate its searching as soon as it finds a row and evaluate the predicate to be 'true'. So, the confusion was all mine. [...] 4130 12 54_Greg Fernholz/Deluxe Corporation is out of the office.13_Greg Fernholz24_Greg_Fernholz@EFUNDS.COM30_Wed, 5 Jul 2000 01:01:18 -0500312_us-ascii I will be out of the office from 07/04/2000 until 07/10/2000.

================================================ 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. 4143 53 40_Question: SELECT MIN(column) is too slow21_Mykhaylo Krasnyanskyy30_krasnyan@SWH600.LANGEN.BULL.DE30_Wed, 5 Jul 2000 12:49:41 +0200348_us-ascii Dear List,

We are DB2 v5.1 on os/390.

We have one table where we are encountering the following situation:

CREATE TABLE TMQ( TIMESTMP TIMESTAMP NOT NULL, SEGMENT SMALLINT NOT NULL, MSG LONG VARCHAR NOT NULL PRIMARY KEY ( TIMESTMP, SEGMENT ) )

CREATE UNIQUE INDEX I1TMQ ON TMQ ( TIMESTMP ASC, SEGMENT ASC ) [...] 4197 67 53_Re: PeopleSoft: Programmer access to production data.18_Teegarden, Stephen22_Teegarden@AEGONUSA.COM30_Wed, 5 Jul 2000 07:36:51 -0500592_- A. NO B. YES C. YES D. YES

Steve Teegarden, DBA AEGON USA 319-398-8066

> -----Original Message----- > From: Mackey, Glenn [SMTP:GMackey@GUIDEMAIL.COM] > Sent: Monday, July 03, 2000 5:03 PM > To: DB2-L@RYCI.COM > Subject: PeopleSoft: Programmer access to production data. > > Hi, > > NB: I have posted this on two list server lists (DB2-l and PeopleSoft) > > We have a security requirement to remove access from production data from > programmers in a PeopleSoft environment. We have no problem with doing > this, > but I have been asked to see if this is common practice [...] 4265 202 28_Re: DB2 Maintenance Strategy15_Billings, Linda30_linda.billings@DOA.STATE.WI.US30_Wed, 5 Jul 2000 07:58:18 -0500611_iso-8859-1 This scenario applies here. We apply maintenance regularly to our subsystems and try very hard to stay current on all subsystems. If there are too many subsystems then, of course, you can spend all of your time applying maintenance. That is why we also have a project in which we are merging our subsystems together into as few as possible. This makes maintaining them much, much easier ( We have three sysprogs/DBAs to support 14 subsystems but will be down to 9 subsystems when all the merging is finally completed.). Maintenance not only includes keeping the software up-do-date but watching [...] 4468 229 28_Re: DB2 Maintenance Strategy15_Billings, Linda30_linda.billings@DOA.STATE.WI.US30_Wed, 5 Jul 2000 08:01:49 -0500389_iso-8859-1 I forgot to mention that even though we stay current we don't stay THAT current. We don't like to be on the "bleeding" edge and usually wait a few months before applying really new maintenance. We would rather that more brave souls than us find all the problems.

Linda Billings Enterprise Systems Programmer Wisconsin Department of Administration Info-Tech Services [...] 4698 212 53_Re: PeopleSoft: Programmer access to production data.15_Andrea Milligan25_Andrea_Milligan@BCBST.COM30_Wed, 5 Jul 2000 09:13:39 -0400683_iso-8859-1 1) No 2) Yes 3) Yes 4) yes

Andrea Milligan DataBase Administrator BlueCross BlueShield of Tennessee



-----Original Message----- From: Mackey, Glenn [mailto:GMackey@GUIDEMAIL.COM] Sent: Monday, July 03, 2000 6:03 PM To: DB2-L@RYCI.COM Subject: [DB2-L] PeopleSoft: Programmer access to production data.



Hi,

NB: I have posted this on two list server lists (DB2-l and PeopleSoft)

We have a security requirement to remove access from production data from programmers in a PeopleSoft environment. We have no problem with doing this, but I have been asked to see if this is common practice for PeopleSoft type environments. [...] 4911 239 28_Re: DB2 Maintenance Strategy12_Hylton Tom P20_Tom.P.Hylton@IRS.GOV30_Wed, 5 Jul 2000 09:09:14 -0400640_iso-8859-1 1) The terms are discussed in the SMP/E User's Guide. If you have Bookmanager look on SMP/E shelf, or ask your SMP/E sysprog to borrow his/her hardcopy, or go to http://www.s390.ibm.com/bookmgr-cgi/bookmgr.cmd/BOOKS/GIMUGD70/CCONTENTS

2/3) We stay relatively current. We have an ESO subscription that brings us PTFs on a quarterly basis, which we APPLY. This means our subsystems are always 3 months behind the current service level. We feel this is a reasonable position; We're not so far behind that it is impossibly complex to apply a necessary fix, but we aren't so "bleeding edge" that we risk applying a lot of [...] 5151 95 44_Re: Question: SELECT MIN(column) is too slow20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM30_Wed, 5 Jul 2000 09:51:46 -0400642_us-ascii Krasnyansky

Since the index is unique the "garbage collection" (removal of pseudo deleted index entry) is not applicable in your case. If your performance slows down after first commit then following could be reasons -

1. Index look-aside does not work beyond commits. Reducing your commit frequency may help improve index look-aside. 2. Check your Virtual Pool VDWQT parameters. Some times VDWQT=0 could cause waits as the DBM1 takes a latch on the page before externalizing the page, while your application may be try to read that page, Check accounting report to see the wait timings distribution. 3. Your pool [...] 5247 101 53_Re: PeopleSoft: Programmer access to production data.12_craig patton21_prgpatton@HOTMAIL.COM28_Wed, 5 Jul 2000 07:32:27 PDT703_- A. NO B. Yes C. Yes D. Yes

Craig Patton President/DBA Consultant Patton Research Group, Inc. "Specializing in Performance and PeopleSoft In DB2 for OS/390"





>From: "Teegarden, Stephen" >Reply-To: DB2 Data Base Discussion List >To: DB2-L@RYCI.COM >Subject: Re: PeopleSoft: Programmer access to production data. >Date: Wed, 5 Jul 2000 07:36:51 -0500 > >A. NO >B. YES >C. YES >D. YES > >Steve Teegarden, DBA >AEGON USA >319-398-8066 > > > -----Original Message----- > > From: Mackey, Glenn [SMTP:GMackey@GUIDEMAIL.COM] > > Sent: Monday, July 03, 2000 5:03 PM > > To: DB2-L@RYCI.COM > > Subject: PeopleSoft: Programmer access to [...] 5349 38 58_Re: Greg Fernholz/Deluxe Corporation is out of the office.19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM30_Wed, 5 Jul 2000 10:05:23 -0500592_iso-8859-1 please don't autoreply to db2-l

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 of this e-mail is strictly prohibited." [...] 5388 137 44_Re: Question: SELECT MIN(column) is too slow14_Michael Hannan24_mhannan@C031.AONE.NET.AU30_Thu, 6 Jul 2000 01:06:24 +1000518_us-ascii Mykhaylo,

You did not mention how many processes are trying to process 1st item off the queue. So if many concurrent let us know so we can address contention issues.

Getting back to the main point. If you pick items off a queue in this manner, after a while you will get lots of getpages to find the first row due to the psuedo-deleted RIDs as you mentioned. Response time and CPU will get progressively worse. I did not really understand you statement of exactly when response gets worse. [...] 5526 22 19_Fluff: I'm Back....14_Favero, Willie21_Willie_Favero@BMC.COM30_Wed, 5 Jul 2000 10:19:27 -0500585_iso-8859-1 Hi all,

I have not been around much lately because of commitments at BMC. However, I am back now and just wanted to say Hi to everyone on the list. I also wanted to let you all know that I am speaking at local User Groups again. If you need a speaker, drop me a note or visit the web site we have put together at BMC listing topics and speakers available to local user groups. The URL is http://www.bmc.com/db2users/ > Hope to start seeing more of you all Willie Favero [...] 5549 43 23_Re: Fluff: I'm Back....20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM30_Wed, 5 Jul 2000 11:34:28 -0400603_us-ascii Welcome back Willie. We missed you on the list-server almost since two years.

-----Original Message----- From: Favero, Willie [SMTP:Willie_Favero@BMC.COM] Sent: Wednesday, July 05, 2000 11:19 AM To: DB2-L@RYCI.COM Subject: Fluff: I'm Back....

Hi all,

I have not been around much lately because of commitments at BMC. However, I am back now and just wanted to say Hi to everyone on the list. I also wanted to let you all know that I am speaking at local User Groups again. If you need a speaker, drop me a note or visit the web site we have put together at BMC listing [...] 5593 53 34_Re: APPC from DB2 Stored procedure15_Toine Michielse18_vndobtm@US.IBM.COM30_Wed, 5 Jul 2000 17:58:46 +0100484_us-ascii Hello Ramesh,

Many moons ago I wrote an SP that talks APPC to IMS. Contact me offline,



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



Ramesh Balasubramanyan @RYCI.COM> on 07/02/2000 07:33:39 PM

Please respond to DB2 Data Base Discussion List [...] 5647 161 53_Re: PeopleSoft: Programmer access to production data.12_troy Coleman20_Troy.Coleman@UAL.COM30_Wed, 5 Jul 2000 10:55:12 -0500462_- A.NO B.YES C.YES D.YES

Troy Coleman Coleman Consulting, Inc. IBM Certified Solutions Expert DB2 UDB V5 Database Administration IBM Certified Advanced Technical Expert DB2 - DRDA mailto: troycci@colemanconsulting.com voice: 847-705-1468Troy Coleman



---------- From: DB2-L(a)RYCI.COM Sent: Wednesday, July 05, 2000 9:32 AM To: DB2-L(a)RYCI.COM Cc: prgpatton(a)HOTMAIL.COM Subject: Re: PeopleSoft: Programmer access to production data. [...] 5809 126 41_Re: DB2 Maintenance Strategy - PDO vs ESO13_Gendron, Dave28_Dave.Gendron@COASTALCORP.COM30_Wed, 5 Jul 2000 12:19:29 -0500277_iso-8859-1 If I may go on a tangent to the original topic...

Can anyone tell me the pros and cons of using PDO vs ESO? Why are some shops using one and not the other? In the past I have used both but I really didn't understand why one may be better than the other. [...] 5936 216 60_Re: DB2 for OS/390 Performance Monitor - which one? - thanks14_Ulrich, Deanna21_Deanna_Ulrich@BMC.COM30_Wed, 5 Jul 2000 12:13:46 -0500669_iso-8859-1 Myron and Gerald:

To answer your concerns regarding the fusion of the MAINVIEW for DB2 and Activity Monitor products from BMC: Because both products contain rich functionality and are well-liked by customers, we have carefully considered how to combine the two products into a more robust and comprehensive performance monitor. We encourage you to consider MAINVIEW for DB2 because we are taking the base MAINVIEW product and adding Activity Monitor functionality to facilitate integration with other BMC products. The MAINVIEW orientation is important because MAINVIEW for DB2 can be used separately or as a part of a complete OS/390 monitoring [...] 6153 42 34_Re: APPC from DB2 Stored procedure0_15_leon@CA.IBM.COM30_Wed, 5 Jul 2000 18:33:01 -0400738_us-ascii Take a look here http://www-4.ibm.com/software/data/db2/os390/spb/exciting/ for the info on how to use APPC in DB2 Stored Procedures.

Leon Katsnelson, DB2 Connect Development Manager mailto:leon@ca.ibm.com



Ramesh Balasubramanyan on 07/02/2000 02:33:39 PM

Please respond to DB2 Data Base Discussion List

To: DB2-L@RYCI.COM cc: Subject: APPC from DB2 Stored procedure







Hello Folks, Has anybody attempted to use APPC(Advanced program to program communication) from DB2 stored procedure to CICS applications?. If so can you please get in touch with me. I am desperate for some help (and sample code). Thanks !!! regards Ramesh [...] 6196 22 25_DB2 version in production6_DB2usa18_db2usa@HOTMAIL.COM30_Wed, 5 Jul 2000 18:53:15 -0500558_- Hi,

I was wondering what DB2 Version (mainframes) do we use in production (Version 3, Version 4, Version 5 or Version 6).

I just set up a DB2 Version poll on the website: http://www.multimania.com/db2usa/eliendb2.htm

Feel free to vote and check the results.

Best regards,

DB2usa.

================================================ 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. 6219 18 5_Row X36_=?iso-8859-1?q?Database=20Research?=19_db2ibm@YAHOO.COM.AU30_Thu, 6 Jul 2000 13:06:58 +1000563_iso-8859-1 Does anyone have a way of retrieving the 1,000,000 th or 2,000,000 th or .... row in a DB2 OS/390 V5 table?

_____________________________________________________________________________ http://movies.yahoo.com.au - Yahoo! Australia & NZ Movies - Find out what's on at the local cinema with Yahoo! Movies

================================================ 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. 6238 160 44_Re: Question: SELECT MIN(column) is too slow7_Sanjeev19_Sanjeev@INTERRA.COM30_Thu, 6 Jul 2000 11:43:14 +0530521_iso-8859-1 Hi , I agree with Michael's and Venkat's suggestions but as far as the application logic is concerned using Michael's way of saving the last timestamp and using ">" to find the next one would be beneficial for avoiding the pseudo deleted entries search as well as reduce the number of getpages and thus the response time. I think there is misunderstanding as far as this question is concerned "When does the response time actually gets poor".My idea says that the scenario something similar to this one : [...] 6399 80 60_AW: DB2 for OS/390 Performance Monitor - which one? - thanks12_Peter, Georg15_G.Peter@DZBW.DE30_Thu, 6 Jul 2000 08:13:51 +0200330_iso-8859-1 Deanna, I am using the Activity Monitor since 1991. You wrote (see your message below) "We encourage you to consider MAINVIEW for DB2 because we are taking the base MAINVIEW product and adding Activity Monitor functionality..."

Does this mean that we have to change to Mainview for DB2 in the next future ? [...] 6480 91 53_Re: PeopleSoft: Programmer access to production data.0_22_The-Thanh.Luong@AGF.BE30_Thu, 6 Jul 2000 09:16:42 +0200536_- a)- NO b)- yes c)- yes d)- yes



LUONG The-Thanh AGF Belgium / ALLIANZ Group, PRODUCTION TECHNICAL SUPPORT DataBase Administration and Security MGR

URL http://www.agf.be

===== Disclaimer: opinions expressed are strictly my own, etc... =====





-----Original Message----- From: Teegarden, Stephen [mailto:Teegarden@AEGONUSA.COM] Sent: Wednesday, July 05, 2000 2:37 PM To: DB2-L@RYCI.COM Subject: Re: PeopleSoft: Programmer access to production data. [...] 6572 63 9_Re: Row X7_Sanjeev19_Sanjeev@INTERRA.COM30_Thu, 6 Jul 2000 13:10:15 +0530518_iso-8859-1 Hi , In DB2 for OS/390 V5 ,the order in which data is kept and retireved is uncertain.Some reflection of this is there in catalog, i.e cluster ratio etc etc.So the order in which the rows are returned is not always certain unless ORDER BY is specified.So ,if you want to locate the Nth row in association with how the data is kept then i do not think there is any way out for this.But if you want to locate the Nth row if the rows are in ascending or descending sequence of some key i.e the Nth row if [...] 6636 53 44_Re: Question: SELECT MIN(column) is too slow21_Mykhaylo Krasnyanskyy30_krasnyan@SWH600.LANGEN.BULL.DE30_Thu, 6 Jul 2000 11:51:11 +0200580_us-ascii Thanks Michael, thanks Venkat, thanks Sanjeev for your suggestions.

Our process runs currently alone. We haven't any contention now on the table in question. We plan to use many processes that read the queue table. For this purpose we add one more column like queue_name. One process will read its own queue and therefore ( we suppose) will have no contention on the queue table with row locking. Our observation was that the elapsed and CPU Time reduce when we increase the commit frequency. I have used ATF of Omegamon and seen for two successive requests: [...] 6690 31 9_BMC doubt11_K Mahadevan28_kmahadevan@CHENNAI.TCS.CO.IN30_Thu, 6 Jul 2000 17:29:35 +0530541_us-ascii Hi,

We are using BMC CA in our project. During one of the CA attempts it gave a severe error saying NO FULL COPY OR LOG AVAILABLE. We found the SYSCOPY entry for BMC full image copy for that Tablespace that is taken just 3 hrs back and its available.

Only difference is that we are using CDB for reorg . This reorg process will do a quiesce before and after reorg and this has added 2 entries in the SYSCOPY for this tablespace with ICTYPE Q. This Reorg has taken place between the full image copy and the Change [...] 6722 61 13_Re: BMC doubt0_27_jim.leask@RS-COMPONENTS.COM30_Thu, 6 Jul 2000 13:01:21 +0100567_iso-8859-1 I believe if the REORG was LOG NO then any image copys taken previously are now unusable for a 'recover to current' process.

-----Original Message----- From: K Mahadevan [mailto:kmahadevan@CHENNAI.TCS.CO.IN] Sent: 06 July 2000 13:00 To: DB2-L@RYCI.COM Subject: BMC doubt



Hi,

We are using BMC CA in our project. During one of the CA attempts it gave a severe error saying NO FULL COPY OR LOG AVAILABLE. We found the SYSCOPY entry for BMC full image copy for that Tablespace that is taken just 3 hrs back and its available. [...] 6784 50 28_Log Archival Stalls - OS/3900_26_ken.george@EDWARDJONES.COM30_Thu, 6 Jul 2000 07:50:29 -0500508_US-ASCII Hi -

This post concerns a problem which we face every month with an application which does a large number of inserts and updates. We are running DB2/OS390 V5. The database that this application runs against has about 30 tables, each partitioned into 16 partitions so that we can run up to 16 jobs at the same time. Each of these iterations of the program inserts or adds between 10 - 14 million rows (total about 190 million rows). This process usually occurs in a 2-3 hour time period. [...] 6835 66 13_Re: BMC doubt23_Jain, Sanjay (Exchange)19_sanjayjain@BEAR.COM30_Thu, 6 Jul 2000 10:14:50 -0400575_- Rerog LOG NO does make the previous imagecopy unusable in a typical recover situation. But one could still use it via DSN1COPY for 'TOCOPY' like recovery.

Sanjay Jain 8-226-7069

> -----Original Message----- > From: K Mahadevan [SMTP:kmahadevan@CHENNAI.TCS.CO.IN] > Sent: Thursday, July 06, 2000 8:00 AM > To: DB2-L@RYCI.COM > Subject: BMC doubt > > Hi, > > We are using BMC CA in our project. During one of the CA attempts it gave > a > severe error saying > NO FULL COPY OR LOG AVAILABLE. We found the SYSCOPY entry for BMC full > image > copy for that [...] 6902 19 13_Re: BMC doubt15_Melvin Demmicks25_melvin_demmicks@YAHOO.COM30_Thu, 6 Jul 2000 07:24:48 -0700496_us-ascii Can't help you here, but

I just think it's cool that BMC has a product named CA.

__________________________________________________ Do You Yahoo!? Send instant messages & get email alerts with Yahoo! Messenger. http://im.yahoo.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. 6922 39 23_Waiting on Dynamic Bind18_Mitchell Michele E22_mitchell.me@MELLON.COM30_Thu, 6 Jul 2000 11:31:58 -0400494_- > Hello all.. > > We have been having a problem for the past couple of month-ends and can > not seem to come up with a reasonable solution. Any suggestions of things > to look at would be greatly appreciated. > > Environment OS/390 V5 > Two-way data sharing > One member read-only, dynamic SQL > Other member - mix - including dynamic SQL SQL statements are complex and involve multi-table joins up to the current limit > Goal mode work load manager Catalog reorged/runstats 3 weeks ago [...] 6962 60 13_Re: BMC doubt13_De Jonge, Jan20_Jan_De_Jonge@BMC.COM30_Thu, 6 Jul 2000 10:48:06 -0500312_iso-8859-1 A REORG LOG(NO) will invalidate an image copy for future accum processing. This is because the log which is required for recovery was not written. Plus, if you are acccum'ing log for indexes, a REORG LOG(YES) will invalidate the image copy for the same reason. REORG'ed indexes are never logged. [...] 7023 73 27_Re: Waiting on Dynamic Bind19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM30_Thu, 6 Jul 2000 10:59:11 -0500455_iso-8859-1 Hi Michele, Haven't experienced your exact problem but I think I see several areas to check out. I'd make sure that the xxxxIRLMs on both members have a higher dispatching priority than the other, related DB2 ASIDs. Check out REGION size on each member's xxxxDBM1 PROCs. Ensure adequate number of buffers in the affected bufferpools. See if VDWQT and DWQT are set too low on the affected pools. Consider larger HPs for the affected pools. [...] 7097 423 56_Re: Custom-Built Offering for the Installation of DB2 V612_Hylton Tom P20_Tom.P.Hylton@IRS.GOV30_Thu, 6 Jul 2000 11:59:50 -0400371_iso-8859-1 Haven't seen an answer, so I'll throw in.

CBPDO is another alternative. There are other delivery alternatives, each varying somewhat in degree of customization available, amount of work and skill involved upon your part, etc..

I'll give you IBM's definition of the 2 delivery methods you mention, and throw in CBPDO, along w/how I see them: [...] 7521 79 9_Triggers.11_Mohamed Beg19_mohamedb@ABSA.CO.ZA30_Thu, 6 Jul 2000 18:21:41 +0200360_iso-8859-1 Hi.

We have a user who would like to know what updates / inserts have taken place on his table. Unfortunately we do not have any Tools to read the logs. So I've done a logprint.

The problem I've encounted is that the log pages I am trying to read are compressed. Why is this ? And is there any way that I can uncompress the logs. [...] 7601 99 12_Primary keys12_Randy Wilson22_Randy_Wilson@BCBST.COM30_Thu, 6 Jul 2000 12:40:35 -0400297_iso-8859-1 Hiya List :o) In UDB/EEE version 6.1 on AIX ..... when you alter a table and create a primary key, does this also create an index on that value? ... I've been under the assumption that this takes place... Thanx, Randy Wilson BlueCross BlueShield of Tennessee, Inc. DRM , ext. 3675 [...] 7701 49 16_Re: Primary keys13_Lynne Flatley17_LFlatley@NEFN.COM30_Thu, 6 Jul 2000 12:47:40 -0400572_us-ascii Yes it does but the name of the index is not very pretty, so you can specify a constraint name when specifying the primary key and that will be its index name...

> -----Original Message----- > From: Randy Wilson [SMTP:Randy_Wilson@BCBST.COM] > Sent: Thursday, July 06, 2000 12:41 PM > To: DB2-L@RYCI.COM > Subject: Primary keys > Sensitivity: Private > > Hiya List :o) > In UDB/EEE version 6.1 on AIX ..... when you alter a table and create a > primary key, does this also create an index on that value? ... I've been > under the assumption that this [...] 7751 117 27_Re: Waiting on Dynamic Bind15_Billings, Linda30_linda.billings@DOA.STATE.WI.US30_Thu, 6 Jul 2000 12:09:24 -0500527_iso-8859-1 I would bet on the IRLM dispatching priority. It sounds like the rest are the result of the first problem.

Linda Billings Enterprise Systems Programmer Wisconsin Department of Administration Info-Tech Services

"Whom computers would destroy, they must first drive mad." -Anon. <../author/1.html>





-----Original Message----- From: DAVIS, RICK (SBCSI) [mailto:RD8246@MOMAIL.SBC.COM] Sent: Thursday, July 06, 2000 10:59 AM To: DB2-L@RYCI.COM Subject: Re: Waiting on Dynamic Bind [...] 7869 222 16_Re: Primary keys12_Randy Wilson22_Randy_Wilson@BCBST.COM30_Thu, 6 Jul 2000 13:22:10 -0400446_iso-8859-1 Thanks for the reply Lynne ! That's what my assumption has been since we started working with this new(to us) database.... however, I have a table with a single column primary key, and when a simple select statement is executed with a where primarykeyfield = 'xxxx' I'm getting a table scan... I'm having a hard time explaining this...it seems to be a bottleneck on a project I'm working on and I need to speed this up. Randy :o) [...] 8092 99 9_Re: Row X13_Mohan, Ashish21_Ashish.Mohan@NIKE.COM30_Thu, 6 Jul 2000 10:32:29 -0700651_us-ascii Just a simpler query (same logic though) :

SELECT * FROM Table T1 WHERE N = (SELECT COUNT(*) FROM Table T2 WHERE T1.col_name >= T2.col_name)

Assumptions being the same as stated by Sanjeev.

Ashish.

-----Original Message----- From: Sanjeev [SMTP:Sanjeev@INTERRA.COM] Sent: Thursday, July 06, 2000 12:40 AM To: DB2-L@RYCI.COM Subject: Re: Row X

Hi , In DB2 for OS/390 V5 ,the order in which data is kept and retireved is uncertain.Some reflection of this is there in catalog, i.e cluster ratio etc etc.So the order in which the rows are returned is not always certain unless ORDER BY is specified.So ,if [...] 8192 32 10_Pl explain15_Kuashal Trivedi17_c_triveka@BAM.COM30_Thu, 6 Jul 2000 13:51:53 -0400298_iso-8859-1 Hi All, I have a job that was modified by a collegue for some SQL enhancements, I am attaching a few numbers below, its just that except the CPU the rest of the times are on the decline , could some one throw some light on this, or am I looking at the right numbers, please advise. [...] 8225 62 14_Re: Pl explain20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM30_Thu, 6 Jul 2000 14:06:12 -0400357_us-ascii Kuashal

Just be aware that the I/O count show on your job output, does not reflect I/O count for DB2 accesses. These are just I/O count for datasets allocated through your JCL.

The increase in CPU could be for many reasons. If it is DB2 heavy hitter then you should look at your DB2 Accounting report rather than job statistics. [...] 8288 71 32_Re: Log Archival Stalls - OS/39014_William Gannon41_wgannon@INET-SERVICES.PALMBEACH.K12.FL.US30_Thu, 6 Jul 2000 14:49:19 -0400393_iso-8859-1 Hi Ken,

Since no on has responded yet - how about a couple questions - queues :o) - to maybe guide you to a solution ...

-Have you looked at the commit interval for the programs involved - if the unit of work spans logs the archival process will wait until the end of unit of recovery. - What is the size of (and/or) are you seeing Output Log Buffer shortages ?? [...] 8360 26 32_Re: Log Archival Stalls - OS/3909_Colin Fay13_cfay2@CSC.COM30_Thu, 6 Jul 2000 15:01:42 -0400515_us-ascii Hi Ken,

Couple of thoughts...

Can you reduce the quantity of logging by:

1) processing the inserts within the partition in partitioning index sequence to allow more updates/page before logging and/or reducing the commit frequency.

2) dropping the non-partitioning indexes for the duration of the processing to reduce index logging

3) possibly the updates are on variable length records , forcing page overflow processing .. would adjusting the free space help this. [...] 8387 102 16_Re: Primary keys13_Lynne Flatley17_LFlatley@NEFN.COM30_Thu, 6 Jul 2000 15:14:06 -0400599_us-ascii Have you done runstats on the table? How many rows are in the table?

> -----Original Message----- > From: Randy Wilson [SMTP:Randy_Wilson@BCBST.COM] > Sent: Thursday, July 06, 2000 1:22 PM > To: DB2-L@RYCI.COM > Subject: Re: Primary keys > Sensitivity: Private > > Thanks for the reply Lynne ! > That's what my assumption has been since we started working with this > new(to us) database.... however, I have a table with a single column > primary key, and when a simple select statement is executed with a where > primarykeyfield = 'xxxx' I'm getting a table scan... I'm having [...] 8490 51 16_Datajoiner & DB218_Whittaker, Stephen26_stephen.whittaker@CPLC.COM30_Thu, 6 Jul 2000 15:23:50 -0400565_iso-8859-1 We are currently at DB2 V5 and have installed Datajoiner on an NT box both for system test and production. The Datajoiner version is based on V2 technology. My question: We seem to be having alot of problems with 'outer-joins' and Datajoiner. Looks to me like Datajoiner trys to bring down all the tables into a temp table in Datajoiner and then do the work down there. I would have thought a push-down method would apply but talking with IBM they say NO. Datajoiner will always bring the data down to Datajoiner on a outer join and form the result [...] 8542 35 28_Piecesize Limit: 32 datasets0_20_Jones.H@GRAINGER.COM30_Thu, 6 Jul 2000 14:39:00 -0500380_- Using version 5. We have a couple of large (defined as large) tablespaces with 250 partitions. We've been having lots of problems with our NPI's during loads. We get an 04E with a -904 on the NPI and a reason code of 00C9009D

The book says a 00C9009D means "Hey, fella, you can't have an index bigger than 64 gig." To which I said, "Hey, DB2, I'm only using 3 gig." [...] 8578 74 32_Re: Log Archival Stalls - OS/39013_Gendron, Dave28_Dave.Gendron@COASTALCORP.COM30_Thu, 6 Jul 2000 14:53:09 -0500468_us-ascii Ken,

At the risk of stating the obvious, have you made sure that the logs are distributed on volumes so that you are never archiving a log that resides with the currently active log? Have you tried archiving to dasd? Just a few more ideas for the pile.

Dave

Dave Gendron - Database Systems Support Colorado Interstate Gas Co. (Coastal Corp.), 2 North Nevada Ave., Colorado Springs, CO 80903 719-520-4760 dave.gendron@coastalcorp.com [...] 8653 90 32_Re: Piecesize Limit: 32 datasets15_Billings, Linda30_linda.billings@DOA.STATE.WI.US30_Thu, 6 Jul 2000 15:18:51 -0500552_iso-8859-1 I haven't had experience in this area but my colleague has and has explained it to me. A nice learning experience.

This is the story. It looks like IBM is positioning NPIs for eventual partitioning of NPIs. (Has your brain fallen out yet???) I know this is strange but bear with me. What happens is that DB2 will allocate a data set for your NPI and then use that data set up to the size specified in the piecesize. At that point, whether the primary quantity is used up or not, DB2 will allocate another data set for your index [...] 8744 62 32_Re: Piecesize Limit: 32 datasets20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM30_Thu, 6 Jul 2000 16:29:49 -0400426_us-ascii In my work place, we happen to be discussing the same topic. 32 datasets is the limit for the index. In order to use up to 254 datasets, your tablespace must be defined with either DSSIZE or LARGE.

Regards, Venkat Pillay

-----Original Message----- From: Jones.H@GRAINGER.COM [SMTP:Jones.H@GRAINGER.COM] Sent: Thursday, July 06, 2000 3:39 PM To: DB2-L@RYCI.COM Subject: Piecesize Limit: 32 datasets [...] 8807 14 36_Recall: Piecesize Limit: 32 datasets20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM30_Thu, 6 Jul 2000 16:30:09 -0400337_us-ascii Pillay, Venkat (PCA) would like to recall the message, "Piecesize Limit: 32 datasets".

================================================ 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. 8822 63 32_Re: Piecesize Limit: 32 datasets20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM30_Thu, 6 Jul 2000 16:35:17 -0400346_us-ascii The limit for 32 datasets for the index is only for non-large tablespaces. I have tested with LARGE tablespace and I was able to reach 254 partitions without any problem. I am surprised that IBM said 32 datasets limit for a LARGE tablespace !!! Well.. I didn't get problem in my testing, may be there is a PTF missing in your case. [...] 8886 70 32_Re: Piecesize Limit: 32 datasets10_Ron Graham23_rgraham@CDBSOFTWARE.COM30_Thu, 6 Jul 2000 15:59:51 -0500339_ISO-8859-1 If the tablespace is defined as LARGE you should be able to get up to 128 data sets. I have seen a NPI with 91 data sets, so I know it is possible.

Ron



-----Original Message----- From: Jones.H@GRAINGER.COM Sent: Thursday, July 06, 2000 2:43 PM To: DB2-L@RYCI.COM Subject: Piecesize Limit: 32 datasets [...] 8957 80 32_Re: Piecesize Limit: 32 datasets10_Ron Graham23_rgraham@CDBSOFTWARE.COM30_Thu, 6 Jul 2000 15:59:52 -0500391_ISO-8859-1 If you are one DB2 V6 the limit is 254 data sets for a LARGE NPI.

Ron

-----Original Message----- From: Ron Graham Sent: Thursday, July 06, 2000 2:57 PM To: 'DB2-L' Subject: RE: Piecesize Limit: 32 datasets

If the tablespace is defined as LARGE you should be able to get up to 128 data sets. I have seen a NPI with 91 data sets, so I know it is possible. [...] 9038 32 43_Current thoughts on table/plan/package auth11_Cliff Boley32_Maurice.C.BOLEY@ODOT.STATE.OR.US30_Thu, 6 Jul 2000 14:05:55 -0700347_iso-8859-1 Hi all,

I'm curious as to what the current thoughts are and what others are doing concerning granting access to tables (via static SQL).

One of my departments has created some new tables and a there will be a new group of users (and old users) that will be accessing the new tables along with some of the old tables. [...] 9071 28 84_August 9th - Performance Estimation Class - Tim Lenahan - Central PA DB2 Users Group10_Cathy Peck23_cathy.peck@HIGHMARK.COM30_Thu, 6 Jul 2000 17:33:13 -0400389_us-ascii If you are interested in Performance Estimation, you might want to consider this class!

Tim Lenahan is coming to Central PA to teach it.

Check out the details on Tim & the course on the Central PA DB2 Users Group webpage at: http://www.db2parug.org

Having Tim teach you Performance Estimation for an entire day for $100 per person is an absolute steal!! [...] 9100 38 18_RC on package bind10_Mary Olson19_Mary.Olson@WCOM.COM30_Thu, 6 Jul 2000 17:39:20 -0700336_iso-8859-1 Greetings everyone

I am getting a RC 4 when I bind a package, and I'm trying to determine if that is acceptable, or if there is something I can do to eliminate this rc 4. Maybe you can help!

Our applications are set up in a way that programs run in one subsystem, and access a table on a remote subsystem. [...] 9139 122 32_Re: Piecesize Limit: 32 datasets0_19_Tim.Lowe@STPAUL.COM30_Thu, 6 Jul 2000 16:44:32 -0500346_us-ascii For DB2 version 5, this limit is documented in InfoApar II11393, "CHANGES TO DB2 ADMINISTRATION GUIDE VOL. 2 SC26-8957-00 THAT DID NOT MAKE DB2 R510 GA PUBS":

Remember that 32 pieces is the limit if the underlying tablespace is not defined as LARGE and that 128 is the limit if the underlying tablespace is defined as LARGE. [...] 9262 86 32_Re: Log Archival Stalls - OS/39019_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM30_Thu, 6 Jul 2000 09:53:02 -0500561_iso-8859-1 Hi Ken, I've never encountered an archive "stall." If all logs are full DB2 suspends all activity until an active log is archived. I worked in a shop with extremely high logging rates; filled 600 cyl active logs every 1 - 2 minutes. We used DASD for the archive logs and then a IEBCOPY job to tape. If the IEBCOPY was successful, we deleted the DASD log and catalogued the tape. We had 12 active log pairs but only about 3/4 that amount in the archive "pool." The IEBCOPY job was triggered by the 'archive log successful' messages with OPS/MVS. [...] 9349 24 26_Re: Migrating DB2 V5 to V612_Roger Miller19_millerrl@US.IBM.COM30_Thu, 6 Jul 2000 17:16:03 -0500371_- Here is a slight twist on what you wrote.

There are several possibilities for failures. When you run the precompiler on source that is missing the colon, you can get an error. There is no complete and general technique for finding all of your source.

The other problem is binding with an old DBRM (prior to V2.3). That's what the REXX exec can find. [...] 9374 112 32_Re: Log Archival Stalls - OS/39014_Roy R Reynolds20_Roy_Reynolds@GAP.COM30_Thu, 6 Jul 2000 15:51:00 -0700573_us-ascii Consider archiving to DASD then let HSM migrate it to ML2. Configure HSM to migrate 'ARCHLOG*' at whatever span of time you want. If you recover a lot you might want to keep them on DASD for an hour or two. Here is what was neat....--> A particularly savvy sysprog added an OPS/MVS-type routine to watch for the DB2 log message about the last DB2 log being 10% full and, when encountered, triggered an aggressive HSM migration of archlogs to free up DASD for waiting archive jobs. Initially, I was skeptical (DBA trait), but it worked great. Best thing since [...] 9487 89 22_FW: RC on package bind14_James Campbell29_James.Campbell@HANCORP.COM.AU30_Fri, 7 Jul 2000 15:07:28 +1000539_iso-8859-1 Mary,

Why are you binding this package locally? If all the program does locally is a CONNECT to the remote location, then it doesn't need to be bound locally. There is a table in the Appl Prog and SQL Guide in 'Bind the Application' which has the comment "You do not have to bind a DBRM whose SQL statements all come from this list:" It should really be headed "You do not have to bind a DBRM at a location where the only SQL statements executed at that location come from this list:" - which would be a little more [...] 9577 18 20_IMS to DB2 Migration17_Vijay Subramanyam33_vijay.subramanyam@HPSBLR.SOFT.NET30_Fri, 7 Jul 2000 10:38:16 +0500388_- Hello All,

Can anybody throw some light on IMS to DB2 database migration. What are the implications,problems etc...

Regards, VIJAY

================================================ 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. 9596 74 13_Re: Triggers.14_James Campbell29_James.Campbell@HANCORP.COM.AU30_Fri, 7 Jul 2000 15:15:31 +1000381_iso-8859-1 The logs are recording the actual updates to the tablespace pages - ie after any editprocs, compression etc.

You ___could___ decode the logs by extracting the compression dictionary and either - get the prior IC output, and manually apply the redo records; or - get the current page data and manually apply the undo records decompressing each row as you go. [...] 9671 141 9_FW: Row X14_Purcell, Terry30_Terry.Purcell@TEAM.TELSTRA.COM30_Fri, 7 Jul 2000 15:55:45 +1000421_iso-8859-1 If you wish to retrieve the 1 million'th or 2 million'th row......Do not use this technique!!!!!!

The correlated subquery will be executed 1 million times (or 2 million), and each time it will add 1 more to the count; ie. COUNT(*) = 1,2,3,4,5...............999997,999998,999999,1000000. I'm told the formula for this is ((N+1) * N) / 2), which equates to 500,000,500,000 rows read by the subquery. [...] 9813 167 9_Re: Row X7_Sanjeev19_Sanjeev@INTERRA.COM30_Fri, 7 Jul 2000 12:13:17 +0530519_iso-8859-1 Hi Terry, I agree with you that the number of touches will be very high in the query and i also believe that using a cursor and fetching it for the same number of time should be a cheaper method.What i am concerned with if at all DB2 does the sort and materializes the result in the OPEN cursor time then it might be costly and you cann't also use the OPTIMIZE FOR clause which might require some heavy random I/Os for doing this job. So what i can say is that it should be better checked by using both [...] 9981 38 24_Re: IMS to DB2 Migration32_Cuneyt Goksu (Garanti Teknoloji)22_CuneytG@GARANTI.COM.TR30_Fri, 7 Jul 2000 09:44:59 +0300474_- Try to use DL/2 Product. It allows you to use DL/I pgms access to DB2 database w/o any modification.

It's developed by Circle Sowtware...

Cuneyt GOKSU GARANTİ TEKNOLOJİ Istanbul / TURKEY Tel : 0090-212-657 0404 Ext: 1218 mailto:Cuneytg@garanti.com.tr





-----Original Message----- From: Vijay Subramanyam [mailto:vijay.subramanyam@HPSBLR.SOFT.NET] Sent: Friday, July 07, 2000 8:38 AM To: DB2-L@RYCI.COM Subject: IMS to DB2 Migration [...] 10020 19 44_Pamela Harris/IT/SEARS is out of the office.0_17_pharri1@SEARS.COM30_Fri, 7 Jul 2000 02:16:53 -0500362_us-ascii I will be out of the office from 07/06/2000 until 07/10/2000.

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 or Ginutis Matutis.

DB2 pin # 7270526, Informix pin # 7270525, Bill Pugh pin # 7270063, Ginutis Matutis pin # 7270300. [...] 10040 20 41_Piecesize, pqty, and sqty choices for NPI0_22_KMulcahy@CHECKFREE.COM30_Fri, 7 Jul 2000 03:26:57 -0400379_us-ascii A non-partitioned index on a LARGE partitioned table is 20 GB and growing. Thoughts on choices for PRIQTY, SECQTY, and PIECESIZE? Please copy me so your response arrives sooner.

And safer! The discussion list emails are getting truncated. If anyone has recent discussions on piecesize I would appreciate the info! This NPI is getting pieced on this weekend! [...] 10061 27 50_Your Request to join the DB2-L mailing List -Reply37_Rob Edlin - White (Rob Edlin - White)28_rob.edlin-white@EXPERIAN.COM30_Fri, 7 Jul 2000 10:27:36 +0000523_- John (& list users),

thank you for the information about how to subscribe. I have subscribed now and find some of the contributions useful and informative.

Would it be possible for the list software to be configured so that the subject field in any mail sent via the list gets prefixed with "(DB2-L)" or something of the kind? Another list I subscribe to (to do with croquet!) does this, and it makes it much easier to distinguish which mail comes from where. I feel this would be a great enhancement. [...] 10089 68 69_Re: Tools for comparing table contents before and after test scri pts7_Jim Lee25_JLee@PRINCETONSOFTECH.COM30_Fri, 7 Jul 2000 06:42:38 -0400582_- Hugh

I work for a software vendor, Princeton Softech, that sells a DB2 relational Compare facility, Compare for DB2. It provides the ability to do a "relational" compare, which compares a referential subset of data in a single step. It will analyze the data following all relationships of dependents to N-levels and comparing all related data at one time. Changes for all rows are identified as well as related changes which will identify the parents of the changed data. Online viewing allows you to "join" from the parent to a dependent to view the related changes. [...] 10158 41 54_Re: Your Request to join the DB2-L mailing List -Reply32_Cuneyt Goksu (Garanti Teknoloji)22_CuneytG@GARANTI.COM.TR30_Fri, 7 Jul 2000 13:38:04 +0300405_- I had the same problem. You can set rules for your E-mail program, for example Outlook, so that it can distinguish your emails and put into different folders.



-----Original Message----- From: Rob Edlin - White (Rob Edlin - White) [mailto:rob.edlin-white@EXPERIAN.COM] Sent: Friday, July 07, 2000 1:28 PM To: DB2-L@RYCI.COM Subject: Your Request to join the DB2-L mailing List -Reply [...] 10200 68 54_Re: Your Request to join the DB2-L mailing List -Reply0_19_mike.holmans@BT.COM30_Fri, 7 Jul 2000 11:51:59 +0100637_- There's actually an option you can set in your subscription. Use the form on the website, or send a listserv command to the command address specifying SUBJECTHDR.

Works for me.

Mike Holmans BT ISE Technical Design mike.holmans@bt.com

> -----Original Message----- > From: Cuneyt Goksu (Garanti Teknoloji) [SMTP:CuneytG@GARANTI.COM.TR] > Sent: Friday, July 07, 2000 11:38 AM > To: DB2-L@RYCI.COM > Subject: Re: [DB2-L] Your Request to join the DB2-L mailing List > -Reply > > I had the same problem. > You can set rules for your E-mail program, for example Outlook, so that it > can distinguish your emails and [...] 10269 353 16_Re: Primary keys12_Randy Wilson22_Randy_Wilson@BCBST.COM30_Fri, 7 Jul 2000 08:18:28 -0400373_iso-8859-1 uhmm... call me a little embarrassed :) this table wasn't in my runstat script, which I had 'assumed' it was.... running runstats on the table fixed it all. Thanx!!! Randy

-----Original Message----- From: Lynne Flatley [mailto:LFlatley@NEFN.COM] Sent: Thursday, July 06, 2000 3:14 PM To: DB2-L@RYCI.COM Subject: Re: Primary keys Sensitivity: Private [...] 10623 21 25_DB2 UDB V6 or V7 Monitors0_25_Baranowski.P@GRAINGER.COM30_Fri, 7 Jul 2000 07:37:00 -0500331_- Anyone recommendations on a monitor for UDB V6 or V7 running on UNIX that can execute a predefined script when an event has occured?

And here is the kicker:

All this while I'm at home trying to get a good nights sleep.



I am aware of CA's ManageIT product but I need it NOW! No vaporware please. [...] 10645 93 47_Re: Current thoughts on table/plan/package auth41_=?iso-8859-1?Q?Cl=E1udio_Meluzzi_Mendes?=18_P000002@BEG.COM.BR30_Fri, 7 Jul 2000 10:02:53 -0300663_iso-8859-1 I would say the standard "it depends". I believe the best is to discuss the appropriate policy with your DBAs and system analysts/programmers, in order to define a standard which supply security needs of your company. In our case, grants are as follows:

Analyst/programmers cannot execute DDL nor add a new plan/package. This task is only for DBAs, which can carefully maintain data integrity (at least better than programmers would do). Analyst/programmers cannot modify production data as well. Each project is analysed following a methodology and tables are created after project approval. A centralised approach would be better for the [...] 10739 109 24_Re: IMS to DB2 Migration50_Knight, Robert A. (Cameron of Pittsburgh at Alcoa)23_Robert.Knight@ALCOA.COM30_Fri, 7 Jul 2000 09:46:12 -0400321_-

I have done this on occassions. For both HDAM & HIDAM databases. I have heard a number of presentations in the late 80's and early 90's on the best approach. But I found that in most cases it was pure speculation. Each case is somewhat different.

Are you planning to use case tools? Bachman, Erwin? [...] 10849 222 32_Re: Log Archival Stalls - OS/3900_26_ken.george@EDWARDJONES.COM30_Fri, 7 Jul 2000 09:02:55 -0500581_US-ASCII Thanks all who responded with helpful suggestions and hints, I appreciate them. I would like to answer some of the questions raised by the responses and provide more information about our environment. We have 16 active logs, each at 830 cylinders. We archive direct to dasd and then copy to tape for local and disaster backups. When the log archive process stalls, we have plenty of room on the archive volumes for the archives to take place. During the height of activity, we are filling active logs at the average rate of one every 2 minutes. Our output log buffer [...] 11072 31 24_Help needed with problem13_Shauna Hadden24_shauna.hadden@BCBSKS.COM30_Fri, 7 Jul 2000 08:48:14 -0500316_- We have a really strange thing going on. Many of our distributed DB2 applications (OS/390 - both V5.1 and V6) are experiencing the same problems. We are using DB2 Connect PE v6.1 with FP2. Our applications are written in both Access and VB. This is happening in three DB2 OS/390 Subsystems (or so I am told). [...] 11104 80 24_Re: IMS to DB2 Migration0_18_damcon2@US.IBM.COM30_Fri, 7 Jul 2000 10:00:44 -0400302_us-ascii Vijay,

We had an IMS to DB2 effort about 2 years ago. There were some folks that said month-end processing will never complete in a sufficient amount of time because DB2 isn't as fast as IMS. Well, month-end processing time was cut in half. It went from taking 3 days to just 1.5. [...] 11185 118 28_Re: Help needed with problem13_Mackey, Glenn21_GMackey@GUIDEMAIL.COM30_Fri, 7 Jul 2000 09:23:43 -0500419_iso-8859-1 Shauna, I have no solution, but our site has had similar problems. Here is a post I sent to this group a few weeks back. You can search the archives on "DB2 Connect Error: SQL30081N", Hugh Beighton-Dykes gave a reply.

I have some questions for you.

1. Do you get these error intermittently? (We do) 2. Can you reproduce the errors. (We cannot) 3. Have these errors just started to happen? [...] 11304 86 28_Re: Help needed with problem12_Myron Miller22_myronwmiller@YAHOO.COM30_Fri, 7 Jul 2000 07:29:41 -0700340_us-ascii My client had these messages occur last year and it was a real chore chasing down the problems. From everything we could tell, it was a timing issue from DB2 writing to the client. We eventually resolved the issues but it was a real shotgun approach. We did a number of things, anyone of which could have been the actual fix. [...] 11391 26 28_Re: Help needed with problem14_Philip Gunning18_pgunning@PHEAA.ORG30_Fri, 7 Jul 2000 10:31:06 -0400550_us-ascii Shauna, I had this problem at another location. Check the timeout parameter set in VB. We saw this as a cause at one time. Also, when running DB2 Connect on NT, with interrupt enabled, which is the default when you catalog the database, this allows the client program to generate an interrupt and DB2 connect detects this and drops the connection. This is further explained in the DB2 Connect users guide which I would recommend that you review when you get a chance as it covers a lot on performance and some common problems. HTH Phil [...] 11418 73 28_Re: Help needed with problem13_Shapiro, Dave26_Shapiro.Dave@PRINCIPAL.COM30_Fri, 7 Jul 2000 09:49:04 -0500544_iso-8859-1 Shauna,

We have seen the 10053 messages. The 10053 code signifies: "The virtual circuit was aborted due to timeout or other failure. The network has dropped the connection on reset."

In VB code that does not specify a connection timeout value the default is 30 seconds. If the code is specifying a value, which would seem to be sufficient, it may not be in isolated cases. The host query time can be the culprit, but network traffic delays can as well (not to mention the 'other' category from the 10053 message). [...] 11492 58 28_Re: Help needed with problem15_Tjader, Polly A20_patjader@EASTMAN.COM30_Fri, 7 Jul 2000 11:05:52 -0400420_- We have had this problem, also. I believe this information will help us.

But I have a different twist on the TCP/IP problem. This is the error we have been getting on our e-business site only since 6/30. The application does not work at all. Before 6/30 it was fine. All other applications , both web based and client/server, are working fine except for the intermittent problem that was described before. [...] 11551 69 20_Re: Datajoiner & DB222_Ramesh Balasubramanyan15_RAMBAL@SAIF.COM30_Fri, 7 Jul 2000 08:04:51 -0700573_iso-8859-1 Stephen, Datjoiner is a product that is used to provide SQL access from Mainfarme to disparate RDBMS platforms like Oracle and MS SQLSERVER. It can also be used by Oracle forms (with CAE) to access DB2 data. It is like many gateways in one, a one stop shop for all your data. But if you are only using for DB2, then it doesn't seem like a good idea. You probably need only DB2 connect enterprise edition. W evaluated Datajoiner extensivley for our shop. We have DB2, Oracle and SQLSERVER and seems like a good product for integrating these RDBMS's. Regards [...] 11621 50 28_Re: Help needed with problem10_Park, Stan17_SPark@STATE.NM.US30_Fri, 7 Jul 2000 09:30:29 -0600671_iso-8859-1 Check out (if you haven't already) the IBM DB2UDB Troubleshooting Guide (I use V7) db2p0e70.pdf It has some great suggestions for connectivity problems!!

-----Original Message----- From: Shauna Hadden [mailto:shauna.hadden@BCBSKS.COM] Sent: Friday, July 07, 2000 7:48 AM To: DB2-L@RYCI.COM Subject: Help needed with problem



We have a really strange thing going on. Many of our distributed DB2 applications (OS/390 - both V5.1 and V6) are experiencing the same problems. We are using DB2 Connect PE v6.1 with FP2. Our applications are written in both Access and VB. This is happening in three DB2 OS/390 Subsystems (or so I am told). [...] 11672 69 28_AW: Help needed with problem16_Roland Schiradin28_SchiradinR@ALTE-LEIPZIGER.DE30_Fri, 7 Jul 2000 17:56:34 +0200630_iso-8859-1 RC=140 means EPIPE The pipe is broken.

RSN 7242 means JRConnTCBNotFound A TCB was not found for the specified socket. Action: Use a valid socket descriptor.

I suggest to run a TCPI socket trace roland

> -----Ursprüngliche Nachricht----- > Von: Shauna Hadden [SMTP:shauna.hadden@BCBSKS.COM] > Gesendet am: Freitag, 7. Juli 2000 15:48 > An: DB2-L@RYCI.COM > Betreff: Help needed with problem > > We have a really strange thing going on. Many of our distributed DB2 > applications (OS/390 - both V5.1 and V6) are experiencing the same > problems. > We are using DB2 Connect PE v6.1 with FP2. Our [...] 11742 17 29_SQLVAR length in DB2/OS390 V616_JOHN G. MATTHEWS26_HC#JGM@MVS.MEDCTR.UCLA.EDU30_Fri, 7 Jul 2000 09:06:00 -0700409_- When the V4 DSNHPC generates assembly code for a SELECT stmt, there are multiple 12 bytes entries(1 per column) generated in SQLDSECT. The V6 DSNHPC generates multiple 44 byte entries. I think these are the SQLVAR entries. The V4 DB2 manual and V6 manual both indicate a 44 byte length. However, the V6 code is causing problems because too many base registers are needed for a table with many columns. [...] 11760 22 30_DB2 OS/390 V5 TEMPORARY TABLES12_HARBRY ARIZA20_harbry_a@HOTMAIL.COM28_Fri, 7 Jul 2000 09:17:32 PDT361_- Hi Folks

Here I come with a quick question. I have a doubt and I want to clarify it. I have an user that requiere create a Temporary table which needs to be reference by several users. I told him to create it and then give the necesary autority to access it. Is it anything else that I have to do about it or it is all that the user needs. Thanks [...] 11783 34 33_Re: SQLVAR length in DB2/OS390 V620_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM30_Fri, 7 Jul 2000 12:17:47 -0400652_us-ascii My guess is that the change is to accommodate for future 30 bytes long column names.

-----Original Message----- From: JOHN G. MATTHEWS [SMTP:HC#JGM@MVS.MEDCTR.UCLA.EDU] Sent: Friday, July 07, 2000 12:06 PM To: DB2-L@RYCI.COM Subject: SQLVAR length in DB2/OS390 V6

When the V4 DSNHPC generates assembly code for a SELECT stmt, there are multiple 12 bytes entries(1 per column) generated in SQLDSECT. The V6 DSNHPC generates multiple 44 byte entries. I think these are the SQLVAR entries. The V4 DB2 manual and V6 manual both indicate a 44 byte length. However, the V6 code is causing problems because too many base registers [...] 11818 16 27_Re: Waiting on Dynamic Bind12_Roger Miller19_millerrl@US.IBM.COM30_Fri, 7 Jul 2000 11:42:47 -0500331_- In addition to the other suggestions, please check that your IRLM is up to current service level. Quite a few of the changes that were made to improve scaling for the IRLM could affect this. If you are running goal mode, then you will need to look at the latest version of the books in soft copy to get the proper settings. [...] 11835 19 20_Re: Datajoiner & DB212_Roger Miller19_millerrl@US.IBM.COM30_Fri, 7 Jul 2000 11:52:35 -0500394_- Here are a couple of thoughts:

The latest DataJoiner to ship is part of DB2 Version 7 for Unix, WIndows, OS/2. You might find some relief in the optimizer improvements.

If you are passing through an outer join to DB2 for OS/390, then you may find that you need the performance improvements in Version 6, or some of those that moved back to V5 (but require a zparm change). [...] 11855 36 28_Re: Help needed with problem13_Shauna Hadden24_shauna.hadden@BCBSKS.COM30_Fri, 7 Jul 2000 11:57:38 -0500610_- On Fri, 7 Jul 2000 09:23:43 -0500, Mackey, Glenn wrote:

>Shauna, >I have no solution, but our site has had similar problems. Here is a post I >sent to this group a few weeks back. You can search the archives on "DB2 >Connect Error: SQL30081N", Hugh Beighton-Dykes gave a reply. > >I have some questions for you. > >1. Do you get these error intermittently? (We do) >2. Can you reproduce the errors. (We cannot) >3. Have these errors just started to happen? > Answers: These problems just started happening yesterday. These applications have been running in production for [...] 11892 226 32_Re: Log Archival Stalls - OS/39022_Piontkowski Michael ML35_michael.piontkowski@ASTRAZENECA.COM30_Fri, 7 Jul 2000 13:13:52 -0400346_iso-8859-1 It's my experience that the archival of an active log is not delayed because a UOW spans more that one active log.

Here's a possibility:

What happens when a UOW must be rolled back while the log records required for the rollback have been moved (or are they copied) to an active log data set that is being archived? [...] 12119 48 29_Re: DB2 UDB V6 or V7 Monitors22_Piontkowski Michael ML35_michael.piontkowski@ASTRAZENECA.COM30_Fri, 7 Jul 2000 13:31:01 -0400662_- Maybe Database-GUYS products can help. They specialize in DB2 UDB on UNIX.

http://www.databaseguys.com/



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

> ---------- > From: Baranowski.P@GRAINGER.COM[SMTP:Baranowski.P@GRAINGER.COM] > Sent: Friday, July 07, 2000 8:37 AM > To: DB2-L@RYCI.COM > Subject: [DB2-L] DB2 UDB V6 or V7 Monitors > > Anyone recommendations on a monitor for UDB V6 or V7 running on UNIX that > can > execute a predefined script when an event has occured? > > And here is the kicker: > > All this while I'm at home trying to get a good nights sleep. > > > I am aware of [...] 12168 45 34_Re: DB2 OS/390 V5 TEMPORARY TABLES20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM30_Fri, 7 Jul 2000 13:18:19 -0400422_us-ascii Looks ok as far as user is concerned. Check for sufficient work tablespace (DSNDB07) priqty, especially if they are using protected threads from CICS. The space used by temporary tables is not released until the thread terminates.

-----Original Message----- From: HARBRY ARIZA [SMTP:harbry_a@HOTMAIL.COM] Sent: Friday, July 07, 2000 12:18 PM To: DB2-L@RYCI.COM Subject: DB2 OS/390 V5 TEMPORARY TABLES [...] 12214 40 16_Re: Primary keys21_Helio Pimpinato Rocha20_hrocha@TELESP.COM.BR30_Fri, 7 Jul 2000 14:59:15 -0300111_us-ascii Hi

Did you run the runstats and bind plan again?.

Helio

Telefonica/Brasil

12255 47 34_Re: DB2 OS/390 V5 TEMPORARY TABLES13_OConner, John26_John.OConner@COURTS.WA.GOV30_Fri, 7 Jul 2000 11:28:08 -0700369_iso-8859-1 Harbry

It's not clear whether your "several users" need to share that data in the Temporary table. They won't be able to do that. As you probably already know, it will appear to them as though they each have their own table.

John O'Conner DB2 DBA Office of the administrator for the courts (360) 705-5210 Mailto:john.oconner@courts.wa.gov [...] 12303 96 38_Re: DB2 Developer's Guide - V6 Edition12_Gene Tilelli16_geetee52@USA.NET30_Fri, 7 Jul 2000 16:00:31 -0400438_iso-8859-1 Hi All...

I ordered the book and I'm glad I did so. Terrific examples in it...nicely organized and worded, etc.

However for the benefit of anyone else that may be purchasing it...take note. I purchased it from Barnes and Noble...and it does not come with a CD despite the fact that it clearly states that there is one in the description. I've cut and pasted the description directly from the B&N site below. [...] 12400 118 38_Re: DB2 Developer's Guide - V6 Edition11_Ray H Janes27_ray_h_janes@EMAIL.MOBIL.COM30_Fri, 7 Jul 2000 15:08:31 -0500434_us-ascii I noticed that also. But the one with the CD is the old third edition and cost more. I guess the CD is not available for the fourth edition.







Gene Tilelli on 07/07/2000 03:00:31 PM

Please respond to DB2 Data Base Discussion List

To: DB2-L@RYCI.COM cc: (bcc: Ray H Janes/C/Dallas/Mobil-Notes) Subject: Re: DB2 Developer's Guide - V6 Edition [...] 12519 264 32_Re: Log Archival Stalls - OS/39019_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM30_Fri, 7 Jul 2000 14:59:05 -0500462_iso-8859-1 Hi Michael, Ken, Michael, good idea. Ken, what is the DISP for the archive dataset in your job that moves an archive from DASD to tape? How often are you checkpointing during this peak period. I don't know if checkpointing suspends archiving -- it could be you may need to increase LOGLOAD. I wish you'd check closely to ensure that logging and archiving isn't happening on a single volume at the same time; it would certainly cause a slow down. [...] 12784 24 39_Porting Net.Data macros between systems11_Joe Luthman22_jluthma@BGNET.BGSU.EDU30_Fri, 7 Jul 2000 16:27:34 -0400473_us-ascii I work at a university that historically has 6-12 month spending freezes about every 6 years. To accommodate this, I'm preparing an internal marketing campaign that our web presentation of MVS data should include three different approaches.

1) Crossplex, a product they are just testing now. It presents CICS screens to the WEB. This is currently the favored method, probably because no applications changes need to be made to our mostly non-DB2 shop. [...] 12809 37 29_DSN1COPYing Image copy files.12_MARTIN WOLFF31_martin_wolff@GLOBALCROSSING.COM30_Fri, 7 Jul 2000 16:27:54 -0400559_us-ascii Hi all, hopefully this is a quick one.

I recently had a requirement to make a database copy of a table that had some rows erronously deleted so the application people could work out what had gone. I created a shadow copy of the table and selected the image copy file that fitted the time best. I then did a DSN1COPY of the image copy to the shadow table (with all the OBIDXLAT stuff) but for some reason the DSN1COPY ran and ran until it eventually failed with a max secondaries defined failure (ie no DB2 initiated error). Regardless of [...] 12847 79 68_Re: Tools for comparing table contents before and after test scripts5_Riyaz15_RMomin@FHSC.COM30_Fri, 7 Jul 2000 16:26:48 -0400611_US-ASCII Hugh I use Princeton's Compare for DB2 everyday and it's a great tool to comapre DB2 Tables. Good Luck, Riyaz. Hugh

I work for a software vendor, Princeton Softech, that sells a DB2 relational Compare facility, Compare for DB2. It provides the ability to do a "relational" compare, which compares a referential subset of data in a single step. It will analyze the data following all relationships of dependents to N-levels and comparing all related data at one time. Changes for all rows are identified as well as related changes which will identify the parents of the changed data. Online [...] 12927 70 33_Re: DSN1COPYing Image copy files.20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM30_Fri, 7 Jul 2000 16:45:45 -0400506_us-ascii DSN1COPY should be done with following parameters Inline copy, which gave you trouble -- Use "INLCOPY" keyword for the DSN1COPY parameters Usual copy, which worked ok -- You must have used "FULLCOPY" keyword.

If these parameters are not specified correctly then you could see absurd errors.



-----Original Message----- From: MARTIN WOLFF [SMTP:martin_wolff@GLOBALCROSSING.COM] Sent: Friday, July 07, 2000 4:28 PM To: DB2-L@RYCI.COM Subject: DSN1COPYing Image copy files. [...] 12998 94 33_Re: DSN1COPYing Image copy files.15_Di Carlo, Donna23_Donna_Bermender@BMC.COM30_Fri, 7 Jul 2000 15:58:10 -0500355_iso-8859-1 An inline copy can contain duplicate pages. This can greatly increase the image copy size for a SHRLEVEL CHANGE REORG, depending on the number of log records encountered during the log phase. Without the INLCOPY parameter, DSN1COPY probably does not have the intelligence to apply only the last page encountered, as the COPY utility does. [...] 13093 60 34_Re: DB2 OS/390 V5 TEMPORARY TABLES14_Richard Yevich21_ryevich@ATTGLOBAL.NET30_Fri, 7 Jul 2000 19:03:05 -0400437_iso-8859-1 Harbry,

Global temporary tables or Declared temporary tables cannot be shared. They are unique to each individual thread.

For what you require, you will have to create a normal table to be used as a shared temporary table.

Hope this helps, Richard Yevich +======+======+======+ Richard_Yevich@attglobal.net www.YLAssoc.com -- DB2 Performance Journal www.ryci.com -- Knowledge Base www.DB2-Portal.com [...]