1 WWW.IDUGDB2-L.ORG /home/listserv/home/db2-l February 2000, week 2 2 76 25_Re: regarding datasharing22_Leslie Pendlebury-Bowe43_Leslie.Pendlebury-Bowe@CCMAIL.EU.SONY.CO.JP30_Tue, 8 Feb 2000 06:19:10 +0000351_ISO-8859-1 Ravi where do you start ... this subject is big....big....BIG...

I am sure you will get many responses but I would suggest that if you do not already have a copy you get a copy of Craig Mullins book (DB2 Developers Guide) - from Amazon.com is as good a place as any. This will explain just about everything you need to know ... [...] 79 13 19_Re: Non-DB2 Utility14_Tom Decabooter27_Tom.Decabooter@HONDA-EU.COM30_Tue, 8 Feb 2000 08:50:44 +0100353_us-ascii Why not trying to ZAP the coding that does the date checking so that the check is no longer performed.

================================================ 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. 93 26 4_Hi !0_24_ssethi@LOT.TATASTEEL.COM28_Tue, 8 Feb 2000 08:33:58 GMT368_us-ascii Hi Kartik, How are u ? No mail or fone since a long time.I have applied in an US organisation for DBA for Os390 but they told me that their company is not providing the H1B. I getting quite a lot of response from here and there.Here i am in DB2 training in the last phase.It was a nice exp. to get that training .Its really nice.I got many informations. [...] 120 15 0_0_24_ssethi@LOT.TATASTEEL.COM28_Tue, 8 Feb 2000 08:39:51 GMT376_us-ascii I am really very sorry about sending personal mail by mistake to the list.

Sorry for the inconvenience.

Sanjeev

================================================ 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. 136 24 17_Special Registers20_Baskaran Subramaniam27_subramaniam.baskaran@DB.COM30_Tue, 8 Feb 2000 13:42:38 +0530326_us-ascii HI All

Is there any web site where I can get the complete set of special registers (like current server,etc.,)? or if somebody has a list of special registers, could they mail it to me.

Also in which catalog table these special registers will be stored? If it is not stored anywhere? why it is so? [...] 161 114 42_Re: Program tests in emergency situations?13_McDonald, Ian32_Ian.McDonald@ABBEYNATIONAL.CO.UK30_Tue, 8 Feb 2000 09:32:00 -0000442_iso-8859-1 Peter I think there may be more than one problem if you continue down the path you have already. I will assume you have no checkpoint or restart ability in your systems at present.

1. If a program abends in DB2 it rolls back. I always assume it takes twice as long to roll back as it does to 'do' anything, i.e. if a job has run for 1 hour it will take 2 hours to roll back. Pessimistic or what? No, bitter experience. [...] 276 314 58_Re: db2 os/390 v5.1 - Escalation of Incr Copy to full copy0_18_mebert@AMADEUS.NET30_Tue, 8 Feb 2000 10:48:46 +0100561_us-ascii The basic problem here is that the file sizes for the incremental image copies are too small. You should generally use a DD SPACE specification like SPACE=(CYL,(250,50),RLSE) where the primary quantity should be at least as large as the largest file size you expect. It is better not to have the step start at all, because of insufficient space, than having it abend later. If you use a primary size that is too small, the file might get allocated on a disk that doesn't have enough room for growth, even if there is enough space on another disk. [...] 591 48 30_Can we use CAF for two plans ?27_Thakral, Deepak Kumar (CTS)24_TDeepak@CHN.CTS-CORP.COM30_Tue, 8 Feb 2000 05:24:15 -0500373_iso-8859-1 Hi !

In my COBOL batch environment for 0S/390 we are using DB2 V5 . We usually use Call Attach Facility (CAF) for executing DB2 subroutines from a calling program . This calling program(say AAAA) does not have any DB2 code (no SQL statements ) but the subroutine it calls is a DB2 subroutine (say BBBB). So when we execute the proc it 's like this : [...] 640 46 39_Naming standards for DB2 load libraries0_19_mike.holmans@BT.COM30_Tue, 8 Feb 2000 11:49:57 -0000271_- We are about to start setting up data sharing. We've got a nice clean LPAR which doesn't even have OS/390 on it yet, let alone DB2, and we're still able to fiddle about with naming standards before we get going after the IML of the box in a couple of weeks' time. [...] 687 15 35_DataPropagator Version 6 -- Capture12_Martin, Paul22_Paul.Martin@ECOLAB.COM30_Tue, 8 Feb 2000 06:34:21 -0600421_iso-8859-1 I'm having some problems getting Capture V6 up and running

I would like to take a poll -- any company have Capture V6 up and running? on DB2 for OS/390 V5 or V6?

================================================ 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. 703 82 43_Re: Naming standards for DB2 load libraries13_Ulrich Althen25_Ulrich.Althen@NATWEST.COM30_Tue, 8 Feb 2000 12:12:21 -0000629_- Mike,

have you considered using dataset name aliases ? Having the member specific alias pointing to a dataset which contains your current version of software. Put the new maintenance in a new dataset (cycle etc) and change the member specific alias.

Good luck Uli Althen

> ---------- > From: mike.holmans@BT.COM[SMTP:mike.holmans@BT.COM] > Reply To: DB2 Data Base Discussion List > Sent: 08 February 2000 11:49 > To: DB2-L@RYCI.COM > Subject: Naming standards for DB2 load libraries > > We are about to start setting up data sharing. We've got a nice clean LPAR > which doesn't even have OS/390 on it [...] 786 81 32_Re: Determining DB2 Object Sizes11_Hayden, Lee18_Lee_Hayden@CSX.COM30_Tue, 8 Feb 2000 08:32:23 -0500372_iso-8859-1 Chuck,

You can use a combination of rexx and Ispf services. Here is an example that uses the same ispf services that 3.4 uses. You can pass the same kind of dataset mask to it. (i.e. db26.dsndbd.dsndb06.dsn*)

If you run this in batch you must provide a minimum ispf batch jcl as documented in the ispf Dialog Developer's Guide and Reference. [...] 868 70 43_Re: Naming standards for DB2 load libraries0_23_Mike_Levine@TEKHELP.NET30_Tue, 8 Feb 2000 08:35:31 -0500516_us-ascii Hi Mike,

Another option to consider is having two sets of libraries per data sharing group. This would give you the ability to apply maintenance to one member at a time but not have to have a separate set of libraries per group. Following your naming standard this could become: SYSD0.DB2.grpname.xxxx.SDSNLOAD where xxxx could be PROD1/PROD2 for example. You can then apply maintenance to one set of libraries and stop and start one member at a time to point to the alternate set of libraries. [...] 939 23 30_Subject: CA/Platinum Education13_Thomas, Janis19_Janis.Thomas@CA.COM30_Tue, 8 Feb 2000 09:13:47 -0500628_iso-8859-1 Yes, information on DB2 classes can be found here: http://www.cai.com/education/descriptions/db2.htm

Janis Thomas DB2 Product Management

-----Original Message----- From: valeriet@US.IBM.COM [mailto:valeriet@US.IBM.COM] Sent: Monday, February 07, 2000 6:46 PM Subject: CA/Platinum Education

Does CA/Platinum still have its DB2 education program/offerings?

================================================ 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. 963 25 27_Restart/Recover from Reorg.9_Ted Pesta18_tpesta@AMQUEST.COM30_Tue, 8 Feb 2000 09:13:21 -0500510_iso-8859-1 Environ : OS/390, DB2 V5 We had a problem here with a Reorg. We got an I/O error on one of the tapes and the job crashed. When I displayed the utility status it showed it to be in the SORT phase. I checked the sysout and the RELOAD phase was completed successfully. I was told that we could just recover the indexes at this point on go on our merry way. Is this correct ? Are we going to have problems with this approach, can I check anything to ensure we don't have problems? I did a COUNT on [...] 989 53 57_Re: CA's Technology Forum for DB2 for OS/390 - World Tour13_Thomas, Janis19_Janis.Thomas@CA.COM30_Tue, 8 Feb 2000 09:14:55 -0500317_iso-8859-1 Rohn,

The coordinators of the DB2 Tech Forum can add shows/cities to the 2000 North American schedule if the need arises. If there is some interest in a seminar in Winnipeg (25-30 attendees would be mininum required), they can definely get one scheduled for possibly the end of July or August. [...] 1043 25 17_DB2 V6 migrations16_Amy C. Whitehead14_B06ACW@FDS.COM30_Tue, 8 Feb 2000 09:26:12 -0500582_us-ascii We have a test sub-system that is V6 & 2-way data sharing. Our production region is V5, non-datasharing. We are migrating our V5 applications into the V6 sub-system. (It is a long story why we took this approach.) Our problem is that the developers/end-users are concerned with the compatiablity of the V5 and V6 sub-systems during production migration. What V6 problems have you encountered? Did you migrate many applications from a V6 DB2 to a V5 DB2 and what kind of problems did you have? How long did you test V6 prior to moving V6 into production? DB2 is pretty [...] 1069 109 31_Re: Restart/Recover from Reorg.0_18_mebert@AMADEUS.NET30_Tue, 8 Feb 2000 15:29:14 +0100508_us-ascii Just do a RECOVER INDEX(ALL) TABLESPACE dbname.tsname and you should be all set (actually we should start calling this REBUILD, in preparation for V6...) You should also check the SORTKEYS option to do away with SYSUT1 and SORTOUT, and SORTDEVT/SORTNUM to get rid of SORTWKnn (your I/O error had to be on one of those, right?). This will improve performance. The same goes for SORTDATA. Don't use NOSYSREC except with SHRLEVEL CHANGE, or if you're really sure you always have a good imagecopy. [...] 1179 168 31_Re: Restart/Recover from Reorg.9_Ted Pesta18_tpesta@AMQUEST.COM30_Tue, 8 Feb 2000 09:36:10 -0500706_iso-8859-1 Thank you Dr. Ebert. I will research the suggestions you made.

Ted



-----Original Message----- From: mebert@AMADEUS.NET [mailto:mebert@AMADEUS.NET] Sent: Tuesday, February 08, 2000 9:29 AM To: DB2-L@RYCI.COM Subject: Re: Restart/Recover from Reorg.

Just do a RECOVER INDEX(ALL) TABLESPACE dbname.tsname and you should be all set (actually we should start calling this REBUILD, in preparation for V6...) You should also check the SORTKEYS option to do away with SYSUT1 and SORTOUT, and SORTDEVT/SORTNUM to get rid of SORTWKnn (your I/O error had to be on one of those, right?). This will improve performance. The same goes for SORTDATA. Don't use NOSYSREC except [...] 1348 117 43_Re: Naming standards for DB2 load libraries19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM30_Tue, 8 Feb 2000 08:43:14 -0600602_iso-8859-1 Mike, This item just recently came up on the list and an IBM ref. was cited indicating IBM's recommendations. Can't remember the ref.; you might check recent archives. 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 314-235-6854 and delete this message immediately from your [...] 1466 67 21_Re: DB2 V6 migrations11_John Wynton21_jwynton@THEMISINC.COM30_Tue, 8 Feb 2000 09:31:53 -0500522_windows-1252 Amy:

If you haven't already done so, please visit our web site for information on the following training classes (I'm assuming you are located in the USA, but didn't recognize the domain name - fds.com):

DB2 UDB for OS/390 V6 Migration & Implementation - http://www.themisinc.com/training/db2/db2v6.htm DB2 UDB for OS/390 Data Sharing - http://www.themisinc.com/training/db2/db2v6ds.htm DB2 UDB for OS/390 Data Sharing Recovery/Restart - http://www.themisinc.com/training/db2/db2v6ds_r.htm [...] 1534 36 40_Re: DB2 Connect Personal Edition and MTS13_Schluter John24_john.schluter@DATA.SJ.SE30_Tue, 8 Feb 2000 08:51:19 -0600387_- No, DB2 Connect Personal does not support MTS. To my knowledge, you have to use DB2 Connect Enterprise Edition.

The reason is that MTS support uses X-open two-phase commit handling and the Connect Enterprise Product works as a synchpoint logger for units- of-work where DB2 OS/390 i involved.

So, you have to use DB2 Connect Enterprise Edition as a gateway product. [...] 1571 91 34_Re: Can we use CAF for two plans ?14_Harvey Wachtel22_babybaby@UAPC.CUNY.EDU30_Tue, 8 Feb 2000 10:08:27 -0500579_us-ascii If AAAA is a COBOL program written in-house, I assume it must be interpreting the PARM on the EXEC statement. You'll have to change it to accept parameters for both connections. It's not clear from your description which program actually executes the connection calls. Either:

AAAA itself is calling DSNALI, in which case you have to modify it to close the first connection and open the second before calling CCCC, or AAAA is passing the parameters on to BBBB to call DSNALI, in which case you have to modify it to pass the second set of parameters to CCCC. [...] 1663 32 22_Db2 UDB Backup pending30_Szeszycki, Matthew (WTC/Lisle)22_mszeszycki@WALLACE.COM30_Tue, 8 Feb 2000 09:26:12 -0600521_ISO-8859-1 I have two UDB questions. UDB 5.2 fixpack 9 on NT

1) Can you turn off Backup Pending (equivalent to Repair in DB2/OS390)

2) I have a table that I need to increase the blob column size from 1 mg to 10 mg. 150000 rows I am planning on doing an export from the old table and load to the new table. (I dont have the space to create another tablespace at this time) The problem I have is, if I do an export without storing LOBs in seperate files the export truncates at 32700. If I do store LOBs [...] 1696 23 43_Adding a Column to a Table with an EDITPROC0_32_Nancy.J.CARLSON@ODOT.STATE.OR.US30_Tue, 8 Feb 2000 07:22:02 -0800330_iso-8859-1 Does anyone know if there is a way to add a column to a table with an EDITPROC on it without dropping the table?

A review of the EDITPROC shows that we do not need to modify it. The EDITPROC populates the value of a column based on the value in another column. Neither of these two columns are in an index. [...] 1720 71 34_Re: Can we use CAF for two plans ?11_rick creech18_ykcirc@HOTMAIL.COM28_Tue, 8 Feb 2000 07:37:30 PST598_- FYI, If you use packages, as long as the appropriate entry is in the pklist of the plan, it does not matter which plan you use. This is one of the many benefits of using packages. Regards, Rick Creech



>From: "Thakral, Deepak Kumar (CTS)" >Reply-To: DB2 Data Base Discussion List >To: DB2-L@RYCI.COM >Subject: Can we use CAF for two plans ? >Date: Tue, 8 Feb 2000 05:24:15 -0500 > >Hi ! > >In my COBOL batch environment for 0S/390 we are using DB2 V5 . We usually >use Call Attach Facility (CAF) for executing DB2 subroutines from a [...] 1792 22 26_db2 datasets > 119 extents10_Tom Taylor17_ttaylor@CHUBB.COM30_Tue, 8 Feb 2000 10:52:08 -0500303_us-ascii Hi all

Just came across what I think maybe a problem, correct me if I'm wrong. I have 2 db2 datasets , one is stogrouped and sms managed in 252 extents across multiple volumes. The second is stogrouped and non-sms managed on multiple volumes and in 184 extents. How could this be.? [...] 1815 46 30_Re: db2 datasets > 119 extents13_Martin Kenney25_martin.kenney@RAILINC.COM30_Tue, 8 Feb 2000 10:54:16 -0500689_us-ascii The allowable number of extents went from 119 to 252, I think I noticed this when we went to OS390 2.4 OR 2.5.







Tom Taylor on 02/08/2000 10:52:08 AM

Please respond to DB2 Data Base Discussion List

To: DB2-L@RYCI.COM cc: (bcc: Marty R Kenney/RICC/AAR)

Subject: db2 datasets > 119 extents







Hi all

Just came across what I think maybe a problem, correct me if I'm wrong. I have 2 db2 datasets , one is stogrouped and sms managed in 252 extents across multiple volumes. The second is stogrouped and non-sms managed on multiple volumes and in 184 extents. [...] 1862 37 30_Re: db2 datasets > 119 extents13_Scott Lindsey29_sflindsey@HIGHLIGHTS-CORP.COM30_Tue, 8 Feb 2000 10:46:44 -0500349_us-ascii When you have multiple volumes assigned either SMS or not and when the 119 extent limit is met, DB2 looks to one of the other volumes and takes the primary allocation and then continues to take secondary allocations until either the pack is full or the extent limit is reached again. HTH. Scott Lindsey Sr. DBA Highlights for Children [...] 1900 35 30_Re: db2 datasets > 119 extents14_Massimo Scarpa16_mscarpa@CESVE.IT30_Tue, 8 Feb 2000 17:06:49 +0100455_us-ascii If you have DFSMS v 1.4 (as seen in Red book sg244892):



- Five-extent limit in DADSM for new data set allocation and for extends to new volumes has been removed.

- Data sets can grow to either 59 volumes (which is the limit imposed for multivolume data sets by the task input/output table (TIOT)) or the number of volumes in the storage group (as long as a volume in the storage group has space), whichever is smaller. [...] 1936 41 13_SIGNOFF DB2-L15_Lentine Chuck O23_Chuck.O.Lentine@IRS.GOV30_Tue, 8 Feb 2000 08:42:24 -050028_iso-8859-1 > SIGNOFF DB2-L > 1978 363 34_Re: Can we use CAF for two plans ?27_Thakral, Deepak Kumar (CTS)24_TDeepak@CHN.CTS-CORP.COM30_Tue, 8 Feb 2000 22:10:53 +0530519_iso-8859-1 Hi Harvey,

The program AAAA is an inhouse program and yes, it does interpret the parm on the exec statement. The progam AAAA is the one which actually executes the connection call.

Now how do I close the first connection and open the second before calling CCCC. Again you are right, I don't need to open it concurrently ,but can I use two different plans and code the parm in such a way that according to the program logic it will establish a thread to either of them ??? Just wondering ! [...] 2342 202 19_Re: Non-DB2 Utility12_Hylton Tom P20_Tom.P.Hylton@IRS.GOV30_Tue, 8 Feb 2000 11:17:04 -0500400_iso-8859-1 Suggestions off the top of my head:

A) As someone said, ZAP around the broken piece. If it ain't working now, you surely have nothing to lose by trying it.

B) Re: PDS search May be able to find something that does your PDS info at http://www.cbttape.org or RYO using the SEARCH-FOR utility (Option 3.14). It can be executed in batch. Don't know the String limits on it. [...] 2545 186 58_Re: db2 os/390 v5.1 - Escalation of Incr Copy to full copy12_Hayes, Chris19_CHRIS.HAYES@SSA.GOV30_Tue, 8 Feb 2000 12:58:49 -0500387_iso-8859-1 Did you terminate the utility before rerunning? Since IBM will not allow an incremental to follow a terminated copy, it might have automajically issued a full copy.

-----Original Message----- From: acaldwe2@CSC.COM [mailto:acaldwe2@CSC.COM] Sent: Monday, February 07, 2000 7:09 PM To: DB2-L@RYCI.COM Subject: db2 os/390 v5.1 - Escalation of Incr Copy to full copy [...] 2732 26 16_No subject given15_Sherri Sterling23_ssterling@BPD.TREAS.GOV30_Tue, 8 Feb 2000 13:31:38 -0500421_US-ASCII We are on UDB for AIX version 6.1.

We have a character column defined in a DB2 table that we would like to force the first character to upper case and the remaining character in lower case. Can we define the table to ensure that the application does not violate this rule through an editproc, trigger, check constraint, etc.? Could we translate the data some way, to conform, on inserts and updates? [...] 2759 18 35_QUALIFIER option of BIND going away12_Greg Popwell27_greg.popwell@MOTION-IND.COM30_Tue, 8 Feb 2000 12:30:19 -0600483_iso-8859-1 I have been told that IBM plans at some point in the future to do away with the QUALIFIER option of the BIND command, at least for OS/390. Has anyone else heard this, or can anyone speak to this?

Thanks,

Greg Popwell

================================================ 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. 2778 48 39_Re: QUALIFIER option of BIND going away0_22_BILL_GALLAGHER@PHL.COM30_Tue, 8 Feb 2000 13:33:49 -0500301_us-ascii I sure hope that this is not the case, because our shop standards are to use unqualified SQL in all our batch programs, with the QUALIFIER parameter being used to resolve the qualifiers at bind time. I would have a lot of trouble believing that IBM would plan to take this away from us. [...] 2827 210 58_Re: db2 os/390 v5.1 - Escalation of Incr Copy to full copy18_Bennett, Darden A.24_Darden.A.Bennett@SSA.GOV30_Tue, 8 Feb 2000 13:29:41 -0500328_iso-8859-1 Chris,

Speaking of the imagecopy subject. When do you want to talk? Specifically re: RDARS as you indicated in yesterday's staff meeting. You have been busy as I have not seen you in your cube much, so I thought it would be best if you set a time.

Other than tomorrow's 10:30, my schedule is open. [...] 3038 35 39_Re: QUALIFIER option of BIND going away9_Rob Crane22_racrane@CONCENTRIC.NET30_Tue, 8 Feb 2000 11:50:52 -0700417_us-ascii It would seem to be a large task for them to undertake and counter productive. The use of mirror tables, set current package set, etc., all currently need the qualifier functionality of the bind statement. The largest percentage of application code running in production is still using static SQL and not coding the table owner in the individual sSQL statements, I don't see that changing anytime soon. [...] 3074 46 20_Re: No subject given9_Mark Ruhe15_mruhe@QUEST.COM30_Tue, 8 Feb 2000 10:56:52 -0800457_iso-8859-1 Sherri, This could be done using a stored procedure fired by insert and update triggers. The stored procedure would convert the data into the upper/lower case requirements.

Mark Ruhe Developer Quest Software

-----Original Message----- From: Sherri Sterling [mailto:ssterling@BPD.TREAS.GOV] Sent: Tuesday, February 08, 2000 12:32 PM To: DB2-L@RYCI.COM Subject: No subject given



We are on UDB for AIX version 6.1. [...] 3121 53 39_Re: QUALIFIER option of BIND going away13_George Peters22_gpeters@COMPUSERVE.COM30_Tue, 8 Feb 2000 14:02:59 -0500661_iso-8859-1 The QUALIFIER option introduced in DB2/MVS V2.3 enabled one to split table/view, etc qualification from PLAN/PACKAGE OWNERship. Unqualified table/views etc. were implicitly qualified by the value for OWNER (which then served double duty, in the absence of such a parameter). Addition of the new parm of QUALIFIER removed this double duty. It also removed the need for SYNONYMS (which were the workaround). With CURRENT PACKAGESET, it enables easy use of mirror tables. In conjunction with COLLECTION, it also enabled easy maintenance of multiple testing environments in one DB2 sub-system. I am sceptical that removal of this option is in IBM's [...] 3175 213 34_Re: Can we use CAF for two plans ?14_Harvey Wachtel22_babybaby@UAPC.CUNY.EDU30_Tue, 8 Feb 2000 14:50:27 -0500396_us-ascii Actually, I wasn't very precise in what I said. As long as you're connecting to the same subsystem, it's not the connection that has to be closed and reopened, but only the thread. If AAAA has been written with explicit CONNECT and OPEN, its sequence should look something like this: CONNECT to subsystem OPEN thread to plan CALL BBBB CLOSE thread to plan DISCONNECT from subsystem [...] 3389 137 31_Re: Restart/Recover from Reorg.16_Milligan, Andrea25_Andrea_Milligan@BCBST.COM30_Tue, 8 Feb 2000 16:01:22 -0500597_- You will be fine to term the utility and recover the indexes. Don't forget to delete any dangling datasets so you don't get caught the next time you reorg.

Andrea Milligan DataBase Administrator BlueCross BlueShield of Tennessee

> -----Original Message----- > From: Ted Pesta [SMTP:tpesta@AMQUEST.COM] > Sent: Tuesday, February 08, 2000 9:13 AM > To: DB2-L@RYCI.COM > Subject: Restart/Recover from Reorg. > > Environ : OS/390, DB2 V5 > We had a problem here with a Reorg. We got an I/O error on one of the > tapes > and the job crashed. When I displayed the utility status it [...] 3527 202 39_Re: QUALIFIER option of BIND going away16_Milligan, Andrea25_Andrea_Milligan@BCBST.COM30_Tue, 8 Feb 2000 16:14:20 -0500612_- ditto

Andrea Milligan DataBase Administrator BlueCross BlueShield of Tennessee

> -----Original Message----- > From: BILL_GALLAGHER@PHL.COM [SMTP:BILL_GALLAGHER@PHL.COM] > Sent: Tuesday, February 08, 2000 1:34 PM > To: DB2-L@RYCI.COM > Subject: Re: QUALIFIER option of BIND going away > > I sure hope that this is not the case, because our shop standards are to > use > unqualified SQL in all our batch programs, with the QUALIFIER parameter > being > used to resolve the qualifiers at bind time. I would have a lot of > trouble > believing that IBM would plan to take this away from us. > > [...] 3730 236 34_Re: Can we use CAF for two plans ?16_Swinski, Kenneth32_KSwinski@INTERNAL.MASSMUTUAL.COM30_Tue, 8 Feb 2000 16:20:47 -0500567_us-ascii My advice is to code an EXEC SQL SET working storage field = CURRENT DATE in AAAA so that AAAA becomes the plan name for the implicit connect. Then bind the DBRMs or packages for BBBB and CCCC into plan AAAA.

> -----Original Message----- > From: Harvey Wachtel [SMTP:babybaby@UAPC.CUNY.EDU] > Sent: Tuesday, February 08, 2000 2:50 PM > To: DB2-L@RYCI.COM > Subject: Re: Can we use CAF for two plans ? > > Actually, I wasn't very precise in what I said. As long as you're > connecting to the same subsystem, it's not the connection that has to be [...] 3967 161 16_Re: explain this10_Robert Ord21_robertord@HOTMAIL.COM28_Tue, 8 Feb 2000 21:32:16 GMT655_- Have you tried using the outer join syntax instead ?

Rob



>From: "Trivedi, Kaushal J, BGM" >Reply-To: DB2 Data Base Discussion List >To: DB2-L@RYCI.COM >Subject: explain this >Date: Fri, 4 Feb 2000 14:12:08 -0500 > >Hi All. > >I have this query and i could really use some suggestions here. I am trying >to look at it many ways but another opinion would be great. > SELECT > A.CUST_ARGT_ID > , A.CMTMT_ADJ_FCTR > , A.CMTMT_HIER_CD > , A.CMTMT_TYPE_CD > , A.DUPL_SUBP_ELG_IND > , A.MLTPL_PNLTY_CD > , A.PNLTY_NET_GRS_CD > , A.RAMP_UP_ELGBL_IND > , A.RTRG_NET_GRS_CD > , A.SHRTFL_METH_CD > , [...] 4129 31 31_NFS mounted drives and UDB v5.212_Janet Rolsma12_janet@NM.NET30_Tue, 8 Feb 2000 14:41:44 -0700413_us-ascii Hi,

We are using UDB v5.2 on AIX.

We are having some trouble with getting enough disk space for our databases on the local drives for our servers. We would like to move the tablespaces to containers residing on a net appliance which is NFS mounted to the server. My question is has anyone done this? Have you had any problems with data integrity? What was the actual performance like? [...] 4161 220 34_Re: Can we use CAF for two plans ?14_Susan Birgeles25_Susan.Birgeles@ALLTEL.COM30_Tue, 8 Feb 2000 16:55:00 -0600395_us-ascii ctually, I wasn't very precise in what I said. As long as you're connecting to the same subsystem, it's not the connection that has to be closed and reopened, but only the thread. If AAAA has been written with explicit CONNECT and OPEN, its sequence should look something like this: CONNECT to subsystem OPEN thread to plan CALL BBBB CLOSE thread to plan DISCONNECT from subsystem [...] 4382 31 19_db2diag.log records17_Gabriel Collantes24_gcollantes@HQ.SEQUEL.NET30_Tue, 8 Feb 2000 15:43:29 -0600631_- Hi,

The following records are regularly generated by db2 in db2diag.log. Is anyone familiar with what the records means, and the information they carry? ---snippet -------------------------------------------------------------- 2000-02-08-11.29.36.339629 Instance:mrl5p1 Node:000 PID:25544(db2bp_s) Appid: oss_2 sqlo_access_listenq Probe:1

0000 0002 ....

2000-02-08-08.23.58.082624 Instance:mrl5p1 Node:000 PID:20732(java) Appid: oss_2 sqlo_access_listenq Probe:1

0000 0002 .... -----------------------------------------------------------------------

Any info would be greatly appreciated. [...] 4414 17 25_Working with DB2 instance17_Gabriel Collantes24_gcollantes@HQ.SEQUEL.NET30_Tue, 8 Feb 2000 15:46:45 -0600416_- Hi,

How do one find out the FencedId used when a DB2 instance was created?

I can't seem to find a facility to get that info.

Regards, Gabriel Collantes

================================================ 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. 4432 73 34_Re: Can we use CAF for two plans ?14_Campbell James30_James.Campbell@ENERGYIS.COM.AU30_Wed, 9 Feb 2000 09:55:47 +1000555_us-ascii Deepak

What is important is what you mean by "another database different from"

- if they are both in the same subsystem, then, as Rick said, use packages. If you can't get a suitable PKLIST in the plan , you'll need to use SET CURRENT PACKAGESET to ensure that each program can find the correct package. - if they are in different DB2's, then how about remote packages (ie bind package for BBBB on DB21 and CCCC on DB22; then program CCCC does a SET :hv = CURRENT SERVER; CONNECT TO "DB22" on entry, and CONNECT TO :hv on exit [...] 4506 51 21_Re: Special Registers14_Campbell James30_James.Campbell@ENERGYIS.COM.AU30_Wed, 9 Feb 2000 10:16:10 +1000436_us-ascii See the SQL REFERENCE. They are not stored in a catalog. If you think through their usage you will understand why.

/* standard disclaimer */ James Campbell DBA Hansen Corporation, Doncaster +61 3 9840 3864 james.campbell@energyis.com.au -----Original Message----- From: Baskaran Subramaniam [mailto:subramaniam.baskaran@DB.COM] Sent: Tuesday, February 08, 2000 6:13 PM To: DB2-L@RYCI.COM Subject: Special Registers [...] 4558 55 54_db2 os/390 v5.1 - Escalation of Incr Copy to full copy28_Dan Courter (1-408-463-3870)27_daniel_courter@VNET.IBM.COM28_Tue, 8 Feb 2000 15:49:31 PST378_- Here are the scenarios that cause the U402 message. Maybe this will help you locate the one you are encountering. Failing this, you will need to open a PMR so we can give you a modification that will add diagnostics to this module. Ask them to contact me in the PMR. The message is issued after the fact so there is no straight forward way to tell which case is failing. [...] 4614 31 22_Re: Checkpoint/Restart0_18_Richar3749@AOL.COM28_Tue, 8 Feb 2000 19:51:46 EST620_US-ASCII I have looked at a number of packages for this and ended up recommending Smart/Restart from Relational Archiitects. We purchased this product and installed it over a year ago, also adding it to about 20 'long' running programs(10 minutes plus). We are currently adding it to an additional dozen or so of another application just brought over to the IBM. Package has been pretty good. Relatively easy installation into the programs. Most difficult part may be the determination of the LUW(if you have spaghetti code anyway, as here). The Self Repositioning Cursors can be a challenge also. Although we have [...] 4646 53 26_Re: DB2 Row Level Security12_BOB JEANDRON21_BOB.JEANDRON@USDA.GOV30_Tue, 8 Feb 2000 12:07:00 -0500496_- How about this: CREATE TABLE TSTGRP.APPL_TABLE (AUTHCODE CHAR(18) NOT NULL, Access code COL1 CHAR(10), Data COL2 CHAR(10), Data COL3 CHAR(10)); Data

CREATE TABLE USER_AUTH (USERID CHAR(30), Userid USER_CODE CHAR(18)); Access code CREATE VIEW TST_VIEW AS SELECT AUTHCODE,COL1,COL2,COL3 FROM TSTGRP.APPL_TABLE WHERE EXISTS (SELECT 'X' FROM TSTGRP.USER_AUTH WHERE USERID = USER AND SUBSTR(TSTGRP.APPL_TABLE.AUTHCODE,1,LENGTH(TSTGRP.USER_AUTH.USER_CODE)) = TSTGRP.USER_AUTH.USER_CODE); [...] 4700 30 41_Replication / DB2 UDB V6 (Solaris) / 24x713_Philip Nelson21_teamdba@ATTGLOBAL.NET30_Tue, 8 Feb 2000 12:59:56 +0000329_us-ascii We're setting up DB2 UDB V6 on Solaris to support our web service which must run 24 x 7. The database will be with updates.

We have to provide a database at all times but will need time out for (e.g.) reorgs and system maintenance.

Is it possible to use replication for this purpose, something like - [...] 4731 19 44_DB2 for OS/390 V6 vs. DB2 for Sun Solaris V617_Scarcella, Gary V21_ScarcellaGV@AETNA.COM30_Wed, 9 Feb 2000 07:20:25 -0500338_- I've been mainly working with DB2 for OS/390 and am now starting to work with DB2 for Sun Solaris V6. I know there's a lot of differences (at least from the DBA perspective) and keep running into these differences one at a time. Does anyone have or know where to find a list of the differences between DB2 in these 2 environments? [...] 4751 63 34_Using DSN1COPY for Backup/Recovery12_David Wilson31_DWILSON@CEDELGLOBALSERVICES.COM30_Wed, 9 Feb 2000 14:03:09 +0100347_- > G'Day All, > We have been asked to provide a process that use DSN1COPY to backup and > restore segmented tablespaces that contain zeroed pages. > Using the parameter 'SEGMENT' the backup appears to work OK but (judging > by the number of pages processed) actually only copies non zeroed pages. > The restore then fails with the following [...] 4815 12 16_Grants to Public13_Shauna Hadden24_shauna.hadden@BCBSKS.COM30_Wed, 9 Feb 2000 07:56:21 -0600423_- How many of you have policies against granting to "Public"? What have you done instead? If using secondary authids, how have you set them up? Do you have one that replaces Public?

================================================ 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. 4828 88 54_Just what in the heck do you think you are doin', huh?14_Greg DiGiorgio29_gdigior@CI.NEWPORT-NEWS.VA.US30_Wed, 9 Feb 2000 09:52:06 -0500611_us-ascii Hi,

I *really* (on my knees begging) need your insight about what you are doing in the area of PC networking and DBMS applications.

First, we are a municipal government and do 98% of all programming in-house, serving the Sheriff's office with Jail Management systems, the Police Dept with offense management systems, the Tax dept with personal property tax management systems, the school systems with student administration systems, Real Estate Assessor, Waterworks, City Attorneys, Parks 'n Rec, and on and on, entwining every fiber of City government with mainframe-based systems. [...] 4917 28 14_FOR FETCH ONLY19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM30_Wed, 9 Feb 2000 08:48:27 -0600329_iso-8859-1 Hi all and TIA, Been through the list archives and didn't really find any direct hits. Are the following statements correct? First, FOR FETCH ONLY is for cursor processing only. Second, FOR FETCH ONLY is used to allow BLOCK FETCHing and possible LOCK avoidance for remote calls and has no effect for local calls. [...] 4946 23 18_Re: FOR FETCH ONLY14_Massimo Scarpa16_mscarpa@CESVE.IT30_Wed, 9 Feb 2000 15:58:16 +0100355_us-ascii I use FOR FETCH ONLY (as well as FOR READ ONLY and WITH UR) instruction in dinamyc

SQL (SPUFI; QMF, DSNTIAUL...) to avoid ambiguous cursors and so to obtain lock avoidance.

No declare cursor (explicit) needed. And I think it works even for local calls.

Hoping this is the answer you need...

Max Scarpa Data Admin [...] 4970 64 18_Re: FOR FETCH ONLY0_24_db46@DAIMLERCHRYSLER.COM30_Wed, 9 Feb 2000 10:03:34 -0500535_us-ascii FOR FETCH ONLY (or FOR READ ONLY) as I understand it is only used for cursors, but it eliminates the possibility of ambiguous cursors. These are cursors that may be used for UPDATE WHERE CURRENT OF.

Now if you put a ORDER BY on a cursor it is defaulted to FOR FETCH ONLY because you cannot do an UPDATE WHERE CURRENT OF later on after you open the cursor. It does help with reduced locking if your cursor does not have an ORDER BY on it because being ambiguous, it has to hold the appropriate locks in case you do [...] 5035 25 27_migration from db2 v5 to v615_Sherri Sterling23_ssterling@BPD.TREAS.GOV30_Wed, 9 Feb 2000 10:17:24 -0500535_US-ASCII We have just migrated from os/390 db2 version 5 to db2 version 6 and have encountered a few problems. One, we are getting 00E70005 on all plans bound as degree(any). When we rebind them as degree(1) they work ok. Also of some reason CICS is getting the same error. Two, we have a simple sql SELECT MAX(PR-DATE) INTO :HV. When the results are returned from spufi or qmf they are fine but when they are returned to a cobol program of pic x(10) it is packed decimal and not a good date. We have opened PMR's with IBM on the [...] 5061 35 20_Re: Grants to Public0_17_JTonchick@AOL.COM28_Wed, 9 Feb 2000 10:32:38 EST437_US-ASCII Shauna,

We have taken the direction of granting execution of plans and packages to PUBLIC and let the external security subsystem (RACF, ACF2, or Top Secret) handle the security. This is especially true of IMS & CICS transactions where the access to the trancode is secured so why add the extra layer of security to the plan or package. This also works for batch programs where the submission of the job is secured. [...] 5097 57 58_Re: Just what in the heck do you think you are doin', huh?14_Massimo Scarpa16_mscarpa@CESVE.IT30_Wed, 9 Feb 2000 16:33:23 +0100428_us-ascii It'll be long to answer to your questions, so I put only some consideration.

I worked in a company where the top management decided to develope a distributed

architecture with 103 servers, VERY APART each others.

With 5-8 specialists we managed a mainframe (1,650,000 transaction/day); to manage 103

servers we needed ten times personnel. Change management and software/products update [...] 5155 69 31_Re: migration from db2 v5 to v616_Amy C. Whitehead14_B06ACW@FDS.COM30_Wed, 9 Feb 2000 11:07:11 -0500385_us-ascii Sherri, Thank you for your response. To date, this is the only response that I have gotten. I guess there are not many shops that have migrated to V6. Thanks again and I am very interested in any other issues that you encounter.

Thanks, Amy













Sherri Sterling on 02/09/2000 10:17:24 AM [...] 5225 31 21_I could use your help20_Inter*Link (C.Baker)24_cbaker@INTERLINK-INC.NET30_Wed, 9 Feb 2000 11:27:23 -0500325_us-ascii Sorry to intrude on the mail list but I am in search of a DB2 DBA for a long term contract in Florida. If you or someone you know is interested please contact me for further details. We also offer a $1000.00 referral fee if we successfully place your referral with our client more than 90 days. Thanks for your help. 5257 48 25_Re: I could use your help14_Manas Dasgupta25_manas.dasgupta@PFSFHQ.COM30_Wed, 9 Feb 2000 11:28:20 -0500565_iso-8859-1 where in FL?

> -----Original Message----- > From: DB2 Data Base Discussion List [mailto:DB2-L@RYCI.COM]On Behalf Of > Inter*Link (C.Baker) > Sent: Wednesday, February 09, 2000 11:27 AM > To: DB2-L@RYCI.COM > Subject: I could use your help > > > Sorry to intrude on the mail list but I am in search of a DB2 DBA for a > long term contract in Florida. If you or someone you know is interested > please contact me for further details. We also offer a $1000.00 referral > fee if we successfully place your referral with our client more than 90 > [...] 5306 199 58_Re: Just what in the heck do you think you are doin', huh?9_Don Alden26_Don.Alden@PKSIS.LEVEL3.COM30_Wed, 9 Feb 2000 09:33:47 -0700404_iso-8859-1 Howdy,

Greg, I can identify with you on the concerns of attempts to replace OS/390 mainframe applications with UNIX based PC and Networking DBMS applications.

Having worked for a large worldwide semiconductor company for eight years, I witnessed what could happen to a 4 million transaction a day (IMS DB/DC, DB2) OS/390 mainframe operation that kept the corporation alive. [...] 5506 34 25_APPROVED FOR TRTCH CHANGE11_Carl Nelson20_Carl_Nelson@AICI.COM30_Wed, 9 Feb 2000 11:51:54 -0600489_US-ASCII Listserver:

Over the weekend while archiving our logs, the following message occurred a couple of times. This appears to be simply an informational message. We did not have any problems with the archiving. I have asked the systems and tape staff but nobody can tell me what causes the message and what actually is going on. One person suggested that the tape was originally mounted as an 18-track, scratched, and later remounted as a 36-track. What is happening here? [...] 5541 172 29_Re: APPROVED FOR TRTCH CHANGE27_Radonjic, Bora (Consultant)23_BRadonjic@ACNIELSEN.COM30_Wed, 9 Feb 2000 11:01:44 -0600461_- Carl Nothing to worry about:

************ Text below copyright 1997 Computer Associates, Inc. ************* IECTMS6E dddd,vvvvvv IS APPROVED FOR LABEL CHANGE

IECTMS6E dddd,vvvvvv IS APPROVED FOR DENSITY CHANGE

IECTMS6E dddd,vvvvvv IS APPROVED FOR TRTCH CHANGE

IECTMS6O dddd,vvvvvv IS APPROVED FOR LABEL CHANGE

IECTMS6O dddd,vvvvvv IS APPROVED FOR DENSITY CHANGE

IECTMS6O dddd,vvvvvv IS APPROVED FOR TRTCH CHANGE [...] 5714 67 21_Re: Special Registers14_Mullins, Craig21_Craig_Mullins@BMC.COM30_Wed, 9 Feb 2000 11:16:57 -0600614_ISO-8859-1 Baskaran:

As of DB2 V6, DB2 provides the following special registers:

CURRENT DATE or CURRENT_DATE CURRENT DEGREE CURRENT LOCALE LC_CTYPE or CURRENT_LC_CTYPE CURRENT OPTIMIZATION HINT CURRENT PACKAGESET CURRENT [FUNCTION] PATH or CURRENT_PATH CURRENT PRECISION CURRENT RULES CURRENT SERVER CURRENT SQLID CURRENT TIME or CURRENT_TIME CURRENT TIMESTAMP or CURRENT_TIMESTAMP CURRENT TIMEZONE USER

You can find this information on page 102 of the SQL Reference Guide which can be downloaded from IBM's web site. The URL is http://www-4.ibm.com/software/data/db2/os390/v6books.html [...] 5782 21 18_SCLM in production12_Andy Wheeler25_Andy.Wheeler@GRIZZARD.COM30_Wed, 9 Feb 2000 12:26:44 -0500479_iso-8859-1 Is anyone using SCLM for program maintenance in a production environment. We have two systems (test and production). The developers want to have their stored procedures NOT RESIDENT so they can make changes and test them easily. When the load module is migrated to production, I would like to make them resident for performance sake. Is it just a matter of changing the "stayresident" column in sysprocedures to 'y', or will I need to relink the modules as well. [...] 5804 36 35_PK and FK names, definitive source?19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM30_Wed, 9 Feb 2000 11:33:02 -0600566_iso-8859-1 Hi all and TIA, If you remember back a little, I asked how you handled duplicate column names when PKs and FKs were involved.

"How do you name columns that will be inherited as foreign keys or part of primary keys when both the parent and child table could have the same column name. Ex: tablea has a primary key column ident dec(3,0) and tableb has a primary key column ident smallint where tablea is parent to tableb. Tableb inherits ident from tablea and it now has two columns with same name. How would you handle this?" The consensus was [...] 5841 19 39_Re: QUALIFIER option of BIND going away12_Roger Miller19_millerrl@US.IBM.COM30_Wed, 9 Feb 2000 11:32:57 -0600407_- I'd like to know who told you that too, so we can do a better job of stopping counterproductive, inaccurate rumors. While I certainly don't know everything, and am often reminded of how inaccurate my crystal ball is, I've never heard it before.

We struggled for a long time before making much smaller incompatible changes, like making the colon mandatory in Version 6. Even that was not easy. [...] 5861 198 58_Re: Just what in the heck do you think you are doin', huh?0_22_Rohn.Solecki@MTS.MB.CA30_Wed, 9 Feb 2000 12:09:45 -0600585_us-ascii We are a telephone company. We have been using home grown PC based apps since about 93-94. We use Powerbuilder connecting to RS/6000 Oracle, DB/2 V5 OS/390 (via CICS COBOL transactions, and IMS on OS/390 (also via CICS COBOL trnx). We do have some functions/screens that display/update data from all 3 DBMS. We currently have 25-30 PB apps with up to 1000+ users. As you've noticed, a lot of things that are taken care of by vendors on the mainframe become your responsiblity in PC apps. Lets face it, the mainframe has almost 40 years of commercial and user application [...] 6060 49 36_Contract DB2 UDB Openings - Illinois11_Chris Beaty18_cbeaty@NCSJOBS.COM30_Wed, 9 Feb 2000 11:38:50 -0800373_iso-8859-1 We have the following contract jobs open for our end-client in Bloomington Illinois:

1. Database Administrator - DB2 for Windows Individual will be responsible for database administration and design support of a production database delivered to NT servers. Excellent DB2 WIN/NT Design and Admin skills Strong NT Server Admin skills Strong SQL skills [...] 6110 101 38_Re: Using DSN1COPY for Backup/Recovery19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM30_Wed, 9 Feb 2000 12:53:20 -0600632_iso-8859-1 David, DB2 for OS/390 V5. Maybe I don't understand your question, but the best method to recover tablespaces is to use the RECOVER utility with FULL IMAGE COPIES created by the COPY utility. Please give more information. Are you talking about copying tablespaces/indexes from one DB2 subsystem to another? If so, the tablespaces/indexes must match, i.e. SEGMENTed, SEGSIZE, PAGESIZE, PIECESIZE, etc. Could you explain the term "zeroed pages"? DSN1COPY, for example, copies page for page from whatever is on an image copy tape created by the COPY utility. We need more to go on here. Regards, Rick Davis "This e-mail [...] 6212 54 20_Re: Grants to Public0_23_Mike_Levine@TEKHELP.NET30_Wed, 9 Feb 2000 14:19:38 -0500642_us-ascii Hi,

Be careful granting PUBLIC EXECUTE on batch plans. If the plan is granted PUBLIC how can you guarantee that no one will run it? On-line plans can be granted public but only enabled for the appropriate environments.

Regards,

Michael Levine Premier Data Services, Inc.



>Shauna, > >We have taken the direction of granting execution of plans and packages to >PUBLIC and let the external security subsystem (RACF, ACF2, or Top Secret) >handle the security. This is especially true of IMS & CICS transactions >where the access to the trancode is secured so why add the extra layer of >security [...] 6267 14 45_Modify Recovery - Additional index on SYSCOPY14_Cathy Puzzuoli24_cathy.puzzuoli@SABRE.COM30_Wed, 9 Feb 2000 13:37:19 -0600322_- There is lots of information in DB2-L about shops having problems with contention on SYSIBM.SYSCOPY when the modify utility executes. Has anyone added an index to SYSCOPY to eliminate the contention when the modify utility runs? If so, what columns did you use in your index? Any help would be greatly appreciated! [...] 6282 94 34_Trigger Calling a Stored Procedure12_Steve Grimes30_Steve_Grimes@AISMAIL.WUSTL.EDU30_Wed, 9 Feb 2000 14:01:17 -0600535_us-ascii



Hello. Any takers for this one...?

This is a problem I wrote about before. Richard Yevich provided some insight, but I'm still stuck. I have tested everything I can think of.

I have put together a MS Word document which decribes the problem in detail. It shows (what I think are) the relevant outputs. However, it's 27KB so I didn't want to send it as an attachment to the list, (but I could be persuaded!) I'll probably look into registering this as a problem with IBM using this document. [...] 6377 21 31_Re: migration from db2 v5 to v60_29_STANLEY_GOODWIN@MECH.DISA.MIL30_Wed, 9 Feb 2000 15:23:41 -0500538_us-ascii I have looked at IBMLINK and I am unable to find any APRAS with the error code of e70005 and problem description concerning Paralellism, Can anyone point me in the right direction. Stan Goodwin DAC MBG DB2 Support

Stanley_goodwin@mech.disa.mil DSN: 430-4335 Outside:- 717-605-4335

================================================ 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. 6399 54 27_Re: SYSLOG message trapping24_Humphris,Richard P.(NXI)24_Richard.Humphris@CNA.COM30_Wed, 9 Feb 2000 14:29:17 -0600524_- We don't have much automated at this time... but we do flag/hilight the following message id's: dsnj107i dsnj108i and dsnj126i (these are all critical messages that should be acted upon by the DB2 support staff ASAP).

Also when DB2 comes up on our production box, we trap msg: DSNR002i and check to see is some CICS regions are already up. For those CICS(s) with DB2 connections, we wanted operations to be reminded (when the DB2 sub-system came back up); that they now need to reconnect those CICS(s) to DB2. [...] 6454 19 22_DB2/Oracle comparisons0_14_whomer@CSC.COM31_Thu, 10 Feb 2000 09:28:03 +1000513_us-ascii Hi all,

I remember reading DB2-L postings (almost 12 months ago) in regard to DB2/Oracle comparisons. Could anyone please be able to dig up these postings and send them to me or let me know how I can dig them up myself. Many thanks.

Regards, Warren

================================================ 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. 6474 56 35_Disaster Recover using EMC products0_54_Shan_Leatherman/MO/americancentury@AMERICANCENTURY.COM30_Wed, 9 Feb 2000 16:43:26 -0600309_us-ascii BACKGROUND: We have DB2 Version 5 running in a two-way data sharing environment (365 days x 24 hours) on OS/390 2.8. We are talking with EMC about using remote DASD mirroring and EMC Timefinder to provide our disaster recovery (DR) site with a DASD mirror image of our MVS and DB2 environments. [...] 6531 29 22_Merging DB2 subsystems0_22_Rohn.Solecki@MTS.MB.CA30_Wed, 9 Feb 2000 17:05:32 -0600561_us-ascii We currently have 2 LPARS, each with a separate production DB2 subsystem. In the next few months we are going to be merging the LPARS. We are at DB2 V5 on OS/390 V2.8. Ideally (I think) we would like to merge the 2 production DB2 subsystems because we have an every increasing need for single programs to read/update data in both subsytems. But it is awkward. Our initial investigaion into the merge: create TS, Load data, rebind, edit JCL to point to new subsys does not appear to be feasible. The bind step alone was estimated to take 70 hours. [...] 6561 55 26_Re: Merging DB2 subsystems14_Wayne Driscoll19_wdriscoll@QUEST.COM30_Wed, 9 Feb 2000 15:16:02 -0800453_iso-8859-1 In order to convert to Datasharing you will have to first merge the 2 catalogs so datasharing won't let you avoid the problem. Wayne Driscoll Product Developer Quest Software Inc. wdriscoll@quest.com Note: All opinions are strictly my own.

-----Original Message----- From: Rohn.Solecki@MTS.MB.CA [mailto:Rohn.Solecki@MTS.MB.CA] Sent: Wednesday, February 09, 2000 5:06 PM To: DB2-L@RYCI.COM Subject: [DB2-L] Merging DB2 subsystems [...] 6617 66 38_Re: Trigger Calling a Stored Procedure14_Campbell James30_James.Campbell@ENERGYIS.COM.AU31_Thu, 10 Feb 2000 10:15:45 +1000475_us-ascii Steve,

Appl Prog and SQL Guide Sect 3.5.3 Invoking stored procedures and user-defined functions from triggers: "Because a before trigger must not modify any table, functions and procedures that you invoke from a trigger cannot include INSERT, UPDATE, or DELETE statements that modify the triggering table."

I wonder what is really meant by "cannot". Obviously it doesn't mean "... and if you try the SQL will be rejected with a negative SQLCODE." [...] 6684 20 54_db2 os/390 v5.1 - Escalation of Incr Copy to full copy28_Dan Courter (1-408-463-3870)27_daniel_courter@VNET.IBM.COM28_Wed, 9 Feb 2000 15:31:14 PST464_- Having a SYSCOPY record ICTYPE T with STYPE blank can happen if the COPY fails before the switch to the COPY or REPORT phase, before the COPY has really done anyting. So, if the phase is UTILINIT, its not a bug. Otherwise it may be, especially if the phase is COPY or REPORT. The phase name switch and the setting of the STYPE occur in the same commit scope and the STYPE should never be blank in that case. This was not change in V6, as far as I can tell. [...] 6705 61 20_Re: Grants to Public12_Kirk Hampton16_khampto1@TXU.COM30_Wed, 9 Feb 2000 17:32:20 -0600391_us-ascii We have created a RACF group called DBPUB, to which virtually all logon IDs are connected (becoming a DB2 secondary AUTHID). In our production systems, we have gone through a retrofit over the last year-and-a-half, REVOKEing grants from PUBLIC and granting the auth back to DBPUB. All this was done out of fear of PUBLIC access via the Web and/or desktop client (DB2 Connect). [...] 6767 80 26_Re: Merging DB2 subsystems9_Rob Crane22_racrane@CONCENTRIC.NET30_Wed, 9 Feb 2000 17:26:38 -0700571_us-ascii There are several approaches to merging the catalogs that do not require unloading and loading of the data. If you have access to prior IDUG material there have been several presentations on doing this. If you don't have access to those let me know and I can forward on the basics of what it will take. I used these techniques to merge to a data sharing environment in a previous life and I believe the same basic process has been enhanced to include clists and rexx execs to simplify the process and increase the automation. This particular shop used this [...] 6848 101 39_Re: Disaster Recover using EMC products11_Roland Chua32_rolandchua@SINGAPOREEXCHANGE.COM31_Thu, 10 Feb 2000 08:44:16 +0800378_us-ascii Hi,

We are running DB2 Version 4 and using EMC SRDF to move the data from production site to DR site. We used synchronous Mode. (ie. all Reads and writes to the volume to which a write has been performed is queued until the write has been acknowledged at the remote site.) This will ensure that recover of DB2 is similar to recovering from a power failure. [...] 6950 19 38_Re: Trigger Calling a Stored Procedure12_Steve Grimes30_Steve_Grimes@AISMAIL.WUSTL.EDU30_Wed, 9 Feb 2000 22:32:56 -0600373_us-ascii Yes - I recall reading this Jim. However, my Stored Procedure contains only a SELECT statement against another table. Also I specified "READS SQL DATA" in my CREATE PROCEDURE. Thus, by my understanding, I should be "legal" for the "Before" trigger.

But thanks for the suggestion!

Steve Grimes (from home, where I don't have my fancy sigNETure.) [...] 6970 138 38_Re: Trigger Calling a Stored Procedure0_24_markus.schaub@SYSTOR.COM31_Thu, 10 Feb 2000 07:43:07 +0100579_us-ascii hi steve

as you are using a BEFORE trigger the triggered statements you are executing (change value from 'A' to 'X' and the call of the stored procedure) are done before db2 will actually insert the new row into the table. and so db2 inserts the row with the value 'X' as your trigger has changed the transition variable. the desired change of the value to '2' (from the stored procedure) can only be achieved if your stored procedure returns the correct value to the trigger and the trigger will set the transition variable to the value ('2') returned from [...] 7109 127 16_Grants to Public0_22_The-Thanh.Luong@AGF.BE31_Thu, 10 Feb 2000 09:19:15 +0100485_- Hi Shauna,

"Grant to Public" ? Oh no! It's too obvious that I am not going to develop the reasons why.

Our environment is DB2 V4.1, OS/390 V1.3. The Security system is basic ACF2 (I mean without ACF2/DB2 facility like Top Secret has). We introduce a field in ACF2's user profile that we call "SQLID" and we use it as Secondary AuthId, by activating the DB2 exits (DSN3@ATH and DSN3@SGN). We make a general use of it and never grant something to a userid directly. [...] 7237 58 20_Re: Grants to Public14_Ruediger Kurtz28_ruediger.kurtz@HUK-COBURG.DE31_Thu, 10 Feb 2000 09:33:08 +0100580_us-ascii It's the same here, and we have granted public access to our production exception tables as well since we follow the rule that data do not 'belong' to us, and if data get written out to the exception tables it's not up to us to 'correct' that data.

Ruediger Kurtz

JTonchick@AOL.COM schrieb: > > Shauna, > > We have taken the direction of granting execution of plans and packages to > PUBLIC and let the external security subsystem (RACF, ACF2, or Top Secret) > handle the security. This is especially true of IMS & CICS transactions > where the access [...] 7296 112 52_Re: db2 os/390 v5.1 - STYPE x'00'; RUNSTATS 00E402130_18_mebert@AMADEUS.NET31_Thu, 10 Feb 2000 09:37:23 +0100448_us-ascii I think this is in response to my question... two points: 1. the STYPE column was hex'00' (nullbyte) - not NULL, not blank; 2. the abend was due to a B37 (disk full) during a full IC and happened in the COPY phase. Furthermore, we have no occurrence of this in either of our V5 subsystems (all correctly have STYPE 'F'), while all occurrences in our V6/SAP subsystem have STYPE x'00'. I did not find any problems on IBMLINK for this. [...] 7409 41 49_Re: Modify Recovery - Additional index on SYSCOPY13_McDonald, Ian32_Ian.McDonald@ABBEYNATIONAL.CO.UK31_Thu, 10 Feb 2000 08:45:00 -0000448_- Cathy I too would be interested.

The only extra index I've put on the catalog is one I called DSNDTX03 on SYSTABLES of TSNAME.NAME.

It was so long ago I can't remember why, but it saved a lot of time whatever it was and it's still there.

HTH

Ian McDonald DB2 DBA Abbey National ---------- From: Cathy Puzzuoli To: DB2-L@RYCI.COM Subject: Modify Recovery - Additional index on SYSCOPY Date: 09 February 2000 19:37 [...] 7451 175 38_Re: Using DSN1COPY for Backup/Recovery12_David Wilson31_DWILSON@CEDELGLOBALSERVICES.COM31_Thu, 10 Feb 2000 13:28:56 +0100571_iso-8859-1 > Rick, > In this case we want to use DSN1COPY to backup (please don't ask) and > DSN1COPY to restore. The restore could be to the original tablespace, to > another tablespace (with the same characteristics) in the same subsystem, > or to a different subsystem. In this case we are restoring to the original > tablespace, so there is no question of differing segment sizes etc. The reasons why we need to use DSN1COPY for this backup are simply too involved to list here - suffice it to say that it involves EMC Snap dumps and over-enthusiastic technical [...] 7627 92 39_Re: Disaster Recover using EMC products19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Thu, 10 Feb 2000 06:46:22 -0600553_iso-8859-1 Shan, I'm not familiar with Timefinder and it might make following inaccurate. For Mirroring, your concerns are correct. If using SRDF in Async mode, there use to be some restrictions on how the database(s) at the hot site could be defined. Example, it use to be that no RI was allowed due to the manner in which the updates were applied. Don't know if this is still a restriction, check it out. If still true, to get your RI definitions to the hot site, have some process running that puts it all in a table whenever a RI DDL statement [...] 7720 156 39_Re: Disaster Recover using EMC products11_Ron Haupert23_rhaupert@GOLDENGATE.NET31_Thu, 10 Feb 2000 06:58:12 -0600544_iso-8859-1 EMC Symmetrix Remote Data Facility (SRDF) provides remote mirroring to disaster recovery or to disaster restart sites. When SRDF synchronous mode is used, the state of the data at the remote site is exactly the state of the data at the local site at any instant in time. If the local site fails, the operating system, DBMS, and all applications can be restarted at the remote site. The restart procedures at the restart site are identical to the restart procedures used for local system restart in the event of a power failure. [...] 7877 36 12_Mass deletes11_Eric Robida22_Eric_Robida@BCBSME.COM31_Thu, 10 Feb 2000 08:36:00 -0500531_us-ascii I would like some help with a minor problem. There are four large tables, 24 - 54 millions rows each. We are in the process of moving older data to a set of history tables then deleting it from the main tables. Our current process has one program that determines and marks the rows to be deleted in all four tables. This takes 18 hours elapsed, 20 - 40 minutes of CPU. We then run four jobs, one per table, concurrently. These use the DB2 utilities to move the marked rows to the history tables and then issue a mass [...] 7914 185 38_Re: Using DSN1COPY for Backup/Recovery19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Thu, 10 Feb 2000 07:44:16 -0600609_iso-8859-1 David, Jeez, you got more issues involved in one db2-l item than I've ever seen.

Think I'll leave them all alone but one. I think the operative phase in the following statement from the Utility Gde. "Zeroed pages in the table space are copied, but no error messages are issued." is the error part. Could this be the problem? Please verify the following, going on memory here and might be on wrong track. Pre-formatting pages in a segmented tablespace simply means that the spacemap pages are formatted at the appropriate intervals throughout the dataset. I don't believe data pages are [...] 8100 113 39_Re: Disaster Recover using EMC products19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Thu, 10 Feb 2000 07:50:21 -0600597_iso-8859-1 Oops, meant RRDF in previous note. 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 of this e-mail is strictly prohibited." [...] 8214 54 16_Re: Mass deletes0_19_mike.holmans@BT.COM31_Thu, 10 Feb 2000 13:52:56 -0000580_- I would have thought that this was a prime case for the DISCARD option of REORG as enhanced in the V5 updates.

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

> -----Original Message----- > From: Eric Robida [SMTP:Eric_Robida@BCBSME.COM] > Sent: Thursday, February 10, 2000 1:36 PM > To: DB2-L@RYCI.COM > Subject: Mass deletes > > I would like some help with a minor problem. There are four large > tables, 24 - 54 millions rows each. > We are in the process of moving older data to a set of history tables then > deleting it from the main tables. > [...] 8269 58 16_Re: Mass deletes14_Ruediger Kurtz28_ruediger.kurtz@HUK-COBURG.DE31_Thu, 10 Feb 2000 15:02:53 +0100525_us-ascii Hi Eric,

what about unloading the data you want to keep and use the LOAD-Utility with REPLACE ?

Eric Robida schrieb: > > I would like some help with a minor problem. There are four large > tables, 24 - 54 millions rows each. > We are in the process of moving older data to a set of history tables then > deleting it from the main tables. > Our current process has one program that determines and marks the rows to > be deleted in all four tables. > This takes 18 hours elapsed, 20 - 40 minutes of [...] 8328 33 16_Re: Mass deletes14_Massimo Scarpa16_mscarpa@CESVE.IT31_Thu, 10 Feb 2000 15:08:10 +0100512_us-ascii We had a similar problem with a huge table. Every night we deleted index(es) on the table, we deleted rows, then we re-created index(es) with DEFER YES option+ recover(s) index(es) in parallel. It's well-known that a delete without index(es) is much faster (verrry fast.....3rd party companies developed tools for this...). It's the only way to perform mass delete with programs, I think, in a reasonable time. Of course it depends on the batch window size, but I think your window is wide enough. [...] 8362 145 16_Re: Mass deletes0_22_The-Thanh.Luong@AGF.BE31_Thu, 10 Feb 2000 15:10:28 +0100412_- Hi Eric,

Could you consider 1.- unload the "good" rows and load them to the history tables 2.- unload the "bad" rows then load/replace them back into the regular tables. We do it all the time here. Pay attention to the Referential Integrity, if any.

Just a thought. HTH.

LUONG The-Thanh AGF Belgium Insurance, DataBase Administration and Security MGR [...] 8508 31 47_Mapping table ASCII or EBCDIC - does it matter?13_Adrian Savory24_adrian.savory@ZURICH.COM31_Thu, 10 Feb 2000 14:11:03 +0000563_us-ascii Hi folks,

We have an interesting situation where we have a mix of ASCII and EBCDIC applications in the same subsystem (don't ask why - not my idea!). We are now starting to run online reorgs and it occurred to me that the mapping table might be a problem. We planned on creating two map table tablespaces (1 ASCII, 1 EBCDIC) and handle it that way. But what happens if you reorg an ASCII tablespace using an EBCDIC mapping table, or vice versa? I tried this, fully expecting the reorg to blow up, but it didn't. It completed successfully and [...] 8540 19 26_Re: Mass delete - addendum14_Massimo Scarpa16_mscarpa@CESVE.IT31_Thu, 10 Feb 2000 15:19:33 +0100460_us-ascii Of course all discarded rows may be unloaded in a flat file compressed by SMS dataclass

(using hardware compression feature), so you can save DASD space avoiding B37 errors....

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. 8560 53 22_Re: SCLM in production0_54_Shan_Leatherman/MO/americancentury@AMERICANCENTURY.COM31_Thu, 10 Feb 2000 08:24:34 -0600536_us-ascii Andy, This is exactly how we have our stored procedures set up. We don't use SCLM (we use Endeavor for our change management) and for us, we wrote a rexx exec that our change management folks execute whenever a new stored procedure is being moved to production environment. The rexx exec will update the STAY RESIDENT flag from NOT RESIDENT to RESIDENT in the Sysprocedures table. It also issues the stop and start commands as well as issuing the VARY commands to refresh the module. (We're WLM Stored procedures, DB2 V5) [...] 8614 184 39_Re: Disaster Recover using EMC products19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Thu, 10 Feb 2000 08:52:51 -0600528_iso-8859-1 Ron, Seems EMC has come a long way in this area. I'm confused on a couple areas.

Snip>> When SRDF synchronous mode is used, the state of the data at the remote site is exactly the state of the data at the local site at any instant in time. Unsnip<<

Are there I/O elongation issues related to this when greater than ESCON channel distances are involved? Storage processor (controller) path length due to mirroring at greater than ESCON distances? When is an application told a COMMIT is successful? [...] 8799 57 20_Re: Grants to Public14_Mullins, Craig21_Craig_Mullins@BMC.COM31_Thu, 10 Feb 2000 09:11:32 -0600324_iso-8859-1 Jim:

You should also consider limiting the execution environment for your CICS plans granted to PUBLIC. Do this by using the ENABLE(CICS) parameter on the BIND for the plan. Doing so will prohibit anyone from using the plan in a non-CICS environment where the security on the trancode would not help. [...] 8857 1670 40_V6 for OS/390 book available (the first)15_Gabrielle_Assoc29_Gabrielle_Assoc@EMAIL.MSN.COM31_Thu, 10 Feb 2000 09:14:45 -0800634_- The second edition of the DB2 for OS/390 Development for Performance book is the first to include V6 features and facilities. We certainly have a myriad of impressive features to use to our advantage in 2000. Here is a list of the chapters.

Table of Contents 3 Concepts and Components 28 Creating STOGROUPs, Databases, and Buffer Pools 72 Creating Tablespaces 108 Creating Tables and Views 163 Index Usage for Performance 227 Index Design for Performance 252 The Basics of SQL Data Manipulation 293 Concurrency Control 338 Program Development 371 Batch Processing 428 Triggering Actions in DB2 455 Stored Procedures 481 [...] 10528 166 40_CA-Platinum Quickcopy creating bad tapes10_Mark Labby16_mlabby@PHEAA.ORG31_Thu, 10 Feb 2000 10:26:22 -0500528_us-ascii We have just found out that the backup tapes we have been creating using CA-Platinum Quickcopy are corrupted. The problem is not with Quickcopy, nor even DB2, but with a PTF we applied to MVS to support the 3590-E tape units. However, the problem is not limited to backups on the 3590, all of our errors have been found on 3490 drives. IBM has an APAR on it and a fix is available. I wanted to throw out a heads up to anyone else who may be using Quickcopy so they can check out their backups before you need them [...] 10695 61 39_Re: PK and FK names, definitive source?19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Thu, 10 Feb 2000 09:21:33 -0600611_iso-8859-1 Hi all, Finally found a couple reference books related to this topic. "The Data Modeling Handbook" by Michael C. Reingruber and William W. Gregory. "Handbook of Relational Database Design" by Candace C. Flemimg and Barbara von Halle.

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 [...] 10757 85 16_Re: Mass deletes9_Rob Crane22_racrane@CONCENTRIC.NET31_Thu, 10 Feb 2000 08:45:55 -0700610_us-ascii Consider the following:

1) Partition your table based on this delete criteria, then use utilities to clean up old partitions, etc. You could have a rolling effect where you have a reference table that represents which partitions correspond to certain periods of data. You could also make use of large tablespace to accomplish the same thing with possibly less hassle. If the old data is ever needed again, you could unload both the data and table structure (prior to purging it) to a flat file, again create another reference table that would keep track of this information incase you had [...] 10843 45 51_Re: Mapping table ASCII or EBCDIC - does it matter?14_Wayne Driscoll19_wdriscoll@QUEST.COM31_Thu, 10 Feb 2000 08:21:17 -0800377_iso-8859-1 Adrian, I don't think that it will really matter, because even though the columns are defined as NOT NULL, it is my understanding that the reorg process treats the columns as NOT NULL FOR BIT DATA (ie no CCSID translation occurs). Hope this helps. Wayne Driscoll Product Developer Quest Software Inc. wdriscoll@quest.com Note: All opinions are strictly my own. [...] 10889 29 29_Update partitioning key in V50_20_Gerhard.Heiss@RUV.DE31_Thu, 10 Feb 2000 17:16:31 +0100465_iso-8859-1 Dear List,

we would like to update partitioning keys of earlier created tables in V. 5. Unfortunately those still have UPDATES=N in syscolumns. Is there any other way instead of unload, drop, create and reload?

kind regards

Gerhard Heiss

__________________________________________________ R+V Versicherung AG Phone: +49 611-533-6488 Abt. ZI-SG-IM Fax: +49 611-533-2796 Abraham-Lincoln-Str. 34 Mailto:Gerhard.Heiss@ruv.de [...] 10919 54 26_Re: Merging DB2 subsystems16_Swinski, Kenneth32_KSwinski@INTERNAL.MASSMUTUAL.COM31_Thu, 10 Feb 2000 11:21:55 -0500366_us-ascii We did such a move in chunks by moving several databases at a time, binding all the dependent packages, and then binding plans in the "sending" subsystems with PKLISTs specifying the receiving locations. If you have a DBRM that references all the databases in your shop, this technique won't work. All the databases in any one DBRM must move together. [...] 10974 75 16_Re: Mass deletes12_HARBRY ARIZA20_harbry_a@HOTMAIL.COM29_Thu, 10 Feb 2000 08:22:47 PST498_- Eric:

How many rows we are talking about?.

> > > -----Original Message----- > > > From: Eric Robida [SMTP:Eric_Robida@BCBSME.COM] > > > Sent: Thursday, February 10, 2000 1:36 PM > > > To: DB2-L@RYCI.COM > > > Subject: Mass deletes > > > > > > I would like some help with a minor problem. There are four large > > > tables, 24 - 54 millions rows each. > > > We are in the process of moving older data to a set of history tables >then > > > deleting it from the main tables. > > [...] 11050 59 33_Re: Update partitioning key in V514_Larry D Bolick24_ldbolick@DUKE-ENERGY.COM31_Thu, 10 Feb 2000 11:27:28 -0500497_us-ascii Create the table under a different name (same qualifier) in a new partitioned tablespace, unload from original table, load new table, rename original table to some other name, and then rename the new table to the original name. This is a non-destructive procedure for structure modifications that require a drop/recreate. This also provides a quick and simple backout procedure. Don't forget to restrict updates to the original table while you are unloading and loading the new one. [...] 11110 39 23_Migrating from VSE/VSAM12_Davis, Diane22_DDavis@COR.STATE.PA.US31_Thu, 10 Feb 2000 11:23:16 -0500373_iso-8859-1 Hello, We are planning to migrate from VSE/VSAM using DB2 VSAM Transparency for VSE/ESA. We are currently using CICS. We have the redbook for this, but are now concentrating on the changes we may have to do to the data definitions. We are not sure what transparency allows us to keep the same and what will have to be changed even to run with transparency. [...] 11150 81 33_Re: Update partitioning key in V519_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Thu, 10 Feb 2000 10:49:25 -0600408_iso-8859-1 Gerhard, Dean, Based on the amount of data, and the related logging, you could use Dean's idea but instead of doing an UNLOAD/LOAD do INSERT INTO new_partitioned_table SELECT * FROM old_partitioned_table. Also, another thing you might try is to get the "Rebalancing data partitions using ALTER INDEX" capability in V6, via PTF or some other means, from IBM and apply it to your V5 subsystem. [...] 11232 49 33_Re: Update partitioning key in V520_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM31_Thu, 10 Feb 2000 11:59:51 -0500376_us-ascii The update of partitioning key is available in V5 with APAR but your tablespace must be created after APAR was applied. This means if you two tablespaces TABLE_A created before APAR and TABLE_B created after apar then you will be able to update partitioning key of TABLE_B only. If you drop and recreate TABLE_A then you could update that partitioning key also. [...] 11282 16 33_JAVA with DB2 Version 6 on OS/3909_Hari Bawa18_Hari.Bawa@SLMA.COM31_Thu, 10 Feb 2000 12:11:50 -0500350_us-ascii Hi LIst;

Is there any info on Java stored procedures using DB2 OS/390?

Thanks... Hari

================================================ 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. 11299 54 16_Re: Mass deletes16_Duane Lee - ATCX22_DLee@MAIL.MARICOPA.GOV31_Thu, 10 Feb 2000 10:12:29 -0700568_- Why not UNLOAD the data you want to keep and then do a LOAD REPLACE?? You could also use this process to migrate your data to your history tables, i.e., UNLOAD the data to go to history and then LOAD into the history tables.

Duane

> -----Original Message----- > From: Eric Robida [SMTP:Eric_Robida@BCBSME.COM] > Sent: Thursday, February 10, 2000 6:36 AM > To: DB2-L@RYCI.COM > Subject: Mass deletes > > I would like some help with a minor problem. There are four large > tables, 24 - 54 millions rows each. > We are in the process of moving older [...] 11354 34 37_Re: JAVA with DB2 Version 6 on OS/39012_Jim Herrmann12_b3d@CUOL.NET31_Thu, 10 Feb 2000 11:29:44 -0600519_us-ascii You can try these links:

Getting started with Java and DB2 for OS/390 http://www.s390.ibm.com/corner/javadb2.pdf

Java for OS/390 http://www.s390.ibm.com/java/

DB2 UDB for OS/390 V6 Application Programming Guide and Reference for Java(TM) http://www.s390.ibm.com:80/bookmgr-cgi/bookmgr.cmd/BOOKS/DSNJV0G1/CCONTENTS

Hope they help.

Jim Herrmann

Hari Bawa wrote:

> Hi LIst; > > Is there any info on Java stored procedures using DB2 OS/390? > > Thanks... Hari [...] 11389 22 19_Use of Views in DB212_Greg P Mucha24_Greg_P_Mucha@CARADON.COM31_Thu, 10 Feb 2000 17:36:58 +0000308_us-ascii Here's a dumb question that wasn't covered much at all in my sojourn to Dallas for 4 days....

What are the benefits/pitfalls of using Views in DB2? When, if ever, SHOULD they be used? When should they NEVER be used?

More to the point...why would a DBA group REFUSE to use Views?? [...] 11412 135 23_Re: Use of Views in DB215_Macleod, Robert23_Robert.Macleod@ICBC.COM31_Thu, 10 Feb 2000 09:54:27 -0800482_iso-8859-1 Greg :

Craig S. Mullins has a good article out title View on Views. I found it at http://www.platinum.com/craigm/viewnw.htm



Robert N. MacLeod Highlander,There can be only one. Remember, don't sweat the petty things, and don't pet the sweaty things...





-----Original Message----- From: Greg P Mucha [mailto:Greg_P_Mucha@CARADON.COM] Sent: Thursday, February 10, 2000 9:37 AM To: DB2-L@RYCI.COM Subject: Use of Views in DB2 [...] 11548 98 16_Re: Mass deletes18_Kartik Subramanian37_kartik.k.subramanian@BELLATLANTIC.COM31_Thu, 10 Feb 2000 13:05:20 -0500345_us-ascii Eric,

This is how I would do it

1) UNLOAD all the data from the tables. 2) LOAD the history tables with this data 3) Do a LOAD REPLACE on the main tables with the SYSREC statement having a DUMMY //SYSREC DD DUMMY This is like loading a empty dataset into the tables , as good as wiping out all the info in the table. [...] 11647 23 12_SQL Question13_Kurt Sahlberg26_Kurt.Sahlberg@EXPERIAN.COM31_Thu, 10 Feb 2000 12:31:27 -0600481_US-ASCII Greetings, In DB2 OS/390 V5 is possible to get the count of a substr select. If so, what would the syntax look like. ex. SELECT SUBSTR (CHAR (xxxxxxxx),1,7), COUNT(*) FROM ..... WHERE ...... GROUP BY xxxxxxxx

TIA Kurt



===============================================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. 11671 10 26_Re: Mass delete - addendum10_Bill Brown25_bill_brown@SPRINGMAIL.COM31_Thu, 10 Feb 2000 13:38:05 -0500484_- If you need to keep the data in the original table available 24x7 you could unload the data that you want to move to history and load it into the history table. Then using the same file as input issue keyed deletes from a batch application.

================================================ 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. 11682 47 16_Re: SQL Question24_Leblanc, Francis C - CNF23_Leblanc.Francis@CNF.COM31_Thu, 10 Feb 2000 10:59:15 -0800334_- You can do this first by creating a view which incorporates the SUBSTR function and gives it a name. Your code would then select the named SUBSTR column and the count, and the group by clause would then have a name. This works with version 4.

You might also try giving the SUBSTR column a name similar to the following: [...] 11730 57 16_Re: SQL Question17_Patricia Candella21_pcandella@METLIFE.COM31_Thu, 10 Feb 2000 14:05:08 -0500325_us-ascii Kurt, I am not familiar with v5 (yet).. but I know that in V4, you could accomplish this with the use of in-line views.

Something like this ==> select a.colx, count(*) from (sELECT SUBSTR (CHAR (xxxxxxxx),1,7) as colx FROM ..... WHERE ...... ) as a group by a.colx;

Hope this helps you out. Patty [...] 11788 38 16_Re: SQL Question13_Helen Johnson25_Helen_Johnson@RAC.RAY.COM31_Thu, 10 Feb 2000 13:04:39 -0600666_us-ascii It would have to look like this:

SELECT FIRST_SEVEN, COUNT(*) FROM (SELECT SUBSTR (CHAR (xxxxxxxx),1,7) AS FIRST_SEVEN FROM ..... WHERE ...... ) GROUP BY FIRST_SEVEN





Kurt Sahlberg wrote:

> Greetings, > In DB2 OS/390 V5 is possible to get the count of a substr select. > If so, what would the syntax look like. > ex. > SELECT SUBSTR (CHAR (xxxxxxxx),1,7), COUNT(*) FROM ..... > WHERE ...... > GROUP BY xxxxxxxx > > TIA > Kurt > > > ============================================== > 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 [...] 11827 18 19_DB2 5.1 and DB2 6.10_20_John_Lendman@FPL.COM31_Thu, 10 Feb 2000 14:17:12 -0500456_us-ascii Can DB2 5.1 run in the same LPAR as DB2 6.1? Also does anyone know the OS390 required for DB2 6.1?

I know that I could get this information from IBM, but I thought this would be quicker.

John

================================================ 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. 11846 49 16_Re: SQL Question14_Walter Schrott21_walter.schrott@VOL.AT31_Thu, 10 Feb 2000 20:17:42 +0100684_iso-8859-1 Hello,

i think you should use a nested table expression

SELECT F1, COUNT(*) FROM (SELECT SUBSTR(............) AS F1 FROM ..... WHERE ......) AS TEMPTABLE GROUP BY F1

Bye Walter

-----Ursprüngliche Nachricht----- Von: Kurt Sahlberg Newsgroups: bit.listserv.db2-l An: Gesendet: Donnerstag, 10. Februar 2000 19:31 Betreff: SQL Question



> Greetings, > In DB2 OS/390 V5 is possible to get the count of a substr select. > If so, what would the syntax look like. > ex. > SELECT SUBSTR (CHAR (xxxxxxxx),1,7), COUNT(*) FROM ..... > WHERE ...... > GROUP BY xxxxxxxx > > TIA > Kurt > > > To [...] 11896 54 16_Re: SQL Question13_Kurt Sahlberg26_Kurt.Sahlberg@EXPERIAN.COM31_Thu, 10 Feb 2000 13:23:48 -0600686_US-ASCII Thanks Helen, That's what I was looking for. Kurt





>>> Helen Johnson 02/10/00 01:04PM >>> It would have to look like this:

SELECT FIRST_SEVEN, COUNT(*) FROM (SELECT SUBSTR (CHAR (xxxxxxxx),1,7) AS FIRST_SEVEN FROM ..... WHERE ...... ) GROUP BY FIRST_SEVEN





Kurt Sahlberg wrote:

> Greetings, > In DB2 OS/390 V5 is possible to get the count of a substr select. > If so, what would the syntax look like. > ex. > SELECT SUBSTR (CHAR (xxxxxxxx),1,7), COUNT(*) FROM ..... > WHERE ...... > GROUP BY xxxxxxxx > > TIA > Kurt > > > ============================================== > To change your [...] 11951 62 23_Re: Use of Views in DB216_Michael McMorrow25_michael.g.mcmorrow@AIB.IE31_Thu, 10 Feb 2000 20:03:59 -0000638_ISO-8859-1 Greg, We started out using views but have mainly discontinued. On our first DB2 system many years ago the consultancy firm we hired proposed that application sql should only access views (one per table with all columns) to provide a bufferzone between applications and base tables. In theory, tables could subsequently change with new columns added etc. with the option to leave those views/applications untouched and set up new expanded ones for the applications interested in the new stuff. We quickly learned that the complexity and overhead of spawning lots of such evolving views over time created more problems than [...] 12014 35 23_Re: DB2 5.1 and DB2 6.140_Andersen, Cindy (CC-Application Support)26_Cindy.Andersen@CONAGRA.COM31_Thu, 10 Feb 2000 13:59:47 -0600393_iso-8859-1 For DB2 6.1, the minimum level of OS/390 is 2.6, according to our contact at IBM.

-----Original Message----- From: John_Lendman@FPL.COM [mailto:John_Lendman@FPL.COM] Sent: Thursday, February 10, 2000 1:17 PM To: DB2-L@RYCI.COM Subject: DB2 5.1 and DB2 6.1



Can DB2 5.1 run in the same LPAR as DB2 6.1? Also does anyone know the OS390 required for DB2 6.1? [...] 12050 85 16_Re: Mass deletes12_HARBRY ARIZA20_harbry_a@HOTMAIL.COM29_Thu, 10 Feb 2000 12:17:28 PST532_- Eric:

I've got almost one activity like you do and the way find to do that is executing selected unload with the rows that you want to move to the history tablespaces and then load it with resume(yes) on it and now you need to delete those rows from the source table , well execute a selected unload with the rows that you want to keep and load it againg with resume(no) replace log(no) . Before that executed a quiesce point just in case that you find any problem. I could be better than executing a mass delete on a [...] 12136 32 13_TMON/DB2 Help11_Suresh Sane21_data_arch@HOTMAIL.COM29_Thu, 10 Feb 2000 14:36:08 CST575_- Need some help from any experienced TMON/DB2 users. My clients is a small/medium-sized DB2 OS/390 V5 shop using TMON 3.2 .

1. How large do you define the archive log datasets? How often do they switch during the day? 2. Do you use the load history option (sql inserts using dynamic sql provided by Landmark)? In our case, this is very resource-intensive, requiring hours cpu just for 1 day to make it unusable. 3. If you use a home-grown method of summarizing historical data, how far back do you keep the thread detail history? 4. We have found that filtering [...] 12169 23 23_Re: DB2 5.1 and DB2 6.10_29_STANLEY_GOODWIN@MECH.DISA.MIL31_Thu, 10 Feb 2000 15:47:15 -0500574_us-ascii I don't believe the min level of OS/390 is 2.6 for DB2 6.1, if one reads the program directory for 6.1 it says on page 24 that DB@ UDB operates on any processor supported by OS/390 version 1 release 3.



Stan Goodwin DAC MBG DB2 Support

Stanley_goodwin@mech.disa.mil DSN: 430-4335 Outside:- 717-605-4335

================================================ 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. 12193 64 27_DB2/MVS distributed problem11_Hilton Tina15_HiltonT@TCE.COM31_Thu, 10 Feb 2000 15:57:10 -0500501_iso-8859-1 We've only used distributed process from PC to the mainframe before, but now I'm trying to use it between the DB2's and I'm having inconsistent results. I believe I've checked everything, but still can't figure out what's wrong. I have 3 DB2 systems (V4) on 3 different lpars. All systems are set up the same for distributed work as far as I can see, but they all behave differently. If anyone can tell me what's wrong or point me in the right direction, I would really appreciate it. [...] 12258 86 31_Re: DB2/MVS distributed problem0_14_dcreed@CSC.COM31_Thu, 10 Feb 2000 15:59:42 -0600687_us-ascii Did you grant select to public at all locations? or grant select to public? Regards Danny







HiltonT@tce.com on 02/10/2000 02:57:10 PM

Please respond to DB2-L@ryci.com

To: DB2-L@ryci.com cc: (bcc: Daniel S Creed/HI/CSC) Subject: DB2/MVS distributed problem





We've only used distributed process from PC to the mainframe before, but now I'm trying to use it between the DB2's and I'm having inconsistent results. I believe I've checked everything, but still can't figure out what's wrong. I have 3 DB2 systems (V4) on 3 different lpars. All systems are set up the same for distributed work as far as I can see, [...] 12345 62 23_Re: DB2 5.1 and DB2 6.112_Kirk Hampton16_khampto1@TXU.COM31_Thu, 10 Feb 2000 16:52:24 -0600391_us-ascii John, Certainly you can have multiple versions of DB2 coexisting on the same LPAR, but you would have to have TSO users and batch jobs allocate the proper release level libraries corresponding to the DB2 subsystem they intend to access. Also the highest version of SDSNLINK would need to be in LNKLST and all DB2's would use this same ERLY code, but it is downward-compatible. [...] 12408 247 39_Re: Disaster Recover using EMC products11_Ron Haupert23_rhaupert@GOLDENGATE.NET31_Thu, 10 Feb 2000 17:11:53 -0600715_iso-8859-1 I/O elongation can occur when synchronously remotely mirroring over long distances (hundreds of miles). Technology is wonderful but we have been unable to change the speed of light - so far. To overcome performance degradation for synchronous remote mirroring, we provide configurations that use combinations of synchronous and "asynchronous" mirroring together. Multi-hop SRDF configurations allow synchronous mirroring for short distances and then data is moved long distances (thousands of miles) using asynchronous methods. These configurations minimize I/O elongation while allowing long distance mirroring. All of the data is protected by ConGroup so data is consistent when it arrives at the [...] 12656 38 30_SMF 42 reporter product naming15_Jim Lewandowski22_jlewand@STARNETINC.COM30_Mon, 7 Feb 2000 07:48:15 -0600567_us-ascii For our SMF 42 reporter product, we have the following suggest names/variations. I have added my own scattered within to protect myself from possible marketing-faux-pas (sp?) embarrassment.

DB2IOSPY or I/O Spy for DB2 I/O Sentry or I/O Sentry for DB2 I/O Sentinel or I/O Sentinel for DB2 DB2IOSOURCE or I/O Source for DB2 DB2 I/O Advisor or I/O Advisor for DB2 I/O Miner or I/O Miner for DB2 I/O Analyzer or I/O Analyzer for DB2 Show I/O Info or ? I/O Query or I/O Query for DB2 I/O Quarry or I/O Quarry for DB2 FIRE (Find I/O Response Events or [...] 12695 18 8_DSNV086E17_suthep thampranee18_suthepth@SCB.CO.TH31_Fri, 11 Feb 2000 10:39:17 +0700411_Windows-874 Hello Have anybody seen this message(DSNV086E) before? I really want to know why DB2 down and how to fix it. Any suggestions much appreciated. Thank you. JOE

================================================ 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. 12714 83 12_Re: DSNV086E0_29_Greg.Palgrave@BANKWEST.COM.AU31_Fri, 11 Feb 2000 12:02:23 +0800683_us-ascii Joe,

According to the manual, there should be a reason code in the message, which may help you. Check your MSTR , DBM1 & IRLM address spaces for other messages if the cause isn't clear.

DSNV086E DB2 ABNORMAL TERMINATION REASON=xxxxxxxx

Explanation: The DB2 subsystem is being abended, because an error that cannot be corrected has occurred. This message, which is not automatically deleted from the operator console, is issued during DB2 abnormal termination. 'REASON' is the subsystem termination reason code. If DB2 abnormal termination is invoked multiple times, the subsystem termination reason code that accompanies this message is the reason [...] 12798 23 39_Re: Disaster Recover using EMC products6_TaoTao19_taotao@ONLINE.SH.CN31_Fri, 11 Feb 2000 13:23:29 +0800502_us-ascii My question is where SRDF and ConGroup run. On mainframe? On any open server connected to the storage or on the storage controller? What is the difference between EMC SRDF and IBM XRC? Becuase XRC has a address space on the mainfram, it can not handle the open system data on the storage server (IBM SHARK or EMC Symmextrix 5830/5630) in asynchronous disaster recovery method. If use SRDF and ConGroup, the volumes defined to ConGroup are just like as in synchronous method. Is it right? [...] 12822 168 39_Re: Disaster Recover using EMC products6_TaoTao19_taotao@ONLINE.SH.CN31_Fri, 11 Feb 2000 14:01:56 +0800600_us-ascii Ron,

If the rolling disaster is recovered, will the mirror continue work? if yes, how does SRDF know the I/O sequence? Does it has a log or a journal?

Thanks

TaoTao



Ron Haupert wrote:

> EMC Symmetrix Remote Data Facility (SRDF) provides remote mirroring to > disaster recovery or to disaster restart sites. When SRDF synchronous mode > is used, the state of the data at the remote site is exactly the state of > the data at the local site at any instant in time. If the local site fails, > the operating system, DBMS, and all applications can [...] 12991 23 34_Re: SMF 42 reporter product naming14_Massimo Scarpa16_mscarpa@CESVE.IT31_Fri, 11 Feb 2000 08:35:32 +0100559_us-ascii SMF SURVEYOR & REPORTER for DB2 >>>>>>> SSR for DB2

(you can start product with command TSO SSRx.....it sounds good for me.....)

I think you need 'for DB2' because it's the product target.

Good luck

Max Scarpa Data & System admin (hoping in a free copy of your product....:-)))) )

================================================ 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. 13015 22 23_Re: Use of Views in DB214_Massimo Scarpa16_mscarpa@CESVE.IT31_Fri, 11 Feb 2000 08:45:01 +0100529_us-ascii If a DBA group refuse view it's an big error, in my opinion. Views are pratical and if there are

errors you can delete/define them easily. In our previous shop views were widely used.

I don't see any big problem in using views.

Regards Max Scarpa Data admin

================================================ 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. 13038 41 16_DB2 SQL Analysis41_=?iso-8859-1?Q?Cl=E1udio_Meluzzi_Mendes?=18_P000002@BEG.COM.BR31_Fri, 11 Feb 2000 10:23:17 -0200590_iso-8859-1 Dear list,

recenty I saw a Brazilian software for DB2 which works as a "filter" for DB2 applications, reducing the SQL statement and program analysis process done by DBAs. It has some 45 rules such as "program which updates DB2 data must have at least one commit", "usage of SELECT * is forbidden" or "Table Space Scan is not allowed". This software uses DB2 catalog, plan_table results and program scan (for COBOL only) in the analysis. Every program receives a status - green if it compliant with all rules, yellow for warnings and red if some rule is broken. Thus [...] 13080 43 12_Re: DSNV086E19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Fri, 11 Feb 2000 06:30:16 -0600647_Windows-874 Joe, What was reason code, circumstances, environment? Any ABEND messages issued prior to this one that may apply?

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 of [...] 13124 103 31_Re: DB2/MVS distributed problem11_Hilton Tina15_HiltonT@TCE.COM31_Fri, 11 Feb 2000 07:40:10 -0500422_iso-8859-1 Thanks Danny. That took care of the -551 errors. Now if I can figure out the -904 resource not available...

Tina

-----Original Message----- From: dcreed@CSC.COM [mailto:dcreed@CSC.COM] Sent: Thursday, February 10, 2000 5:00 PM To: DB2-L@RYCI.COM Subject: Re: DB2/MVS distributed problem



Did you grant select to public at all locations? or grant select to public? Regards Danny [...] 13228 102 31_Re: DB2/MVS distributed problem19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Fri, 11 Feb 2000 06:47:09 -0600628_iso-8859-1 Tina, Was xxxxDIST started and connected in each subsystem when you attempted to go from DB2T to DB2D?

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 314-235-6854 and delete this message immediately from your computer. Any other use, retention, dissemination, forwarding, printing, or copying [...] 13331 61 17_Re: TMON/DB2 Help11_Hilton Tina15_HiltonT@TCE.COM31_Fri, 11 Feb 2000 07:49:44 -0500486_iso-8859-1 Suresh,

My answers are below your questions.

Tina

-----Original Message----- From: Suresh Sane [mailto:data_arch@HOTMAIL.COM] Sent: Thursday, February 10, 2000 3:36 PM To: DB2-L@RYCI.COM Subject: TMON/DB2 Help



Need some help from any experienced TMON/DB2 users. My clients is a small/medium-sized DB2 OS/390 V5 shop using TMON 3.2 .

1. How large do you define the archive log datasets? How often do they switch during the day? [...] 13393 36 12_Re: DSNV086E12_HARBRY ARIZA20_harbry_a@HOTMAIL.COM29_Fri, 11 Feb 2000 04:56:19 PST690_- Suthep:

DB2 goes down with this message but what is the reason code?



>From: suthep thampranee >Reply-To: DB2 Data Base Discussion List >To: DB2-L@RYCI.COM >Subject: DSNV086E >Date: Fri, 11 Feb 2000 10:39:17 +0700 > >Hello > Have anybody seen this message(DSNV086E) before? > I really want to know why DB2 down and how to fix it. >Any suggestions much appreciated. >Thank you. > JOE > >================================================ >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. [...] 13430 114 31_Re: DB2/MVS distributed problem11_Hilton Tina15_HiltonT@TCE.COM31_Fri, 11 Feb 2000 07:59:10 -0500486_iso-8859-1 Yes, the distributed address spaces are all started. The only connection that doesn't work now is from DB2T to DB2D, where I get the -904. I can connect from DB2D to DB2T with no problem.

Changing the grant to 'public at all locations' fixed the -551 problem.

Tina

-----Original Message----- From: DAVIS, RICK (SBCSI) [mailto:RD8246@MOMAIL.SBC.COM] Sent: Friday, February 11, 2000 7:47 AM To: DB2-L@RYCI.COM Subject: Re: DB2/MVS distributed problem [...] 13545 133 31_Re: DB2/MVS distributed problem11_Hilton Tina15_HiltonT@TCE.COM31_Fri, 11 Feb 2000 08:06:58 -0500335_iso-8859-1 In DB2T, I deleted the row in syslocations and syslunames pointing to LUDB2D. Then I stopped/started DDF on DB2T and DB2D. Now everything works. Either hadn't stopped/started DDF before (although I thought I did that to all 3 DB2's) or there was some garbage in the CDB on DB2T.

Thanks to everyone for the help. [...] 13679 56 13_TMON/DB2 Help11_Sam Knutson21_sknutson@LANDMARK.COM31_Fri, 11 Feb 2000 08:08:42 -0500654_us-ascii Suresh Sane wrote: >Need some help from any experienced TMON/DB2 users. My clients is a >small/medium-sized DB2 OS/390 V5 shop using TMON 3.2 .

>1. How large do you define the archive log datasets? How often do they >switch during the day? See this article on the customer service web site for Landmark (you will need a userid and password but you can get one by contacting Customer Service without any wait) http://www.support.landmark.com/Common/LGGNGED.htm also the command F lfsname,DISPLAY SOURCES will give you a nice display of the record types and the time ranges currently in each logical file. This [...] 13736 126 31_Re: DB2/MVS distributed problem19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Fri, 11 Feb 2000 07:12:59 -0600581_iso-8859-1 Tina, Isn't DSNESPCS SPUFI? 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." [...] 13863 256 39_Re: Disaster Recover using EMC products19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Fri, 11 Feb 2000 07:40:49 -0600572_iso-8859-1 Hi Ron, Thanks for clearing up a couple items. I'd like to follow up on Shan's original concern. Many sites, in order to improve DB2 availability/reliability put BSDS01 on one controller and BSDS02 on another. Same for the active logs. Based on your explanation and probably my lack of understanding, my concern with SRDF/ConGroup is that when I arrive at the hotsite the BSDSs and/or Active logs don't match. How does SRDF/ConGroup ensure this doesn't occur? Or can it? And its just another "inflight situation" we would have to be prepared to deal with. [...] 14120 37 31_Re: DB2/MVS distributed problem15_Gordon Fishwick39_Gordon.Fishwick@SCOTTISH-SOUTHERN.CO.UK31_Fri, 11 Feb 2000 13:48:40 +0000717_us-ascii Rick (Sorry to jump in Tina),

DSNESPCS is indeed the SPUFI Cursor Stability plan. Thought I'd better get in while I can as this is something I do actually know!

Best regards,

Gordon Fishwick DB2 Systems Programmer Scottish and Southern Energy



********************************************************************** The information in this E-Mail is confidential and may be legally privileged. It may not represent the views of Scottish and Southern Energy plc. It is intended solely for the addressees. Access to this E-Mail by anyone else is unauthorised. If you are not the intended recipient, any disclosure, copying, distribution or any action taken or omitted to [...] 14158 69 22_extent numbers and SMS0_27_BoSDBA@BANKOFSCOTLAND.CO.UK31_Fri, 11 Feb 2000 13:59:18 +0000537_us-ascii Dear list, I know this has been asked before, but I don't think there was a consensus. I wanted to check what the feeling was now. We are on DB2 v 5.1 OS/390 2.6. We use RAID technology. We have some older Db2 systems which have user defined datasets; other systems are DB2 stogroup managed, with a list of DASD devices named in the storage group. We DBAs had some problems managing space with this system, so for our latest system we decided to use a DB2 stogroup defined with Volumes (*) and got a rule set up within SMS [...] 14228 30 31_Tmon DB2 / Supertrace question.32_Cuneyt Goksu (Garanti Teknoloji)22_CuneytG@GARANTI.COM.TR31_Fri, 11 Feb 2000 16:18:26 +0200364_- Tmon for DB2 does not show the SQL statements in supertrace in statement PREPARE.

The explanation says

TMDB21302I - NO EXPLAIN INFO. IFCID 22 (MINI PLAN) NOT IN TRACE

PLAN : PRODFRNI LOCATION : PDB2DDF STMT TYP: PREPARE

PACKAGE : FRNMSP01 COLLID : FRNIBMCP STMT Ö : 2453



But IFCID 22 is ýncluded in the Trace Set. [...] 14259 260 44_Re: CA-Platinum Quickcopy creating bad tapes13_Thomas, Janis19_Janis.Thomas@CA.COM31_Fri, 11 Feb 2000 10:10:56 -0500740_iso-8859-1 Below is the information hiper from Computer Associates on this issue:

To: Computer Associates Quick Copy Users

From: Technical Support Computer Associates, Inc.

Date: February 4, 2000

Subject: Possible invalid copies generated by Quick Copy following IBM maintenance, APAR OW34071.

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

This memo describes a potential problem that has been discovered within Quick Copy. This problem can potentially exist on any version of Quick Copy. IBM has indicated that with the PTFs represented by APAR OW34071, there can be a valid return code issued from within the IBM low-level EXCP write process when the write was not successful. Customers who [...] 14520 15 22_DB2 Security thro RACF8_Kurian B23_Bejoy_Kurian@MAY-CO.COM31_Fri, 11 Feb 2000 08:54:40 -0600418_us-ascii I would like to hear the pros and cons on DB2 security through RACF. Did any shops implemented this?? Could you please share your experiences?. Thanks in Advance Bejoy

================================================ 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. 14536 27 23_DB2 Row Archive Manager15_Jones, Scott A.15_scjones@CSW.COM31_Fri, 11 Feb 2000 09:13:06 -0600484_iso-8859-1 Has any one out there had any experiece with IBM's DB2 Row Archive Manager? Have you looked at it and decided to go with a competitor? Are there other products that function the same, i.e. Archive for DB2 from Princeton/SOFTECH, other vendors?

Thanks.



Scott Jones Data Base Consultant

Central and South West Services, Inc. Voice: (918) 594-2624 Fax: (918) 594-3877 www.csw.com scjones@csw.com [...] 14564 51 27_Re: DB2 Row Archive Manager13_Thomas, Janis19_Janis.Thomas@CA.COM31_Fri, 11 Feb 2000 10:48:58 -0500544_iso-8859-1 Computer Associates has a new product, RC/Extract, currently in Beta status, which is scheduled for GA release at the end of March. Among other functionality, the GA release of RC/Extract will contain an archival function for DB2 on OS/390.



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



-----Original Message----- From: Jones, Scott A. [mailto:scjones@CSW.COM] Sent: Friday, February 11, 2000 9:13 AM Subject: DB2 Row Archive Manager [...] 14616 46 20_REORG option DISCARD12_Claude Birtz24_Claude.Birtz@CIE.ETAT.LU31_Fri, 11 Feb 2000 17:09:58 +0100426_us-ascii I'm testing the new parameter DISCARD of REORG and I have problems when I use the expression CURRENT DATE. Here is my statement:

REORG TABLESPACE CIXXXDB.CIXXXS05 DISCARD FROM TABLE CIDB2CB.TEST WHEN ( DATUM1 = CURRENT_DATE )

I get the message:

DSNU049I +DB2T DSNUUGMS - INVALID OPERAND '' FOR KEYWORD 'DATUM1'

When I use the statement CURRENT DATE (without an underscore) like this: [...] 14663 48 26_Re: DB2 Security thro RACF0_22_Rohn.Solecki@MTS.MB.CA31_Fri, 11 Feb 2000 10:11:00 -0600607_us-ascii We use it with no problem. We currently have 2 minor issues. We are coming to the conclusion that on the development side, we have too many RACF groups. Our current thinking is that in development maybe a better approach would be, "if you are a developer, you have update access to all dev datasets and DBs." Only in exceptional circumstances would access be restricted in development, ie testing with confidential data. Currently we have RACF setup for each separate application and development environment, so every developer ends up with dozens of RACF authorizations. It is a real pain to [...] 14712 26 51_DRDA COMMIT/ROLLBACK after calling stored procedure11_Joe Luthman22_jluthma@BGNET.BGSU.EDU31_Fri, 11 Feb 2000 11:24:20 -0500556_us-ascii We have JAVA programs calling stored procedures on v6 of DB2 for OS390 through a DB2 Connect machine (AIX). When the calling program sends a ROLLBACK, data is still being externalized. The stored procedure packages are bound with RELEASE(COMMIT). Thread reuse and connection pooling are in effect.

1) Before the programmer knew that COMMIT/ROLLBACK had to be done from the calling program and could NOT be issued from within the stored procedure, we were still seeing data get externalized (w/o explicit COMMIT) How could this happen? [...] 14739 19 32_DB2 UDB on RS/6000 HACMP cluster17_Gabriel Collantes24_gcollantes@HQ.SEQUEL.NET31_Fri, 11 Feb 2000 10:14:09 -0600499_- Hi,

Can anyone point me to some guide/reference material on getting DB2 UDB (5.2/6.1) into a RS/6000 HACMP cluster.

I found a redbook on the subject for Oracle, but none for DB2 UDB (duh!).



Thanks in advance, Gabriel Collantes

================================================ 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. 14759 64 24_Re: REORG option DISCARD0_19_mike.holmans@BT.COM31_Fri, 11 Feb 2000 16:24:12 -0000661_- Have you implemented *all* the PTFs to V5 for the updates to REORG? See APAR PQ23219 which added the capability to interpret special registers, which wasn't in the first version of the enhancements.

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

> -----Original Message----- > From: Claude Birtz [SMTP:Claude.Birtz@CIE.ETAT.LU] > Sent: Friday, February 11, 2000 4:10 PM > To: DB2-L@RYCI.COM > Subject: REORG option DISCARD > > I'm testing the new parameter DISCARD of REORG and I have problems when > I use the expression CURRENT DATE. Here is my statement: > > REORG TABLESPACE CIXXXDB.CIXXXS05 > DISCARD FROM TABLE CIDB2CB.TEST [...] 14824 53 55_Re: DRDA COMMIT/ROLLBACK after calling stored procedure10_Rao, Peavy26_Peavy.Rao@TRANSAMERICA.COM31_Fri, 11 Feb 2000 10:29:03 -0600619_- What is COMMIT_ON_RETURN value for the DB2 SP. On DB2 V5 SP this should be 'N' for NO.

> ---------- > From: Joe Luthman[SMTP:jluthma@BGNET.BGSU.EDU] > Reply To: DB2 Data Base Discussion List > Sent: Friday, February 11, 2000 10:24 AM > To: DB2-L@RYCI.COM > Subject: DRDA COMMIT/ROLLBACK after calling stored procedure > > We have JAVA programs calling stored procedures on v6 of DB2 for OS390 > through a DB2 Connect machine (AIX). When the calling program sends a > ROLLBACK, data is still being externalized. The stored procedure packages > are bound with RELEASE(COMMIT). > Thread reuse and connection [...] 14878 84 38_Re: Using DSN1COPY for Backup/Recovery15_Di Carlo, Donna23_Donna_Bermender@BMC.COM31_Fri, 11 Feb 2000 10:40:22 -0600481_iso-8859-1 David,

You are correct, DSN1COPY only copies those pages that are non-zero. When laying this copy back down, try using the FULLCOPY parm. I believe this will tell DSN1COPY that it is OK to have a break in the page numbers.

Donna Di Carlo BMC Software

-----Original Message----- From: David Wilson [mailto:DWILSON@CEDELGLOBALSERVICES.COM] Sent: Wednesday, February 09, 2000 7:03 AM To: DB2-L@RYCI.COM Subject: Using DSN1COPY for Backup/Recovery [...] 14963 49 36_Re: DB2 UDB on RS/6000 HACMP cluster14_Philip Gunning20_pgunning@BOSCOVS.COM31_Fri, 11 Feb 2000 11:53:19 -0500463_iso-8859-1 Gabriel, I just got the manuals in; they are: SC23-4276-01 HACMP V4.3 Concepts and Facilities SC23-4277-01 " " PlanningGuide SC23-4278-01 "" Installation SC23-4279-01 ""Admin guide SC23-4280-01 ""Troublshooting Guide SC23-4281-01 ""Program Locking Apps SC23-4282-01 "" Program Client Apps SC23-4285-01 "" Master Glossery SC23-4306-00 Installation Guide vol 2.

You should be able to find these on www.software.ibm.com via a search. HTH Phil [...] 15013 83 55_Re: DRDA COMMIT/ROLLBACK after calling stored procedure13_Vaughan, Mike26_Vaughan.Mike@PRINCIPAL.COM31_Fri, 11 Feb 2000 10:54:26 -0600654_iso-8859-1 Another thing that could cause this would be the "autocommit" setting on the JAVA connection. Have your JAVA developer check the connection autocommit setting immediately prior to making the stored procedure call (depending on what type of connection is being used, there should be a getautocommit() or isautocommit() method, or something along those lines). I would take a guess that autocommit is turned on, causing a commit to be issued after the stored procedure is called. They may need to turn autocommit off before making the stored procedure call. ...which leads to some interesting implications when using connection pooling. My [...] 15097 135 26_Re: extent numbers and SMS10_Bill Welch27_spgwdw@UCLINK4.BERKELEY.EDU31_Fri, 11 Feb 2000 09:08:21 -0800615_us-ascii Claire,

I've never heard of this 30,30 strategy. We try to size our PRIQTY to fit the expected size of the tablespace or indexspace.

We have RAID technology, so we aren't as concerned about number of extents for the reasons you've mentioned. Since our database is growing rather rapidly, we find it useful to monitor extents and ALTER the PRIQTY of indexspaces or tablespaces that go over a certain number of extents, then REORG or RECOVER without the REUSE parameter. If we don't do this, we sometimes run into the maximum extents (especially indexes) and get called by the programmers [...] 15233 70 20_Re: DB2 SQL Analysis11_Suresh Sane21_data_arch@HOTMAIL.COM29_Fri, 11 Feb 2000 11:53:25 CST335_- Platinum has a host of products - Plan Analyzer, SQl-Ease and an Index Analyzer. I have used Plan Analyzer extensively and use it to weed out the trivial and focus on the real problem SQL. I think BMC also has a simila product (Explorer?). Candle's DB explain, Landmark's explain within TMON are some of the other alternatives. [...] 15304 31 26_Re: extent numbers and SMS15_Jim Lewandowski22_jlewand@STARNETINC.COM30_Mon, 7 Feb 2000 23:56:51 -0600398_us-ascii BoSDBA@BANKOFSCOTLAND.CO.UK wrote:

To support their case, they' ve said that some DB2 shops define all > their tablespaces as priqty 30, seqty 30. In this way, when datasets reach 119 > extents, they also reach the 2 gig dataset limit and another primary extent is > allocated. > > We in DBA have a number of issues with this: > 1. Do number of extents still matter with RAID? [...] 15336 41 24_DB2 hardware compression15_Jim Lewandowski22_jlewand@STARNETINC.COM30_Tue, 8 Feb 2000 00:02:37 -0600409_us-ascii Could someone out there run a test for us on the DB2-L list?

Create 2 tables (duplicate) data. One table with compression, the other without. Make sure they are decent in size (50 cyls or maybe larger). Ideally without ANY indexes over them.

Then, against EACH table, run:

SELECT COUNT(*) FROM table WHERE some-data-column-that-is-not-in-an-index = 'EWREWREWR' WITH UR!!!!! [...] 15378 64 23_Re: Use of Views in DB211_Suresh Sane21_data_arch@HOTMAIL.COM29_Fri, 11 Feb 2000 12:03:11 CST631_- I disagree.

For end users, views that are row-subsets (using WHERE), column subsets (for security) and those with a pre-defined join or grouping do indeed make sense. But for application developers the widespeard use SELECT * FROM views causes more trouble than the benfit. Specifically:

1. Adding a column to a table (not null with default or nulls allowed) would required no outage. With views, the view must be dropped, re-created and all plans/packages rebound. 2. View Materialization is a danger that lurks around the corner. PeopleSoft has a tenedency to define 1 view per panel ("a logical table"). In [...] 15443 58 28_Re: DB2 hardware compression14_Joel Goldstein29_joel_goldstein@COMPUSERVE.COM31_Fri, 11 Feb 2000 13:32:12 -0500368_ISO-8859-1 Unless you are running in a complete stand-alone environment there will be enough small variation in CPU times to obviate any meaningful measurement of I/O costs. You would also need a very large number of I/Os, and then a substantial difference (reduction) of I/Os. Running a meaningful benchmark is not a trivial undertaking. Regards, Joel Goldstein [...] 15502 75 28_Re: DB2 hardware compression15_Jim Lewandowski22_jlewand@STARNETINC.COM30_Tue, 8 Feb 2000 00:45:18 -0600374_us-ascii I'm not quite following.

Are you saying that the CPU related to I/O times WILL be small enough to not "worry" about?

I'm just trying to see, all things relatively being equal, what the CPU for accessing the rows on the data pages. Asynch prefetch is charged to DBM1 anyway so I assume this would NOT show up under IN-DB2-TIME for the SQL/thread. [...] 15578 92 23_Re: Use of Views in DB211_Paul Mayhew15_mayhewp@CIBC.CA31_Fri, 11 Feb 2000 13:30:50 -0500466_us-ascii Where I used to work.........

We did use views very sucessfully, (select * from whatever) and found no problem. Mostly we liked the extra layer between the application and the data.... (true below though).

We eventually changed this to use alias definitions instead, which also worked quite nicely.....



Regards,







From: "Suresh Sane" [SMTP:data_arch@HOTMAIL.COM] AT NABB on 02/11/2000 01:03 PM [...] 15671 16 12_Mass deletes11_Eric Robida22_Eric_Robida@BCBSME.COM31_Fri, 11 Feb 2000 11:47:32 -0500433_us-ascii I'd like to thank those who responded to my question about mass deletes. It has not been determined the method that we'll use yet. Thanks for providing all the food for thought. Eric

================================================ 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. 15688 39 12_RLIMIT Setup41_=?iso-8859-1?Q?Cl=E1udio_Meluzzi_Mendes?=18_P000002@BEG.COM.BR31_Fri, 11 Feb 2000 17:00:46 -0200713_iso-8859-1 I am trying to set up the resouce limit (RLIMT) feature in our development subsystem. For the first test I populated the SYSIBM.DSNRLST01 table with the folowing:

---------+---------+---------+---------+---------+---------+--------- AUTHID PLANNAME ASUTIME LUNAME RLFFUNC RLFBIND RLFCOLLN ---------+---------+---------+---------+---------+---------+--------- M084353 0

The Administration Guide says that if I have a line like this one in the table the "limit applies to a specific AUTHID for all PLANNAMES at the local location". After that a START RLIMIT ID=01 was issued and a DISPLAY RLIMIT returned "DSNT700I @ SYSIBM.DSNRLST01 IS THE ACTIVE RESOURCE LIMIT SPECIFICATION TABLE". [...] 15728 165 26_Re: extent numbers and SMS15_Loria, Susan C.27_Susan_C_Loria@TVRATINGS.COM31_Fri, 11 Feb 2000 14:07:13 -0500551_iso-8859-1 Claire,

We are a shop using OS/390 V2.6, DB2 V4.1 (soon to go to DB2 V6.1) and using DFSMS 1.4. We are a small DB2 shop running one application -- largest production table about 4 million rows, approx 1800 cylinders of data.

A few days ago I was reading in the Redbooks about DB2 and DFSMS storage to help me prepare better for a meeting with our storage admins about some similar issues. We're using a mixture of RAMAC and RVA for some of our DB2 volumes. In the near future, the storage group will be moving us to 100% [...] 15894 84 20_Re: DB2 SQL Analysis14_Holman, Vivian27_Vivian_Holman@COMPUWARE.COM31_Fri, 11 Feb 2000 14:05:08 -0500424_iso-8859-1 Compuware has File-AID for DB2 which will execute and/or explain your SQL directly from your source code without having to compile. You can explain a single statement or a set of statements grouped by DBRM, plan, collection or package, online or in batch. Also, the rules are totally customizable.

If you want more information try here: http://www.compuware.com/products/file/products/fadb2/index.htm [...] 15979 42 36_Re: DB2 UDB on RS/6000 HACMP cluster19_Philip Nelson (DBA)21_teamdba@ATTGLOBAL.NET31_Fri, 11 Feb 2000 13:03:24 -0600692_us-ascii There is a whole section in the V6.1 Admin Guide on setting up HACMP clusters for DB2.

We have these as well, so if you need any help ...

Phil Nelson Senior DBA Scottish Widows (teamdba@attglobal.net) (philip.nelson@scottishwidows.co.uk)



On Fri, 11 Feb 2000 10:14:09 -0600, Gabriel Collantes wrote:

>Hi, > >Can anyone point me to some guide/reference material on getting DB2 UDB >(5.2/6.1) into a RS/6000 HACMP cluster. > >I found a redbook on the subject for Oracle, but none for DB2 UDB (duh!). > > >Thanks in advance, >Gabriel Collantes > >================================================ >To change your subscription options or to cancel [...] 16022 96 55_Re: DRDA COMMIT/ROLLBACK after calling stored procedure11_Joe Luthman22_jluthma@BGNET.BGSU.EDU31_Fri, 11 Feb 2000 14:18:00 -0500456_us-ascii Thank you, Mike. The "autocommit" setting on the JAVA connection was the trick! Now, for the first time, we feel comfortable with the data integrity of the approach that we are using. The connection pooling will not pose a problem for us because we never want any of the web users to autocommit. If we want that, we can set the individual stored procedure to COMMIT_ON_RETURN = 'Y'. Especially, thank you for the quick response. -joe luthman [...] 16119 70 28_Re: DB2 hardware compression14_Richard Yevich21_ryevich@ATTGLOBAL.NET31_Fri, 11 Feb 2000 13:14:46 -0500456_iso-8859-1 Jim,

There was a very comprehensive analysis and presentation on this at several IDUG's last year by Bob Omerza from UPS. It was very complete, and presented the results of all their analysis to determine the impacts, etc. Then they used it to great advantage. Strongly suggest that you try and get a copy of the presentation, and the tape of the presentation from IDUG (www.idug.org). It will answer any questions you have on this. [...] 16190 57 16_Re: RLIMIT Setup13_Moss, William15_MossW@AETNA.COM31_Fri, 11 Feb 2000 14:25:33 -0500557_- Since QMF is bound with packages, I think you need RLFFUNC=2.

Also, If MO84353 has SYSADM or SYSCTL authority, it will not use the RLIMIT.



Bill Moss Aetna/US Healthcare.



-----Original Message----- From: Claudio Meluzzi Mendes [mailto:P000002@BEG.COM.BR] Sent: Friday, February 11, 2000 2:01 PM To: DB2-L@RYCI.COM Subject: RLIMIT Setup



I am trying to set up the resouce limit (RLIMT) feature in our development subsystem. For the first test I populated the SYSIBM.DSNRLST01 table with the folowing: [...] 16248 67 36_Re: Read Only Access to DB2 Clusters24_Humphris,Richard P.(NXI)24_Richard.Humphris@CNA.COM31_Fri, 11 Feb 2000 14:26:15 -0600610_- Read Chuck Hoover's "Everything you ever wanted to know about DB2 buffering (but were afraid to ask ). It's online (in PowerPoint97) at: http://www.compuware.com/products/db2.htm this will explain why you don't want to read DB2 clusters directly.

But, since you seem to be willing to sacrifice consistency for performance; why don't you try just binding your SQL with ISOLATION(UR), uncommitted read. The advantage this has is that UR can read a logically consistent view of the data (even if some of it is wrong) without having to spend much time in lock wait. See Chuck Hoover's discussion. [...] 16316 22 38_Using the DB2 log for billing purposes12_Andy Wheeler25_Andy.Wheeler@GRIZZARD.COM31_Fri, 11 Feb 2000 15:38:56 -0500462_iso-8859-1 We are a new DB2 shop and are working on the ever present dilemma of how to bill customers. Does anyone know of any products on the market that will utilize the DB2 log to produce billing information (or any other method of billing that is in use with DB2)? We do not want to spend megabucks on acquiring new software, nor do we want to develop the system inhouse (sound familiar?) Any information reqarding billing would be greatly appreciated. [...] 16339 40 42_Re: Using the DB2 log for billing purposes24_Leblanc, Francis C - CNF23_Leblanc.Francis@CNF.COM31_Fri, 11 Feb 2000 12:47:26 -0800607_- I'm currently looking at Komand from Pace Software and at CIMS Lab. Both look very good so far, and we are continuing our evaluations. You can get more information from their web sites at http://www.pace.com, and http://www.cimslab.com.

> -----Original Message----- > From: Andy Wheeler [SMTP:Andy.Wheeler@GRIZZARD.COM] > Sent: Friday, February 11, 2000 12:39 PM > To: DB2-L@RYCI.COM > Subject: Using the DB2 log for billing purposes > > We are a new DB2 shop and are working on the ever present dilemma of how > to > bill customers. Does anyone know of any products on the market that will > [...] 16380 38 42_Re: Using the DB2 log for billing purposes12_Reid, Louise25_ReidL@CI.FORT-WORTH.TX.US31_Fri, 11 Feb 2000 14:49:11 -0600406_iso-8859-1 We use CIMs Integrated Chargeback system. (Used to be Platinum - I guess its CA now) It is pretty flexible and also allows us to import our network type charges and project based billable hours.

-----Original Message----- From: Andy Wheeler [mailto:Andy.Wheeler@GRIZZARD.COM] Sent: Friday, February 11, 2000 2:39 PM To: DB2-L@RYCI.COM Subject: Using the DB2 log for billing purposes [...] 16419 62 42_Re: Using the DB2 log for billing purposes24_Leblanc, Francis C - CNF23_Leblanc.Francis@CNF.COM31_Fri, 11 Feb 2000 13:09:23 -0800282_- Actually, it's not CA. It separated from Platinum about a week before Platinum sold out to CA. CA also has a product in this area, but I did not mention it earlier because after repeated attempts to get more information about it, I finally gave up. You may have better luck. [...] 16482 19 45_Modify Recovery - Additional index on SYSCOPY28_Dan Courter (1-408-463-3870)27_daniel_courter@VNET.IBM.COM29_Fri, 11 Feb 2000 13:37:32 PST454_- Cathy, I would not expect any additional index on SYSCOPY to make a significant difference. MODIFY accesses SYSCOPY with a selection expression of DBNAME=x and TSNAME=y. An index which contained *ONLY* these two columns might yield a small improvement. I can try an experiment.

The biggest improvement you can make to MODIFY can be made by increasing the PCTFREE for SYSCOPY itself, and the indexes DSNUCH01 and DSNUCX01 and running REORG. [...] 16502 40 49_Re: Modify Recovery - Additional index on SYSCOPY0_22_Rohn.Solecki@MTS.MB.CA31_Fri, 11 Feb 2000 15:58:38 -0600760_us-ascii Dan, sounds like you are a 'voice of experience'. So, what PCTFREE would you suggest for the table and indexes.









"Dan Courter (1-408-463-3870)" on 2000/02/11 03:37:32 PM

Please respond to DB2 Data Base Discussion List

To: DB2-L@RYCI.COM cc: (bcc: Rohn Solecki/MTSCommunications/MTS) Subject: Modify Recovery - Additional index on SYSCOPY







Cathy, I would not expect any additional index on SYSCOPY to make a significant difference. MODIFY accesses SYSCOPY with a selection expression of DBNAME=x and TSNAME=y. An index which contained *ONLY* these two columns might yield a small improvement. I can try an experiment. [...] 16543 248 17_Re: Catalog stats10_Wang.James24_Wang.James@AAA-CALIF.COM31_Fri, 11 Feb 2000 14:06:38 -0800525_us-ascii DB2I2 by JRH GoldenState Software Inc. can do it too. WWW.DB2I2.COM

James Wang Sr. Systems Programmer Automobile Club of Southern California (714) 850-2851 Wang.James@aaa-calif.com

-----Original Message----- From: Adrian Savory [SMTP:adrian.savory@ZURICH.COM] Sent: Wednesday, January 05, 2000 04:16 AM To: DB2-L@RYCI.COM Subject: Re: Catalog stats

David,

CA-Migrator does this, I believe BMC Change Manager can also do this. [...] 16792 98 17_Re: Catalog stats9_Mark Ruhe15_mruhe@QUEST.COM31_Fri, 11 Feb 2000 14:20:59 -0800476_iso-8859-1 CA/Platinum Plan Anaylzer also has a stats migration tool called Statistics Manager which will migrate stats from one subsystem to another, allow full editing of stats, and allow propagation of stats by percentages(ie, migrate from DB2A to DB2B increase stats across all objects by 150%).

-----Original Message----- From: Wang.James [mailto:Wang.James@AAA-CALIF.COM] Sent: Friday, February 11, 2000 4:07 PM To: DB2-L@RYCI.COM Subject: Re: Catalog stats [...] 16891 316 39_Re: Disaster Recover using EMC products11_Ron Haupert23_rhaupert@GOLDENGATE.NET31_Fri, 11 Feb 2000 15:53:58 -0600623_iso-8859-1 Hi Rick,

The state of the data at the restart site after a ConGroup trip is exactly the same as state of the data at the source when the ConGroup was tripped. Restart issues at the restart site are identical to restart issues at the source for a power failure. Many customers have databases spanning many storage processors. Many customers have transactions that span multiple DBMS'. When the ConGroup is tripped, all databases can be restarted and in-flight transactions are resolved. Transaction managers must be warm started to resolve active threads for transactions that cross database systems. [...] 17208 18 34_Using DSN1COPY for Backup/Recovery28_Dan Courter (1-408-463-3870)27_daniel_courter@VNET.IBM.COM29_Fri, 11 Feb 2000 14:37:13 PST307_- What you are describing is a bug in DSN1COPY and is fixed in APAR PQ34731 which is being packaged now and should be available in about two weeks.

In the mean time, DSN1COPY of segmented tablespaces will bypass this problem only if you copy VSAM to VSAM.

Sorry you had to run into this. [...] 17227 103 52_Re: Merging DB2 subsystems - Overview of Basic Steps9_Rob Crane22_racrane@CONCENTRIC.NET31_Fri, 11 Feb 2000 16:28:30 -0700313_us-ascii Here is a high level overview of one approach that can be used to bring a database(s) form one subsystem to another in preparation for data sharing or consolidating subsystems, etc. The idea is to minimize the outage by utilizing the existing DB2 VSAM datasets tied to the objects you need to move. [...] 17331 10 28_Re: DB2 hardware compression10_Bill Brown25_bill_brown@SPRINGMAIL.COM31_Fri, 11 Feb 2000 19:39:40 -0500586_- I have been using data compression in DB2 for about 8 years. First with BMCs software compression ant then with IBMs when it became available. Of all the tools we have used, the least appreciated is probably DB2 compression. For starters I usually estimate that you wil conservatively achieve 50% compression on your data. Of course your milage may vary. We actually achieve in the 75% to 80% range on some tablespaces. This equates to reduced DASD required (also reduced $ for DASD), Smaller image copy files since you will be backing up compressed pages (reduced $ for tape or [...] 17342 10 26_Re: extent numbers and SMS10_Bill Brown25_bill_brown@SPRINGMAIL.COM31_Fri, 11 Feb 2000 19:44:11 -0500505_- Another thought on priqty/secqty allocation. We always try to set up priqty/secqty to be either on track or cylinder boundaries. For track boundaries the qty should be evenly divisible by 48 and for cylinder boundaries the qty should be evenly divisible by 720.

================================================ 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. 17353 196 53_Re: what constitutes a unit of work (UDB EEE on AIX)?12_David Callen18_Callend@BCBSKC.COM31_Fri, 11 Feb 2000 18:50:12 -0600423_US-ASCII We've had similar problems. What we found, using UDB EE on AIX, was that while load does not write records to the log as it processes, it does write some control information. As it loaded the rows, until it reached build index phase, one UOW was opened and our logs filled We changed our scheduling so that our very large loads do not execute concurrently with our large updates (and increased our log size). [...] 17550 13 0_16_Krystal McCaskey28_princess_krystal@HOTMAIL.COM29_Sat, 12 Feb 2000 04:11:17 GMT375_- DO NOT MAIL ME AGAIN! ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.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. 17564 60 52_Re: Regarding OPERATIVE column in sysibm.syspackage.12_Ranjay Sinha24_rsinha@LOT.TATASTEEL.COM29_Sat, 12 Feb 2000 15:08:28 GMT333_us-ascii hello ravi, The operative column in the sysibm.syspackage or sysibm.sysplan tell whether the plan or package is operative or not. To change the status of a plan from operative=n to opreative = y rebind the plan. Yes it is true that whenever you drops a table the status of all the related plan is set to operative = N. [...] 17625 116 28_Re: DB2 hardware compression12_Isaac Yassin23_yassin@NETVISION.NET.IL31_Sat, 12 Feb 2000 20:05:12 +0200383_x-user-defined Hi Jim,

I've put the following numbers sometime last century ;) It's not exactly what you asked for and I may some more time to do your test as I'm in a very tight schedule just now. Hope it helps.



Hi,

Today I digged out an excel spreadsheet from about a year ago when we last checked compression. The numbers are from a small table. [...] 17742 609 18_SHARE Focus on ERP19_Freeling, Marv, ITD25_mfreelin@CO.ALAMEDA.CA.US31_Thu, 10 Feb 2000 12:21:53 -0800558_iso-8859-1 The ERP Focus track at SHARE will provide the edge your company needs to stay ahead of the curve...

ERP Focus Sessions at SHARE

Mark your calendars to attend the SHARE in Anaheim from March 5 - 10, 2000. Wednesday at SHARE will be an ERP Focus day sponsored by the DB2 project. There will be sessions focusing on PeopleSoft and SAP solutions running on DB2 on an OS390 platform, and how to get started. There will be sessions on internet access, creating a recovery strategy, performance tuning for ERPs, and much, much more... [...] 18352 94 52_Re: Regarding OPERATIVE column in sysibm.syspackage.11_Suresh Sane21_data_arch@HOTMAIL.COM29_Sat, 12 Feb 2000 14:10:20 CST525_- This has been discussed at great length before yet I see incorrect statements like "dropping a table make the plans operative = N".

Here is a summary as I understand it:

time 1: valid = y operative = y time 2: object dropped valid = n operative = y time 3: explicit bind or autobind at next exec (depending on zparm) ...successful? y >>> valid = y operative = y .. n >>> valid = n operative = n At this stage, a rebind will NOT fix the problem (unless the table now does exist) - it generally requires a [...] 18447 30 45_Modify Recovery - Additional index on SYSCOPY28_Dan Courter (1-408-463-3870)27_daniel_courter@VNET.IBM.COM29_Sat, 12 Feb 2000 13:37:39 PST523_- Rohn,

To be candid, being a utility developer I have very little of the "experience" you are talking about...just lots of theory. The DB2 listers may help more than I can. We ship SYSCOPY with 5% and the indexes with 10%. However, due to a bug in product packaging the DBDs are shipped with 0% in both cases and that's the number we use building new pages. (I've discussed this before on this list) The fix is to ALTER the PCTFREE and REORG. Even re-using the stock values will cause the DBD to be corrected. [...] 18478 17 20_REORG option DISCARD28_Dan Courter (1-408-463-3870)27_daniel_courter@VNET.IBM.COM29_Sat, 12 Feb 2000 15:48:18 PST527_- Claude, I don't see anything wrong with your syntax. CURRENT_DATE is correct. The utility parser "should" be converting this to a timestamp. Check your maintenance level because this is working for me. What release are you on?

Dan dcourter@us.ibm.com DB2 Utilities Development

================================================ 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. 18496 15 20_REORG option DISCARD28_Dan Courter (1-408-463-3870)27_daniel_courter@VNET.IBM.COM29_Sat, 12 Feb 2000 15:51:29 PST531_- Claude, Check the datatype for DATUM1. It must be date or time compatible. Even if the datatype is wrong, the U049 message should not be showing you an invalid value of ' ' but something like: '2000-02-12-00.00.00.000000' Try it with a character datatype and see what you get.

Dan

================================================ 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. 18512 16 14_DB/2 vs ADABAS9_Tio Vital20_tiovital@HOTMAIL.COM31_Sat, 12 Feb 2000 23:33:24 -0600386_- Hi DB/2 Guru..

Can Any one give me please a list of features in DB/2 that are lacking in ADABAS and vice versa?

Best Regrads Tio

================================================ 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. 18529 223 53_Re: what constitutes a unit of work (UDB EEE on AIX)?11_Scott Hayes34_faster.databases@DATABASE-GUYS.COM31_Sun, 13 Feb 2000 01:35:08 -0500510_iso-8859-1 Precisely. I've been watching this thread with interest as I get caught up on DB2-L. LOAD does make entries in the log, and may actually do logging during the index build phase... to delete rows per a unique index. I covered this in a session "At 200GB, Every Mistake is a BIG ONE" at DB2 Tech and IDUG a couple years ago. The "marketing" on LOAD is that it operates independently of the DB2 Manager, but the intrusion of the LOAD utility on the DB2 UDB logs sometimes sneaks up and bites you. [...] 18753 63 35_Re: NFS mounted drives and UDB v5.211_Scott Hayes34_faster.databases@DATABASE-GUYS.COM31_Sun, 13 Feb 2000 02:01:56 -0500356_iso-8859-1 Last I tried something like that, DB2 denied the create tablespace request because the media was not local media. In short, * DENIED *

Regards, Scott

DB2 UDB UNIX ... It's all we think about! (SM) If it's speed you need, check out the Database-GUYS(TM) Performance Monitoring and Management tools and win the e-Business race. [...] 18817 64 47_Re: No subject given / DB2 UDB "Character Case"11_Scott Hayes34_faster.databases@DATABASE-GUYS.COM31_Sun, 13 Feb 2000 02:18:46 -0500487_iso-8859-1 Sherri,

You would do this with a UDF. Frank Fillmore and The Fillmore Group have a really cool collection of very useful UDFs that they call "SQL Expander". I believe they have a UDF that would meet your needs. Find more information at:

http://www.thefillmoregroup.com/sqlexpand.htm

When you want to see what is going on inside DB2 UDB AIX performance and ensure that performance of new applications will meet expectations, I hope you'll contact us. [...] 18882 70 25_Re: recovery history file11_Scott Hayes34_faster.databases@DATABASE-GUYS.COM31_Sun, 13 Feb 2000 03:37:28 -0500532_iso-8859-1 Hao,

1) Yes, you can restore a database without the recovery history file. 2) Yes, the recovery history file is a UNIX file system file. It is neatly tucked away under the database's directory, perhaps: /home/db2inst1/db2inst1/NODE0000/SQL00001/db2rhist.asc There is also a db2rhist.bak in case the first file had a "bad day". 3) Yes, the db2 udb backup command backs up these files. To prove it to myself, I backed up my database, dropped it, recovered it, and ran command: db2 "list history all for DBNAME" [...] 18953 190 55_Re: Listing Databases for an Instance (UDB v6.1 on AIX)11_Scott Hayes34_faster.databases@DATABASE-GUYS.COM31_Sun, 13 Feb 2000 04:03:33 -0500487_iso-8859-1 Hi Bill,

Why not write a script that would take a list of database names as arguments? Then you could run this for each instance passing a list of DBNAMEs according to what is appropriate for a given instance.

root could run a script like: #!/bin/ksh su - db2inst1 -c "db2doit DBNAME1 DBNAME2" su - db2inst2 -c "db2doit DBNAME1 DBNAME3 DBNAME4" su - db2inst3 -c "db2doit DBNAME5 DBNAME6" su - db2inst4 -c "db2doit DBNAME1 DBNAME2 DBNAME3 DBNAME4 ..." # end [...] 19144 47 44_Re: sql function simil initcap of Oracle sql11_Scott Hayes34_faster.databases@DATABASE-GUYS.COM31_Sun, 13 Feb 2000 04:13:17 -0500797_iso-8859-1 SQL Expander from The Fillmore Group, a collection of UDFs to meet many needs. Initcap is one of them. See http://www.thefillmoregroup.com/sqlexpand.htm

Regards, Scott

<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> Scott R. Hayes, President * IBM GOLD Consultant * Database-GUYS Inc. Pager: http://www.database-guys.com/pgform.shtml Consulting and Performance Management Tools for DB2 Universal Database Check out the Database-GUYS(TM) at http://www.database-guys.com/ <><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> -----Original Message----- From: Giuseppe Newsgroups: bit.listserv.db2-l To: DB2-L@RYCI.COM Date: Friday, January 28, 2000 9:50 AM Subject: sql function simil initcap of Oracle sql [...] 19192 215 15_Re: DB recovery11_Scott Hayes34_faster.databases@DATABASE-GUYS.COM31_Sun, 13 Feb 2000 04:26:00 -0500795_iso-8859-1 When you FTP the backup file from Server1 to Server2, make sure you do a binary transfer. Beyond this additional tip, Bruce hit the nail right on the head.

HTH, Scott

<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> Scott R. Hayes, President * IBM GOLD Consultant * Database-GUYS Inc. Pager: http://www.database-guys.com/pgform.shtml Consulting and Performance Management Tools for DB2 Universal Database Check out the Database-GUYS(TM) at http://www.database-guys.com/ <><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> -----Original Message----- From: ALLEN,Bruce Newsgroups: bit.listserv.db2-l To: DB2-L@RYCI.COM Date: Tuesday, February 01, 2000 5:00 PM Subject: Re: DB recovery [...] 19408 113 28_Re: DB2 hardware compression15_Jim Lewandowski22_jlewand@STARNETINC.COM30_Wed, 9 Feb 2000 17:41:33 -0600562_us-ascii Very good results. I have no reason to doubt these statistics. It still begs the question: why not compress everything?

Jim Lewandowski





Isaac Yassin wrote:

> Number of records - 864432 > Number of pages - before 33317 after 14010 => 42.05% > Select * with CS elapse - before 8.5 min. after 8.4 => 98.82% > Select * with UR elapse - before 6.9 min. after 6.0 => 86.96% > Select * with CS CPU - before 232.8 sec. after 238.66 => 102.52% > Select * with UR CPU - before 231.6 sec. after 233.81 => 100.95% > Update [...] 19522 157 28_Re: DB2 hardware compression14_Richard Yevich21_ryevich@ATTGLOBAL.NET31_Sun, 13 Feb 2000 10:25:01 -0500330_iso-8859-1 Jim,

Many reasons: 1 - Because of something called indirect references Fixed length rows become variable length rows Violatility of certain types of tables get ugly 2 - Because it does not improve I/O in some cases 3 - Because it can hurt performance in some cases 4 - Because there is no always in DB2 :-) [...] 19680 210 55_Re: Listing Databases for an Instance (UDB v6.1 on AIX)19_Philip Nelson (DBA)21_teamdba@ATTGLOBAL.NET31_Sun, 13 Feb 2000 11:39:26 -0600456_us-ascii I'm confused about this conversation - the command "LIST DB DIRECTORY" will only list databases for the instance you are currently attached to (based on the setting of the DB2INSTANCE environment variable. Unless you've cataloged them cross instance.

One thing I have noticed is that I sometimes have to logoff and logon again and set the DB2INSTANCE variable before issuing any DB2 commands, otherwise it doesn't seem to take effect. [...] 19891 74 49_Get educated at the DB2 Symposium Europe May 8-1013_DB2 Symposium22_KBrant@XANADU.DEMON.NL31_Mon, 14 Feb 2000 03:03:16 +0100682_iso-8859-1 Dear DB2 Professional,

We proudly announce the European DB2 Symposium 2000 in Veldhoven Holland 8-10 May 2000.

The European DB2 Symposium has 3 parallel tracks available on application/performance, systems programming and trends in technology. Have a look at www.db2-symposium.com Click on the Dutch flag and view the complete program.

We think that we have composed a symposium that is attractive to everybody (designers, developers, sysprogs etc.) Are you planning for e-business then this an event that you have to see. Sessions on middleware, Inter/Intranet, stored procedures and more. But also bufferpools, high performance/tuning and V6. [...] 19966 57 12_Re: DSNV086E17_suthep thampranee18_suthepth@SCB.CO.TH31_Mon, 14 Feb 2000 10:27:16 +0700593_Windows-874 HELLO ALL DB2 USERS I TRIED TO FIND OUT THE MESSAGE CODE FOR 2 DAYS IN SYSLOG BUT I COULD NOT FIND IT. BECAUSE IT WAS HAPPENED 5 WEEKS AGO.IF IT HAPPEN AGAIN I WILL LET YOU KNOW. ANYWAY THANK YOU FOR YOUR CONCERN. THANKS JOE

---------- > From: HARBRY ARIZA > To: DB2-L@RYCI.COM > Subject: Re: DSNV086E > Date: 11 ¡ØÁÀҾѹ¸ì 2543 19:56 > > Suthep: > > DB2 goes down with this message but what is the reason code? > > > >From: suthep thampranee > >Reply-To: DB2 Data Base Discussion List > >To: DB2-L@RYCI.COM [...] 20024 21 12_SUPRA to DB210_ajay kumar19_db2v5r1@HOTMAIL.COM29_Mon, 14 Feb 2000 09:36:09 IST471_- hi all,

Can somebody throw some light on migrating SUPRA database to DB2 database ?..

TIA. Ajay.

______________________________________________________ Get Your Private, Free Email at http://www.hotmail.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. 20046 78 21_IMS to DB2 conversion12_Jacob Ganzel19_jacobg@SEMECH.CO.IL31_Mon, 14 Feb 2000 09:56:06 +0200506_x-user-defined Hello all.

We need information about tools, methods, etc. to convert a large COBOL CICS/IMS-DB (OS/390) customer to DB2?

We look for references of success stories, especially in banking environment.

Thanks for your help. Jacob. --------------------------------------------------------------- Jacob Ganzel Database Products Mgr. SEMECH Software Marketing Ltd.

3b Netanyahu St., OR YEHUDA, ISRAEL Email: jacobg@semech.co.il Tel.: +972-3-5333144 Fax.: +972-3-5333132 20125 92 23_Re: Use of Views in DB214_Massimo Scarpa16_mscarpa@CESVE.IT31_Mon, 14 Feb 2000 09:28:37 +0100606_us-ascii It's a widespread production rule (in Italy and abroad as someone said to me) that SELECT *

FROM is not a good choice so it should not be used by programmers.

It's 'easy', fast, error-prone and a problem for software maintenance .

********************************************************

For end users, views that are row-subsets (using WHERE), column subsets (for security) and those with a pre-defined join or grouping do indeed make sense. But for application developers the widespeard use SELECT * FROM views causes more trouble than the benfit. Specifically: [...] 20218 31 19_Infotel MASTER-UTIL25_CAPLIN Philippe (UA 1101)32_philippe.caplin@BANQUE-FRANCE.FR31_Mon, 14 Feb 2000 09:48:37 +0100739_- Hi, Did any shops tested or implemented the Infotel Product 'MASTER-UTIL' ? If yes, could you share your feedback ? TIA.

PhilCap.



************************************************************** Ce mel est a l'attention exclusive des destinataires designes. Il peut contenir des informations confidentielles. Si vous le recevez par erreur, merci d'en informer sans delai l'expediteur.

Le contenu de ce mel ne pourrait engager la responsabilite de la Banque de France que s'il a ete emis par une personne dument habilitee agissant dans le strict cadre des fonctions auxquelles elle est employee et a des fins non etrangeres a ses attributions. ************************************************************* [...] 20250 20 13_FTP SQL error16_Swaminaathan K.R33_swaminaathan.rajgopalan@WIPRO.COM31_Mon, 14 Feb 2000 14:43:09 +0530547_us-ascii Greeting list:

I am trying to execute a SQL ( simple select statement) using FTP DB2 SQL features by connecting to mainframe ,during which I am getting00F30040 on plan EZAFM320. Can anyone shed some light on whatneeds to be set up for executing SQL on mainframe through FTP?

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. 20271 66 16_Re: SUPRA to DB213_Marcus Davage29_Marcus.Davage@LLOYDSTSB.CO.UK31_Mon, 14 Feb 2000 10:06:42 -0000526_iso-8859-1 Certainly - you have as many options as you can afford! 1. Pay some consultants lots of money to do it for you (less effort for you and more money for them!) 2. Pay a vendor lots of money for software to intercept your RDM calls and translate them to SQL (but why bother when you can...) 3. Rewrite the application from scratch and migrate your data as you like it, or 4. Develop your new (DB2) application, slowly replacing your legacy system as you go, using nifty things like CICS transaction routing, etc. [...] 20338 49 28_Re: DB2 hardware compression0_19_mike.holmans@BT.COM31_Mon, 14 Feb 2000 10:33:43 -0000671_- -----Original Message----- From: Jim Lewandowski [SMTP:jlewand@STARNETINC.COM] Sent: Wednesday, February 09, 2000 11:42 PM To: DB2-L@RYCI.COM Subject: Re: DB2 hardware compression

Very good results. I have no reason to doubt these statistics. It still begs the question: why not compress everything?

Jim Lewandowski



One reason is that it makes the log difficult to read. We have Grafton's DB2/Audit product, which allows you to produce audit trails and such like from the DB2 log. We have DATA CAPTURE CHANGES switiched on for the tables we are auditing, and D/A is then able to produce the interesting things by reading the log after [...] 20388 131 26_Re: extent numbers and SMS12_Steve Thomas29_sherpa@HORNST.FREESERVE.CO.UK31_Mon, 14 Feb 2000 10:58:09 -0000523_iso-8859-1 My 2 cents worth on this is that while multiple extents on the disks themselves may not matter in a RAID environment, as far as OS/390 it still has a slight overhead as you have to go through the extent chain in the DBM1 address space when reading it. I would always try to keep the DB2 objects to a small number of extents as far as OS/390 is concerned. If the RAID disks decide to fragment it as the disk level, and they can handle it, then that's fine. Of course, you may have so much spare CPU capacity [...] 20520 34 26_Re: extent numbers and SMS14_Massimo Scarpa16_mscarpa@CESVE.IT31_Mon, 14 Feb 2000 12:18:09 +0100328_us-ascii I agree.

RAID disks can span datasets among multiple physical disks and they can managed this with

cache size & many algorithms but from 'the DB2 point of view' if you have 'dasd extents' (as you

can see from ISPF panels for instances) this is not take in account . I've never read of any DB2 [...] 20555 62 38_FW: Using DSN1COPY for Backup/Recovery12_David Wilson31_DWILSON@CEDELGLOBALSERVICES.COM31_Mon, 14 Feb 2000 13:04:01 +0100587_- Many thanks Dan. We will order this fix when available.

Donna, I tried this but ended up with a DSN1983I abend. It looks like there is a bug here after all.

> ---------- > From: Dan Courter > (1-408-463-3870)[SMTP:daniel_courter@VNET.IBM.COM] > Reply To: DB2 Data Base Discussion List > Sent: 11 February 2000 23:37 > To: DB2-L@RYCI.COM > Subject: Using DSN1COPY for Backup/Recovery > > What you are describing is a bug in DSN1COPY and is fixed in APAR PQ34731 > which is being packaged now and should be available in about two weeks. > > In the mean time, DSN1COPY [...] 20618 48 38_Re: Using DSN1COPY for Backup/Recovery19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM31_Mon, 14 Feb 2000 06:44:42 -0600620_iso-8859-1 Hi Dan, Thanks for info, however, I didn't get a hit on PQ34731 in APAR DB at website www.ibm.com/Search. 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 [...] 20667 125 31_Re: DB2/MVS distributed problem0_25_alexander.andrade@CAI.COM31_Mon, 14 Feb 2000 06:58:22 -0600550_- Tina Re 904, only thing I can think of is whether the Spufi Plan for DSNESPCS is actually bound with CS and not RR, I know I've done it myself the other way round deliberateley to catch apps dev out. From the information you've provided there doent seem to be any real consistency in any of the ssid always being the "offender", though DB2D seems to feature in most cases, being V4 you are obviously using SNA yes, maybe one of the vtam parameters has an inconsistency with the other 2 ssids major nodes or applid's convlimit maybe or pacing. [...] 20793 68 57_Grafton DB2 audit tool (was Re: DB2 hardware compression)0_23_Mike_Levine@TEKHELP.NET31_Mon, 14 Feb 2000 08:11:19 -0500697_us-ascii Hi Jim,

Could you please share information on the Grafton Audit tool? I Couldn't find a website for them. Are there any other tools that perform a similar function (i.e., produce audit reports from the log)?

Thanks,

Michael Levine Premier Data Services, Inc.



>-----Original Message----- >From: Jim Lewandowski [SMTP:jlewand@STARNETINC.COM] >Sent: Wednesday, February 09, 2000 11:42 PM >To: DB2-L@RYCI.COM >Subject: Re: DB2 hardware compression > >Very good results. I have no reason to doubt these statistics. It >still begs the question: why not compress everything? > >Jim Lewandowski > > >One reason is that it makes the log difficult to read. [...] 20862 107 61_Re: Grafton DB2 audit tool (was Re: DB2 hardware compression)12_Steve Thomas29_sherpa@HORNST.FREESERVE.CO.UK31_Mon, 14 Feb 2000 13:42:56 -0000343_iso-8859-1 Michael,

Grafton are a relatively small company here in the UK. I may be out of step, and I'm prepared to be corrected, but I don't think they market their product outside of the UK.

The best known tools for that do this are Resolve LogMaster from BMC and Log Analyzer from CA/Platinum (both of which I've used). [...] 20970 113 61_Re: Grafton DB2 audit tool (was Re: DB2 hardware compression)0_19_mike.holmans@BT.COM31_Mon, 14 Feb 2000 14:02:37 -0000602_- Michael,

My name isn't Jim, you know.

Grafton Database Consultants Ltd are a UK company who produce a suite called Data Commander, which is an umbrella name for the DC/Audit and DC/Propagator products. Audit post-processes offloaded logs and populates a database with its findings, which can be in summary or detail format. It can, optionally, consolidate all the changes made to a row between processing runs so that you get one row showing the final result of, say, an INSERT followed by four UPDATES (shown as an INSERT of the final version of the row) rathger than one row for [...] 21084 306 55_Re: Listing Databases for an Instance (UDB v6.1 on AIX)0_22_BILL_GALLAGHER@PHL.COM31_Mon, 14 Feb 2000 09:52:55 -0500430_us-ascii Phil,

I used to notice the same thing about changing the DB2INSTANCE variable until I discovered that doing a "db2 terminate" takes care of that problem. Make sure you do a terminate before you try to change the DB2INSTANCE variable, and you should be OK.

Also, based on a number of peoples suggestions (thanks everybody!), here is a copy of the script that I finally got to work the way I want it to: [...] 21391 48 17_Re: FTP SQL error16_Michael McCarthy18_mmccarthy@DTCC.COM31_Mon, 14 Feb 2000 10:06:13 -0500279_us-ascii Hi,

The problem is that you need to bind the EZAFM320 plan. The DBRM to use should be in your TCPIP.SEZADBRM pds. Bind this and you should have no problems. By the way, I don't know your current OS level but the plan at my shop is EZAFTPMQ. Hope this helps, [...] 21440 129 61_Re: Grafton DB2 audit tool (was Re: DB2 hardware compression)0_23_Mike_Levine@TEKHELP.NET31_Mon, 14 Feb 2000 11:18:28 -0500634_us-ascii Sorry Mike! You would think I would get a fellow Mike correct! Thanks for the info.

Regards, Michael (AKA 'Jim') Levine

P.S. No offense to any other Jims on the list!



>Michael, > >My name isn't Jim, you know. > >Grafton Database Consultants Ltd are a UK company who produce a suite called >Data Commander, which is an umbrella name for the DC/Audit and DC/Propagator >products. Audit post-processes offloaded logs and populates a database with >its findings, which can be in summary or detail format. It can, optionally, >consolidate all the changes made to a row between processing runs so [...] 21570 53 44_Strange results while using the DB2 Governor18_Richard Mccutcheon32_Richard_McCutcheon@HCM.HONDA.COM31_Mon, 14 Feb 2000 12:01:51 -0500330_us-ascii Good day. I have a question that is odd but not a mission critical thing. I am using the DB2 Governor on the following hardware/software: AIX 4.3.1, UDB 5.2

The following is a copy of the Governor config file: {wake up once every 30 seconds and check autoinq for offending UOW's} interval 30; dbname autoinq; [...] 21624 19 44_DB2/400 and CAE (Client Application Enabler)0_18_jchimes@PARKER.COM31_Mon, 14 Feb 2000 12:18:35 -0500333_us-ascii Help! Anybody using these 2 products together? When I query the catalog tables (from the command line processor from within WIN NT), I get a code page error. Even though the default AS/400 code page is '65535', it appears that the catalog tables use code page 1027. The target code page is '1252'. Is there a solution? [...] 21644 42 48_Re: DB2/400 and CAE (Client Application Enabler)0_15_leon@CA.IBM.COM31_Mon, 14 Feb 2000 12:46:06 -0500723_us-ascii DB2 Connect can not work with data tagged as 65535. 65535 means "do not convert".

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



jchimes@PARKER.COM on 02/14/2000 12:18:35 PM

Please respond to DB2 Data Base Discussion List

To: DB2-L@RYCI.COM cc: Subject: DB2/400 and CAE (Client Application Enabler)







Help! Anybody using these 2 products together? When I query the catalog tables (from the command line processor from within WIN NT), I get a code page error. Even though the default AS/400 code page is '65535', it appears that the catalog tables use code page 1027. The target code page is '1252'. Is [...] 21687 26 45_Where to find good benchmarking info for UDB?18_Richard Mccutcheon32_Richard_McCutcheon@HCM.HONDA.COM31_Mon, 14 Feb 2000 13:23:30 -0500459_us-ascii Does anyone know of a website, or document library that would contain any info on memory usage of UDB on AIX with respect to the number of direct connections to the database ?

I want to trend memory usage as we scale up a few inhouse applications. We are going from about 40 connections to potentially well over 1500 very shortly, and I am trying decide what needs we will have at a hardware and software level to accomodate this growth. [...] 21714 10 19_CA buying Sterling?12_BOB JEANDRON21_BOB.JEANDRON@USDA.GOV31_Mon, 14 Feb 2000 12:27:00 -0500312_- Any truth to this? We have Sterling CoolGen and Connect Direct here.

================================================ 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. 21725 32 23_Re: CA buying Sterling?12_Steve Thomas29_sherpa@HORNST.FREESERVE.CO.UK31_Mon, 14 Feb 2000 18:41:15 -0000657_iso-8859-1 Yes it's true. There's a press release knocking abuot somewhere. 4 billion dollars in a stock based acquisition.



----- Original Message ----- From: "BOB JEANDRON" Newsgroups: bit.listserv.db2-l To: Sent: 14 February 2000 17:27 Subject: CA buying Sterling?



> Any truth to this? We have Sterling CoolGen and Connect Direct here. > > ================================================ > 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. > [...] 21758 10 23_Re: CA buying Sterling?13_Shauna Hadden24_shauna.hadden@BCBSKS.COM31_Mon, 14 Feb 2000 12:39:18 -0600299_- It is true. Visit Sterling's web page www.sterling.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. 21769 32 23_Re: CA buying Sterling?18_Stritzinger, David24_StritzingerD@NABISCO.COM31_Mon, 14 Feb 2000 14:07:33 -0500759_- CA to buy Sterling for $4B



Computer Associates plans to purchase Sterling Software in what the companies call the software industry's largest-ever acquisition.

http://www.computerworld.com/home/print.nsf/CWFlash/000214ca





> -----Original Message----- > From: Shauna Hadden [SMTP:shauna.hadden@BCBSKS.COM] > Sent: Monday, February 14, 2000 1:39 PM > To: DB2-L@RYCI.COM > Subject: Re: CA buying Sterling? > > It is true. Visit Sterling's web page www.sterling.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. [...] 21802 55 23_Re: CA buying Sterling?18_Whittaker, Stephen26_stephen.whittaker@CPLC.COM31_Mon, 14 Feb 2000 14:15:46 -0500590_iso-8859-1 Well there's goes the rest of the neighborhood....

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



-----Original Message----- From: Stritzinger, David [mailto:StritzingerD@NABISCO.COM] Sent: Monday, February 14, 2000 2:08 PM To: DB2-L@RYCI.COM Subject: Re: CA buying Sterling?



CA to buy Sterling for $4B



Computer Associates plans to purchase Sterling Software in what the companies call the software industry's largest-ever acquisition. [...] 21858 615 27_ERP Focus Sessions at SHARE19_Freeling, Marv, ITD25_mfreelin@CO.ALAMEDA.CA.US31_Mon, 14 Feb 2000 11:25:18 -0800558_iso-8859-1 The ERP Focus track at SHARE will provide the edge your company needs to stay ahead of the curve...

ERP Focus Sessions at SHARE:

Mark your calendars for the SHARE in Anaheim March 5 - 10, 2000. Wednesday at SHARE will be an ERP Focus day, sponsored by the DB2 project at SHARE. There will be sessions focusing on PeopleSoft and SAP solutions running on DB2 on an OS390 platform, and how to get started. There will be sessions on internet access, creating a recovery strategy, performance tuning for ERPs, and much, much more... [...] 22474 73 23_Re: CA buying Sterling?16_Duane Lee - ATCX22_DLee@MAIL.MARICOPA.GOV31_Mon, 14 Feb 2000 12:26:04 -0700673_- Well, not exactly the rest of the neighborhood. BMC is still not a subsidiary of CA.

Duane

> -----Original Message----- > From: Whittaker, Stephen [SMTP:stephen.whittaker@CPLC.COM] > Sent: Monday, February 14, 2000 12:16 PM > To: DB2-L@RYCI.COM > Subject: Re: CA buying Sterling? > > Well there's goes the rest of the neighborhood.... > > Steve Whittaker > DB2/DBA Carolina Power & Light > Mail stop: CPB 17A4 - Raleigh > Email: stephen.whittaker@cplc.com > Phone: (919) 546-7267 > > > -----Original Message----- > From: Stritzinger, David [mailto:StritzingerD@NABISCO.COM] > Sent: Monday, February 14, 2000 2:08 PM > To: DB2-L@RYCI.COM > Subject: Re: [...] 22548 124 23_Re: CA buying Sterling?16_Galambos, Robert29_Robert_Galambos@COMPUWARE.COM31_Mon, 14 Feb 2000 14:39:21 -0500560_iso-8859-1



And we can add Compuware to the list of those companies who are not CA.

Robert Galambos robert_galambos@compuware.com

Compuware Corp. Of Canada 1-905-803-8600 1-800-348-8299 (Canada & US) 1-905-803-8603 (fax) 1-800-447-1662 (Quebec)



-----Original Message----- From: Duane Lee - ATCX [mailto:DLee@MAIL.MARICOPA.GOV] Sent: February 14, 2000 2:26 PM To: DB2-L@RYCI.COM Subject: Re: CA buying Sterling?



Well, not exactly the rest of the neighborhood. BMC is still not a subsidiary of CA. [...] 22673 113 23_Re: CA buying Sterling?12_Isaac Yassin23_yassin@NETVISION.NET.IL31_Mon, 14 Feb 2000 21:35:59 +0200630_x-user-defined NOT or NOT YET ???? ;)







"Galambos, Robert" wrote: > > And we can add Compuware to the list of those companies who are not > CA. > > Robert Galambos > robert_galambos@compuware.com > > Compuware Corp. Of Canada > 1-905-803-8600 > 1-800-348-8299 (Canada & US) > 1-905-803-8603 (fax) > 1-800-447-1662 (Quebec) > > -----Original Message----- > From: Duane Lee - ATCX [mailto:DLee@MAIL.MARICOPA.GOV] > Sent: February 14, 2000 2:26 PM > To: DB2-L@RYCI.COM > Subject: Re: CA buying Sterling? > > Well, not exactly the rest of the neighborhood. BMC is still not a > subsidiary of CA. > > [...] 22787 107 21_Can anyone EXPLAIN...16_Peter G Backlund15_pbacklu@IBM.NET31_Mon, 14 Feb 2000 20:37:49 +0100397_us-ascii Can anyone explain...

I have a table and two indexes defined as follows,

CREATE TABLE QQ.TCPV (PRSLANR SMALLINT NOT NULL ,VERNR SMALLINT NOT NULL ,PRSFDAT DATE NOT NULL ,PRSTDAT DATE NOT NULL ,PRODDAT DATE NOT NULL ,UPDSIGN CHARACTER(6) NOT NULL ,PRIMARY KEY (PRSLANR ,VERNR) )

CREATE TYPE 2 UNIQUE INDEX DPETRA.ICPV ON QQ.TCPV (PRSLANR ASC ,VERNR ASC ) CLUSTER [...] 22895 94 23_Re: CA buying Sterling?12_Isaac Yassin23_yassin@NETVISION.NET.IL31_Mon, 14 Feb 2000 21:36:54 +0200580_x-user-defined Just you wait Henry Higgins, just you wait .... ;)









Duane Lee - ATCX wrote: > > Well, not exactly the rest of the neighborhood. BMC is still not a > subsidiary of CA. > > Duane > > > -----Original Message----- > > From: Whittaker, Stephen [SMTP:stephen.whittaker@CPLC.COM] > > Sent: Monday, February 14, 2000 12:16 PM > > To: DB2-L@RYCI.COM > > Subject: Re: CA buying Sterling? > > > > Well there's goes the rest of the neighborhood.... > > > > Steve Whittaker > > DB2/DBA Carolina Power & Light > > Mail stop: CPB 17A4 - [...] 22990 54 23_Re: CA buying Sterling?22_Piontkowski Michael ML35_michael.piontkowski@ASTRAZENECA.COM31_Mon, 14 Feb 2000 14:40:27 -0500749_- Is CA buying Sterling Commerce as well?



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

> ---------- > From: Stritzinger, David[SMTP:StritzingerD@NABISCO.COM] > Sent: Monday, February 14, 2000 2:07 PM > To: DB2-L@RYCI.COM > Subject: Re: [DB2-L] CA buying Sterling? > > CA to buy Sterling for $4B > > > Computer Associates plans to purchase Sterling Software > in what the companies call the software industry's > largest-ever acquisition. > > http://www.computerworld.com/home/print.nsf/CWFlash/000214ca > > > > > -----Original Message----- > > From: Shauna Hadden [SMTP:shauna.hadden@BCBSKS.COM] > > Sent: Monday, February 14, 2000 1:39 PM > > To: DB2-L@RYCI.COM > > Subject: Re: CA [...] 23045 12 23_Re: CA buying Sterling?0_31_truman.g.brown@BELLATLANTIC.COM31_Mon, 14 Feb 2000 14:42:09 -0500291_us-ascii CA also has not yet acquired Tupperware.

================================================ 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. 23058 55 48_Re: DB2/400 and CAE (Client Application Enabler)12_Myron Miller22_myronwmiller@YAHOO.COM31_Mon, 14 Feb 2000 11:04:50 -0800577_us-ascii Here's what you need to do straight from a PMR I opened with IBM. We're doing that at a client site and DB2 Connect works great with the AS400.

"The default CCSID on AS/400 systems is 65535. DB2 Connect does not have a translation table for this codepage. You have to change the individual user profiles to contain a CCSID that is known to DB2 Connect, for example, 037 for US English, or 850 for International. To check the system CCSID, issue: "DSPSYSVAL QCCSID" To change the CCSID in the individual user profiles, issue: "CHGUSRPRF userid CCSID(037)." [...] 23114 506 30_Re: DB2 V5.1 & Parallelism....17_Hynes, Kenneth J.17_khynes@TRIGON.COM31_Mon, 14 Feb 2000 15:40:19 -0500594_iso-8859-1 Dr. Ebert (et. al.),

Based on your request for input -- it just so happened a question/problem fell into my lap last week involving parallel I/O. While your correct that in Most cases you won't see a reduction in CPU -- I have in fact witnessed it -- although not in this instance. I am surprised at the high % increase sited in your message as it seems somewhat excessive. Here is the information you requested -- Catalog information, Explain, and Accounting information (taken via screen snapshots from an online monitor). As you can see the difference in Elapsed time [...] 23621 16 56_Re: DB2-L Digest - 13 Feb 2000 to 14 Feb 2000 (#2000-47)14_Jennifer Libby23_jennifer@ICIAMERICA.COM31_Mon, 14 Feb 2000 16:12:30 -0500309_US-ASCII please take me off of your mailing list.

Thank you.

================================================ 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. 23638 91 35_Re: NFS mounted drives and UDB v5.212_Janet Rolsma12_janet@NM.NET31_Mon, 14 Feb 2000 14:21:55 -0700216_us-ascii Scott,

Actually, it let me create the database on an NFS mounted drive. I've even been able to create some tables.

I am going to call IBM to see if they support it.

Thanks, Janet

23730 84 15_Re: REXX SMF 4224_Humphris,Richard P.(NXI)24_Richard.Humphris@CNA.COM31_Mon, 14 Feb 2000 15:46:53 -0600643_- Here are some suggestions:

De-Fog DB2 I/O Performance Uncover DB2 I/O Performance (by a SolarFlare) DB2 I/O Relief Map'ping (by a SolarFlare)





> -----Original Message----- > From: Jim Lewandowski [SMTP:jlewand@STARNETINC.COM] > Sent: Thursday, February 03, 2000 12:57 PM > To: DB2-L@RYCI.COM > Subject: REXX SMF 42 > > I wish to respond to all those that either received a copy of a REXX > EXEC that processes SMF 42, subtype 6 records (dataset I/O performance) > in the past -or- responded most recently concerning another of my > offers. SolarFlare software has had some discussions lately that we > wish [...] 23815 133 25_Re: Can anyone EXPLAIN...17_Linda F. Claussen18_lindafc@NETINS.NET31_Mon, 14 Feb 2000 15:52:17 -0600563_ISO-8859-1 Peter,

This part of your predicate VERNR = (SELECT COALESCE(MAX(VERNR),:STCPVVERNR) is now an expression, thus, to my understanding, stage 2. You have lost indexability for the VERNR column for query block #1.

Linda F. Claussen Claussen & Associates DB2 Consulting and Training lindafc@netins.net http://www.netins.net/showcase/lclaussen/ Phone: (319) 343-3216 Office: (319) 847-1985

---------- From: Peter G Backlund To: DB2-L@RYCI.COM Subject: Can anyone EXPLAIN... Date: Monday, February 14, 2000 1:37 PM [...] 23949 154 49_Re: Where to find good benchmarking info for UDB?11_ALLEN,Bruce37_bruce.allen@EMPLOYMENTNATIONAL.COM.AU31_Tue, 15 Feb 2000 09:50:25 +1100303_- Hi Richard, The only documentation I am aware of for this is the V6 Quick Beginnings for Unix. It says 96 Mb for 25 clients and 186 for 50 clients. That increment calculates to 3.6Mb per concurrent connection, but as the manual says it depends on function. That is consistent with our experince. [...] 24104 128 35_Re: NFS mounted drives and UDB v5.211_Scott Hayes34_faster.databases@DATABASE-GUYS.COM31_Mon, 14 Feb 2000 21:07:10 -0500312_iso-8859-1 Darn. My lovely bride does the same thing... tell her it can't be done and about an hour later she gives me a demo of how she did it. I keep a supply of old shoes in the closet just for these very occasions.

As I often like to say, because you can do something, does that mean you should? [...] 24233 362 55_Re: Listing Databases for an Instance (UDB v6.1 on AIX)11_Scott Hayes34_faster.databases@DATABASE-GUYS.COM31_Mon, 14 Feb 2000 21:26:00 -0500540_iso-8859-1 You know Bill, between our sample scripts, I think we've just about got all the DB2 OS/390 people thinking that UNIX is indeed a scary place.

For those in the studio audience wanting to run runstats for each table in a database schema:

$ db2 connect to DBNAME $ db2 list tables [for all] | grep SCHEMANAME | while read tb sc ty ti > do > db2 connect to DBNAME # only needed on some UNIX platforms > db2 -v "runstats on table $sc.$tb and detailed indexes all" > db2 terminate # only use if connect used above > [...] 24596 163 25_Re: Can anyone EXPLAIN...14_Purcell, Terry37_Terry.Purcell@CORPMAIL.TELSTRA.COM.AU31_Tue, 15 Feb 2000 13:29:22 +1100526_iso-8859-1 Linda,

That's incorrect. The COALESCE solved the mismatch of nullability, thereby making the statement stage 1 and indexable (for a non-correlated subquery). This is exactly how Peter had explained it in his original mail below.

Peter,

The optimizer has decided (based on stats of course) that a matchcols 2 on ICPV plus data access, is more efficient than a matchcols 1 on ICPV2 plus index screening with no data access. Which is more efficient is dependant on your column cardinalities. [...] 24760 123 63_Re: NFS mounted drives and UDB v5.2 / Part 2, what the doc says11_Scott Hayes34_faster.databases@DATABASE-GUYS.COM31_Mon, 14 Feb 2000 21:42:29 -0500878_iso-8859-1 Janet,

Found this in the DB2 UDB V5.2 doc, SQL Reference, under Create Tablespace:

"Remote resources (such as LAN-redirected drives on OS/2, Windows 95 and Windows NT or NFS-mounted file systems on AIX) are not supported."

Cheers, Scott

<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> Scott R. Hayes, President * IBM DB2 GOLD Consultant * Database-GUYS Inc. Pager: http://www.database-guys.com/pgform.shtml Consulting and Performance Management Tools for DB2 Universal Database Check out the Database-GUYS(TM) at http://www.database-guys.com/ <><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> -----Original Message----- From: Janet Rolsma Newsgroups: bit.listserv.db2-l To: DB2-L@RYCI.COM Date: Monday, February 14, 2000 4:28 PM Subject: Re: NFS mounted drives and UDB v5.2 [...] 24884 45 28_Re: DB2 hardware compression15_Jim Lewandowski22_jlewand@STARNETINC.COM31_Fri, 11 Feb 2000 02:04:13 -0600338_us-ascii I began to think more about this today. The CPU timings below are not simply SQL entrance/exit, correct? We have tons of other CPU wrapped around all this DB2 SQL-related nucleus code.

That is why I wanted a single SQL COUNT(*) statement to be resolved since it will only be 1 instance of SQL entrance/exit into DB2. [...] 24930 90 48_Re: Strange results while using the DB2 Governor11_Scott Hayes34_faster.databases@DATABASE-GUYS.COM31_Mon, 14 Feb 2000 23:54:39 -0500906_iso-8859-1 Negative Idle time, now that is getting work done at a high rate of speed. I bet there's some employers out there that would love to harness that technology.

Sorry Richard, I'm stumped on that one. I'd sure like to know if you find anything out.

Regards, Scott

<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> Scott R. Hayes, President * IBM GOLD Consultant * Database-GUYS Inc. Pager: http://www.database-guys.com/pgform.shtml Consulting and Performance Management Tools for DB2 Universal Database Check out the Database-GUYS(TM) at http://www.database-guys.com/ <><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> -----Original Message----- From: 'Richard Mccutcheon' Newsgroups: bit.listserv.db2-l To: DB2-L@RYCI.COM Date: Monday, February 14, 2000 12:05 PM Subject: Strange results while [...] 25021 25 32_QUery on clustering of DB2 on NT21_Pramod S ( Benchmark)19_pramods@WIPRO.CO.IN31_Tue, 15 Feb 2000 10:29:04 +0530273_- Hi, I am a new member to the list. I want to know if it is possible to use DB2 in a Microsoft clustering(wolfpack) scenario. I.e. Is DB2 on NT, cluster aware. If so can you please tell me where I can find more information on it, so that I can implement the cluster. [...]