1 WWW.IDUGDB2-L.ORG /home/listserv/home/db2-l November 2006, week 2
2 151 28_Re: Alias on the DB2 Catalog13_Michael Ebert18_mebert@AMADEUS.COM30_Wed, 8 Nov 2006 08:26:56 +0100517_US-ASCII We had a shadow catalog when I started here 8 years ago. It was about the first thing I got rid of. It's never been missed. I don't see the point of having ALIASes or SYNONYMs either. I guess clone catalogs were introduced 20 years ago because the first versions of DB2 had locking problems if anyone was running a query on the catalog. I'd guess further that this was fixed 18 years ago and people have kept them because it's always been done that way (numbers are for illustrational purposes only)... [...]
154 21 11_DB2 V8 z/OS12_Adam Baldwin19_y0027528@ES.IBM.COM30_Wed, 8 Nov 2006 01:48:44 -0600283_- Fellow Listers - a quick straw poll please:
1: How long have you spent / are planning to spend in V8CM in development? 2: How long have you spent / are planning to spend in V8NFM in development? 3: How long have you spent / are planning to spend in V8CM in production? [...]
176 15 15_Re: DB2 V8 z/OS12_Adam Baldwin19_y0027528@ES.IBM.COM30_Wed, 8 Nov 2006 02:57:28 -0600666_- Question 2 should of course read:
2: How long have you spent / are planning to spend in V8NFM in development before going to CM in production?
Given that it's a one way ticket....
--------------------------------------------------------------------------------- Welcome to the IDUG DB2-L list. To unsubscribe, go to the archives and home page at http://www.idugdb2-l.org/archives/db2-l.html. From that page select "Join or Leave the list". The IDUG DB2-L FAQ is at http://www.idugdb2-l.org. The IDUG List Admins can be reached at DB2-L-REQUEST@www.idugdb2-l.org. Find out the latest on IDUG conferences at http://conferences.idug.org/index.cfm
192 63 15_Re: DB2 V8 z/OS10_Sysdba AHE18_Sysdba.AHE@TNT.COM30_Wed, 8 Nov 2006 09:34:04 +0000855_US-ASCII Adam,
I suggest your list should read
1) CM in development 2) CM in production 3) NFM in development 4) NFM in production
This has been much discussed previously.
Regards
Neil Price TNT Express, UK
--------------------------------------------------------------------------------------------------------------- This message and any attachment are confidential and may be privileged or otherwise protected from disclosure. If you are not the intended recipient, please telephone or email the sender and delete this message and any attachment from your system. If you are not the intended recipient you must not copy this message or attachment or disclose the contents to any other person. --------------------------------------------------------------------------------------------------------------- [...]
256 13 15_Re: DB2 V8 z/OS12_Adam Baldwin19_y0027528@ES.IBM.COM30_Wed, 8 Nov 2006 03:30:52 -0600720_- Thanks Neil - you are of course right! And I know that this has been discussed on numerous occasions but I am hoping that now that there are far more sites with experience of migrating to V8 it might be possible to get some more realistic figures.
--------------------------------------------------------------------------------- Welcome to the IDUG DB2-L list. To unsubscribe, go to the archives and home page at http://www.idugdb2-l.org/archives/db2-l.html. From that page select "Join or Leave the list". The IDUG DB2-L FAQ is at http://www.idugdb2-l.org. The IDUG List Admins can be reached at DB2-L-REQUEST@www.idugdb2-l.org. Find out the latest on IDUG conferences at http://conferences.idug.org/index.cfm
270 76 15_Re: DB2 V8 z/OS10_Sysdba AHE18_Sysdba.AHE@TNT.COM30_Wed, 8 Nov 2006 10:07:42 +0000436_US-ASCII Adam,
To answer your questions, we're planning the following:
- development/test subsystems to stay CM for 6 weeks before going NFM - but not until production is at CM. - production to go NFM 6 weeks after development goes NFM - production to go NFM 7 weeks after going CM
and the one you didn't ask - production to go CM 5 weeks after development goes CM, and in any case before development goes NFM [...]
347 73 15_Re: DB2 V8 z/OS13_Bell, Raymond22_raymond.bell@LANDG.COM30_Wed, 8 Nov 2006 12:01:05 -0000330_- Adam,
We took our first Sandpit subsystems to V8 CM early August and will convert Prod to V8 CM this weekend, so about 3 months from first to last. We'll stay in CM in Dev until probably 1st quarter next year, mostly because of other workload commitments, and will covert them all to ENFM over 2-3 months next year. [...]
421 155 36_Re: Finding out who issued a command13_Bell, Raymond22_raymond.bell@LANDG.COM30_Wed, 8 Nov 2006 12:03:05 -0000559_- Hugh,
If you have IFCID 191 being captured you can extract the records and see what DSN commands were issued. Omegamon does a nice IFCID 191 trace report. 'Course, it kinda relies on you having these records captured already...
Cheers,
Raymond Bell Database Administrator PS. Love the background colour...
_____
From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Hugh Lapham Sent: 07 November 2006 19:29 To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] Finding out who issued a command [...]
577 19 15_Re: DB2 V8 z/OS12_Adam Baldwin19_y0027528@ES.IBM.COM30_Wed, 8 Nov 2006 06:12:22 -0600338_- Raymond - thanks for that. I can see that we (my DB2 group here) are going to have to put some rockets under people. Given the timescales being mentioned and the work load that we have (30 - 40 subsystems) it's all getting a bit tight.
We've already missed the chance of being able to put year end processing through in CM. [...]
597 66 15_Re: DB2 V8 z/OS13_Bell, Raymond22_raymond.bell@LANDG.COM30_Wed, 8 Nov 2006 12:37:57 -0000309_- Have fun - I believe V7 may go out of service March '08, so you have some time yet. It's only taking us this long because we've other z/OS commitments and we want to sit on CM thru a financial quarter. Plus, once we're CM, it gives us time to prepare presentations to the world on how white-hot V8 is. [...]
664 185 15_Re: DB2 V8 z/OS13_Robert Tilkes22_TILKESR@NATIONWIDE.COM30_Wed, 8 Nov 2006 07:30:12 -0600382_US-ASCII Adam,
I suggest you find Roger Miller's post about how long to stay in ENFM prior to going to NFM. It may help in your planning.
Time line:
Ordered - Late 3/06 Installed on Sandbox - 04/06 Installed on Dev - 06/06 Installed on Prod - 07/06 Installed ENFM/NFM on Sandbox - 10/06 Installed ENFM/NFM on Dev - 11/06 Inatalled ENFM/NFM on Prod - 12/06 [...]
850 210 15_Re: DB2 V8 z/OS13_Roberts Colin32_colin.roberts@BRITISH-ENERGY.COM30_Wed, 8 Nov 2006 13:46:02 -0000483_us-ascii We prepared very thoroughly and, I believe, were quite lucky in terms of encountering few problems. That being said, our timings were as follows:
Received DB2v8 May 05 Went to Test Bed (CM) June 05 Went to Test Bed (NFM) July 05 Went to Dev (CM) Sept 05 Went to Dev (NFM) Nov 05 Went to User Test (CM) Sept 05 Went to User Test (NFM) Dec 05 Upgraded QMF to 8.1 in Prod Oct 05 Went to Prod (CM) Dec 05 Went to Prod (NFM) Feb 06 Major Application Upgrade March 06 [...]
1061 72 22_DB2 V7.1 and z/OS V1.415_Foweather, Iris30_Iris.Foweather@UK.EXPERIAN.COM30_Wed, 8 Nov 2006 14:17:32 -0000538_iso-8859-1 DB2 V7.1 and z/OS V1.4
We are getting timeouts:
DSNL511I &D2LB DSNLIENO TCP/IP CONVERSATION FAILED TO LOCATION xxx.xxx.xxx.xxx IPADDR=xxx.xxx.xxx.xxx PORT=xxx SOCKET=RECV RETURN CODE=1127 REASON CODE=00000000
TCP/IP Keepalive is 2 hours TCP/IP Keepalive (TCPKPALV) is ENABLE Idle thread timeout (IDTHTOIN) is 1200 DDF Threads (CMTSTAT) is INACTIVE
I am struggling to understand how TCP/IP and DB2 work together regarding keepalive and timeouts etc. and determine the cause of the 1127 timeout. [...]
1134 30 36_UDB Windows 8.2.3 Timestamp question12_Matthew Rhea24_matthew_rhea@CONSECO.COM30_Wed, 8 Nov 2006 08:22:59 -0600459_- Hi, List. UDB Windows 8.2.3 here. We've noticed some time discrepancies in DB2 messages regarding time.
This message is from the LOAD utility, which was run at 9:09 AM: SQL3515W The utility has finished the "BUILD" phase at time "11/08/2006 07:51:57.525135".
This is the message in the db2diag.log. Notice the two different timestamps. LOADID: 2584.2006-11-08-09.09.50.676000.0 (83;2) Completed BUILD phase at 11/08/2006 07:51:57.525135. [...]
1165 21 15_Re: DB2 V8 z/OS12_Adam Baldwin19_y0027528@ES.IBM.COM30_Wed, 8 Nov 2006 09:13:01 -0600376_- Thanks for your input - both on and off list. From what I'm seeing / hearing, we've got time spans ranging from 9 days the lot (maybe I read those dates incorrectly - ordered late on 3rd june and installed in prod in NFM by the 12th june?? Impressive or what!) to one year. With Q1 2008 being talked about for V7 support ending....the available window is fast closing. [...]
1187 95 15_Re: DB2 V8 z/OS0_19_mike.holmans@BT.COM30_Wed, 8 Nov 2006 15:25:13 -0000309_us-ascii We're one of those worrying places, then.
We have one production app in NFM, and the subsystems on that app's development LPAR are NFM, but everything else is in CM, as it has been since about February. We won't be bringing the other 30-odd CM subsystems up to NFM until early next year. [...]
1283 78 15_Re: DB2 V8 z/OS0_17_sjvagnier@AEP.COM30_Wed, 8 Nov 2006 10:41:56 -0500389_US-ASCII Adam, I am in the process of migrating to V8 CM, ENFM, and NFM. Call me if you need details on what the problems (minor) I have encountered.
**************************************** Steve Vagnier American Electric Power Database Administration One Riverside Plaza - 7th Floor Columbus, Ohio 43215 Email: sjvagnier@aep.com Phone: 614-716-3677 Audinet: 200-3677 [...]
1362 200 50_DB2 zOS reporting and ad hoc architecture question15_Levine, Michael23_mike.levine@CGI-AMS.COM30_Wed, 8 Nov 2006 11:23:47 -0500618_us-ascii Hi,
I have a slight difference of opinion with our non-mainframe database architecture team for a new mainframe DB2 application. Basically without getting into too much detail, we are using Web Logic and Business Objects and the OLTP will reside on DB2 zOS V8 with a 1200 MIP machine all to itself. The architecture team had a 'bad experience' once trying to run reporting and Ad Hoc access against a production DB2 OLTP database. So they are basically recommending that we put triggers on all 500+ tables to capture updates, inserts and deletes and write them to shadow tables on [...]
1563 192 54_Re: DB2 zOS reporting and ad hoc architecture question14_Mark E Vickers26_MarkVickers@GROCERYBIZ.COM30_Wed, 8 Nov 2006 11:13:58 -0600534_ISO-8859-1 If you have a z9 and a zIIP engine, SQL via DRDA/TCP-IP will run on the zIIP and by the sounds of moderate on-line updates (and commits), locking shouldn't be a problem.
Depending on need for concurrent data, have you considered building data marts for the ad-hoc queries ?
"Levine, Michael" Sent by: DB2 Data Base Discussion List 11/08/2006 10:23 AM Please respond to DB2 Database Discussion list at IDUG [...]
1756 273 54_Re: DB2 zOS reporting and ad hoc architecture question13_Bell, Raymond22_raymond.bell@LANDG.COM30_Wed, 8 Nov 2006 17:14:27 -0000332_- If you can guarantee your BO bods a) know what WITH UR means, and b) swear on their first born that they'll use it, go for it. Imagine the DASD, CPU cycles and general PITA you'll save yourself if you can successfully get MI-type queries to run off your operational systems. I guess you have, which is why you wanna do this. [...]
2030 110 54_Re: DB2 zOS reporting and ad hoc architecture question12_Robert Smith24_robert.smith@NCFBINS.COM30_Wed, 8 Nov 2006 12:21:20 -0500
2141 69 54_Re: DB2 zOS reporting and ad hoc architecture question11_Hugh Lapham26_hugh.lapham@RCMP-GRC.GC.CA30_Wed, 8 Nov 2006 12:35:28 -0500381_US-ASCII This must be another classic case of "your mileage may vary".
However here are a few things to consider: typically, tuning for an ad-hoc reporting database requires significantly different indexes and buffer pools than the OLTP version
relative priorities can prevent ad-hoc reporting from impacting OLTP, but then the reporting can become UNresponsive [...]
2211 56 33_[z/OS v7] Sorting 8K & 16K tables14_Mark E Vickers26_MarkVickers@GROCERYBIZ.COM30_Wed, 8 Nov 2006 11:45:20 -0600319_US-ASCII 1. Since you cannot specify 8K & 16K bufferpools for DSNDB07 tablespaces, where do 8K & 16K tables get sorted ?
If they do get sorted in a 32K TS/BP, then three more questions:
2. If you have 8/16/32K tablespaces, MUST you have at least one tablespace in DSNDB07 linked to a 32K bufferpool? [...]
2268 88 50_Using the DB2 log as an application tracking tool.13_Brendan Friel28_Brendan_Friel@NOTES.NTRS.COM30_Wed, 8 Nov 2006 11:54:32 -0600575_US-ASCII There seem to two schools of thought for this. The goal (for example) would be to find out who changed K-Fed's marital status in DB2 in the last two weeks.
The options seem to be :
A) Set up triggers to write updates to DB2 application type audit tables that can be read through any medium. Disadvantage : We need a DBA or DB2 application programmer to maintain the trigger and audit table definitions. Advantage : Anyone with the correct access can easily query the tables. B) Use the DB2 logs - it's in there somewhere. Disadvantages : The data [...]
2357 67 54_Re: Using the DB2 log as an application tracking tool.10_Dave Nance16_dav1mo@YAHOO.COM30_Wed, 8 Nov 2006 11:11:34 -0800414_iso-8859-1 Brendan, You've pretty much summed it up. I prefer the log as it cuts down on CPU consumption and logging(have to insert/log to another table). There can be some issue with keeping logs around and space, but they can be farmed off to tape. Some of the third party products out there for reading the logs are fairly easy to use and you could set up a process to easily run a job to look up changes. [...]
2425 176 54_Re: Using the DB2 log as an application tracking tool.11_Hugh Lapham26_hugh.lapham@RCMP-GRC.GC.CA30_Wed, 8 Nov 2006 14:25:26 -0500489_US-ASCII OTOH (isn't it always the case) since this sounds like an applications requirement, why not make it a part of the application? The audit log itself doesn't necessarily have to be in DB2 (and some would even argue that it should be elsewhere -- kinda like double-entry bookkeeping ;-))
Of course, in an uncontrolled environment when "anyone" could have updated the marital status via "any" connection mechanism, then the solution pretty much has to be within the DBMS. [...]
2602 40 43_Covert Cobol DB2 programs for DB2 V7 to V8.16_mallik vemugunta19_kris_vemu@YAHOO.COM30_Wed, 8 Nov 2006 12:46:45 -0800418_iso-8859-1 Hi,
We have a online application that uses cobol as the host language and DB2 as the database, can some one please tell me if I have to make any application changes for the DB2 version upgrade from V7 to V8.
Thanks, Kris.
--------------------------------- Sponsored Link
Get an Online or Campus degree - Associate's, Bachelor's, or Master's - in less than one year. [...]
2643 60 47_Re: Covert Cobol DB2 programs for DB2 V7 to V8.9_Mike Bell21_mbell11a1@VERIZON.NET30_Wed, 8 Nov 2006 15:40:50 -0600324_US-ASCII The general answer is no.
What V8 removed was the precompiler support for the old COBOL II programs. IE PARM=('HOST(COB2)' on the precompiler. If you still have any programs that use the old COBOL II compiler, you will have to convert them to something more current before you can make changes to them. [...]
2704 130 47_Re: Covert Cobol DB2 programs for DB2 V7 to V8.13_Willie Favero21_wfavero@ATTGLOBAL.NET30_Wed, 8 Nov 2006 17:37:47 -0600399_us-ascii Almost sounds like a trick question...
First, what COBOL are you using? The DB2 V8 precompiler only supports the current COBOL compilers.
However, if you are asking "Will a load module produced in V7 run on V8?" Yes, it should.
All of the COBOL stuff you hear about connected to a V8 upgrade usually has to do with changing, precompiling, and compiling a COBOL program.
2835 35 80_How to bypass DSNJ252E-forDB2 DisasterRecovery with systemClock set to 1day ago?32_G=?ISO-8859-1?Q?=FCray_Meri=E7?=22_mericguray@HOTMAIL.COM30_Thu, 9 Nov 2006 00:23:49 +0000587_- Hi to all, currently we get DB2 catalog backups in primary site in T1 time, and restore&COLD start DB2 in T1+2 time in disastery site. We stopped DB2 in disastery site in T1+3 time. When we try to set the zOS clock to T1-24 (1 day ago) in disastery, because datasharing uses LRSN, we normally get DSNJ252E error and DB2 ABENDS, can not start. DSNJ252E +DB1P DSNJW009 CURRENT DATA SHARING LRSN BFAA5C85A8BD IS LESS THAN HIGH WRITTEN LRSN BFAB84EE05B3. At this point, we tried another COLD START but we saw critical STOPEs in SYSLGRNX,SYSUTILX,etc...We tried to solve them by START [...]
2871 55 53_DB2 Presentations at CMG Conference Dec 4-8, Reno, NV35_Joel Goldstein - Responsive Systems26_joel@RESPONSIVESYSTEMS.COM30_Wed, 8 Nov 2006 20:18:19 -0500637_iso-8859-1 There are several DB2 presentations, and many that would be to those involved with DB2 performance and tuning issues. The complete program can be found at www.cmg.org
Monday: Measurement and Modeling of DB2 zIIP workloads Ned Diehl
Tuesday: DB2 for z/OS Performance and Tuning Bonnie Baker DB2 for z/OS Stored Procedures Performance Hot Topics Peggy Zagelow zIIPs and zAAPS - How Special are They Kathy Walsh Tuning your system with the Buffer Pool Tool for DB2 Joel Goldstein A Simple Approach to DB2 Index Redesign Martin Hubel Panel: DB2 Q&A Baker, Goldstein, Halinski, Hubel, Wingfeild, Zagelow [...]
2927 216 37_Re: [z/OS v7] Sorting 8K & 16K tables13_Taddei, Cathy27_Cathy.Taddei@PACIFICORP.COM30_Wed, 8 Nov 2006 17:44:20 -0800514_us-ascii Chuck Hoover wrote an excellent white paper called "Understanding the DB2 for z/OS Sort" -- I highly recommend it to anyone interested in sorting. DB2 will use a 32K DSNDB07 tablespace for anything that doesn't fit in a 4K page. To answer your other questions:
2. I may be wrong, but you could have need of a 32K DSNDB07 tablespace even if you have no other tablespaces > 4K in your subsystem. When joining lots of tables and sorting on lots of columns, the result might not fit in a 4K page. [...]
3144 59 9_[ASK] DB213_Reza Agustian19_agustianr@GMAIL.COM30_Thu, 9 Nov 2006 08:35:38 +0700428_ISO-8859-1 Hi, I'm reza, and I'm a newbie in DB2 currently I'm using DB2 v8.1.7.445 on development system on Windows XP Pro i want to ask a few question:
1. i want to alter my table using control center, but when it execute the error was :
[IBM][CLI Driver][DB2/NT] SQL0443N Routine "SYSPROC.ALTOBJ" (specific name "ALTOBJ") has returned an error SQLSTATE with diagnostic text "DBA7904, DBAD". SQLSTATE=38553 [...]
3204 39 25_[MVS] DSNTEP2 distributed14_Matthews, John25_JMatthews@MEDNET.UCLA.EDU30_Wed, 8 Nov 2006 18:07:51 -0800415_iso-8859-1 I am just trying to access one DB2 for z/OS V7 subsystem(DB2A) from another z/OS DB2 subsystem(DB2B) and vice versa(via DSNTEP2 initially). I think the communications DB is set up OK but I am confused about the packages/plans.
on DB2A do I need to do the following pkg BIND using the DB2A DBRMLIB? Won't this pkg BIND overlay the DSNTEP2 pkg in DB2B since the DSNTEP2 pkgs are not VERSION'd? [...]
3244 78 29_Re: [MVS] DSNTEP2 distributed9_Mike Bell21_mbell11a1@VERIZON.NET30_Wed, 8 Nov 2006 20:19:08 -0600337_us-ascii Since the default jcl for the DSNTEP2 install does not precompile with versioning, you will have to setup different collection id's for each version and then list each one in PKLIST with location information.
The other possible fix is to change DSNTEJ1P (the DSNTEP2 install jcl) to specify VERSION(AUTO) and rerun. [...]
3323 36 37_Re: [z/OS v7] Sorting 8K & 16K tables10_Max Scarpa28_Massimo.Scarpa@PHOENIXSPA.IT30_Thu, 9 Nov 2006 08:46:31 +0100345_US-ASCII Tag sort oblige you to have 32K BPs. As Cathy mentioned, in the same white paper is explained why a 'sorted input' in DB2 RDS sort is fastest (say a perfect clustered input), there's always a 'winner'.
Max Scarpa Returning from z/OS 1.7 war
---------------------------------------------------------------------------------
3360 246 29_Re: [MVS] DSNTEP2 distributed13_Michael Ebert18_mebert@AMADEUS.COM30_Thu, 9 Nov 2006 09:45:04 +0100871_US-ASCII No need for versioning. Just run a BIND step like this on all subsystems:
//*--------------------------------------------------------------------- //BIND EXEC PGM=IKJEFT1B,COND=(4,LT),PARM='DSN S(ssid)' //DBRMLIB DD DISP=SHR,DSN=DB2.SDSNSAMP //SYSTSPRT DD SYSOUT=* //SYSPRINT DD SYSOUT=* //SYSTSIN DD * BIND PACKAGE(DSNTEP2) OWNER(ssidadm) MEM(DSN@EP2L) - ACT(REP) ISO(CS) ENCODING(EBCDIC) CURRENTDATA(NO) BIND PLAN(DSNTEP2) OWNER(ssidadm) PKLIST(*.DSNTEP2.DSN@EP2L) - ACT(REP) ISO(CS) ENCODING(EBCDIC) CURRENTDATA(NO) SQLRULES(DB2)
Note the preceding * in the PKLIST. Similar for other programs where you want distributed access, like DSNTIAD, DSNTIAUL and SPUFI. If an assembly/compilation is needed, do this on the first subsystem only, then just copy the DBRMs and LOADMODs to the other systems - otherwise you'll get contoken mismatches. [...]
3607 23 37_Re: [z/OS v7] Sorting 8K & 16K tables13_Martin Packer24_martin_packer@UK.IBM.COM30_Thu, 9 Nov 2006 13:07:25 +0000770_US-ASCII On 4) I personally would recommend a 32K sort buffer pool to be dedicated - at the very least from an instrumentation perspective. Just like I would for 4K.
Martin Packer Performance Consultant IBM United Kingdom Ltd +44-20-8832-5167 +44-7802-245-584 martin_packer@uk.ibm.com
--------------------------------------------------------------------------------- Welcome to the IDUG DB2-L list. To unsubscribe, go to the archives and home page at http://www.idugdb2-l.org/archives/db2-l.html. From that page select "Join or Leave the list". The IDUG DB2-L FAQ is at http://www.idugdb2-l.org. The IDUG List Admins can be reached at DB2-L-REQUEST@www.idugdb2-l.org. Find out the latest on IDUG conferences at http://conferences.idug.org/index.cfm
3631 51 15_Re: DB2 V8 z/OS13_David Simpson22_DSIMPSO@TRANSUNION.COM30_Thu, 9 Nov 2006 09:08:42 -0600518_US-ASCII It took us 9 months to go through all the modes for 15 subsystems, so it sounds typical of the experience being reported.
One other thing that was a huge benefit was borrowing a project manager to manage the administrative and communication issues for testing and implementing in all the modes. We actually plotted it in MS Project and had meetings with the app teams every 2 weeks to discuss testing efforts and adjust the schedule as necessary. This helps people take the effort more seriously. [...]
3683 436 46_Rough notes from IBM IOD conference (Oct15-20)11_Rick Butler22_richard.butler@BMO.COM30_Thu, 9 Nov 2006 09:06:45 -0600466_ISO-8859-1 Below are and notes and presentation materials from the IBM Information on Demand Conference, which was actually Six conferences in one, combining: BI Customer Solutions Summit,CM Technical Conference,DB2 IM Technical Conference,IMS Technical Conference,Information Integration Live Master Data Management Conference
If you have a short attention span, :) , here is an IBM ¡§commercial¡¨ on the new ¡§IBM Information server¡¨ on youtube go to [...]
4120 58 33_Websphere UDB performance Problem0_24_db46@DAIMLERCHRYSLER.COM30_Thu, 9 Nov 2006 10:28:35 -0500412_US-ASCII We have a UDB database on an AIX box that is used three different applications. It is also used through some web services calls for sending information to and from this database.
Our production box was at version 8 fixpack 6 and was upgraded to fixpack 12. It started out fine, but all of a sudden one of the webservices calls started to degraded in performance to actually stop responding. [...]
4179 211 22_DB2 V7.1 and z/OS V1.415_Foweather, Iris30_Iris.Foweather@UK.EXPERIAN.COM30_Thu, 9 Nov 2006 15:38:19 -0000610_us-ascii Sorry to resend so soon but I was really hoping some kindly soul out there can help remove some of my ignorance of TCP/IP keepalive and DB2 and how they work together...
> DB2 V7.1 and z/OS V1.4 > > We are getting timeouts: > > DSNL511I &D2LB DSNLIENO TCP/IP CONVERSATION FAILED > TO LOCATION xxx.xxx.xxx.xxx > IPADDR=xxx.xxx.xxx.xxx PORT=xxx > SOCKET=RECV RETURN CODE=1127 REASON CODE=00000000 > > TCP/IP Keepalive is 2 hours > TCP/IP Keepalive (TCPKPALV) is ENABLE > Idle thread timeout (IDTHTOIN) is 1200 > DDF Threads (CMTSTAT) is INACTIVE > > I am struggling to understand how TCP/IP [...]
4391 301 26_Re: DB2 V7.1 and z/OS V1.413_Mertens, Bart18_bart.mertens@CZ.NL30_Thu, 9 Nov 2006 17:05:56 +0100587_US-ASCII The most likely cause of this problem is a network issue. You can't really make adjustments to ddf which would impact this problem.
________________________________
Van: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] Namens Foweather, Iris Verzonden: donderdag 9 november 2006 16:38 Aan: DB2-L@WWW.IDUGDB2-L.ORG Onderwerp: [DB2-L] DB2 V7.1 and z/OS V1.4
Sorry to resend so soon but I was really hoping some kindly soul out there can help remove some of my ignorance of TCP/IP keepalive and DB2 and how they work together... [...]
4693 279 26_Re: DB2 V7.1 and z/OS V1.414_Sniatecki, Jim31_Jim.Sniatecki@ERIEINSURANCE.COM30_Thu, 9 Nov 2006 11:33:21 -0500571_us-ascii I refer you to the REDBOOK "Distributed Functions of DB2 for z/OS and os/390. See pages 192 ff.
________________________________
From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Foweather, Iris Sent: Thursday, November 09, 2006 10:38 AM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] DB2 V7.1 and z/OS V1.4
Sorry to resend so soon but I was really hoping some kindly soul out there can help remove some of my ignorance of TCP/IP keepalive and DB2 and how they work together... [...]
4973 256 46_Web site performance is critical for retailers35_Joel Goldstein - Responsive Systems26_joel@RESPONSIVESYSTEMS.COM30_Thu, 9 Nov 2006 12:47:46 -0500436_iso-8859-1 DB2 V7.1 and z/OS V1.4I got this from my BBC news. It's interesting that it mirrors - somewhat - the way I deal with web sites. I may have a bit more patience with the initial loading, but have a rather short fuse with lookup, search, and checkout times. I probably cancel several transactions per week during the search or checkout process for anything where there are multiple vendor choices - when it takes too long. [...]
5230 87 24_[z/OS DB2v8] SQLCODE 8029_Sam Baugh24_Sam.Baugh@MOTION-IND.COM30_Thu, 9 Nov 2006 12:21:56 -0600421_us-ascii Just curious if there was a change in SQLCODE returned for "zero divide" exception between DB2v7 and DB2v8. An application failed with an +802 error and supposedly it did not fail because of this before. We have migrated all our systems to DB2v8, so I can't verify this. Would someone still on DB2v7 let me know what return code you get with the following SQL? Thanks! SELECT 10.0 / 0 FROM SYSIBM.SYSDUMMY1 [...]
5318 162 28_Re: [z/OS DB2v8] SQLCODE 80215_Pradeep Gunjala31_Pradeep_Gunjala@TAX.STATE.NY.US30_Fri, 1 Dec 2006 13:32:23 -0500767_ISO-8859-1 yes, getting the same error in V7 and V8(CM).
Thanks, Pradeep
Sam Baugh Sent by: DB2 Data Base Discussion List 11/09/2006 01:21 PM Please respond to DB2 Database Discussion list at IDUG
To DB2-L@WWW.IDUGDB2-L.ORG cc
Subject [DB2-L] [z/OS DB2v8] SQLCODE 802
Just curious if there was a change in SQLCODE returned for ?zero divide? exception between DB2v7 and DB2v8. An application failed with an +802 error and supposedly it did not fail because of this before. We have migrated all our systems to DB2v8, so I can?t verify this. Would someone still on DB2v7 let me know [...]
5481 112 28_Re: [z/OS DB2v8] SQLCODE 80212_Bill Shipley28_william_shipley@VANGUARD.COM30_Thu, 9 Nov 2006 13:38:53 -05001012_UTF-8
Sam, I ran your select in DB2v7 and got the following: ---------+---------+---------+---------+---------+---------+---------+---------+ SELECT 10.0 / 0 FROM SYSIBM.SYSDUMMY1 00010000 ---------+---------+---------+---------+---------+---------+---------+---------+ ---------+---------+---------+---------+---------+---------+---------+---------+ ????????????????? DSNT404I SQLCODE = 802, WARNING: EXCEPTION ERROR ZERO DIVIDE HAS OCCURRED DURING DIVISION OPERATION ON DECIMAL DATA, POSITION 001 DSNT418I SQLSTATE = 01519 SQLSTATE RETURN CODE DSNT415I SQLERRP = DSNXRDEC SQL PROCEDURE DETECTING ERROR DSNT416I SQLERRD = 1009 0 0 -1 0 0 SQL DIAGNOSTIC INFORMATION DSNT416I SQLERRD = X'000003F1' X'00000000' X'00000000' X'FFFFFFFF' X'00000000' X'00000000' SQL DIAGNOSTIC INFORMATION DSNE610I NUMBER OF ROWS DISPLAYED IS 1 DSNE616I STATEMENT EXECUTION WAS SUCCESSFUL, SQLCODE IS 100 ---------+---------+---------+---------+---------+---------+---------+---------+ DSNE617I COMMIT PERFORMED, SQLCODE [...]
5594 95 28_Re: [z/OS DB2v8] SQLCODE 80215_Chris Hoelscher21_choelscher@HUMANA.COM30_Thu, 9 Nov 2006 13:50:40 -0500693_US-ASCII V7.1 z/OS
SELECT 10.0 / 0 FROM SYSIBM.SYSDUMMY1 ; DSNT404I SQLCODE = 802, WARNING: EXCEPTION ERROR ZERO DIVIDE HAS OCCURRED DURING DIVISION OPERATION ON DECIMAL DATA, POSITION 001 ROLLBACK WORK ; DSNT400I SQLCODE = 000, SUCCESSFUL EXECUTION CONTROL PROCESS CARD:
Chris Hoelscher IDMS & DB2 Database Administrator Humana Inc 502-476-2538 choelscher@humana.com
The information transmitted is intended only for the person or entity to which it is addressed and may contain CONFIDENTIAL material. If you receive this material/information in error, please contact the sender and delete or destroy the material/information. [...]
5690 14 28_Re: [z/OS DB2v8] SQLCODE 8029_Sam Baugh24_sam.baugh@MOTION-IND.COM30_Thu, 9 Nov 2006 15:48:22 -0600326_- Thanks for the responses. I guess the problem runs deeper than just the simple divide by zero example. The application uses functions SUM, DEC, FLOAT, ROUND, and VALUE on the dividend and divisor. Perhaps the VALUE function was overriding the exception somehow. I will just say they got away with this exception in v7. [...]
5705 156 28_Re: [z/OS DB2v8] SQLCODE 8027_J Anand34_AnandJayachandran@FREIGHTLINER.COM30_Thu, 9 Nov 2006 15:15:37 -0800452_us-ascii I executed the SQL using DB2 connect & BMC catalog manager against z/OS database(s), SQLCODE = 0
In SPUFI, SQLCODE = 802
Against AIX DB2, I receive error. SQL0801N Division by zero was attempted. SQLSTATE=22012
From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Sam Baugh Sent: Thursday, November 09, 2006 10:22 To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] [z/OS DB2v8] SQLCODE 802 [...]
5862 167 36_Re: Finding out who issued a command15_nguyen duc tuan17_ndt.db2@GMAIL.COM31_Fri, 10 Nov 2006 11:55:12 +0100302_ISO-8859-1 I remember that it is also possible to scan the log (dsn1logp) searching for this command if you didn't have an active trace. The problem i met when i tested this , is that the user associated with the command was SYSOPR , not my userid ... I suppose that there was a bug on my DB2 ... [...]
6030 55 49_[z/OS v8 CM] IBM DB2 Administration Tool for z/OS35_=?iso-8859-1?Q?Kurtz=2C_R=FCdiger?=28_Ruediger.Kurtz@HUK-COBURG.DE31_Fri, 10 Nov 2006 14:07:51 +0100326_iso-8859-1 Hi all,
do any of you know how to change the subsystem while working with IBM's Administration Tool for z/OS (V7.1.0) without having to exit and reinvoke the product ? I just can't find anything in the manual, so it's either not there (and thus most likely not possible) or my glasses need replacement. [...]
6086 27 40_DB2 v7 zOS - Production Control Turnover11_Bear, Brian23_Brian.Bear@CHARMING.COM31_Fri, 10 Nov 2006 09:49:57 -0500479_US-ASCII Dear Listers;
Our procedures for promoting DB2 programs include Production Control staff pre-compiling and binding the DB2 program/paakage into production. Currently these Production Control staff have been granted SYSADM authority to facilitate the authority requirements for updating/deleting/inserting, within the packages.
I am looking to change the process in some way that will no longer require granting SYSADM to the Production Control staff. [...]
6114 18 46_Re: Using the log as application tracking tool12_Phil Gunning22_pgunning@GUNNINGTS.COM31_Fri, 10 Nov 2006 08:48:04 -0600420_- -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
I've seen both application and using the DB2 log solution used. With the application there was concern regarding ability to provide requisite response time but that turned out to not be a problem (CICS/DB2 version of state National Crime Information Center (NCIC) implementation. However, I lean toward using the logs and software to read them later. HTH Phil [...]
6133 49 44_Re: DB2 v7 zOS - Production Control Turnover15_Robert Lawrence21_rlawrence@BOSCOVS.COM31_Fri, 10 Nov 2006 09:57:23 -0500684_us-ascii BINDADD authority should do it
Bob Lawrence DBA Boscovs Dept Stores LLC
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Bear, Brian Sent: Friday, November 10, 2006 9:50 AM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] DB2 v7 zOS - Production Control Turnover
Dear Listers;
Our procedures for promoting DB2 programs include Production Control staff pre-compiling and binding the DB2 program/paakage into production. Currently these Production Control staff have been granted SYSADM authority to facilitate the authority requirements for updating/deleting/inserting, within the packages. [...]
6183 53 44_Re: DB2 v7 zOS - Production Control Turnover37_=?iso-8859-2?B?UG9zcO25aWwgRHW5YW4=?=17_DPospisil@CSAS.CZ31_Fri, 10 Nov 2006 16:12:54 +0100439_iso-8859-2 1. we don't pre-compile at production; we only move load-modules and dbmr and bind them in production 2. the bind is done by dbadm
Regards,
dp
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Bear, Brian Sent: Friday, November 10, 2006 3:50 PM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] DB2 v7 zOS - Production Control Turnover [...]
6237 74 44_Re: DB2 v7 zOS - Production Control Turnover9_Mike Bell21_mbell11a1@VERIZON.NET31_Fri, 10 Nov 2006 09:24:34 -0600655_us-ascii Aren't you thinking of bind agent?
Mike
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Robert Lawrence Sent: Friday, November 10, 2006 8:57 AM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] DB2 v7 zOS - Production Control Turnover
BINDADD authority should do it
Bob Lawrence DBA Boscovs Dept Stores LLC
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Bear, Brian Sent: Friday, November 10, 2006 9:50 AM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] DB2 v7 zOS - Production Control Turnover [...]
6312 71 44_Re: DB2 v7 zOS - Production Control Turnover15_Robert Lawrence21_rlawrence@BOSCOVS.COM31_Fri, 10 Nov 2006 10:22:42 -0500441_us-ascii Oops BINDAGENT not BINDADD
Bob Lawrence DBA Boscovs Dept Stores LLC
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Robert Lawrence Sent: Friday, November 10, 2006 9:57 AM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] DB2 v7 zOS - Production Control Turnover
BINDADD authority should do it
Bob Lawrence DBA Boscovs Dept Stores LLC [...]
6384 75 65_Re: [Maybe Spam] [DB2-L] DB2 v7 zOS - Production Control Turnover13_Bell, Raymond22_raymond.bell@LANDG.COM31_Fri, 10 Nov 2006 15:32:25 -0000604_- BINDAGENT is the privilege that'll do what you want. Your Sysadm could grant Bindagent to your Prod Control bods and they could specify this ID in the OWNER clause of the bind statements - problem half solved.
A better solution is to create a bogus package owner (something like PRODPACK - up to you), grant PRODPACK select, insert, update, delete on all your Production tables/views to PRODPACK, have PRODPACK grant Bindagent to your Prod Control bods and get your Prod Control bods to specify OWNER(PRODPACK) in the binds. Bingo! They can bind the statements but don't have access to the [...]
6460 96 44_Re: DB2 v7 zOS - Production Control Turnover15_Robert Lawrence21_rlawrence@BOSCOVS.COM31_Fri, 10 Nov 2006 10:41:46 -0500402_us-ascii Yes and I sent a correction soon as I saw that my fast fingers had a mind of their own
Bob Lawrence DBA Boscovs Dept Stores LLC
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Mike Bell Sent: Friday, November 10, 2006 10:25 AM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] DB2 v7 zOS - Production Control Turnover [...]
6557 94 48_ENC: [DB2-L] Websphere Default Isolation for DB219_Luiz Roberto Foroni27_lforoni@CAIXASEGUROS.COM.BR31_Fri, 10 Nov 2006 15:18:57 -0200634_iso-8859-1 Hello Bart,
Is this property suitable for websphere acessing DB2 Z/OS ? Since the isolation level is a column of sysibm.syspackage table, what is the relation between them ?
Thanks in advance
Foroni Administrador de Banco de Dados GEINF - Gerência de Engenharia das Infras-estruturas Tecnológicas CAIXA SEGUROS TEL 55-61-2192-2939 FAX 55-61-3328-2108
-----Mensagem original----- De: Mertens, Bart [mailto:bart.mertens@CZ.NL] Enviada em: quarta-feira, 11 de outubro de 2006 06:13 Para: DB2-L@WWW.IDUGDB2-L.ORG Assunto: Re: [DB2-L] Websphere Default Isolation for DB2 [...]
6652 143 65_Re: [Maybe Spam] [DB2-L] DB2 v7 zOS - Production Control Turnover11_Bear, Brian23_Brian.Bear@CHARMING.COM31_Fri, 10 Nov 2006 13:32:24 -0500617_US-ASCII Thanks too all who replied. BINDAGENT it is.
Raymond, to apply your better solution, are you saying SET CURRENT SQLID='PRODPACK"; GRANT BINDAGENT TO authid; then as authid: BIND PACKAGE .... OWNER(PRODPACK); assuming PRODPACK has sysadm or any appropriate required authorties to exec SQL in the package?
Thanks Brian Bear
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] Sent: Friday, November 10, 2006 10:32 AM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] [Maybe Spam] [DB2-L] DB2 v7 zOS - Production Control Turnover [...]
6796 39 23_DBA Performance Metrics10_Ravi Reddy19_dbadb2udb@YAHOO.COM31_Fri, 10 Nov 2006 15:51:31 -0800344_iso-8859-1 Hello All,
My Boss says DBA group are not representing themselves in the right way to senior mgmt in our organization. This is because we do not have formal metrics/measurements for DBA work to showcase and publish as a monthly/quartely report.
And iam asked to come up with metrics/measurements for our DBA work. [...]
6836 75 27_Re: DBA Performance Metrics14_Andy Lankester26_alankester@CDBSOFTWARE.COM31_Fri, 10 Nov 2006 18:09:51 -0600618_windows-1250 Total GB of data managed? Number of objects? Both tend to increase!
Walk tall!
Andy Lankester
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Ravi Reddy Sent: Friday, November 10, 2006 11:52 PM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] DBA Performance Metrics
Hello All,
My Boss says DBA group are not representing themselves in the right way to senior mgmt in our organization. This is because we do not have formal metrics/measurements for DBA work to showcase and publish as a monthly/quartely report. [...]
6912 73 27_Re: DBA Performance Metrics12_Gerald Hodge26_ghodge@HLSTECHNOLOGIES.COM31_Fri, 10 Nov 2006 18:51:15 -0800522_US-ASCII If you are a publicly traded company then your management publishes a 10K, 10Q and an annual report. In there are the things that upper management measures. Read the report and try and restate your service in those terms.
For example state the number of customers supported and the average amount of data required per customer and then report those growth metrics. DB2 does not have transactions, but there should be some measure of the number of data access, again restated in terms of the business. [...]
6986 54 36_On-Line REORG Elapsed Time Increases15_Barbara Jo Nigh26_barbara.j.nigh@US.HSBC.COM31_Fri, 10 Nov 2006 17:10:53 -0800367_US-ASCII We have weekly scheduled conditional online REORGs for a large partitioned tablespace. There are no NPIs defined on the table. I have noticed that in the last month or so, the elapsed time has been increasing from 3 hours to between 5 and 6 hours. The CPU time is still around 15 minutes. The SORTKEYS parameter is being specified as part of the REORG. [...]
7041 195 40_Re: On-Line REORG Elapsed Time Increases13_Michael Ebert18_mebert@AMADEUS.COM31_Sat, 11 Nov 2006 08:42:05 +0100599_US-ASCII DB2 uses the HI-U-RBA of the VSAM file along with RECLENGTH, PAGESAVE and AVGROWLEN and possibly additional information - IBM hasn't published the algorithm, as far as I know. That DB2 over-estimates the number of rows so much indicates that something is out of whack; I've only ever had the opposite situation (which leads to x37 ABENDs). However, it is not related to runtimes. Such a long elapsed time (whether 3 or 6 hours) for such a small table most likely indicates that your job is not getting CPU, either because of an incorrect WLM setup/priorities assignment or because the [...]
7237 207 40_Re: On-Line REORG Elapsed Time Increases13_Horacio Villa17_hvilla@AR.IBM.COM31_Sat, 11 Nov 2006 10:54:40 -0300795_US-ASCII Hi Barbara,
you say 4.17 millions in each partition. 27.6 million records should be for the entire tablespace.
Cheers,
Horacio Villa
Barbara Jo Nigh Sent by: DB2 Data Base Discussion List 10/11/2006 22:10 Please respond to DB2 Database Discussion list at IDUG
To DB2-L@WWW.IDUGDB2-L.ORG cc
Subject [DB2-L] On-Line REORG Elapsed Time Increases
We have weekly scheduled conditional online REORGs for a large partitioned tablespace. There are no NPIs defined on the table. I have noticed that in the last month or so, the elapsed time has been increasing from 3 hours to between 5 and 6 hours. [...]
7445 176 65_Re: [Maybe Spam] [DB2-L] DB2 v7 zOS - Production Control Turnover9_Mike Bell21_mbell11a1@VERIZON.NET31_Sat, 11 Nov 2006 08:43:47 -0600432_US-ASCII One minor note - when it comes time to disentangle the production migration group from their sysadm authority, you do not want to just revoke the sysadm unless you don't mind all the existing plans and packages being invalidated.
The Administration guide has a section on how safely remove SYSADM by making them an install sysadm, then doing the revoke, then setting it back to normal. Also check the archives. [...]
7622 175 65_Re: [Maybe Spam] [DB2-L] DB2 v7 zOS - Production Control Turnover13_Bell, Raymond22_raymond.bell@LANDG.COM31_Sun, 12 Nov 2006 08:32:53 -0000525_- Yeah, basically. You grant PRODPACK the table/view privileges it needs to successfully be the package owner (you know, anything the program is trying to do, PRODPACK should be authorised for, so ordinariily just SIUD - or CRUD as it was called at SOA thing I went to a few weeks ago), then PRODPACK grants Bindagent to the bods going the binding (authid in your example) and authid does the binds. Not sure if authid can be a group or not; knowing plan/package privileges probably not, but it shouldn't be a big deal. [...]
7798 21 59_Disaster Recovery : How to create archive from active log ?15_Daniel Cremieux17_c300501@YAHOO.COM31_Sun, 12 Nov 2006 04:22:34 -0600281_- Dear all, Our DB2 Z/OS signaled a B37 when trying to archive the full active log, then it continues processing without archiving this active log. I try to create manually an archive , but don't know how to (get some abends) due to the specific format of the archive log ... [...]
7820 44 63_Re: Disaster Recovery : How to create archive from active log ?15_Marcel Harleman25_marcel.harleman@HCCNET.NL31_Sun, 12 Nov 2006 16:35:20 +0100577_us-ascii Haven't tested it, but my guess would be: adjust the DSNZPARM for the archive log allocation, do a -SET SYSPARM(LOAD) (I think it's online changeable) en issue a -ARCHIVE LOG command.
Hope this helps.
Marcel.
>Dear all, >Our DB2 Z/OS signaled a B37 when trying to archive the full active log, then > it continues processing without archiving this active log. >I try to create manually an archive , but don't know how to (get some >abends) due to the specific format of the archive log ... > >How to achieve this ? > >We are going to the last [...]
7865 106 27_Re: DBA Performance Metrics12_Ayalew Kassa20_ayalakassa@YAHOO.COM31_Sun, 12 Nov 2006 09:22:05 -0800367_iso-8859-1 You many need to mention what your team does to your managment. Such as
1. Production issues that you resolved --- what will be the impact if the DB2 DBAs were not involved in resolving the production problem ?
2. 24 x 7 support of the database == if the DBAs do not support the database in 24 X7 what will be the impact on the company. [...]
7972 186 53_Antwort: [DB2-L] On-Line REORG Elapsed Time Increases11_Roy Boxwell16_R.Boxwell@SEG.DE31_Mon, 13 Nov 2006 07:54:46 +0100302_us-ascii I would add the SUBPROCESS paramter to the utility JCL so that you can easily see where all the elpased time is going. From that you could then try and "tune" the reorg parameters so that more/less bigger/smaller work files are allocated or any other reorg parms could/should be changed. [...]
8159 135 52_Re: ENC: [DB2-L] Websphere Default Isolation for DB213_Mertens, Bart18_bart.mertens@CZ.NL31_Mon, 13 Nov 2006 09:00:33 +0100659_iso-8859-1 Foroni,
DB2 connect (or type4 java driver) knows which package to execute to use the desired isolation level (and settings like cursor hold).
More info: http://www-1.ibm.com/support/docview.wss?rs=180&uid=swg21224492 http://www-1.ibm.com/support/docview.wss?uid=swg21162747 http://www-1.ibm.com/support/docview.wss?rs=180&uid=swg21224492
Regards, Bart
-----Oorspronkelijk bericht----- Van: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] Namens Luiz Roberto Foroni Verzonden: vrijdag 10 november 2006 18:19 Aan: DB2-L@WWW.IDUGDB2-L.ORG Onderwerp: [DB2-L] ENC: [DB2-L] Websphere Default Isolation for DB2 [...]
8295 56 63_Re: Disaster Recovery : How to create archive from active log ?11_Mike Turner29_michael_turner@COMPUSERVE.COM31_Mon, 13 Nov 2006 09:16:56 -0000404_iso-8859-1 Daniel
You may need to add another active log dataset to keep DB2 running in the short term, and you must also resolve the issue that caused the B37. But you do not need to manually archive the active log dataset. Every time the DB2 archive process is triggered (by a full active log dataset or by -ARCHIVE LOG command) the archive process will retry any failed previous archives. [...]
8352 35 28_Locking and Tablespace Scans4_Tony15_tmoore@IKON.COM31_Mon, 13 Nov 2006 06:51:03 -0600340_- Yo Listers, I know this is, like, DB2 101, but it's been about a decade since I've been in a class. I have an static (bound) COBOL program (stored procedure) declaring and using a cursor with a "FOR UPDATE OF". Nothing fancy, it declares the cursor, opens it, fetches through the rows and updates certain ones and closes the cursor. [...]
8388 14 53_Re: [z/OS v8 CM] IBM DB2 Administration Tool for z/OS12_Adam Baldwin19_y0027528@ES.IBM.COM31_Mon, 13 Nov 2006 07:56:33 -0600578_- You can always use the distributed option:
DD - Distributed DB2 systems
from the main menu.
--------------------------------------------------------------------------------- Welcome to the IDUG DB2-L list. To unsubscribe, go to the archives and home page at http://www.idugdb2-l.org/archives/db2-l.html. From that page select "Join or Leave the list". The IDUG DB2-L FAQ is at http://www.idugdb2-l.org. The IDUG List Admins can be reached at DB2-L-REQUEST@www.idugdb2-l.org. Find out the latest on IDUG conferences at http://conferences.idug.org/index.cfm
8403 40 61_AW: [DB2-L] [z/OS v8 CM] IBM DB2 Administration Tool for z/OS35_=?iso-8859-1?Q?Kurtz=2C_R=FCdiger?=28_Ruediger.Kurtz@HUK-COBURG.DE31_Mon, 13 Nov 2006 15:52:45 +0100448_iso-8859-1 Adam,
guess my glasses want replacing after all.
Ruediger
-----Ursprüngliche Nachricht----- Von: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] Im Auftrag von Adam Baldwin Gesendet: Montag, 13. November 2006 14:57 An: DB2-L@WWW.IDUGDB2-L.ORG Betreff: Re: [DB2-L] [z/OS v8 CM] IBM DB2 Administration Tool for z/OS
You can always use the distributed option:
DD - Distributed DB2 systems [...]
8444 99 89_DB2 Websphere Information Integrator v8.2 (may be known by another name at this time)....18_Whittaker, Stephen29_stephen.whittaker@PGNMAIL.COM31_Mon, 13 Nov 2006 09:55:22 -0500470_us-ascii We are currently using this product on a Windows 2003 machine with a 32-bit Operating System. Lately we have been having performance issues and IBM's recommendation is that we upgrade to a 64-Bit operating system.
Question: Has anyone out there used this product/version and either upgraded to a 64-bit OS or installed it on Windows 64-bit OS?? If so were you able to stay with DB2 UDB v8.2 ?? We're accessing both DB2 and Oracle data at this time. [...]
8544 74 32_Re: Locking and Tablespace Scans12_Isaac Yassin20_yassini@BEZEQINT.NET31_Mon, 13 Nov 2006 18:14:01 +0200409_us-ascii Hi,
As you say - it's CURSOR FOR UPDATE which means U-lock on each row read (it get changed to X on rows that you really update.
Isaac Yassin
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Tony Sent: Monday, November 13, 2006 2:51 PM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] Locking and Tablespace Scans [...]
8619 26 53_Command Editor giving tw different query result sets.19_Ayalew "Mark" Kassa20_ayalakassa@YAHOO.COM31_Mon, 13 Nov 2006 10:28:24 -0600827_- I used command editor to execute same SQL statment in UDB and z/OS database managment system. However, when I get the result it is different.
The SQL statment result from UDB is as shown
+1.00000E+000 +3.93700E-002 +3.93700E-001 +3.28080E+000 +1.09360E+000 +6.21370E-001 +1.00000E+000 +1.00000E+000 +1.00000E+000 +1.00000E+000 +1.00000E+000
--------------------------------------------------------------------------------- Welcome to the IDUG DB2-L list. To unsubscribe, go to the archives and home page at http://www.idugdb2-l.org/archives/db2-l.html. From that page select "Join or Leave the list". The IDUG DB2-L FAQ is at http://www.idugdb2-l.org. The IDUG List Admins can be reached at DB2-L-REQUEST@www.idugdb2-l.org. Find out the latest on IDUG conferences at http://conferences.idug.org/index.cfm
8646 46 89_Command Editor giving two different query result sets between UDB and z/OS database table19_Ayalew "Mark" Kassa20_ayalakassa@YAHOO.COM31_Mon, 13 Nov 2006 10:34:13 -0600322_- I used command editor to execute same SQL statment in UDB and z/OS database managment system. However, the result of this SQL statment is different between this two subsystems. The table which is in UDB and z/OS has the same rows. I just created the table in z/OS and loaded the data using from UDB database table. [...]
8693 167 106_Antwort: [DB2-L] Command Editor giving two different query result sets between UDB and z/OS database table11_Roy Boxwell16_R.Boxwell@SEG.DE31_Mon, 13 Nov 2006 18:05:01 +0100516_us-ascii looks like you're not using a FLOAT column definition on z/OS to me!
Roy Boxwell SOFTWARE ENGINEERING GMBH -Product Development- Robert-Stolz-Strasse 5 40470 Duesseldorf/Germany Tel. +49 (0)211 96149-0 Fax +49 (0)211 96149-35 E-mail r.boxwell@seg.de Homepage www.seg.de
"Ayalew \"Mark\" Kassa" Gesendet von: DB2 Data Base Discussion List 13.11.2006 17:34 Bitte antworten an DB2 Database Discussion list at IDUG [...]
8861 30 32_Re: Locking and Tablespace Scans14_Peter Vanroose26_peter_vanroose@YAHOO.CO.UK31_Mon, 13 Nov 2006 18:06:36 +0100306_iso-8859-1 > it's CURSOR FOR UPDATE which means U-lock on each row read > it get changed to X on rows that you really update.
Actually, the rows being effectively locked also depends on the isolation level. With RR, the whole table gets locked, even with a LOCKSIZE ROW in the tablespace decl. [...]
8892 102 110_Re: Antwort: [DB2-L] Command Editor giving two different query result sets between UDB and z/OS database table12_Ayalew Kassa20_ayalakassa@YAHOO.COM31_Mon, 13 Nov 2006 09:26:01 -0800518_iso-8859-1
I am doing this from the command editor window... May be the setting of the command editor or JDBC acts differently when receiving SQL results in z/OS and UDB Dist.
Roy Boxwell wrote:
looks like you're not using a FLOAT column definition on z/OS to me!
Roy Boxwell SOFTWARE ENGINEERING GMBH -Product Development- Robert-Stolz-Strasse 5 40470 Duesseldorf/Germany Tel. +49 (0)211 96149-0 Fax +49 (0)211 96149-35 E-mail r.boxwell@seg.de Homepage www.seg.de [...]
8995 212 40_Re: On-Line REORG Elapsed Time Increases15_Barbara Jo Nigh26_barbara.j.nigh@US.HSBC.COM31_Mon, 13 Nov 2006 10:11:59 -0800478_ISO-8859-1 Horacio,
There are 10 partitions. That would equal 41 million rows not 27 million.
Barbara J. Nigh - CAL261 Consultant, Database Management Systems - North America Phone: 831.754.1400 Ext. 2048 Pager: 877.406.7868 barbara.j.nigh@us.hsbc.com
Horacio Villa To Sent by: DB2 Data DB2-L@www.idugdb2-l.org Base Discussion cc List Re: [DB2-L] On-Line REORG Elapsed Time Increases [...]
9208 36 13_DB2 v8 on zOS14_Ramon Santiago28_ramon_santiago@SBCGLOBAL.NET31_Mon, 13 Nov 2006 12:31:00 -0800333_iso-8859-1 I have a segmented table in which we only do inserts and no updates. Multiple batch jobs are inserting to the same table. I am getting lock escalations.
Would LOCKMAX 0 be wise choice to prevent lock escalation? Currently the locksize is ANY and LOCKMAX SYSTEM
ZPARMS NUMLKUS ==> 10000 NUMLKTS ==> 1000 [...]
9245 36 35_cost for DB2 to back out SQL errors9_Jim Stoll20_stollj@MECH.DISA.MIL31_Mon, 13 Nov 2006 14:39:14 -0600538_- DB2 listers,
We are trying to determine the cost/number of instructions it takes DB2 to roll back SQL errors. Below is a typical one hour snapshot of the number of SQL errors one of our customers receives. Does anyone have a matrix and or formula to determine what it takes for DB2 to back out errors? We would like to provide this information to the programming department to indicate the extra work DB2 must do to clear these errors. Hopefully this information would help push the programmers into cleaning up their code. [...]
9282 54 17_Re: DB2 v8 on zOS10_Dave Nance16_dav1mo@YAHOO.COM31_Mon, 13 Nov 2006 13:32:30 -0800560_iso-8859-1 Ramon, I would say a few items need to be looked at here.
First, an answer to your question about escalation. I have always set lockmax to 0 on systems that I've supported. I have seen others here on the list and I, believe, someone wrote a paper on it as well, who think the same, why would you ever want locks to get escalated? The higher the lock level escalates the more problems you can run into. Why are these insert processes inserting so much data without commits? Can you change the commit frequency? A lot of applications will [...]
9337 60 17_Re: DB2 v8 on zOS21_wfavero@attglobal.net21_wfavero@ATTGLOBAL.NET31_Mon, 13 Nov 2006 16:42:19 -0500560_iso-8859-1 Have you been able to determine why you are going into lock escalation?
My second comment would be if you don't want escalation, why not just specify LOCKSIZE PAGE?
Willie
Original Message: ----------------- From: Ramon Santiago ramon_santiago@SBCGLOBAL.NET Date: Mon, 13 Nov 2006 12:31:00 -0800 To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] DB2 v8 on zOS
I have a segmented table in which we only do inserts and no updates. Multiple batch jobs are inserting to the same table. I am getting lock escalations. [...]
9398 54 17_Re: DB2 v8 on zOS14_Ramon Santiago28_ramon_santiago@SBCGLOBAL.NET31_Mon, 13 Nov 2006 13:52:54 -0800567_iso-8859-1 Thank you for your input. Dave Nance wrote: Ramon, I would say a few items need to be looked at here.
First, an answer to your question about escalation. I have always set lockmax to 0 on systems that I've supported. I have seen others here on the list and I, believe, someone wrote a paper on it as well, who think the same, why would you ever want locks to get escalated? The higher the lock level escalates the more problems you can run into. Why are these insert processes inserting so much data without commits? Can you [...]
9453 96 17_Re: DB2 v8 on zOS9_Mike Bell21_mbell11a1@VERIZON.NET31_Mon, 13 Nov 2006 15:58:17 -0600448_us-ascii Let me add to that - a lock is a lock is a lock - the only time page locking helps is if you have multiple rows on the same page that are referenced or updated by the same transaction. This usually means a batch process with sorted input that will update in clustering order. Online transactions usually don't work that way. If you have 3 rows on 3 different pages, you still have 3 locks whether you specify row locks or page locks. [...]
9550 84 47_Fw: [DB2-L] cost for DB2 to back out SQL errors0_22_BRIAN_GOLDBERG@QVC.COM31_Mon, 13 Nov 2006 17:12:01 -0500491_US-ASCII Jim,
DB2 does not automatically do ROLLBACK on all SQL errors; that is an application decision to perform the ROLLBACK.
A SQLCODE of -911 will do a ROLLBACK, but I don't see that on your list.
Brian ----- Forwarded by BRIAN GOLDBERG/QVC on 11/13/2006 05:06 PM -----
Jim Stoll To Ext: DB2-L@WWW.IDUGDB2-L.ORG Sent by: DB2 Data cc Base Discussion List Subject SQL errors [...]
9635 138 51_Re: Fw: [DB2-L] cost for DB2 to back out SQL errors12_Carol Sutfin24_carol.sutfin@REGIONS.COM31_Mon, 13 Nov 2006 17:12:59 -0600315_us-ascii The cost will also depend on "how much activity" gets rolled back.
Is it ten units of work, or is it 100 units of work. How many log files are still active or do you need to go to Archive logs.
Also, batch or on-line activity is being rolled back.
Cost and/or time, it "depends". [...]
9774 154 17_Re: DB2 v8 on zOS26_Sorensen Henrik (KSFA 321)33_henrik.sorensen@CREDIT-SUISSE.COM31_Tue, 14 Nov 2006 09:30:06 +0100338_us-ascii Mike Bell wrote: >Third point - segemented tablespace with massive insert volume is going to >have contention on the space map page unless you can preformat the >tablespace and scatter the inserts. This means specifying PREFORMAT on LOAD >and REORG and making sure the primary quantity is big enough for normal >processing. [...]
9929 151 52_Antwort: [DB2-L] cost for DB2 to back out SQL errors11_Roy Boxwell16_R.Boxwell@SEG.DE31_Tue, 14 Nov 2006 12:20:01 +0100401_us-ascii best answer I have heard came from Roger Miller or Terry Purcell who said "DB2 Rollback code has *not* been optimized" - I take that to mean that you should avoid rollbacks......
Roy Boxwell SOFTWARE ENGINEERING GMBH -Product Development- Robert-Stolz-Strasse 5 40470 Duesseldorf/Germany Tel. +49 (0)211 96149-0 Fax +49 (0)211 96149-35 E-mail r.boxwell@seg.de Homepage www.seg.de [...]
10081 109 39_Re: cost for DB2 to back out SQL errors0_19_mike.holmans@BT.COM31_Tue, 14 Nov 2006 12:15:14 -0000512_us-ascii If I were your programmers, I would tell you that I will do no such thing because I don't want to adversely affect the performance of the system.
You can, if you like, precede an INSERT with a SELECT to check whether a row with that key already exists, and you then won't get -803s. But in our shop, we regularly encourage programmers to make the assumption that the row won't exist in 99.999% of cases and code a check to handle the occasional -803 when someone does try duplicating a row. [...]
10191 12 39_Re: cost for DB2 to back out SQL errors33_Walter Jani=?ISO-8859-1?Q?=DFen?=27_walter.janissen@VICTORIA.DE31_Tue, 14 Nov 2006 06:37:41 -0600496_- Mike
Well spoken
--------------------------------------------------------------------------------- Welcome to the IDUG DB2-L list. To unsubscribe, go to the archives and home page at http://www.idugdb2-l.org/archives/db2-l.html. From that page select "Join or Leave the list". The IDUG DB2-L FAQ is at http://www.idugdb2-l.org. The IDUG List Admins can be reached at DB2-L-REQUEST@www.idugdb2-l.org. Find out the latest on IDUG conferences at http://conferences.idug.org/index.cfm
10204 59 25_[DB2 V8 NFM] Why 2 sorts?33_Walter Jani=?ISO-8859-1?Q?=DFen?=27_walter.janissen@VICTORIA.DE31_Tue, 14 Nov 2006 06:42:12 -0600694_- Hi listers
I am pondering, why I get 2 sorts for the following query:
SELECT AVDB2.GES_SL , AVDB.BDAGT , CHAR('!') , CHAR(COUNT(*)) , CHAR('!') , CHAR(SUM(VERT.VERTR_GES_BEITR)) FROM DB2.VETB0040 VERT , DB2.VATB0890 AVDB2 , (SELECT SUBSTR(CHAR(DECIMAL(AVDB.OE_NR)), 4, 9) AS BDAGT , VNR FROM DB2.VATB0860 AVDB WHERE AVDB.V_VERW_SYS_SL = 'SV' AND HIST_LNR = 1 ) AS AVDB WHERE VERT.V_VERW_SYS_SL = 'SV' AND VERT.HIST_LNR = 1 AND VERT.BEARB_GRD = '10' AND VERT.PT_KENN IN ('DASRRS', 'DASRRV', 'DASRSP') AND AVDB.VNR = VERT.VNR AND AVDB2.V_VERW_SYS_SL = 'SV' AND AVDB2.VNR = VERT.VNR AND AVDB2.HIST_LNR = 1 AND AVDB2.GES_SL = '0071' GROUP BY AVDB2.GES_SL , AVDB.BDAGT ORDER BY [...]
10264 125 17_Re: DB2 v8 on zOS13_Carol Broyles21_clbroyles55@YAHOO.COM31_Tue, 14 Nov 2006 05:50:17 -0800596_ascii I would caution anyone with data sharing about setting lockmax to 0. The problem with not allowing escalation is that a couple of rouge programs can bring down your entire data sharing group by blowing out your lock structure. You could elect, I suppose, to set NUMLKUS to a low enough value to abend the program instead, but that doesn't work for us (SAP shop with row level locking). Our lock structure is 1GB, and it's still been exceeded by some badly behaving programs. We set NUMLKUS to 512,000 and NUMLKTS to 500,000. Carol Broyles Infrastructure Management Consultant ACS Inc. [...]
10390 78 44_DB2 V8 z/OS (ENFM) Locks on DSNDB06.SYSSTATS15_Moss, William R15_MossW@AETNA.COM31_Tue, 14 Nov 2006 08:48:06 -0500422_US-ASCII I've run into a problem with the UTILTERM step of the Load Utility when the utility includes updating runstats with inline statistics. The UTILTERM step takes up to five minutes to complete as it tries to acquire an IX lock on DSNDB06.SYSSTATS. I looked at the lock waits in Omegamon and saw that the lock being waited on was an IS lock that was being held by a batch job running dynamic SQL using DSNTIAUL. [...]
10469 279 17_Re: DB2 v8 on zOS13_Bell, Raymond22_raymond.bell@LANDG.COM31_Tue, 14 Nov 2006 14:03:01 -0000498_- Yeah, you gotta watch those red programs. We've got a couple of blau ones that cause us problems. Fortunately I'm not colour-blind so I can usually spot them...
Sorry, caught me at an odd moment (aren't they all? I hear you cry)
Raymond Bell Database Administrator
_____
From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Carol Broyles Sent: 14 November 2006 13:50 To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] DB2 v8 on zOS [...]
10749 100 17_Re: DB2 v8 on zOS10_Max Scarpa28_Massimo.Scarpa@PHOENIXSPA.IT31_Tue, 14 Nov 2006 15:21:26 +0100376_US-ASCII
YES RAYMOND they are ALL ! I could say if for every microsecond.
Even here we have red (or brown) programs with bad performance, expecially about LOCK ESCALATION as programmers do not want to manage 00C90096 error (max lock reached) and I see every day a carnage of transactions caused by batch lock escalation reaching the highest level. [...]
10850 163 48_Re: DB2 V8 z/OS (ENFM) Locks on DSNDB06.SYSSTATS13_Michael Ebert18_mebert@AMADEUS.COM31_Tue, 14 Nov 2006 15:34:17 +0100316_US-ASCII The lock is held only while the cursor is being opened, subsequent FETCHes have no need for these locks, maybe that explains it (Omegamon shows this in "Current SQL statement"). My experience with inline stats has been that they offer no advantage but a lot of serious disadvantages. This is one more. [...]
11014 23 23_DB2 9 LUW XML licensing12_Matthew Rhea24_matthew_rhea@CONSECO.COM31_Tue, 14 Nov 2006 14:32:25 -0600488_- List,
Has anyone heard that the XML feature of DB2 9 on LUW is a separate chargeable license? As is the storage feature that allows you to use table compression? Our IBM sales reps are insisting that there is a separate license for each of those features, yet when we installed DB2 9 AIX we could immediately use the XML features. I would think that the features would be disabled until you had a license. I haven't seen any of the literature mention the separate licenses. [...]
11038 93 27_Re: DB2 9 LUW XML licensing0_28_BILL.GALLAGHER@PHOENIXWM.COM31_Tue, 14 Nov 2006 15:56:15 -0500519_US-ASCII I've never heard anything about a separate charge for either of these features. I've been to a number of IBM presentations and webinars talking about DB2 v9 in general, and these features specifically, and these features have always been touted as being part of the base product.
------------------------------------------------------- Bill Gallagher, Logical DBA IT / Infrastructure and Common Services Phoenix Life Insurance Company (W) 860-403-6327 (C) 860-539-5576 bill.gallagher@phoenixwm.com [...]
11132 25 45_DB2 V8 nfm with z/OS V1.7 and BACKUP function15_Lizette Koehler23_starsoul@MINDSPRING.COM31_Tue, 14 Nov 2006 16:29:20 -0500433_UTF-8 Doing a quick survey.
Is anyone at DB2 V8 nfm with z/OS V1.7 using the DB2 utility BACKUP?
If so,
1) Has it helped in time to backup and recover? 2) Can you use it in a DB2 Data Sharing Environment? Or is it only non DB2 Data Sharing environment? 3) Can you use it with any hardware or specific hardware? For example, only can be used with SHARK or can it be used with TerraData, EMC DMX, or Hitachi? [...]
11158 16 37_DB2 9 XML and other Options Licensing12_Phil Gunning22_pgunning@GUNNINGTS.COM31_Tue, 14 Nov 2006 16:21:30 -0600701_- Yes, there are several separately priced options. See the announcement letter at http://www-306.ibm.com/common/ssi/fcgi-bin/ssialias? subtype=ca&infotype=an&appname=iSource&supplier=897&letternum=ENUS206-128:
Phil
--------------------------------------------------------------------------------- Welcome to the IDUG DB2-L list. To unsubscribe, go to the archives and home page at http://www.idugdb2-l.org/archives/db2-l.html. From that page select "Join or Leave the list". The IDUG DB2-L FAQ is at http://www.idugdb2-l.org. The IDUG List Admins can be reached at DB2-L-REQUEST@www.idugdb2-l.org. Find out the latest on IDUG conferences at http://conferences.idug.org/index.cfm
11175 60 39_Re: cost for DB2 to back out SQL errors9_Jeff Todd18_jjtodd@PACBELL.NET31_Tue, 14 Nov 2006 16:29:42 -0600573_- Jim, I am not sure in your case, the question you are asking is the correct one. Your question on the cost for backing out SQL errors in relation to the number of SQL errors a customer receives within a single hour, does not seem like something that can be calculated just based on just the number of SQL errors. Unless the customer's SQL/application had actually done updates (update or insert within the same commit scope) before encountering the SQL error, there seems like there would be no back out cost. Only if an application did updates/inserts prior to the [...]
11236 33 41_Re: DB2 9 XML and other Options Licensing12_Phil Gunning22_pgunning@GUNNINGTS.COM31_Tue, 14 Nov 2006 16:40:34 -0600788_- On Tue, 14 Nov 2006 16:21:30 -0600, Phil Gunning wrote:
>Yes, there are several separately priced options. See the announcement >letter at >http://www-306.ibm.com/common/ssi/fcgi-bin/ssialias? >subtype=ca&infotype=an&appname=iSource&supplier=897&letternum=ENUS206-128: > > >Phil > >-------------------------------------------------------------------------- ------- >Welcome to the IDUG DB2-L list. To unsubscribe, go to the archives and home page at http://www.idugdb2-l.org/archives/db2-l.html. From that page select "Join or Leave the list". The IDUG DB2-L FAQ is at http://www.idugdb2-l.org. The IDUG List Admins can be reached at DB2-L- REQUEST@www.idugdb2-l.org. Find out the latest on IDUG conferences at http://conferences.idug.org/index.cfm [...]
11270 65 23_Re: Bind / Rebind Blues10_Nick Ridge23_nridge@ISD.SBCOUNTY.GOV31_Tue, 14 Nov 2006 16:34:01 -0600319_- Hello:
Regards - I work for the local government of the County of San Bernardino, CA. We have nearly the exact same Z-OS/DB2/Neon environment as you do. Have you discovered any further enlightenment on the deadlock problem when binding batch jobs? Your error trace is nearly identical to the ones we get: [...]
11336 52 27_Re: DB2 9 LUW XML licensing13_Willie Favero21_wfavero@ATTGLOBAL.NET31_Tue, 14 Nov 2006 19:47:56 -0600575_us-ascii Yes, pureXML is a for cost option of DB2 ( for LUW.
Willie
Matthew Rhea wrote:
>List, > >Has anyone heard that the XML feature of DB2 9 on LUW is a separate >chargeable license? As is the storage feature that allows you to use >table compression? Our IBM sales reps are insisting that there is a >separate license for each of those features, yet when we installed DB2 9 >AIX we could immediately use the XML features. I would think that the >features would be disabled until you had a license. I haven't seen any of >the literature mention [...]