1 WWW.IDUGDB2-L.ORG /home/listserv/home/db2-l May 2008, week 5
2 41 29_Production Problem Page latch14_Renzo razzetti24_renzo.razzetti@GMAIL.COM31_Thu, 29 May 2008 09:07:32 +0800630_ISO-8859-1 Please,
I have an application running under DB2 v7 in datasharing. At one moment all the cics transactions became longer, we found in DB2 they have high page latch. Any idea how to find the cause of the high page latch ???
thank you
The IDUG DB2-L Listserv is only part of your membership in IDUG. The DB2-L list archives, FAQ, and delivery preferences are at http://www.idug.org/lsidug under the Listserv tab. While at the site, you can also access the IDUG Online Learning Center, Tech Library and Code Place, see the latest IDUG conference information and much more. If you [...]
44 568 51_Re: Assembler Validation Exit - DB2 Timestamp Parse35_Joel Goldstein - Responsive Systems26_joel@RESPONSIVESYSTEMS.COM31_Wed, 28 May 2008 21:27:19 -0400553_iso-8859-1 Troy was still with Softbase last week at IDUG.
Regards, Joel
Joel Goldstein Responsive Systems Buffer Pool Tool for DB2, the worldwide industry standard Performance software that works...... Predicts Group Buffer Pool performance too! www.responsivesystems.com tel. (732) 972-1261 fax.(732) 972-9416 ----- Original Message ----- From: Weaver, Rick Newsgroups: bit.listserv.db2-l To: DB2-L@www.idugdb2-l.org Sent: Wednesday, May 28, 2008 3:32 PM Subject: Re: [DB2-L] Assembler Validation Exit - DB2 Timestamp Parse [...]
613 72 33_Re: Production Problem Page latch14_Renzo razzetti24_renzo.razzetti@GMAIL.COM31_Thu, 29 May 2008 11:45:04 +0800466_ISO-8859-1 Hi,
I found this error in the DBM1
IEC070I 209-220,DSN1DBM1,DSN1DBM1,SYS03927,8173,P1D273, 660 660 IEC070I XXXXXX.DSNDBC.DXXXPRD.STTTTTT.J0001.A001, 660 IEC070I XXXXXX.DSNDBC.DXXXPRD.STTTTTT..J0001.A001,xxxxxx
that's means SMS has a problem to extend a table. I need to probe that the extend table cause the page latch, as we know the page lock is in the same table..
Any idea ? how to probe the extend cause page latch ? [...]
686 43 42_Re: DB2 z/OS V9 - Administrative Scheduler15_Douwe van Sluis21_d.b.van.sluis@HOME.NL31_Thu, 29 May 2008 07:02:53 +0000616_UTF-8 I can answer my question 2 myself.
I found PK47893 which states:
-DSNTIJUZ: Now contains a new subsystem parameter, ADMTPROC, under DSN6SPRM in job step DSNTIZA that specifies a name for the JCL procedure that is used to start the DB2 administrative scheduler. See the note for DSNTIJMV above for the naming convention of the address space proc. A null value for ADMTPROC is allowed, in which case no DB2 administrative scheduler will be associated with the DB2 subsystem. The default setting for ADMTPROC is DSNADMT. In data sharing, this job needs to be customized and run for each member [...]
730 62 37_Re: Tracking GRANTS / REVOKES on z/OS0_19_ulrich.boche@SVA.DE31_Thu, 29 May 2008 09:19:13 +0200365_US-ASCII
On Wednesday, 28.05.2008 at 12:16 CST, Thomas Pierce wrote: > Tom Pierce > IBM Corp. > (206) 420-6332 > ----------------- > Sent from my BlackBerry Handheld. >
Nice to know that you can see the DB2-L postings on your Blackberry and even greater that you can post from there - but what was it that you wanted to say? :-)
793 186 18_Re: SQLSTATE 5701114_Avram Friedman25_ibmsysprog@GEEK-SITES.COM31_Thu, 29 May 2008 10:37:08 +0000353_utf-8 Vanitha The DB2 command you need to issue is -DIS GROUP
DB2 commands is an option from DB2I ... the same parent pannel that gets you to SPUFI assuming things are not modified.
I am assuming you have authority to issue commands to DB2 on ZOS. If you do not I would suggest getting someone involved who does or get the authority. [...]
980 191 24_How to access plan data?15_Itschak mugzach26_i_mugzach@SECURITEAM.CO.IL31_Thu, 29 May 2008 13:09:57 +0200702_us-ascii
How can I retrieve information included in the PLAN like database names? I know how to do it in IMS.
Thanks for your help,
Itschak
_____
The IDUG DB2-L Listserv is only part of your membership in IDUG. The DB2-L list archives, FAQ, and delivery preferences are at IDUG.ORG under the Listserv tab. While at the site, you can also access the IDUG Online Learning Center, Tech Library and Code Place, see the latest IDUG conference information , and much more. If you have not yet signed up for Basic Membership in IDUG, available at no cost, click [...]
1172 364 33_Re: Production Problem Page latch13_Hanne Lyssand10_HAN@VPS.NO31_Thu, 29 May 2008 13:34:28 +0200412_iso-8859-1 IEC070I 209-220 means that you don't have space on the volume to the extent, and no secondary volume available. Or haven't any secondary space specified. The volume is P1D273
Do SMS rules allow you to go multivolume? Is there any more volumes available. If I remember this right you have to have the primary allocation available on the next volume. (perhaps this is just a setting in SMS) [...]
1537 36 28_Re: How to access plan data?0_25_jacampbell@ACSLINK.NET.AU31_Thu, 29 May 2008 22:17:30 +1000431_ISO-8859-1 Look in the catalog tables: sysibm.sysplan sysibm.sysplandep
However most plans these days refer to a list of packages: sysibm.syspklist sysibm.syspackage sysibm.syspackdep
James Campbell
On 29 May 2008 at 13:09, Itschak mugzach wrote:
> > > How can I retrieve information included in the PLAN like database names? I know how to do it in > IMS... > > Thanks for your help, > > Itschak > [...]
1574 137 38_DB2 V8 CICS TS 3.2 performance problem12_Asamoto, Roy25_Roy.Asamoto@STATEAUTO.COM31_Thu, 29 May 2008 10:36:51 -0400443_us-ascii Hi,
We recently upgraded from CICS TS 2.2 to CICS TS 3.2 and started experiencing enlongated CICS transaction response times in our CICS/DB2 transactions. Most of the time appeared to be in the waiting to be dispatched category. We are on DB2 V8 in NFM with RSU0706.
The programs are COBOL 3.2 and defined as threadsafe. We did not recompile when switching from TS2.2 to TS 3.2. We realize this is an old compiler. [...]
1712 331 36_AW: [DB2-L] How to access plan data?0_26_Walter.Janissen@ITERGO.COM31_Thu, 29 May 2008 16:46:05 +0200495_iso-8859-1
Hi
That's not easy to do and it depends, how you bind your plans.
1. Option:
You only use DBRMs, which is no longer recommended. Then you can look into SYSIBM.SYSPLANDEP. There is the relationship between tables, indexes, databases and tablespaces to plans.
2.
You bind your plans with a PKLIST with entries (collid.package_name1, collid.package_name2 ....) Then you can look into SYSIBM.SYSPACKDEP for all the packages named in PKLIST [...]
2044 236 18_Question about IMS14_Sevetson, Phil22_PSevetson@FISA.NYC.GOV31_Thu, 29 May 2008 10:47:55 -0400381_us-ascii I know this is off topic. I'll keep it short - can someone point me to a mailing list for IBM IMS? Is there a mailing list like DB2-L for IMS?
Thanks!
--Phil Sevetson, NYCAPS DBA Support
Financial Information Services Agency of The City of New York
450 West 33rd Street, 4th Floor
New York, NY 10001 [...]
2281 137 49_DB2 data requests to another DB2 on separate lpar15_Shannon, Robert24_Robert.Shannon@ALIANT.CA31_Thu, 29 May 2008 11:39:56 -0300325_us-ascii zSeries z/OS 1.8 DB2 V8 NFM
We have a DB2 application on one lpar making data requests to a DB2 on a separate lpar. We are wondering if anyone is doing this and if they have tried. Currently we are using TCP/IP so in order to communicate over TCP/IP using Stored Procedures we have to use DRDA protocol. [...]
2419 19 28_Re: How to access plan data?12_Adam Baldwin22_adambaldwin@ES.IBM.COM31_Thu, 29 May 2008 14:55:37 +0000756_utf-8 And SYSTABAUTH will give you details of the type of DML that the PLAN / PACKAGE performs against a given Table.
______________________________________________________________________
* IDUG 08 Warsaw, Poland * 13-17 October 2008 * http://IDUG.ORG/lsEU * ______________________________________________________________________
The IDUG DB2-L Listserv is only part of your membership in IDUG. The DB2-L list archives, FAQ, and delivery preferences are at http://www.idug.org/lsidug under the Listserv tab. While at the site, you can also access the IDUG Online Learning Center, Tech Library and Code Place, see the latest IDUG conference information and much more. If you have not yet signed up for Basic Membership in [...]
2439 435 22_Re: Question about IMS14_John Caccavale18_jcac@THEMISINC.COM31_Thu, 29 May 2008 11:04:37 -0400706_us-ascii http://www.lsoft.com/scripts/wl.exe?SL1=IMS-L&H=PO.MISSOURI.EDU
Hope this helps.
John Caccavale
Vice President
Themis Education
908-233-8900
http://www.themisinc.com
________________________________
From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Sevetson, Phil Sent: Thursday, May 29, 2008 10:48 AM To: DB2-L@www.idugdb2-l.org Subject: [DB2-L] Question about IMS
I know this is off topic. I'll keep it short - can someone point me to a mailing list for IBM IMS? Is there a mailing list like DB2-L for IMS? [...]
2875 344 22_Re: Question about IMS10_Roger Hecq18_Roger.Hecq@UBS.COM31_Thu, 29 May 2008 11:02:20 -0400382_us-ascii Visit our website at http://www.ubs.com
This message contains confidential information and is intended only for the individual named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately by e-mail if you have received this e-mail by mistake and delete this e-mail from your system. [...]
3220 697 40_Re: AW: [DB2-L] How to access plan data?15_Itschak mugzach26_i_mugzach@SECURITEAM.CO.IL31_Thu, 29 May 2008 17:11:22 +0200558_us-ascii So Walter,
So, there is no construct (Control Block) that holds this information in memory during the plan execution?
Itschak
_____
From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Walter.Janissen@ITERGO.COM Sent: Thursday, May 29, 2008 4:46 PM To: DB2-L@www.idugdb2-l.org Subject: [DB2-L] AW: [DB2-L] How to access plan data?
Hi
That's not easy to do and it depends, how you bind your plans. [...]
3918 550 22_Re: Question about IMS15_Itschak mugzach26_i_mugzach@SECURITEAM.CO.IL31_Thu, 29 May 2008 17:18:10 +0200508_us-ascii Send email to LISTSERV@PO.MISSOURI.EDU with the body text of SUBscribe IMS-L your name
Itschak
_____
From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Roger Hecq Sent: Thursday, May 29, 2008 5:02 PM To: DB2-L@www.idugdb2-l.org Subject: Re: [DB2-L] Question about IMS
There most certainly is and it is a good one.
IMS Data Base Discussion List [IMS-L@PO.MISSOURI.EDU] [...]
4469 97 33_Re: Production Problem Page latch14_Avram Friedman25_ibmsysprog@GEEK-SITES.COM31_Thu, 29 May 2008 15:33:28 +0000372_utf-8 Renzo The unable to extend dataset message from AMS / Media Manager does indicate a problem. It, however is not the problem you asked about.
If you are spending a lot of time waiting on page latches the problem is two diffrent tasks going after the same page in such a way that one must wait (be suspendeded) to maintain the type of integrety requested. [...]
4567 97 22_Re: Question about IMS9_Mike Bell21_mbell11a1@VERIZON.NET31_Thu, 29 May 2008 10:41:36 -0500439_us-ascii from google https://po.missouri.edu/cgi-bin/wa?A0=IMS-L
IMS-L list
It is not as active as DB2 - maybe 10-15 posts a week. I see Avram there occasionally.
Mike HLS Technologies
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Sevetson, Phil Sent: Thursday, May 29, 2008 9:48 AM To: DB2-L@www.idugdb2-l.org Subject: [DB2-L] Question about IMS [...]
4665 686 22_Re: Question about IMS13_Robert Knight28_rknight@CONTEMPORARYTECH.COM31_Thu, 29 May 2008 12:15:35 -0400341_us-ascii Your message to IMS-L-request@PO.MISSOURI.EDU has been forwarded to the "list owners" (the people who manage the IMS-L list). If you wanted to reach a person, then you used the correct procedure and you can ignore the remainder of this message. If you were trying to send a command for the computer to execute, please read on. [...]
5352 79 53_Re: DB2 data requests to another DB2 on separate lpar7_Ed Long19_rdhm99a@PRODIGY.NET31_Thu, 29 May 2008 11:55:42 -0700596_iso-8859-1 Hi. I will assume that the two DB2's are not in a Data Sharing Group.
I have used 3 part names and TCPIP pretty successfully using both dynamic SQL and static SQL in the form of programs and Stored Procedures.
"Shannon, Robert" wrote: zSeries z/OS 1.8 DB2 V8 NFM
We have a DB2 application on one lpar making data requests to a DB2 on a separate lpar. We are wondering if anyone is doing this and if they have tried. Currently we are using TCP/IP so in order to communicate over TCP/IP using Stored Procedures we have to use [...]
5432 880 91_Re: BMC Software - Sr MSM Technical Support Analyst Position -(DB2 Admin Suite of Products)12_Estrada, Amy19_Amy_Estrada@BMC.COM31_Thu, 29 May 2008 15:05:49 -0500367_iso-8859-7 Hi David
I apologize for not getting back to you sooner. Although ideally we would like a candidate located in Austin, we will consider remote candidates. Please apply to the position online and once you have done so we can schedule a time to talk about your, resume, skills and career path. I look forward to speaking with you [...]
6313 174 42_Re: DB2 V8 CICS TS 3.2 performance problem12_Asamoto, Roy25_Roy.Asamoto@STATEAUTO.COM31_Thu, 29 May 2008 16:13:15 -0400669_us-ascii Never mind!
Our CICS programmer found that during the conversion the Omegamon entry was set to Quasireenterant instead of threadsafe by mistake.
Thanks, Roy
________________________________
From: Asamoto, Roy Sent: Thursday, May 29, 2008 10:37 AM To: 'DB2 Database Discussion list at IDUG' Subject: DB2 V8 CICS TS 3.2 performance problem
Hi,
We recently upgraded from CICS TS 2.2 to CICS TS 3.2 and started experiencing enlongated CICS transaction response times in our CICS/DB2 transactions. Most of the time appeared to be in the waiting to be dispatched category. We are on DB2 V8 in NFM with RSU0706. [...]
6488 405 91_Re: BMC Software - Sr MSM Technical Support Analyst Position -(DB2 Admin Suite of Products)13_Michael Ebert18_mebert@AMADEUS.COM31_Thu, 29 May 2008 22:49:28 +0200446_ISO-8859-1 Well, keep us in the distribution list, I'm sure we're all very curious how this will continue!
Dr. Michael Ebert DB2 & Oracle Database Administrator aMaDEUS Data Processing Erding / Munich, Germany
"Estrada, Amy" To DB2-L@www.idugdb2-l.org cc
bcc
Subject Re: [DB2-L] BMC Software - Sr MSM Technical Support Analyst Position -(DB2 Admin Suite of Products) [...]
6894 79 44_DB2 zOS v7.1 Revoking SYSADM takes long time11_Bear, Brian23_Brian.Bear@CHARMING.COM31_Thu, 29 May 2008 18:21:06 -0400456_US-ASCII I am revoking SYSADM auth from some IDs found to be no longer valid in RACF.
I am questioning the "long" response for the revoke to actually return. BMC Mainview was showing me 100's of locks being taken on various DSNDB06 tables (syspackage, etc) during the revoke.
Oddly, the long response is not consistent for every ID I revoke. A couple of the IDs I revoked came back immediately but two other's took 1+ minutes to return. [...]
6974 106 48_Re: DB2 zOS v7.1 Revoking SYSADM takes long time13_David Simpson22_dsimpson@THEMISINC.COM31_Thu, 29 May 2008 18:49:26 -0400413_us-ascii Also look for authorizations granted by the id:
SYSIBM.SYSTABAUTH, etc...
You may have lost some authorizations granted by those ids.
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Bear, Brian Sent: Thursday, May 29, 2008 5:21 PM To: DB2-L@www.idugdb2-l.org Subject: [DB2-L] DB2 zOS v7.1 Revoking SYSADM takes long time [...]
7081 96 42_Re: DB2 V8 CICS TS 3.2 performance problem9_Tony Saul30_generalemaillists@YAHOO.COM.AU31_Fri, 30 May 2008 10:37:46 +1000356_iso-8859-1 Roy, Although you said you found the problem, also remember that CICS TS 3.2 increased the accuracy of the response time metrics from 16 microsecond intervals (< CICS TS 3.2) to microsecond intervals i.e. previously up to 15 microseconds could go unreported, which could add up.
--- "Asamoto, Roy" wrote: [...]
7178 83 48_Re: DB2 zOS v7.1 Revoking SYSADM takes long time15_Chris Hoelscher21_choelscher@HUMANA.COM31_Thu, 29 May 2008 20:52:55 -0400316_US-ASCII
hmmmm - i always thought that if you made the soon-to-be-deleted sysadm the INSTALL SYSADM via ZPARM - revoking SYSADM would bypass ALL checks - does this work on a subsystem using RACF as the DB2 Security repository? it works in am environment using the catalog as the DB2 security repository [...]
7262 100 48_Re: DB2 zOS v7.1 Revoking SYSADM takes long time25_Thanikachalam Sundarrajan21_billysundar@YAHOO.COM31_Thu, 29 May 2008 18:24:20 -0700402_iso-8859-1 If you set the INSTALL SYSADM2 ID to the ID that you are revoking SYSADM from (prior to the REVOKE), then DB2 will not do a CASCASE check/revoke. Otherwise, DB2 will check the authorization tables to see if there is any authority that needs to be cascade revoked. You could unload the AUTH tables prior and after the REVOKE to see if there are changes in tables other than SYSUSERAUTH. [...]
7363 22 48_Re: DB2 zOS v7.1 Revoking SYSADM takes long time28_=?utf-8?B?VGVkIE1hY05FSUw=?=18_eamacneil@YAHOO.CA31_Fri, 30 May 2008 02:11:58 +0000764_- >then DB2 will not do a CASCASE check/revoke.
I assume you mean CASCADE?
- Too busy driving to stop for gas!
______________________________________________________________________
* IDUG 08 Bangalore, India * 21-23 August 2008 * http://IDUG.ORG/lsIN * ______________________________________________________________________
The IDUG DB2-L Listserv is only part of your membership in IDUG. The DB2-L list archives, FAQ, and delivery preferences are at http://www.idug.org/lsidug under the Listserv tab. While at the site, you can also access the IDUG Online Learning Center, Tech Library and Code Place, see the latest IDUG conference information and much more. If you have not yet signed up for Basic Membership in [...]
7386 60 22_Re: Question about IMS0_17_JTonchick@AOL.COM29_Fri, 30 May 2008 00:23:51 EDT951_US-ASCII Yes, IMS-L can be found at: _https://po.missouri.edu/cgi-bin/wa?A0=IMS-L_ (https://po.missouri.edu/cgi-bin/wa?A0=IMS-L)
**************Get trade secrets for amazing burgers. Watch "Cooking with Tyler Florence" on AOL Food. (http://food.aol.com/tyler-florence?video=4&?NCID=aolfod00030000000002)
______________________________________________________________________
* IDUG 08 Bangalore, India * 21-23 August 2008 * http://IDUG.ORG/lsIN * ______________________________________________________________________
The IDUG DB2-L Listserv is only part of your membership in IDUG. The DB2-L list archives, FAQ, and delivery preferences are at http://www.idug.org/lsidug under the Listserv tab. While at the site, you can also access the IDUG Online Learning Center, Tech Library and Code Place, see the latest IDUG conference information and much more. If you have not yet signed up for Basic [...]
7447 61 29_Production Problem Page latch14_Renzo razzetti24_renzo.razzetti@GMAIL.COM31_Fri, 30 May 2008 16:31:07 +0800263_ISO-8859-1 > > > I have an application running under DB2 v7 in datasharing. At one moment > all the cics transactions became longer, we found in DB2 a high IRLM LATCH > SUSPENSIONS at that time and 1 minute after we see a package with a high > page latch. > [...]
7509 76 33_Re: Production Problem Page latch14_Avram Friedman25_ibmsysprog@GEEK-SITES.COM31_Fri, 30 May 2008 10:12:25 +0000734_utf-8 Renzo Are you CPU Starved,
Here are two links
From this lists archives http://www.idugdb2-l.org/cgi-bin/wa?A2=ind0506B&L=DB2- L&D=0&T=0&P=2268
and see page 6 of this presention which is a good broad performance evaluation discussion. http://www.gse-nordic.org/Working% 20Groups/NRTC/Conferences/2007/DB2/S48
Regards Avram Friedman
On Fri, 30 May 2008 16:31:07 +0800, Renzo razzetti wrote:
>> >> >> I have an application running under DB2 v7 in datasharing. At one moment >> all the cics transactions became longer, we found in DB2 a high IRLM LATCH >> SUSPENSIONS at that time and 1 minute after we see a package with a high >> page latch. >> > >Can you [...]
7586 104 33_Re: Production Problem Page latch14_Avram Friedman25_ibmsysprog@GEEK-SITES.COM31_Fri, 30 May 2008 10:21:20 +0000825_utf-8 I found a pointer to a paper / presentation that looks good in a DB2-L listserv posting by Roger Miller:
http://idug-365-lstsrv01.itciss.com/cgi-bin/wa?A2=ind0506B&L=db2- l&D=0&P=4540
Avram Friedman
On Fri, 30 May 2008 10:12:25 +0000, Avram Friedman wrote:
>Renzo >Are you CPU Starved, > >Here are two links > >From this lists archives >http://www.idugdb2-l.org/cgi-bin/wa?A2=ind0506B&L=DB2- >L&D=0&T=0&P=2268 > >and see page 6 of this presention which is a good broad performance >evaluation discussion. >http://www.gse-nordic.org/Working% >20Groups/NRTC/Conferences/2007/DB2/S48 > >Regards >Avram Friedman > >On Fri, 30 May 2008 16:31:07 +0800, Renzo razzetti > wrote: > >>> >>> >>> I have an application running under DB2 v7 in [...]
7691 20 42_Re: DB2 V8 CICS TS 3.2 performance problem12_Adam Baldwin22_adambaldwin@ES.IBM.COM31_Fri, 30 May 2008 10:42:22 +0000772_utf-8 Tony - can you point me in the direction of any documentation regarding the micro second response time metrics? Many thanks, Adam
______________________________________________________________________
* IDUG 08 Bangalore, India * 21-23 August 2008 * http://IDUG.ORG/lsIN * ______________________________________________________________________
The IDUG DB2-L Listserv is only part of your membership in IDUG. The DB2-L list archives, FAQ, and delivery preferences are at http://www.idug.org/lsidug under the Listserv tab. While at the site, you can also access the IDUG Online Learning Center, Tech Library and Code Place, see the latest IDUG conference information and much more. If you have not yet signed up for Basic Membership [...]
7712 131 48_Re: DB2 zOS v7.1 Revoking SYSADM takes long time10_Roger Hecq18_Roger.Hecq@UBS.COM31_Fri, 30 May 2008 08:40:03 -0400442_us-ascii My recollection is that this has been a known problem for some time. Revokes, especially if cascading, simply take an inordinately long time.
I remember pursuing this issue, with IBM, a good decade ago. The end result was that the problem is due to the poor access path being used. We (STL & I) tried creating some indexes to improve the performance, but the access paths are hard coded and could not be changed. [...]
7844 122 48_Re: DB2 zOS v7.1 Revoking SYSADM takes long time11_Bear, Brian23_Brian.Bear@CHARMING.COM31_Fri, 30 May 2008 09:24:45 -0400297_iso-8859-1 Understood, but since these IDs do not own any objects per my SELECTs of the catalog tables, I didn't feel the need to go to that extreme. I feel safe just revoking.
My main concern or question was why revoking one ID takes longer to complete than when revoking another ID. [...]
7967 122 42_Re: DB2 V8 CICS TS 3.2 performance problem12_Asamoto, Roy25_Roy.Asamoto@STATEAUTO.COM31_Fri, 30 May 2008 09:40:01 -0400456_us-ascii Thanks Tony. Our CICS guy also mentioned something about some stats differences and planned to tune or turn off some journaling. He's off today, so I'm not sure exactly what he planned on doing.
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Tony Saul Sent: Thursday, May 29, 2008 8:38 PM To: DB2-L@www.idugdb2-l.org Subject: Re: [DB2-L] DB2 V8 CICS TS 3.2 performance problem [...]
8090 64 26_Case expression in trigger10_Wim Ruarus20_wim.ruarus@GMAIL.COM31_Fri, 30 May 2008 13:37:35 +0000390_utf-8 Hi all,
today I encountered a problem with a trigger containign case-expressions. This trigger was created in DB2 z/OS V8. On our test system we are currently running V9 NFM. When I tried to re-create the trigger today this failed with an SQL-code -104, saying the symbol "END" was illegal. The trigger is an after update type trigger, with 3 triggered actions like this: [...]
8155 115 30_Re: Case expression in trigger10_Roger Hecq18_Roger.Hecq@UBS.COM31_Fri, 30 May 2008 10:08:47 -0400322_us-ascii What terminator did you specify in the --#set terminator statement? The CASE END is being recognized as the BEGIN ATOMIC END. You need to specify a terminator, other than ';' and code that character after the END for the BEGIN ATOMIC
Roger Hecq MF IB USA DB Support 203-719-0492 / 19-337-0492 [...]
8271 159 48_Re: DB2 zOS v7.1 Revoking SYSADM takes long time14_Grainger, Phil20_Phil.Grainger@CA.COM31_Fri, 30 May 2008 15:34:38 +0100377_iso-8859-1 As has already been pointed out, you need to take account of the GRANTS that this ID has done, not just the objects it's created
E.g.
userA has SYSADM userA grants DBADM to userB userB create a tableZ
If you revoke SYSADM from userA ...you also have revoked DBADM from userB ...now userB cannot have created tableZ so that gets dropped too [...]
8431 158 48_Re: DB2 zOS v7.1 Revoking SYSADM takes long time14_Sevetson, Phil22_PSevetson@FISA.NYC.GOV31_Fri, 30 May 2008 12:56:40 -0400573_utf-8 Phil -- Wait, what? I thought REVOKE cascades only revoked authorizations, not data structures?
--Phil Sevetson, NYCAPS DBA Support Financial Information Services Agency of The City of New York 450 West 33rd Street, 4th Floor New York, NY 10001 phone: (212) 857-1688 mailto: psevetson@fisa.nyc.gov
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Grainger, Phil Sent: Friday, May 30, 2008 10:35 AM To: DB2-L@www.idugdb2-l.org Subject: Re: [DB2-L] DB2 zOS v7.1 Revoking SYSADM takes long time [...]
8590 174 48_Re: DB2 zOS v7.1 Revoking SYSADM takes long time13_David Simpson22_dsimpson@THEMISINC.COM31_Fri, 30 May 2008 13:37:19 -0400468_iso-8859-1 That is correct... only authorizations cascade, not structures (just tested it to be sure).
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Sevetson, Phil Sent: Friday, May 30, 2008 11:57 AM To: DB2-L@www.idugdb2-l.org Subject: Re: [DB2-L] DB2 zOS v7.1 Revoking SYSADM takes long time
Phil -- Wait, what? I thought REVOKE cascades only revoked authorizations, not data structures? [...]
8765 188 48_Re: DB2 zOS v7.1 Revoking SYSADM takes long time16_Rasmussen, Steen22_Steen.Rasmussen@CA.COM31_Fri, 30 May 2008 13:36:22 -0400302_iso-8859-1 I think Phil was replacing the word VIEW with the word TABLE. Tables do not get dropped but VIEWS do and all granted plans and packages are invalidated. I've tried this myself - and like we said when I was a "real DBA" - Real DBA's don't do backups - they work the loooong extra hours. [...]
8954 294 48_Re: DB2 zOS v7.1 Revoking SYSADM takes long time11_Nick Dordea20_ndordea@COMPUTER.ORG31_Fri, 30 May 2008 18:12:56 +0000
9249 262 36_[DB2v8 z/OS] One-Part Partitioned TS14_Sevetson, Phil22_PSevetson@FISA.NYC.GOV31_Fri, 30 May 2008 15:38:21 -0400586_us-ascii Anyone know whether creating a one-part partitioned tablespace is supported? I'm about to try it but still working on the DDL.
I have a test environment where we're mimicking production but leaving certain tables, including a handful of partitioned ones, empty. I've got orders not to make the partitioned tablespaces fullsize, to save DASD and dataset handles and DBD size, so I thought I'd try this to preserve as much resemblance as possible to the basic design (and allow the boss to add partitions later without having to drop the tablespace). [...]
9512 130 40_Re: [DB2v8 z/OS] One-Part Partitioned TS11_Mark Kimble25_mark.a.kimble@US.HSBC.COM31_Fri, 30 May 2008 15:05:07 -0500517_UTF-8 Yes, I have created single partition partitioned tablespaces.
Mark Kimble Database Management HTSN - WD2 Phone: 630-521-3287 Pager/Cell: 3124982467@messaging.sprintpcs.com
"Sevetson, Phil" To Sent by: DB2 Data DB2-L@www.idugdb2-l.org Base Discussion cc List [DB2v8 z/OS] One-Part Partitioned TS
05/30/2008 02:38 PM
Please respond to DB2 Database Discussion list at IDUG [...]
9643 257 40_Re: [DB2v8 z/OS] One-Part Partitioned TS16_Rasmussen, Steen22_Steen.Rasmussen@CA.COM31_Fri, 30 May 2008 16:14:50 -0400321_iso-8859-1 Hey Phil - sure it is supported, and in fact thinking about V8 and V9, if you plan on growing, adding or rotating partitions, its probably a good idea since you cannot convert from non-partitioning to partitioning, but once you have a one-partitioned tablespace you can exploit all the good V8 TCP stuff. [...]
9901 88 40_Re: [DB2v8 z/OS] One-Part Partitioned TS13_Willie Favero21_wfavero@ATTGLOBAL.NET31_Fri, 30 May 2008 15:46:39 -0500601_windows-1252 One (1) is a valid value for NUMPARTS on the CREATE TABLESPACE statement.
Willie
Sevetson, Phil wrote: > > Anyone know whether creating a one-part partitioned tablespace is > supported? I’m about to try it but still working on the DDL. > > I have a test environment where we’re mimicking production but leaving > certain tables, including a handful of partitioned ones, empty. I’ve > got orders not to make the partitioned tablespaces fullsize, to save > DASD and dataset handles and DBD size, so I thought I’d try this to > preserve as much resemblance as possible to [...]
9990 467 40_Re: [DB2v8 z/OS] One-Part Partitioned TS13_Seibert, Dave26_Dave.Seibert@COMPUWARE.COM31_Fri, 30 May 2008 16:47:42 -0400640_us-ascii Hi Phil.
Yes you can. I'll happily sell you my ddl if you'd like. I'm real cheap.
In fact, the default tablespace in DB2 v9 is a 1 part Universal tablespace -- a partition by growth tablespace.
-- DROP TABLESPACE BFHDJSDB.TSP0036; -- COMMIT; CREATE TABLESPACE TSP0036 IN BFHDJSDB FREEPAGE 0 PCTFREE 0 NUMPARTS 1 (PART 1 USING STOGROUP STG002 PRIQTY 1 SECQTY 1 ERASE NO FREEPAGE 0 PCTFREE 0) BUFFERPOOL BP0 LOCKSIZE PAGE LOCKMAX SYSTEM CLOSE NO CCSID EBCDIC MAXROWS 255; -- COMMIT; CREATE TABLE STAFFPDSC (ID SMALLINT NOT NULL, NAME VARCHAR(9) NOT NULL, DEPT SMALLINT, JOB CHAR(5), YEARS SMALLINT, [...]
10458 675 40_Re: [DB2v8 z/OS] One-Part Partitioned TS14_Sevetson, Phil22_PSevetson@FISA.NYC.GOV31_Fri, 30 May 2008 17:55:45 -0400370_us-ascii Thanks, all. I did it. It worked without trouble. I'm happy. My boss is happy. I'm going home for the weekend, reading IMS on the bus, and not thinking any more about this until Monday :-)
--Phil Sevetson, NYCAPS DBA Support
Financial Information Services Agency of The City of New York
450 West 33rd Street, 4th Floor [...]
11134 754 40_Re: [DB2v8 z/OS] One-Part Partitioned TS12_Weaver, Rick19_Rick_Weaver@BMC.COM31_Fri, 30 May 2008 16:58:22 -0500446_us-ascii And, if you create that 1 part PBG Universal Tablespace, you get a default SEGSIZE of 4.
Rick Weaver Product Manager DB2 z/OS Solutions BMC Software
________________________________
From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Seibert, Dave Sent: Friday, May 30, 2008 3:48 PM To: DB2-L@www.idugdb2-l.org Subject: Re: [DB2-L] [DB2v8 z/OS] One-Part Partitioned TS [...]
11889 230 40_Re: [DB2v8 z/OS] One-Part Partitioned TS10_Dave Nance16_dav1mo@YAHOO.COM31_Sat, 31 May 2008 04:30:06 -0700509_iso-8859-7 Phil, As quite a few have told you it is possible, but I like to say just because you can do something, you may not want to. What about using the same number of parts as in production, but a very low allocation(1 track) on all but one part? This way you would still be able to copy the production stats into your test region. That way you would be able to tell if your access path has changed prior to moving a new or modified SQL statement into your production environment. [...]
12120 54 40_Re: [DB2v8 z/OS] One-Part Partitioned TS14_Peter Vanroose26_peter_vanroose@YAHOO.CO.UK31_Sat, 31 May 2008 12:11:59 +0000555_iso-8859-1 > In fact, the default tablespace in DB2 v9 is a 1 part > Universal tablespace -- a partition by growth tablespace. > ... > And, if you create that 1 part PBG Universal Tablespace, > you get a default SEGSIZE of 4.
Just two remarks related to this: - for DB2 9 z/OS: the default tablespace (when issueing "CREATE TABLESPACE ts IN d") is segmented, with segsize 4. But indeed the default implicit tablespace (when issueing "CREATE TABLE tb (..) IN DATABASE d") is a 1-part PBG TS. - for DB2 v8 z/OS: although you indeed get "all the [...]
12175 81 40_Re: [DB2v8 z/OS] One-Part Partitioned TS13_Willie Favero21_wfavero@ATTGLOBAL.NET31_Sat, 31 May 2008 10:37:48 -0500583_ISO-8859-1 Since Peter brought losing stuff...
If you are data sharing, remember that you cannot specify MEMBER CLUSTER on a universal table space.
Willie
Peter Vanroose wrote: >> In fact, the default tablespace in DB2 v9 is a 1 part >> Universal tablespace -- a partition by growth tablespace. >> ... >> And, if you create that 1 part PBG Universal Tablespace, >> you get a default SEGSIZE of 4. >> > > Just two remarks related to this: > - for DB2 9 z/OS: > the default tablespace (when issueing "CREATE TABLESPACE ts IN d") is segmented, with segsize 4. [...]