1 WWW.IDUGDB2-L.ORG /home/listserv/home/db2-l February 2006, week 1
2 57 26_Accessing uncommitted rows14_Michael Kaplan25_micaelkp@NETVISION.NET.IL30_Wed, 1 Feb 2006 03:15:44 -0600331_- Hi,
Can anybody explain the following, very strange ( for me ) behavior :
We have long-running task ( under CICS) which makes SQL inserts around a clock. All inserts are divided into group of 50 and each group is finished with COMMIT. Rows come and are put according to TIMESTAMP which is the first key field. [...]
60 162 37_System Managed CF Structure duplexing58_=?iso-8859-9?Q?Serdar_Sabri_=D6zkubulay_=28B.T.S.Y.G.=29?=27_Serdar.Ozkubulay@AKBANK.COM30_Wed, 1 Feb 2006 11:49:34 +02001023_iso-8859-9 Hi,
I want to ask if you prefer or not "System Managed CF Structure Duplexing" for SCA and LOCK structures especially when there is no external CF around?
Thanks for your comments.
Serdar Sabri OZKUBULAY AKBANK, DB2 systems programmer
*************************************************************************************
Bu e-posta ve muhtemel eklerinde verilen bilgiler kişiye özel ve gizli olup, yalnızca mesajda belirlenen alıcı ile ilgilidir.Size yanlışlıkla ulaşmışsa lütfen göndericiye bilgi veriniz, mesajı siliniz ve içeriğini başka bir kişiye açıklamayınız, herhangi bir ortama kopyalamayınız. Bu mesaj aksi sözleşme ile belirtilmedikçe herhangi bir finansal işlem teklifi, alımı, satımı veya herhangi bir havalenin teyidi gibi bankacılık işlemi yapılması amacını taşımamaktadır.Verilen [...]
223 44 24_Logging in DB2 Connect ?12_Peter, Georg15_G.Peter@DZBW.DE30_Wed, 1 Feb 2006 14:49:14 +0100430_- Collegues, environment is DB2 for z/OS version 7 and DB2 Connect version 8.1.
As usual our Java appz are using DB2 Connect to access the big iron. These appz are working with dynamic SQL and each statement has a clause "fetch first 50 rows" (as the programmers told me).
But it seems that this clause is ignored by DB2 and thousands or millions of rows were sent back to the appz server - bad, bad, bad...... [...]
268 106 30_Re: Accessing uncommitted rows7_troycci29_troycci@COLEMANCONSULTING.COM30_Wed, 1 Feb 2006 07:58:15 -0600345_us-ascii I'm not sure what the default isolation level is for SPUFI or how your systems people installed it. What you may want to do is force the isolation level using the "WITH CS" at the end of your query.
SELECT MAX(F1) FROM T1 WITH CS;
Troy Coleman, Support Engineer IBM Certified Solutions Expert [...]
375 138 30_Re: Accessing uncommitted rows5_db2-117_db2-1@NIOI.GOV.IL30_Wed, 1 Feb 2006 16:12:25 +0200673_ISO-8859-1 Hi Michael. Check the isolation of plan DSNESPCS. Maybe, this plan was binded with isolation(ur) and not isolation(cs) during the db2 instaltion.
Gibli
----- Original Message ----- From: "troycci" Newsgroups: bit.listserv.db2-l To: Sent: Wednesday, February 01, 2006 3:58 PM Subject: Re: [DB2-L] Accessing uncommitted rows
> I'm not sure what the default isolation level is for SPUFI or how your > systems people installed it. > What you may want to do is force the isolation level using the "WITH CS" at > the end of your query. > > SELECT MAX(F1) > FROM T1 > WITH CS; [...]
514 101 54_Stored Procedure Problem - Converting from SPAS to WLM13_Wayne Stevens25_twstevens@DHR.STATE.AL.US30_Wed, 1 Feb 2006 08:19:05 -0600492_us-ascii We are having a problem with our stored procedures. We have been running them in the SPAS address space. We are now trying to convert them to WLM. The WLM is defined as: The definition of the application environment to WLM is defined as follows:
Appl Environment Name . . WLMDB2D1 Description . . . . . . . DB2D Application Environment Subsystem type . . . . . DB2D Procedure name . . . . . DB2DWLM1 Start parameters . . . . DB2SSN=DB2D,NUMTCB=5,APPLENV=WLMDB2D1 [...]
616 26 41_Help: DB2 for Unix\Aix tutorials required11_ali hasnain23_me_alihasnain@YAHOO.COM30_Wed, 1 Feb 2006 06:36:17 -0800752_iso-8859-1 Hi;
Can u tell me from where can i download Db2 for unix/Aix tutorials.
Thanks Ali Hasnain
__________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.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
643 18 23_DB2 Z/OS zparm RETVLCFK14_Zobjeck, A. J.21_Allen_Zobjeck@TTX.COM30_Wed, 1 Feb 2006 08:43:30 -0600304_US-ASCII Can anyone tell me any disadvantaged on setting the RETVLCFK Zparm value to YES? It specifies whether a varchar column will be retrieved from an index. Applications must be able to handle padding.
For those of you that have varchar column in indexes do you have it set to YES or NO? [...]
662 36 29_Re: DMS tablespace size error3_lan18_lxw176@HOTMAIL.COM30_Wed, 1 Feb 2006 08:37:17 -0600559_- HI,
The PP size is 128 MB. I got this from the following, Am I right? since I'm not the person to setup the AIX server. # lsvg rootvg VOLUME GROUP: rootvg VG IDENTIFIER: 00c824ff00004c000000010844b0cfb2 VG STATE: active PP SIZE: 128 megabyte(s) VG PERMISSION: read/write TOTAL PPs: 546 (69888 megabytes) MAX LVs: 256 FREE PPs: 506 (64768 megabytes) LVs: 10 USED PPs: 40 (5120 megabytes) OPEN LVs: 9 QUORUM: 2 TOTAL PVs: 1 VG DESCRIPTORS: 2 STALE PVs: 0 STALE PPs: 0 ACTIVE PVs: 1 AUTO ON: yes MAX PPs per VG: 32512 MAX PPs per PV: 1016 MAX PVs: [...]
699 49 27_Re: DB2 Z/OS zparm RETVLCFK12_Higgins John26_HigginsJohnP@JOHNDEERE.COM30_Wed, 1 Feb 2006 08:53:16 -0600361_us-ascii Up thru V7, DB2 pads varchar columns to their full length in indexes. If you set RETVLCFK = YES, you enable index-only access for the indexed varchar columns, but the data retrieved from the index will be the padded value, not the value stored in the table.
We find that our applications do not handle this correctly, so we can not use it. [...]
749 62 27_Re: DB2 Z/OS zparm RETVLCFK9_Mike Bell21_mbell11a1@VERIZON.NET30_Wed, 1 Feb 2006 08:59:06 -0600506_Windows-1252 The data returned from the SQL from an index will always have the length set to the max for the varchar instead of the actual length. For example - varchar(8) value 'ABC' length = 3 SQL from the index will return length=8 and 'ABC ' SQL from data will return length=3 and 'ABC' The program will process the same data UNLESS the program then uses the data value for an UPFATE statement which will change the length value in the data row and the varchar is now a fixed length maximum char. [...]
812 17 27_Re: DB2 Z/OS zparm RETVLCFK11_Ed Mullikin17_EMullikin@SCU.EDU30_Wed, 1 Feb 2006 06:59:16 -0800591_US-ASCII I'll be out of the office on 2/1. For any database issues, please contact Mai Anh at ex.4814.
Thanks, ED
--------------------------------------------------------------------------------- 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
830 112 43_Re: bmc tools install question- db2 z/os v712_craig patton21_prgpatton@HOTMAIL.COM30_Wed, 1 Feb 2006 15:09:32 +0000335_- Shery,
I have found that it is easiest to use a single loadlib at install, but have used seperate libraries for Test, QA and Prod. When we upgrade, we typically upgrade all products at the same time and testing is performed on all products. Aptune may be the exception but I have not installed it and can't say for sure. [...]
943 68 29_Re: DMS tablespace size error9_Mo, Don D12_donmo@BU.EDU30_Wed, 1 Feb 2006 10:08:05 -0500721_US-ASCII Lan,
Run these commands for me please:
Lsvg datavg Lslv ShortSpace1 Lslv -l ShortSpace1 Lslv ShortSpace0 Lslv -l ShortSpace0
-Don
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of lan Sent: Wednesday, February 01, 2006 9:37 AM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] DMS tablespace size error
HI,
The PP size is 128 MB. I got this from the following, Am I right? since I'm not the person to setup the AIX server. # lsvg rootvg VOLUME GROUP: rootvg VG IDENTIFIER: 00c824ff00004c000000010844b0cfb2 VG STATE: active PP SIZE: 128 megabyte(s) VG PERMISSION: read/write TOTAL PPs: 546 (69888 megabytes) MAX [...]
1012 32 33_DB2 OS/390 v7 reson code 00E7008110_Steve Lamb25_steven.lamb@CAPGEMINI.COM30_Wed, 1 Feb 2006 09:55:54 -0600403_- Help!
On Sunday we had a batch job that Step 1. issues an ALTER LOCKSIZE command for a tablespace Step 2. runs some SQL for about 2.5 hours Step 3. issues an ALTER command to put things back how they were. The job failed in step 3 trying to issue the ALTER LOCKSIZE command with a - 904, reason code 00E70081 on the tablespace - TYPE OF RESOURCE 00000200, AND RESOURCE NAME WCSSP.WCSSSWHI [...]
1045 82 29_Re: DMS tablespace size error3_lan18_lxw176@HOTMAIL.COM30_Wed, 1 Feb 2006 09:57:39 -0600589_- Hi Don,
Here is the result I got: # lsvg datavg VOLUME GROUP: datavg VG IDENTIFIER: 00c824ff00004c000000010848ff6b4d VG STATE: active PP SIZE: 128 megabyte(s) VG PERMISSION: read/write TOTAL PPs: 1631 (208768 megabytes) MAX LVs: 256 FREE PPs: 1262 (161536 megabytes) LVs: 16 USED PPs: 369 (47232 megabytes) OPEN LVs: 4 QUORUM: 2 TOTAL PVs: 1 VG DESCRIPTORS: 2 STALE PVs: 0 STALE PPs: 0 ACTIVE PVs: 1 AUTO ON: yes MAX PPs per VG: 32512 MAX PPs per PV: 2032 MAX PVs: 16 LTG size (Dynamic): 1024 kilobyte(s) AUTO SYNC: no HOT SPARE: no BB POLICY: relocatable # lslv ShortSpace1 [...]
1128 89 58_Re: Stored Procedure Problem - Converting from SPAS to WLM13_Johnny Wilder28_Johnny.Wilder@MOTION-IND.COM30_Wed, 1 Feb 2006 10:08:53 -0600436_us-ascii Wayne, all looks good to me, but I did notice the WLM application environment definition had one thing incorrect.
The Environment Subsystem type . . . . . DB2D should be DB2.
-----Original Message----- From: Wayne Stevens [mailto:twstevens@DHR.STATE.AL.US] Sent: Wednesday, February 01, 2006 8:19 AM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] Stored Procedure Problem - Converting from SPAS to WLM [...]
1218 364 58_Re: Stored Procedure Problem - Converting from SPAS to WLM25_Biswal, Manas (Cognizant)26_Manas.Biswal@COGNIZANT.COM30_Wed, 1 Feb 2006 20:27:28 +0530499_iso-8859-1
We have got -471 with 00E7900C before on our SPs running on a WLM environment before but the problem almost always has been the unavailability of the WLM. This does not seem to be so in your case. However, just to be sure, you can try a quiesce and resume of the WLM. One more thing - Just ensure that the RRS task is started on your MVS lpar. WLM managed stored procedures are mandated to connect to DB2 via RRSAF. So, just ensure that the RRS is up and running. Good luck. [...]
1583 84 41_Re: System Managed CF Structure duplexing11_Mike Turner29_michael_turner@COMPUSERVE.COM30_Wed, 1 Feb 2006 16:20:24 -0000565_iso-8859-9 System managed duplexing is more expensive in CPU than the user managed duplexing used for Group Buffer Pools. Because of this cost, I normally only recommend duplexing SCA and LOCK1 when these are placed in an internal coupling facility (ICF) that shares a CPC with an MVS LPAR containing a member of the DB2 data sharing group. In this case a CPC failure will cause a multiple failure scenario in DB2 (SCA and/or LOCK1 and a DB2 member). This will usually bring down the entire data sharing group. If the SCA and LOCK1 are in an ICF that does not [...]
1668 13 49_Db2 V8 stored procedure calling cobol sub program14_Sherry Simmerl20_ssimmerl@CSIHOME.COM30_Wed, 1 Feb 2006 10:51:58 -0600683_- Client is using .net and calls a stored procedure. This stored procedure on the main frame in turn calls a cobol sub program. What is the best way to get the sql error codes from sub program back to the client?
--------------------------------------------------------------------------------- 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
1682 155 30_Re: Accessing uncommitted rows14_Michael Kaplan25_micaelkp@NETVISION.NET.IL30_Wed, 1 Feb 2006 19:17:37 +0200526_windows-1250 It is double-checked: no UR is involved.
Michael Kaplan
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of troycci Sent: Wednesday, February 01, 2006 3:58 PM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] Accessing uncommitted rows
I'm not sure what the default isolation level is for SPUFI or how your systems people installed it. What you may want to do is force the isolation level using the "WITH CS" at the end of your query. [...]
1838 198 30_Re: Accessing uncommitted rows14_Avram Friedman21_avramf@IBMSYSPROG.COM30_Wed, 1 Feb 2006 09:45:44 -0800509_iso-8859-1 This may be an index issue. Generally speaking there is no locking for TYPE 2 indexes. By your discription, your query is index only.
Michael Kaplan wrote: It is double-checked: no UR is involved.
Michael Kaplan
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of troycci Sent: Wednesday, February 01, 2006 3:58 PM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] Accessing uncommitted rows [...]
2037 111 30_Re: Accessing uncommitted rows9_Mike Bell21_mbell11a1@VERIZON.NET30_Wed, 1 Feb 2006 11:43:49 -0600265_Windows-1252 Some guess work coming at you - or maybe things to check. Select MAX can be (and usually is ) index only. DB2 does not lock index rows. I do not remember how DB2 handles the committed vs not data in indexes but I suspect this is the problem area. [...]
2149 262 30_Re: Accessing uncommitted rows14_Michael Kaplan25_micaelkp@NETVISION.NET.IL30_Wed, 1 Feb 2006 20:05:51 +0200387_windows-1250 Hi Avram,
Two points :
1. It is very "stupid" simple table defined in dsndb04 and no PK and no INDEX has been defined. 2. In my real problem ( not test case), you are right, I'm going to access INDEX only to get last inserted row as fast as possible. But I do not want to get the data which might be rolled back. Can you suggest how to solve my problem ? [...]
2412 111 29_Re: DMS tablespace size error9_Mo, Don D12_donmo@BU.EDU30_Wed, 1 Feb 2006 13:10:10 -0500363_US-ASCII Hi Lan,
12 PP with a 128M size has 1,572,864,000 bytes. If you have 8K page database structure, you would use
1,572,864,000 / 8192 = 192,000 8K pg
But you used the number of 196,608 in your set tablespace containers statement "(DEVICE "/dev/rShortSpace1" 196608)" which would be too big for the defined size of your raw device. [...]
2524 41 26_Re: Field level encryption16_Steve Allegretti27_sallegretti@DATAVANTAGE.COM30_Wed, 1 Feb 2006 12:39:17 -0600422_us-ascii Take a look at www.datavantage.com if you are looking for a software utility to encrypt/de-identify sensitive DB2 data. It might support IDMS as well.
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Anne Stout-Delaware Valley UG Leader Sent: Tuesday, January 31, 2006 9:51 AM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] Field level encryption [...]
2566 15 30_DB2 ZOS V8 CPU Usage verses V710_Jones, Jim19_jmjones@COMDATA.COM30_Wed, 1 Feb 2006 12:39:00 -0600300_- We're currently on V7 and considering V8. One of the rumors we've heard is that there is a significant CPU increase running the same workload on V8 as V7. So I'm wondering about the experiences of those that have done the migration. FYI we are not using data sharing just multiple subsystems. [...]
2582 175 34_Re: DB2 ZOS V8 CPU Usage verses V715_Carol L Broyles30_carol.broyles@MEADWESTVACO.COM30_Wed, 1 Feb 2006 13:54:28 -0500516_US-ASCII We saw a significant decrease. The system was a non-data sharing business warehouse system, so YMMV.
Carol Broyles Sr. IT Consultant MeadWestvaco Corp.
"Jones, Jim" Sent by: DB2 Data Base Discussion List 02/01/2006 01:39 PM Please respond to DB2 Database Discussion list at IDUG
To DB2-L@WWW.IDUGDB2-L.ORG cc
Subject [DB2-L] DB2 ZOS V8 CPU Usage verses V7 [...]
2758 19 29_Re: DMS tablespace size error3_lan18_lxw176@HOTMAIL.COM30_Wed, 1 Feb 2006 12:54:16 -0600272_- Hi Don,
Do you mean I shoud reduce the PPs from 12 to 3 maybe of 128MB? to make it 3*128 MB page size. or tell me how should I tell our admin to do this? Actually I feel strange why the Shortspace0 with the same size was working fine in mt previous restore. [...]
2778 62 29_Re: DMS tablespace size error9_Mo, Don D12_donmo@BU.EDU30_Wed, 1 Feb 2006 14:17:39 -0500285_US-ASCII Lan,
No, PP is 12. That you don't need to change. Only thing you want to try is to re-execute your restore command with a different number of 8k pages in your statement to fit the actual size of your raw device rShortSpace1, which is of the size of 12 PP * 128 M. [...]
2841 145 59_Re: Stored Procedure Problem - Converting from SPAS to WL M13_Bell, Raymond22_raymond.bell@LANDG.COM30_Wed, 1 Feb 2006 19:48:45 -0000504_iso-8859-1 Wayne,
It's not something as simple as the proc name in the WLM definition (DB2DWLM1) not matching the name in the proc (//DB2DWLM) is it? Is a 1 missing?
Just an idea.
Raymond Bell Database Administrator
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG]On Behalf Of Wayne Stevens Sent: 01 February 2006 14:19 To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] Stored Procedure Problem - Converting from SPAS to WLM [...]
2987 16 29_Re: DMS tablespace size error3_lan18_lxw176@HOTMAIL.COM30_Wed, 1 Feb 2006 13:49:16 -0600643_- Hi Don, I was told the Db2 page size if 4K instead of 8K, so what's the number you think I should use in my command: still 192000 or should be 19200*2?
Thanks! Lan
--------------------------------------------------------------------------------- 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
3004 67 37_Re: DB2 OS/390 v7 reson code 00E7008110_Dave Nance16_dav1mo@YAHOO.COM30_Wed, 1 Feb 2006 12:13:28 -0800583_iso-8859-1 Steve, On the distributed threads, do the threads stay persistent? I think that may be your issue. Do you really need to change your locksize everytime you run this process or can you just get some commit logic added to the process.
Steve Lamb wrote: Help!
On Sunday we had a batch job that Step 1. issues an ALTER LOCKSIZE command for a tablespace Step 2. runs some SQL for about 2.5 hours Step 3. issues an ALTER command to put things back how they were. The job failed in step 3 trying to issue the ALTER LOCKSIZE command [...]
3072 43 29_Re: DMS tablespace size error9_Mo, Don D12_donmo@BU.EDU30_Wed, 1 Feb 2006 15:25:43 -0500392_US-ASCII Hi Lan,
If page size was 4K then yes, the number would be 192,000 * 2 = 384,000 (4K page) in your command statement.
-Don
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of lan Sent: Wednesday, February 01, 2006 2:49 PM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] DMS tablespace size error [...]
3116 29 38_Re: Sql update Performance problem....15_Patrick Bossman18_bossman@US.IBM.COM30_Wed, 1 Feb 2006 14:28:54 -0600317_- Hello, I just want to add a consideration on this issue for those of you moving to V8.
1) The technique proposed by Terry is necessary to obtain optimal performance in V7, but is no longer necessary in V8 NFM (soon for CM/ENFM watch PK12389) due to data type, length, nullability mismatch enhancements. [...]
3146 43 32_db2 udb v8.2 windows/nt contol-m12_Jeff Frazier26_Jeffrey_Frazier@WENDYS.COM30_Wed, 1 Feb 2006 16:03:20 -0500442_US-ASCII Hello, We are attempting to take a db2 udb backup running on windows/nt by running a batch job using control-m job scheduler. The db2 backup runs fine but the control-m job never stops and we have to end up canceling it. Control never seems to be given back to the batch job to terminate. Is anyone out there doing this type of backup using a batch job with a scheduler? if so how do you get it to terminate properly? TIA, Jeff [...]
3190 21 29_Re: DMS tablespace size error3_lan18_lxw176@HOTMAIL.COM30_Wed, 1 Feb 2006 14:57:22 -0600773_- HI Don,
I still get the same error when I issued the command as:
db2 => set tablespace containers for 5 replay rollforward container operations using (DEVICE "/dev/rShortSpace1" 384000) SQL1422N The size of the container is invalid. SQLSTATE=54039
Any other thoughts? Thanks! Lan
--------------------------------------------------------------------------------- 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
3212 38 29_Re: DMS tablespace size error9_Mo, Don D12_donmo@BU.EDU30_Wed, 1 Feb 2006 16:17:51 -0500428_US-ASCII Lan,
Could you do a 'ls -l /dev/*Short*' command?
Also what's in the last 200 lines of your instance's db2diag.log?
-Don
-----Original Message----- From: lan [mailto:lxw176@HOTMAIL.COM] Sent: Wednesday, February 01, 2006 3:57 PM To: DB2-L@WWW.IDUGDB2-L.ORG; Mo, Don D Subject: Re: DMS tablespace size error
HI Don,
I still get the same error when I issued the command as: [...]
3251 16 29_Re: DMS tablespace size error9_Mo, Don D12_donmo@BU.EDU30_Wed, 1 Feb 2006 17:55:41 -0500555_US-ASCII Lan, May I see the restore db command statement you were using?
-Don
--------------------------------------------------------------------------------- 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
3268 37 34_Re: DB2 ZOS V8 CPU Usage verses V733_Schiradin,Roland HG-Dir itb-db/dc28_SchiradinR@ALTE-LEIPZIGER.DE30_Thu, 2 Feb 2006 00:31:17 +0100283_iso-8859-1 OLTP is about 3% better. Batches are almost equal but some of them increase by +20% up to 30%. I believe it has to do with OPTIMIZED FOR 1 ROW. We still looking at them.
Oh well this is a V8 CM system without data sharing. I don't expect changes after V8 NFM. [...]
3306 24 37_Re: DB2 OS/390 v7 reson code 00E7008110_Steve Lamb25_steven.lamb@CAPGEMINI.COM30_Thu, 2 Feb 2006 03:12:53 -0600572_- Dave,
On this occasion the first distributed thread of the day didn't start till after step3 ran and failed. They would have been my first choice for pointing the finger at, but the server they come through was down... I can't see any SPUFI's either.
The table is normally used by CICS transactions and so has row-level locking - the batch job changes it to ANY because of the large amount of processing it does (it runs while the CICS services are out). The workload's supposed to be coming off the mainframe towards the end of the year, so there'll [...]
3331 125 54_Antwort: Re: [DB2-L] DB2 OS/390 v7 reson code 00E7008111_Roy Boxwell16_R.Boxwell@SEG.DE30_Thu, 2 Feb 2006 10:25:36 +0100527_us-ascii you need to flush out your DSC. Somewhere in there is a SQL that references your table.
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
Steve Lamb Gesendet von: DB2 Data Base Discussion List 02.02.2006 10:12 Bitte antworten an DB2 Database Discussion list at IDUG [...]
3457 31 37_Re: DB2 OS/390 v7 reson code 00E7008110_Steve Lamb25_steven.lamb@CAPGEMINI.COM30_Thu, 2 Feb 2006 03:33:02 -0600555_- A little footnote to this...
We also got some -904's on CICS transactions on the same day - the resource that was unavailable was the package.
DSNT408I SQLCODE = -904, ERROR: UNSUCCESSFUL EXECUTION CAUSED BY AN UNAVAILABLE RESOURCE. REASON 00C90090, TYPE OF RESOURCE 00000801, AND RESOURCE NAME DBP1.SOCC020.17B4A6CB02D77D48
The reason code implies that there was some bind activity taking place, as the resource was a package. However, having checked the DB2 Catalog, no binds took place on this day, which confused us a bit. [...]
3489 114 40_DB2 Z/Os V7 : REORG failing with extents21_Sriramulu, Vijayababu28_vijayababu.sriramulu@EDS.COM30_Thu, 2 Feb 2006 15:28:32 +0530536_us-ascii Dear list,
We are doing a REORG on a table which has gone to extents. Since REORG unloads data from the tablespace and tries to loads it to the shadow dataset (online reorg), my expectation will be that the shadow dataset will also reach 251 extents but will succeed. But the REORG failed in the RELEAD phase. I am surprised since it's the same number of records that is existing already in the table is what is unloaded and reloaded, why should DB2 not accompany those same number of rows in the same 251 extents. [...]
3604 158 57_Antwort: [DB2-L] DB2 Z/Os V7 : REORG failing with extents11_Roy Boxwell16_R.Boxwell@SEG.DE30_Thu, 2 Feb 2006 11:02:28 +0100595_iso-8859-1 do not forget that REORG will use your %FREE and Freepage settings! This means that after a Reorg a TS can be much larger than before....
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
"Sriramulu, Vijayababu" Gesendet von: DB2 Data Base Discussion List 02.02.2006 10:58 Bitte antworten an DB2 Database Discussion list at IDUG [...]
3763 19 30_Re: Accessing uncommitted rows33_Walter Jani=?ISO-8859-1?Q?=DFen?=27_walter.janissen@VICTORIA.DE30_Thu, 2 Feb 2006 03:55:23 -0600470_- Hi
I am following this thread with interest. And I also have no idea, why you see uncommitted data. Even if the access is index only, the underlying data pages will be locked. I only wanted to remark that, to set the record a little bit straight.
Maybe, it has something to do with datasharing and lock avoidance. But I would say, you cannot read uncommitted data, unless you tell DB2 you want (option WITH UR oder ISO UR) or there is a bug in DB2. [...]
3783 343 44_Re: DB2 Z/Os V7 : REORG failing with extents19_Alekos Papadopoulos13_apapad@NBG.GR30_Thu, 2 Feb 2006 12:22:40 +0200407_iso-8859-7 Hello Vijay,
Q1: During reorg DB2 follows PCTFREE and FREEPAGE. Even if you already have the data in your tablespace, the rows will be rearranged to make room for new rows to come, according to the specified PCTFREE and FREEPAGE. Unless you have no insert activity or PCTFREE and FREEPAGE are 0, you typically expect a size increase after REORG. In a typical value of [...]
4127 61 34_Re: DB2 ZOS V8 CPU Usage verses V70_19_mike.holmans@BT.COM30_Thu, 2 Feb 2006 11:33:21 -0000540_us-ascii Hi,
We've just taken one of our main production systems up to compat mode. We did no rebinds, so we are still using the packages optimized and bound under V7. We are experiencing about a 9% increase in CPU load. The system is heavily OLTP-biased.
We didn't do any rebinds because when we did do rebinds on our test system, the system came to a stop. As a result of a background task deciding to do a tablespace scan instead of index-only access, it was locking a pivotal table and everything else got timed out. [...]
4189 49 36_Re: Omegamon V540 on DB2 V8 for Z/os0_19_mike.holmans@BT.COM30_Thu, 2 Feb 2006 11:44:15 -0000482_us-ascii Hi,
We're using V540 with DB2 V8 in both CM and NFM. As yet, we have encountered no issues. But we aren't trying to collect to DB2 tables.
Mike Holmans DB2 Consultant BT Exact OneIT Operational Integrity mike.holmans@bt.com
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Asaf Meller Sent: 31 January 2006 14:05 To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] Omegamon V540 on DB2 V8 for Z/os [...]
4239 350 44_Re: DB2 Z/Os V7 : REORG failing with extents19_Alekos Papadopoulos13_apapad@NBG.GR30_Thu, 2 Feb 2006 12:22:40 +0200407_iso-8859-7 Hello Vijay,
Q1: During reorg DB2 follows PCTFREE and FREEPAGE. Even if you already have the data in your tablespace, the rows will be rearranged to make room for new rows to come, according to the specified PCTFREE and FREEPAGE. Unless you have no insert activity or PCTFREE and FREEPAGE are 0, you typically expect a size increase after REORG. In a typical value of [...]
4590 388 30_Re: Accessing uncommitted rows13_Vaughan, Mike26_Vaughan.Mike@PRINCIPAL.COM30_Thu, 2 Feb 2006 07:51:38 -0600549_us-ascii With no Pk and no index defined then your query is most likely getting a tablespace scan followed by a sort to obtain the max row. With isolation(cs), you are not holding any locks as you scan through the pages. Think of the following scenario (I think you mentioned that there are commits after every 50 inserts) -- rows #1-25 are inserted into page #5 and then your query comes along and reads pags 1-4 before stopping at the locked page #5 which has not filled up from the inserts, at which point DB2 finds some free space in pages [...]
4979 409 96_Db2 for ZOS-Java Multiprocess Application Can't Update Records Simultaneously-Deadlock Situation55_=?iso-8859-9?Q?Kadir_G=FCray_Meri=E7_=28B.T.S.Y.G.=29?=22_Kadir.Meric@AKBANK.COM30_Thu, 2 Feb 2006 16:14:49 +0200355_iso-8859-9 Hi to all,
We have a DB2 database and a DB2 table in our DB2 Zos V7 system. We run our program in a multiprocess way. There are 4 instances of
our application. Our program is implemented in JAVA.our application environment is JDK 1.3.1,JDBC 2.0
The flow of the application is;
1) Connect to db2 [...]
5389 88 27_Re: DB2 Z/OS zparm RETVLCFK7_troycci29_troycci@COLEMANCONSULTING.COM30_Thu, 2 Feb 2006 08:34:09 -0600350_us-ascii John I agree with what you said. What amazes me is that you actually have developers looking at the real size of a varchar field and putting in the LENGTH. In most cases from what I've seen is developers never look at this and just put the max size in anyways. It is to much work to inspect the field and figure out the real length. :) [...]
5478 23 30_Re: Accessing uncommitted rows10_Jorg Lueke21_jorg_lueke@EFUNDS.COM30_Thu, 2 Feb 2006 08:27:45 -0600543_- Michael,
Did you validate you insert application? Say you insert 99 rows and then issue a Rollback, does the SPUFI query always show 0 rows?
If that works as expected could it be possible that you are seeing the pages as they are being commited. The commit would release the locks page by page, so perhaps you were viewing the data while the commit had released part of a batch if inserts. I don't think you can force it to hold all 100 rows once you issue the commit unless you are writing all the data to the same page. [...]
5502 12 61_Re: Antwort: [DB2-L] DB2 Z/Os V7 : REORG failing with extents10_Jorg Lueke21_jorg_lueke@EFUNDS.COM30_Thu, 2 Feb 2006 08:31:58 -0600640_- Freespace is a potential problem. In my experience the DBA forgetting to rebuild the compression dictionary has lead to the system described much more often though :-)
--------------------------------------------------------------------------------- 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
5515 43 28_Re: Logging in DB2 Connect ?10_Jorg Lueke21_jorg_lueke@EFUNDS.COM30_Thu, 2 Feb 2006 08:40:20 -0600507_- On Wed, 1 Feb 2006 14:49:14 +0100, Peter, Georg wrote:
>Collegues, environment is DB2 for z/OS version 7 and DB2 Connect version >8.1. > >As usual our Java appz are using DB2 Connect to access the big iron. These >appz are working with dynamic SQL and each statement has a clause "fetch >first 50 rows" (as the programmers told me). > >But it seems that this clause is ignored by DB2 and thousands or millions of >rows were sent back to the appz server - bad, bad, bad...... [...]
5559 64 34_Re: DB2 ZOS V8 CPU Usage verses V714_Grainger, Phil20_Phil.Grainger@CA.COM30_Thu, 2 Feb 2006 15:03:19 -0000464_US-ASCII To GREATLY paraphrase various comments from IBM (and the expanded words are a lot longer than this!)
1. Running the SAME workload under v8 compared with v7 will have a variable change in CPU load from about -20% (i.e. better) to +30% (i.e. worse) 2. The better tuned your SQL is on v7, the further towards the +30% end you are likely to be 3. The more plans/packages you rebind AFTER moving to v8, the further towards the -20% end you will be [...]
5624 90 26_Re: Field level encryption7_troycci29_troycci@COLEMANCONSULTING.COM30_Thu, 2 Feb 2006 09:07:57 -0600621_us-ascii Are you looking for encryption of columns in production to be used by production transactions or are you looking for a data privacy solution when moving the production data to test?
If you are looking for data privacy for test data management then take a look at Testbase http://www.softbase.com/sbprod1.php
You have to be careful when you use the term encrypt when you are dealing with test data management. Some product use the term data encryption for test data management. Encryption is really used to secure data so if someone hacks into your database they cannot read the data. When the [...]
5715 82 37_Re: DB2 OS/390 v7 reson code 00E700817_troycci29_troycci@COLEMANCONSULTING.COM30_Thu, 2 Feb 2006 09:07:57 -0600560_us-ascii Did you do a -DIS DB(xxx) SPACE(*) USE This will show the CONNID, CORRID, and USERID using the object. I have had situations where I cannot afford to have a distributed thread come in during a given batch process so I've added a step to STOP then START DDF. The problem with this is that you kick everyone off the subsystem. The other thing you could do is write a rexx routine that will issue the -DIS DB(xxx) space(xxx) USE to get the USERID. Then -DIS THREAD(*) TYPE(ACTIVE) and issue the -CANCEL THREAD(token) for each token for this userid. [...]
5798 297 44_Re: DB2 Z/Os V7 : REORG failing with extents18_Prather, Doug, ISD28_Dprather467@WORLDSAVINGS.COM30_Thu, 2 Feb 2006 09:09:30 -0600582_us-ascii Vijay , you need contiguous space and Alters on PQTY to match the existing size of the Tablespace (accounting for a PCTFREE and some growth ) and a SQTY that matches some chunk (10%?) that you would like to take extents in. If your DASD pool doesn't have enough contiguous space in it for your primary Tablespace allocation then you will immediately go into extents (possibly 251 if the DASD pool is extremely fragmented). I'd suggest you talk to your DASD folks and either defragment the pool or add another pack that can hold the Tablespace. While you're asking for [...]
6096 103 34_Re: DB2 ZOS V8 CPU Usage verses V77_troycci29_troycci@COLEMANCONSULTING.COM30_Thu, 2 Feb 2006 09:17:30 -0600402_us-ascii You may want to look at a tool like DBIQ/QA http://www.softbase.com/sbprod2a.php to explain all your DBRM's in V7 then in V8 and compare them for difference. This will help you quickly identify the SQl access paths that have changed so you can fix the problem before it hits production.
Troy Coleman, Support Engineer IBM Certified Solutions Expert [...]
6200 190 61_Re: Antwort: [DB2-L] DB2 Z/Os V7 : REORG failing with extents15_Robert Lawrence21_rlawrence@BOSCOVS.COM30_Thu, 2 Feb 2006 10:20:23 -0500515_iso-8859-1 Vijay, The number of extents depends on the number of discrete areas of DASD that is needed to satisfy the request that DB2 makes for space as it fills its pageset with data. So yes if it takes 3 extents to satisfy the request then the extent count increases by 3.. If all you are attempting is extent consolidation you may be able to do this with DFDSS to volumes which have enough contiguous space. Or you may need to DEFRAG the target volumes(always a problem in a 24 X 7 environment). Good luck [...]
6391 249 44_Re: DB2 Z/Os V7 : REORG failing with extents7_troycci29_troycci@COLEMANCONSULTING.COM30_Thu, 2 Feb 2006 09:31:56 -0600547_us-ascii You have seen some good information from other list members on the fact that REORG will honor the Percent Free and Free Page. I've used a tool in the past from Platinum now CA called Database Analyzer. With a tool like Database Analyzer you can automate the ALTER PRIQTY/SECQTY before the REORG. I used to keep my tablespaces 10 to 20% larger than the current number of rows in the table. This leaves room for growth. A common practice by many DBA's is to ask the development team how many rows will be in this table in a year. Size [...]
6641 114 37_Re: DB2 OS/390 v7 reson code 00E700815_db2-117_db2-1@NIOI.GOV.IL30_Thu, 2 Feb 2006 17:56:04 +0200444_iso-8859-1 Hello Steve, the prepare statement is in the EDM pool , you can clear it by running runstats on the problematic table. In V8 there is a much lighter solution : runstats (update none report no) ----- Original Message ----- From: "troycci" Newsgroups: bit.listserv.db2-l To: Sent: Thursday, February 02, 2006 5:07 PM Subject: Re: [DB2-L] DB2 OS/390 v7 reson code 00E70081 [...]
6756 146 42_DB2PM GUI views for Bufferpool question ??16_Trivedi, Kaushal34_Kaushal.J.Trivedi@CITIZENSBANK.COM30_Thu, 2 Feb 2006 11:05:17 -0500563_us-ascii Hi All,
I am looking through DB2PM's GUI tool and we have a 2 way datasharing setup. You can select either each DB2 or the Group when viewing bufferpool statistics under the statistics detail. I see very high numbers for one of the DB2 and very less for the other DB2. I have been going back and forth with the systems group to verify if WLM was distributing loads to both members . What they say is WLM works at much higher level than DB2 specific requests and it distributes loads from a system throughput perspective. I think I understand [...]
6903 21 37_Re: DB2 OS/390 v7 reson code 00E7008110_Steve Lamb25_steven.lamb@CAPGEMINI.COM30_Thu, 2 Feb 2006 10:21:40 -0600576_- Unfortunately we didn't get called when the problems occured and so nobody did any DISPLAYs at the time - has anybody written a retrospective Display yet?! The reason code says that the prepared statement is "in the prepared statement cache and is in use by an application". It's the "in use" bit that is confusing me, because I don't think that there WAS anything actually running any SQL against the tablespace. I've run several TMON reports and there just doesn't seem to have been anything else running at the same time. No threads at all (not just ones using this [...]
6925 293 46_Re: DB2PM GUI views for Bufferpool question ??12_Higgins John26_HigginsJohnP@JOHNDEERE.COM30_Thu, 2 Feb 2006 10:39:29 -0600563_us-ascii Each user connects to only one datasharing member. All of the user's work will be done by that member -- unless a query invokes sysplex parallelism.
The main balancing of load between the members is to be sure the users are distributed between the members.
John
________________________________
From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Trivedi, Kaushal Sent: Thursday, February 02, 2006 10:05 AM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] DB2PM GUI views for Bufferpool question ?? [...]
7219 45 36_View permission puzzle (z/OS V8 NFM)0_19_mike.holmans@BT.COM30_Thu, 2 Feb 2006 16:52:17 -0000427_us-ascii We've got a bit of a funny with regard to granting permissions on views.
I have a table T1.
I issue GRANT ALL ON TABLE T1 TO USER1 WITH GRANT OPTION
SYSTABAUTH shows that USER1 has Gs in SELECTAUTH, INSERTAUTH, UPDATEAUTH, DELETEAUTH, REFERENCESAUTH and TRIGGERAUTH
Then I issue SET CURRENT SQLID = 'USER1' followed by CREATE VIEW USER1.V1 AS SELECT col1, col2, ... , col-last FROM T1 [...]
7265 19 42_Listserver archives not being kept current18_Sandra Lakenburger34_sandra.lakenburger@MAIL.SPRINT.COM30_Thu, 2 Feb 2006 10:47:39 -0600310_- This is for the list owners,
Is there any chance to get the archives a little more up-to-date? It looks like the last one was from the first week in January. I wanted to point a collegue to the archives so they could look at the threads discussing the zIIP engines without having to do a search. [...]
7285 319 46_Re: DB2PM GUI views for Bufferpool question ??16_Trivedi, Kaushal34_Kaushal.J.Trivedi@CITIZENSBANK.COM30_Thu, 2 Feb 2006 11:58:19 -0500438_us-ascii How does this work from a App server perspective ? Our app servers are setup to utilize SYSPLEX and they go thru our DB2 connect server. The port that DB2 connect goes thru on the MF is setup to use parallel sysplex . This tells me things should be distributed evenly why should I still see such a big difference in the BPOOL usage on both members and should I be looking at something else as well that ay not be turned on. [...]
7605 42 29_Re: DMS tablespace size error3_lan18_lxw176@HOTMAIL.COM30_Thu, 2 Feb 2006 10:56:02 -0600502_- Hi Don,
Here is the command I use:
db2 => create database javaprod on /db2/db2prod1 USING CODESET ISO8859-1 TERRITORY US collate USING SYSTEM
db2 => restore database javaprod from "/db2temp/" taken at 20060112154626 to "/db2/db2prod1/" into javaprod newlogpath "/db2/db2prod1/logs/" with 1 buffers buffer 1024 redirect parallelism 1 without prompting
db2 => SET TableSpace containers for 0 replay rollforward container operations using (PATH "/db2/db2prod1/catalog") [...]
7648 243 46_Re: DB2PM GUI views for Bufferpool question ??12_Higgins John26_HigginsJohnP@JOHNDEERE.COM30_Thu, 2 Feb 2006 11:12:21 -0600360_us-ascii Well, there are all kinds of App servers. I don't know what you have or how they work.
My App servers are SAP. Each App server has a primary connection to one datasharing member and a fail-over connection available to the other datasharing member. Each App server only connects to its primary member unless it fails over to its alternate. [...]
7892 26 13_NPGTHRSH = -114_Mark McCormack27_mamccormack@STATESTREET.COM30_Thu, 2 Feb 2006 12:24:24 -0500343_US-ASCII We run Content Manager On Demand and we are seeing some tables (not small ones) that are being scanned, where we think that access through an index should be better. I know it is not a good idea to try to fool Mother Nature or the DB2 Optimizer. Still, some people suggest changing zparm NPGTHRSH from the default value of zero. [...]
7919 73 29_Re: DMS tablespace size error9_Mo, Don D12_donmo@BU.EDU30_Thu, 2 Feb 2006 12:40:16 -0500588_US-ASCII Lan,
You want to abort your db restore operation by using the "db2 restore db javaprod abort" command every time your restore failed to clear it up; and "db2 restore db javaprod continue" if succeeded;
The diaglog says that the db restore had issues/conflicts with the size of the raw device. I'd recommend you delete the /dev/rShortSpace1 raw device and rebuilt it to the allocated size of 12 PP, and then change the owner of all the raw devices /dev/*Space1 to name of current your DB2 instance (use cmd chown); right now they are all owned by root and they [...]
7993 161 44_Re: DB2 Z/Os V7 : REORG failing with extents12_Isaac Yassin20_yassini@BEZEQINT.NET30_Thu, 2 Feb 2006 20:02:18 +0200441_US-ASCII Hi,
While reloading, db2 respects the freepage/pctfree so your shadow is bigger than your source. (the rest has already been answered)
Isaac Yassin
_____
From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Sriramulu, Vijayababu Sent: Thursday, February 02, 2006 11:59 AM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] DB2 Z/Os V7 : REORG failing with extents [...]
8155 74 34_Re: DB2 ZOS V8 CPU Usage verses V712_Isaac Yassin20_yassini@BEZEQINT.NET30_Thu, 2 Feb 2006 20:03:42 +0200445_US-ASCII Hi,
We were +20% in V8 NFM, no rebind.
Isaac Yassin
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Grainger, Phil Sent: Thursday, February 02, 2006 5:03 PM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] DB2 ZOS V8 CPU Usage verses V7
To GREATLY paraphrase various comments from IBM (and the expanded words are a lot longer than this!) [...]
8230 19 75_Runstats In what phase of DB2 V8 migration is the 40% reduction in RUNSTATS13_Wayne Stevens25_twstevens@DHR.STATE.AL.US30_Thu, 2 Feb 2006 12:51:46 -0600653_us-ascii IBM reports that RUNSTATS TABLESPACE in DB2 V8 has a 40% reduction in time.
What phase does DB2 V8 have to be in to realize that performance improvement?
Wayne
--------------------------------------------------------------------------------- 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
8250 173 59_Re: Stored Procedure Problem - Converting from SPAS to WL M13_Wayne Stevens25_twstevens@DHR.STATE.AL.US30_Thu, 2 Feb 2006 12:47:09 -0600349_us-ascii I want to thank everyone who replied. We had 2 problems. The WLM has TYPE=DB2D, it should have had TYPE=DB2. It was set to the subsystem in error. That was the 1st problem. The second problem was that RRS was not set up. Once we address those 2 issues, we were able to have the WLM being started when the stored procedure was invoked. [...]
8424 47 82_Reorg using Listdef, with a backup, Backup not dynamically allocating proper size.9_Dee Reins24_dee.reins@CENTURYTEL.COM30_Thu, 2 Feb 2006 13:17:22 -0600410_us-ascii Currently in DB2 V7, our reorg's work with a backup, and dynamically allocate the proper backup file size. On our test LPAR we have the same database that is upgraded to V8. Version 8 does not allocate the backup files to the proper size. Version 8 allocates the backup files in 4 trac increments. The current Version 7 allocates the proper backup file sizes. Both systems have current runstats. [...]
8472 139 79_Re: Runstats In what phase of DB2 V8 migration is the 40% reduction in RUNSTATS0_25_LL581@DAIMLERCHRYSLER.COM30_Thu, 2 Feb 2006 15:04:34 -0500453_US-ASCII Wayne,
If by "phase" you mean what Mode, DB2 V8 in Compatibility Mode (CM) makes available this new utility function. In fact, almost all new V8 utility functionality is available in CM, with the exception of BACKUP SYSTEM and RESTORE SYSTEM.
Lock Lyon Data Services/DBA/DW ITM Technology & Architecture DaimlerChrysler Services North America LLC Phone: (248) 427-5881 T/L: 870-5881 Email: LL581@dcx.com Page: (586) 898-9080 [...]
8612 40 16_REORG DB2 V7/zos27_DB2 Database Solutions Port26_db2dba_certified@YAHOO.COM30_Thu, 2 Feb 2006 13:45:02 -0800370_iso-8859-1 Ref: db2 v7 /zos- reorg.
In our client place, we have weekly RUNSTATS set with UPDATE SPACE option. For some reason they do not want to change the access path. Here is my question.
If I set up a generic REORG jobs with OFFPOSLIMIT, INDREFLIMIT options, will it do any help?. Note: CARDF doesn't get updated as we run ONLY space statistics. [...]
8653 52 37_Re: DB2 OS/390 v7 reson code 00E7008125_Thanikachalam Sundarrajan21_billysundar@YAHOO.COM30_Thu, 2 Feb 2006 16:02:41 -0800413_iso-8859-1 Steve,
Check your CURSORHOLD settings in your DB2 connect/DB2 client. You could be holding on to locks even though nothing shows in the monitor. The CURSORHOLD option has a default of 1 (hold cursors across commits and holds locks after the SQL is complete) unless explicitly set in the CLI.INI file. If you have Omegamon/ DB2 PE you can easily find the offending dynamic SQL in the cache. [...]
8706 331 44_Re: DB2 Z/Os V7 : REORG failing with extents21_Sriramulu, Vijayababu28_vijayababu.sriramulu@EDS.COM30_Fri, 3 Feb 2006 08:27:34 +0530452_iso-8859-1 Thank you all for all your responses !
Every response added more information on this topic which provided me a better insight on this particular process.
Thanks again !
Regards Vijay
________________________________
From: DB2 Data Base Discussion List on behalf of Prather, Doug, ISD Sent: Thu 2/2/2006 8:39 PM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] DB2 Z/Os V7 : REORG failing with extents [...]
9038 45 46_Re: Listserver archives not being kept current14_James Campbell25_jacampbell@ACSLINK.NET.AU30_Fri, 3 Feb 2006 18:39:41 +1100511_US-ASCII I have found that this problem is usually the generation of the index page. If you open a week's archives (eg 1st week of Jan 2006), you'll see that the URL is http://www.idugdb2- l.org/adminscripts/wa.exe?A1=ind0601a&L=db2-l
"06" = 2006; "01" = first month; "a" = first week.
Editing this to, say, http://www.idugdb2- l.org/adminscripts/wa.exe?A1=ind0601e&L=db2-l
brings up the archives for the fifth week - even though they are not (currently) listed on the archives index. [...]
9084 37 41_Insert processing and pseudo-deleted rows12_Adam Baldwin19_y0027528@ES.IBM.COM30_Fri, 3 Feb 2006 01:40:21 -0600520_- Fellow Listers - I need help with an explanation of the following:
We have a legacy system with a single table in a simple table space. The tablespace is defined with a 4000Kb/1000 page primary.
The table is used as a "scratch pad".
The table has one index that is clustering by default and not by design. Each month tens of thousands of rows are inserted into the table via various processes. They are then deleted by a batch proccess one row at a time and the table is left with zero rows. [...]
9122 10 45_Re: Insert processing and pseudo-deleted rows12_Adam Baldwin19_y0027528@ES.IBM.COM30_Fri, 3 Feb 2006 01:41:57 -0600542_- Sorry, I forgot to add that we are running DB2 V7 on OS/390 02.10.00.
--------------------------------------------------------------------------------- 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
9133 22 37_Re: DB2 OS/390 v7 reson code 00E7008110_Steve Lamb25_steven.lamb@CAPGEMINI.COM30_Fri, 3 Feb 2006 02:59:56 -0600323_- On this occasion the distributed queries seem to be innocent - there weren't any connections made before the batch job failed, as the server the distributed queries come though was down. DB2 was reloaded just after midnight, the batch job fell over at 06:15 and the first distributed query didn't start until 09:53. [...]
9156 149 44_DB2 for z/OS to Oracle for Unix connectivity12_Jim McAlpine22_Jim.McAlpine@CEDAR.COM30_Fri, 3 Feb 2006 10:20:06 -0000423_iso-8859-1 Is there any way to have a CICS application accessing DB2 tables on z/OS to also have access to Oracle tables sitting on a Unix machine.
Jim McAlpine Email: jim.mcalpine@cedar.com Direct Dial: +44 (0)191 4827856
CedarOpenAccounts PrismTech House 5th Avenue Business Park Team Valley Gateshead Tyne and Wear NE11 ONG United Kingdom Tel 0191 482 7900 Fax 0191 482 7901 www.cedaropenaccounts.com [...]
9306 291 55_R: [DB2-L] DB2 for z/OS to Oracle for Unix connectivity18_Vincenzo Di Franco30_vincenzo.difranco@ISIDE.BCC.IT30_Fri, 3 Feb 2006 13:31:48 +0100469_iso-8859-1 Oracle Gateway for DB2
-----Messaggio originale----- Da: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] Per conto di Jim McAlpine Inviato: venerdì 3 febbraio 2006 11.20 A: DB2-L@WWW.IDUGDB2-L.ORG Oggetto: [DB2-L] DB2 for z/OS to Oracle for Unix connectivity
Is there any way to have a CICS application accessing DB2 tables on z/OS to also have access to Oracle tables sitting on a Unix machine. [...]
9598 163 48_Re: DB2 for z/OS to Oracle for Unix connectivity14_Neff Stephen R22_Stephen.R.Neff@IRS.GOV30_Fri, 3 Feb 2006 07:36:48 -0500485_- Information Integrator with an ORACLE WRAPPER will allow this.
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Jim McAlpine Sent: Friday, February 03, 2006 5:20 AM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] DB2 for z/OS to Oracle for Unix connectivity
Is there any way to have a CICS application accessing DB2 tables on z/OS to also have access to Oracle tables sitting on a Unix machine. [...]
9762 425 60_Re: R: [DB2-L] DB2 for z/OS to Oracle for Unix connectivi ty13_Bell, Raymond22_raymond.bell@LANDG.COM30_Fri, 3 Feb 2006 12:42:36 -0000449_iso-8859-1 Oracle Gateway? Do you mean the OTG (i.e. the Oracle Transparent Gateway)? If you do, stand well back, Jim. The mainframe bit I used, admittedly 4 years ago, was the flakiest (sp?) bit of MF software I'd ever seen. Good luck to your and yours. Apparently the AIX-based version was better - no started tasks required - but never got 'round to trying it out. But we were going 'up' to access z/OS DB2, not 'down' to access Oracle DBs. [...]
10188 293 55_R: [DB2-L] DB2 for z/OS to Oracle for Unix connectivity18_Vincenzo Di Franco30_vincenzo.difranco@ISIDE.BCC.IT30_Fri, 3 Feb 2006 13:52:58 +0100517_iso-8859-1 To complete my suggestion : Oracle AM4CICS and Oracle Gateway for DB2.
-----Messaggio originale----- Da: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] Per conto di Jim McAlpine Inviato: venerdì 3 febbraio 2006 11.20 A: DB2-L@WWW.IDUGDB2-L.ORG Oggetto: [DB2-L] DB2 for z/OS to Oracle for Unix connectivity
Is there any way to have a CICS application accessing DB2 tables on z/OS to also have access to Oracle tables sitting on a Unix machine. [...]
10482 75 83_Baltimore/Washington DB2 Users' Group--HEADS UP!!! BONNIE BAKER IS COMING TO TOWN!!14_Henry L Nalven25_henry.nalven@MARRIOTT.COM30_Fri, 3 Feb 2006 07:00:18 -0600660_ISO-8859-1 HEADS UP!!! BONNIE BAKER IS COMING TO TOWN!! The Baltimore/Washington DB2 Users' Group Special DB2 Course Offering - Wednesday, March 8, 2006 presenting...
Bonnie Baker International DB2 Consultant
• DBA Things a Programmer Should Know
There are times when DBA issues and Developer issues overlap. What does the DBA know that a Developer should know? And how can this knowledge be used to make designing and writing programs easier? What DDL parameters impact program design and data integrity? And what DB2 roadblocks exist that can make the most perfect program perform poorly? These questions and more will be answered. [...]
10558 266 46_Re: Listserver archives not being kept current13_Seibert, Dave26_Dave.Seibert@COMPUWARE.COM30_Fri, 3 Feb 2006 08:49:15 -0500366_- Hi Sandra,
Try this:
From the List Archives page, click Search the Archives. Then click List Archives.
This has always worked for me to update the list of the archives. As James implies, the data is probably always there - it's just the index is off.
Perhaps the site is run from an Oracle database and there is a corrupted index. [...]
10825 288 46_Re: Listserver archives not being kept current13_Bell, Raymond22_raymond.bell@LANDG.COM30_Fri, 3 Feb 2006 14:13:42 -0000485_iso-8859-1 Yay - turns out today is 'Stick the knife in all Oracle products' Day. IBM should lobby, well, everyone to make it an international holiday...
Raymond Bell DB2 Database Administrator and Damn Proud Of It ;o)
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG]On Behalf Of Seibert, Dave Sent: 03 February 2006 13:49 To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] Listserver archives not being kept current [...]
11114 395 53_Re: DB2 for Z/OS Version 8 and OS/VS Cobol and COBOL216_Galambos, Robert29_Robert.Galambos@COMPUWARE.COM30_Fri, 3 Feb 2006 09:34:07 -0500470_iso-8859-1 Rich
May I suggest that you this to the 'Code Place' within the www.idug.org website (insider). This is where we offer code that maybe of help to others.
If you are interested go to www.idug.org and proceed to the 'Insider'. then on the left side you will see a link to the Code Place. This is where you can search and/or submit code that you feel would be nice to offer to the community [...]
11510 98 45_Re: Insert processing and pseudo-deleted rows0_28_Missy.Case@FIRSTDATACORP.COM30_Fri, 3 Feb 2006 09:02:50 -0600372_US-ASCII Adam,
Are you datasharing? Did your tablespace get altered with the member cluster option set to yes? That can sometimes cause the space issue you're seeing. It lets every member of the datasharing group run it's own 'set' of cluster pages, so 2 or 3 members can cause quite a jump in space when the inserts are run on different members of the group. [...]
11609 36 46_Re: Listserver archives not being kept current13_Willie Favero21_wfavero@ATTGLOBAL.NET30_Fri, 3 Feb 2006 09:28:29 -0600692_us-ascii Fixed
Willie
Sandra Lakenburger wrote:
>This is for the list owners, > >Is there any chance to get the archives a little more up-to-date? It looks >like the last one was from the first week in January. I wanted to point a >collegue to the archives so they could look at the threads discussing the >zIIP engines without having to do a search. > >Thanks > >Sandra Lakenburger > >--------------------------------------------------------------------------------- >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 [...]
11646 11 45_Re: Insert processing and pseudo-deleted rows12_Adam Baldwin19_y0027528@ES.IBM.COM30_Fri, 3 Feb 2006 09:31:12 -0600549_- No, the tablespace in question is in a stand alone DB2 on a non sysplex box.
--------------------------------------------------------------------------------- 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
11658 399 54_DB2 V8 Abends caused by WebSphere Application Timeouts15_Schade, Jeffrey15_JSchade@ISO.COM30_Fri, 3 Feb 2006 11:17:38 -0500605_iso-8859-1 We are running DB2 V8 Z/OS in compatibility mode along with WebSphere. The design of WebSphere dictates that when certain timeouts are reached the server automatically re-cycles itself. Something to do with the JVM not being able to clean-up storage. This in itself is bad enough but we have had two instances where the WebSphere re-cycles have caused my production DB2 sub-system to crash. We have a PMR opened with IBM and after the 1st occurrence they pointed out required (very recent) fixes that we needed to apply that we thought corrected the situation but then it happened again 2 [...]
12058 14 17_Re: NPGTHRSH = -110_Jones, Jim19_jmjones@COMDATA.COM30_Fri, 3 Feb 2006 10:57:33 -0600301_- If you're on V8 try the VOLATILE option of the ALTER TABLE. I've had success with that on some tables on the Unix flavor. On the mainframe (we're V7) I've not had the nerve to try a -1 and experimenting with small positive numbers as IBM suggests didn't change things much, but of course YMMV. [...]
12073 59 30_TEMPLATE UTILITY &HOUR. option27_DB2 Database Solutions Port26_db2dba_certified@YAHOO.COM30_Fri, 3 Feb 2006 09:35:28 -0800296_iso-8859-1 I am using TEMPLATE UTILITY to get generic dataset names from the COPY utility. I am using &HO. to get the "hour" and &MI. for the "minutes". I noticed 5 hours difference between ICTIME vs the hour generated by &HO. option. No major issue. But any idea why there is a difference? [...]
12133 264 34_Re: TEMPLATE UTILITY &HOUR. option12_McKown, John26_john.mckown@UICIINSCTR.COM30_Fri, 3 Feb 2006 11:37:59 -060090_us-ascii What timezone are you in? It could be GMT vs. Local Time.
12398 62 34_Re: DB2 ZOS V8 CPU Usage verses V712_Norbert Wolf21_norbert.wolf@DATEV.DE30_Fri, 3 Feb 2006 11:43:46 -0600567_ISO-8859-1 Hi ,
we are just on our migration path to V8. My experience is, that you have to check your access path's before!! you move to V8. You can read this advice also in several article's written by IBM. We already experienced in V7 (production) und V8 (in our test enviroment) critical access path changes ( modification of join sequences and join method's and other serious experiences, strange access path results with using the fetch first clause under V8 and so on), that forced us to precheck our access paths before doing the migration to V8. [...]
12461 89 34_Re: TEMPLATE UTILITY &HOUR. option27_DB2 Database Solutions Port26_db2dba_certified@YAHOO.COM30_Fri, 3 Feb 2006 10:32:21 -0800389_iso-8859-1 John: I am in EST. If I query current time from db2, it shows local time (same as ICTIME value). How can I control this to pick the local time?.
-VG
"McKown, John" wrote: What timezone are you in? It could be GMT vs. Local Time.
-- John McKown Senior Systems Programmer UICI Insurance Center Information Technology [...]
12551 464 34_Re: TEMPLATE UTILITY &HOUR. option12_McKown, John26_john.mckown@UICIINSCTR.COM30_Fri, 3 Feb 2006 12:53:01 -0600427_us-ascii From reading the book, under "Table 123" in "DB2 Universal Database for z/OS: Utility Guide and Reference" you can't:
Note:, All date and time values are set by using the STCK instruction, and they reflect the date and time value in Greenwich Mean Time (GMT). DATE and TIME values are captured in the UTILINIT phase of each utility and remain constant until the utility terminates.
[...]
13016 77 34_Re: DB2 ZOS V8 CPU Usage verses V718_Angela M Arettines28_aarettines@KEYSPANENERGY.COM30_Fri, 3 Feb 2006 14:15:35 -0500601_ISO-8859-1 Would others who have gone through migration to Version 8 already please share with us information about path checker tools they have trialed/used? What were the pros/cons? How does the tool you used differ from other vendors' path checker tools that you are familiar with? Thanks. Angela
Norbert Wolf wrote:
>Hi , > >we are just on our migration path to V8. My experience is, that you have to >check your access path's before!! you move to V8. You can read this advice >also in several article's written by IBM. We already experienced in V7 >(production) und V8 (in our [...]
13094 138 34_Re: DB2 ZOS V8 CPU Usage verses V714_Avram Friedman21_avramf@IBMSYSPROG.COM30_Fri, 3 Feb 2006 12:03:21 -0800421_iso-8859-1 We used the Verizon tool for our V8 upgrade. When the phone rang, we knew we might be in trouble.
Fortunately trouble only came a small number of times for us. The migration to V8 is the cleanest I have ever participated in and I have been installing and maintaining DB2 since the first customer ship of V1 in the early 80's. Note we have NF enabled but do not go out of our way to exploit it yet. [...]
13233 81 53_Re: DB2 for Z/OS Version 8 and OS/VS Cobol and COBOL215_Philip Sevetson22_db2.dba.guru@GMAIL.COM30_Fri, 3 Feb 2006 15:20:59 -0500412_ISO-8859-1 Rich,
*Very* useful. Nice, quick, self-explanatory results. Thanks very much!
On 1/30/06, Gugel, Rich - DOA wrote: > > Here's a bit of SQL I wrote to identify package and plans running with old > Cobol DBRM's. I apologize in advance if someone has already posted this > or presented it at a conference > > "IBM Cobol" is the result you want to see. > >
13315 138 20_Re: REORG DB2 V7/zos15_Philip Sevetson22_db2.dba.guru@GMAIL.COM30_Fri, 3 Feb 2006 16:35:45 -0500624_ISO-8859-1 What do you want your generic REORGs to address? Are you having troubles with unclustered data (CLUSTERRATIOF dropped more than a few percent) or badly fragmented indexes (High Faroffpos)? If these are your problems, REORGs will help.
--Phil
On 2/2/06, DB2 Database Solutions Port wrote: > > Ref: db2 v7 /zos- reorg. > > In our client place, we have weekly RUNSTATS set with UPDATE SPACE option. > For some reason they do not want to change the access path. Here is my > question. > > If I set up a generic REORG jobs with OFFPOSLIMIT, INDREFLIMIT options, [...]
13454 33 50_AW: [DB2-L] DB2 for LUW Static SQL Monitoring tool12_Per Stricker22_per.stricker@ITGAIN.DE30_Fri, 3 Feb 2006 22:42:05 +0100743_ISO-8859-1 Try Speedgain for DB2. The tool will provide you with the information you requested and much more...
http://www.speedgain.de/en/speedgain_trial.html
Per
> >If anyone has experience with a tool that gathers information about static >SQL on the DB2 for Linux, Unix, Windows (LUW) platform, I would be >interested in hearing about it. By gathering information, I mean number of >times the static SQL statement has executed, execution times, CPU cost for >the statement, etc. I know event monitors can capture data, but they don't >aggregate the results. > >------------------------------------------------------------------------------- >-- >Welcome to the IDUG DB2-L list. To unsubscribe, go to the archives [...]
13488 112 26_Re: Field level encryption15_Philip Sevetson22_db2.dba.guru@GMAIL.COM30_Fri, 3 Feb 2006 17:17:17 -0500634_ISO-8859-1 DB2 for z/OS encryption providers: www.protegrity.com www.megacryption.com www.ingrian.com IBM has a row-level encryption solution for DB2 and IMS which uses EDITPROC and the ICSF.
I don't know if any of these vendors supply IDMS encryption, sorry.
What's driving your encryption need? PCI? SOX? DoD? Or just the headlines?
On 1/31/06, Anne Stout-Delaware Valley UG Leader wrote: > > I'm at DB2 V7, and need to be able to encrypt data on a field level. Does > anyone have a method for doing this that they would be willing to share? I > also need a solution for the IDMS [...]
13601 106 36_Re: DB2 V7 z/os v1.4 data encryption15_Philip Sevetson22_db2.dba.guru@GMAIL.COM30_Fri, 3 Feb 2006 17:41:22 -0500628_ISO-8859-1 Jeff,
See my reply to Anne Stout, a few minutes ago, for a list of vendors.
--Phil
On 1/31/06, Jeff Frazier wrote: > > > Hello, > > We currently have an application running on DB2 UDB AIX that encrypts > data. There is a requirement to move this to DB2 V7 z/os. We have plans to > go to v8 but not near future. Are there products that anyone is using that > will encrypt/decrypt the data within db2 v7 that we can currently use? > TIA, > Jeff --------------------------------------------------------------------------------- > Welcome to the IDUG DB2-L list. To [...]
13708 156 34_Re: DB2 ZOS V8 CPU Usage verses V77_Ed Long19_rdhm99a@PRODIGY.NET30_Fri, 3 Feb 2006 14:47:11 -0800553_iso-8859-1 Fascinating discussion.
In a production shop I don't know how you could make a change like upgrade to V8 and intentionally not rebind. You are postponing until the middle of a dark and stormy night finding out that you now have bad access paths.
A better path, so to speak, in my view, is don't go to V8 until you know that all your critical access paths are at least not materially worse than they were on V7. That way when the oncall programmer has to rebind a program at 3am you are unlikely to be awakened at 0400 with [...]
13865 149 43_Re: bmc tools install question- db2 z/os v712_HEPP SHERY C17_schepp@SRPNET.COM30_Fri, 3 Feb 2006 15:53:55 -0700501_us-ascii I wanted to thank Craig, Carol, Sherri and Peter for taking the time to respond to my question. The general consensus is to go with 1 loadlib- which is what we'll move towards.
Again- thanks very much.
Shery
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of craig patton Sent: Wednesday, February 01, 2006 8:10 AM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] bmc tools install question- db2 z/os v7 [...]
14015 25 34_Re: DB2 ZOS V8 CPU Usage verses V714_James Campbell25_jacampbell@ACSLINK.NET.AU30_Sat, 4 Feb 2006 11:16:39 +1100375_US-ASCII OA07685? "Documentation for this new support is delivered in SISPSAMP member ISRUCGDE. It can be downloaded and viewed with Adobe Acrobat(PDF)."
James Campbell
On 3 Feb 2006 at 12:03, Avram Friedman wrote:
> 4. V8 DBRMs seem to be encoded in UNICODE which limits our ability to use the ever popular > IBM utility "IEH-I-BALL" [...]
14041 111 49_DB2 ZOS Rebinding the world due to system service14_Avram Friedman21_avramf@IBMSYSPROG.COM30_Fri, 3 Feb 2006 17:00:28 -0800350_iso-8859-1 Re-titled from:
Re: [DB2-L] DB2 ZOS V8 CPU Usage verses V7
Retitled the thread as to not dilute the original thread due to topic changes.
Also the topic of to rebind or not was covered in a multi week discussion on this forum less than a year ago and I am not sure how worthwhile going through all of this again. [...]
14153 38 30_Re: Accessing uncommitted rows15_Marcel Harleman25_marcel.harleman@HCCNET.NL30_Sat, 4 Feb 2006 16:25:03 +0100373_us-ascii On Wed, 1 Feb 2006 03:15:44 -0600, you wrote:
>In SQL Reference Guide, it is stated clearly that inserted but not >committed rows can not be accessed by other users.
When you run your test, do you use two TSO sessions, so do you issue the insert and the query from two different address spaces, or do you issue them from the same address space? [...]
14192 23 30_Re: Accessing uncommitted rows15_Marcel Harleman25_marcel.harleman@HCCNET.NL30_Sat, 4 Feb 2006 16:34:15 +0100356_us-ascii On Wed, 1 Feb 2006 03:15:44 -0600, you wrote:
>In SQL Reference Guide, it is stated clearly that inserted but not >committed rows can not be accessed by other users.
Sorry, I hit the enter key too soon. I was just trying to check a crazy thought I had and I hadn't finished thinking about it. Ignore my previous posting please. [...]
14216 38 42_Re: DB2 for LUW Static SQL Monitoring tool12_Per Stricker22_per.stricker@ITGAIN.DE30_Sun, 5 Feb 2006 11:42:49 +0100746_ISO-8859-1 Hi Paul,
Try Speedgain for LUW. The tool will provide you with the information you need and more !
www.speedgain.de/en/speedgain_trial.html
Per
>>If anyone has experience with a tool that gathers information about static >>SQL on the DB2 for Linux, Unix, Windows (LUW) platform, I would be >>interested in hearing about it. By gathering information, I mean number of >>times the static SQL statement has executed, execution times, CPU cost for >>the statement, etc. I know event monitors can capture data, but they don't >>aggregate the results. >> >>------------------------------------------------------------------------------ >- >>-- >>Welcome to the IDUG DB2-L list. To unsubscribe, go to [...]
14255 421 30_Re: Accessing uncommitted rows14_Michael Kaplan25_micaelkp@NETVISION.NET.IL30_Mon, 6 Feb 2006 00:05:25 +0200339_windows-1250 Hi Mike,
I'm pleased to confirm: you are absolutely right. This is how rows are resided in page dataset ( thanks to DSN1PRNT) :
page 2: rows 261 - 253, 201 - 252, 262 - 300 pages 3, 4, 5, 6, 7 - holes page 8 : rows 193 - 189, 146 - 188, 194 - 200 pages 9, A - holes page B : rows 22 - 11, 1 - 10, 32 - 145. [...]
14677 302 34_Re: TEMPLATE UTILITY &HOUR. option13_Kornelis, Abe26_Abe.Kornelis@GETRONICS.COM30_Mon, 6 Feb 2006 08:27:01 +0100464_iso-8859-1 We had the same problem. Very confusing DSName result from this 'feature'. Especially when it is not just the time, but also the date that is 'wrong'. I have opened a PMR with IBM to add local date/time variables, which IBM has accepted with number MR0611045824.
The requirement was subsequently rejected. I'm not sure why. My guess is it might have to do with SYSPLEX/DataSharing where different members may live in different time zones... [...]
14980 100 38_High wait times from DSNAPRH/DSNXEEZ ?25_Sorensen Henrik (KBIC 21)33_henrik.sorensen@CREDIT-SUISSE.COM30_Mon, 6 Feb 2006 10:13:35 +0100400_us-ascii Esteemed List,
On DB2 V8 CM, z/OS.
We are seeing some excessive wait times originating from the modules DSNAPRH DSNXEEZ
Can anybody shed some light on what could cause DSNXEEZ / DSNAPRH to wait for a very long time ?
When looking at a strobe report, the wait time is accounted to IEAVEPS1.
Anything in particular we should trace for ? [...]
15081 86 31_Z/Os DB2 V7 : DSSIZE alteration21_Sriramulu, Vijayababu28_vijayababu.sriramulu@EDS.COM30_Mon, 6 Feb 2006 18:52:15 +0530413_us-ascii Dear list users,
Can you please suggest me the different possible ways of altering the DSSIZE for a production tablespace.
I checked in the manuals and found that ALTER TABLESPACE does not support altering this parameter.
Please let me know your thoughts/experiences on doing this in a efficient and effective manner. Outage if required, will have to be as minimum as possible. [...]
15168 103 30_Z/Os DB2 V7 : CLOB column size21_Sriramulu, Vijayababu28_vijayababu.sriramulu@EDS.COM30_Mon, 6 Feb 2006 18:56:32 +0530334_us-ascii Dear List users,
We are planning to use CLOB's in one of our new request. We initially planned to have the size as 2 MB. But the application team has come up with a possibility of reducing this size to 1 MB if and only if it will bring in a BIG ADVANTAGE from the performance point of view. Now my questions are: [...]
15272 125 59_Re: R: [DB2-L] DB2 for z/OS to Oracle for Unix connectivity17_Gugel, Rich - DOA24_Rich.Gugel@WISCONSIN.GOV30_Mon, 6 Feb 2006 07:34:44 -0600747_us-ascii @font-face { font-family: Tahoma; } @font-face { font-family: Verdana; } @page Section1 {size: 595.3pt 841.9pt; margin: 70.85pt 2.0cm 2.0cm 2.0cm; } P.MsoNormal { FONT-SIZE: 12pt; MARGIN: 0cm 0cm 0pt; FONT-FAMILY: "Times New Roman" } LI.MsoNormal { FONT-SIZE: 12pt; MARGIN: 0cm 0cm 0pt; FONT-FAMILY: "Times New Roman" } DIV.MsoNormal { FONT-SIZE: 12pt; MARGIN: 0cm 0cm 0pt; FONT-FAMILY: "Times New Roman" } A:link { COLOR: blue; TEXT-DECORATION: underline } SPAN.MsoHyperlink { COLOR: blue; TEXT-DECORATION: underline } A:visited { COLOR: blue; TEXT-DECORATION: underline } SPAN.MsoHyperlinkFollowed { COLOR: blue; TEXT-DECORATION: underline } P { FONT-SIZE: 12pt; MARGIN-LEFT: 0cm; MARGIN-RIGHT: 0cm; FONT-FAMILY: "Times New Roman" [...]
15398 198 35_Re: Z/Os DB2 V7 : DSSIZE alteration13_Michael Ebert18_mebert@AMADEUS.COM30_Mon, 6 Feb 2006 14:50:06 +0100576_US-ASCII The best strategy depends heavily on how the TS is used. I've often used this general-purpose procedure:
1. Stop TS, restart it in RO. 2. Unload TS on a partition level, several jobs in // 3. Drop TS 4. Create TS with new DSSIZE & partitioning index, start TS & IX in UT mode; RUNSTATS on empty TS/IX for initializing aggregate stats (this prevents warning messages in the following RUNSTATS). 5. Load TS on a partition level, several jobs in //; RUNSTATS for TS & IX partition should follow LOAD job (I don't recommend inline stats). 6. Create NPIs with [...]
15597 17 35_Re: Z/Os DB2 V7 : DSSIZE alteration33_Walter Jani=?ISO-8859-1?Q?=DFen?=27_walter.janissen@VICTORIA.DE30_Mon, 6 Feb 2006 07:46:57 -0600395_- Hi
The official way is: drop and recreate. I don't know if there in any trick, doing it without a drop. To minimize the outage, you can create a second tablespace with a new table, stop your original tablespace for update, crossload your data into the new table, rename your old table, rename the new table and start the new tablespace for update. That's a very brief description. [...]
15615 19 23_Governor and Static SQL11_Matt Steele21_matt.steele@MEDRY.COM30_Mon, 6 Feb 2006 07:55:51 -0600299_- I realize the db2 governor cannot be used for static sql but, what can be used to ensure a static SQL thread does not take too long. We use Stored Procedures on Z/OS and ran into an issue where a thread was running for over 12 hours. Our area is charged back and we took a huge hit for this. [...]
15635 46 34_Re: DB2 ZOS V8 CPU Usage verses V713_Keemle Joan T25_KeemleJoanT@JOHNDEERE.COM30_Mon, 6 Feb 2006 08:06:40 -0600463_us-ascii In z/OS 1.7, the ISPF Unicode browse/find documentation moves into z/OS V1R7.0 ISPF User's Guide Vol II, section 1.3.3.2 Browse Primary Commands -- the sections on DISPLAY and FIND. (SC34-4823)
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of James Campbell Sent: Friday, February 03, 2006 6:17 PM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] DB2 ZOS V8 CPU Usage verses V7 [...]
15682 143 35_Re: Z/Os DB2 V7 : DSSIZE alteration14_lfa@woh.rr.com14_lfa@WOH.RR.COM30_Mon, 6 Feb 2006 09:12:03 -0500422_iso-8859-1 I am currently performing this same process on a 210 partition tablespace. I am using the same method described in Michael's response.
I too, have a tablespace with rolling partitons (weekly), so I am able to unload all of the inactive partitions ahead of time, and only unload the active partiton during our outage window. This allows for minimal downtime for the application. Works very smoothly. [...]
15826 31 25_DB2 for Z/OS V8 Migration0_17_sjvagnier@AEP.COM30_Mon, 6 Feb 2006 09:16:17 -0500360_US-ASCII Has anyone successfully completed a migration to DB2 Z/OS V8 (Compatibility Mode) in a data sharing environment without taking a complete outage? If so, what kinds of problems have you encountered when migrating one member at a time? I am trying to avoid a complete group outage on a 24X7 environment if at all possible during a migration to V8. [...]
15858 134 27_Re: Governor and Static SQL13_Richard Fazio21_rfazio@TRANSUNION.COM30_Mon, 6 Feb 2006 08:15:50 -0600553_US-ASCII Check with your DB2 monitor. Most monitors have event based triggers to notify you of something based upon a threshold. Some even will allow you to kill a thread automagically.
Be warned, however about specifying the event response properly.
Example: DEFINITION: If CPU_TIME > 100 Seconds....Then, send email.
Ok, your task is running and it has a CPU_TIME of 101 seconds...so, the monitor sends an email. Your task is still running and it has a CPU_TIME of 102 seconds...so, another email is sent....get the picture. [...]
15993 32 32_DB2 V7 z/os v1.4 data encryption12_Ulrich Boche19_ulrich.boche@SVA.DE30_Mon, 6 Feb 2006 17:12:49 +0100424_US-ASCII On Tuesday, 31.01.2006 at 15:48 EST, Jeff Frazier wrote: > Hello, > > We currently have an application running on DB2 UDB AIX that encrypts data. > There is a requirement to move this to DB2 V7 z/os. We have plans to go to v8 > but not near future. Are there products that anyone is using that will > encrypt/decrypt the data within db2 v7 that we can currently use? > TIA, > Jeff [...]
16026 19 49_Executing an Oracle stored procedure from DB2 zOS15_Cristian Molaro18_cristian@MOLARO.BE30_Mon, 6 Feb 2006 12:11:53 -0600741_- Hello list,
In a configuration consisting of a DB2 II connecting a DB2 for zOS V7.1 and a Oracle DB... Is it possible to execute an Oracle stored procedure from the DB2 for zOS?
Thanks a lot in advance for your help!
Cristian Molaro MConsulting.be
--------------------------------------------------------------------------------- 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
16046 91 29_Re: DB2 for Z/OS V8 Migration14_Avram Friedman21_avramf@IBMSYSPROG.COM30_Mon, 6 Feb 2006 10:44:39 -0800560_iso-8859-1 Although the IBM documentation says a migration of a data sharing environment to a newer version of DB2 is possible without stopping all members but 1 for CATMAINT. my personal experience with a migration to V7 suggests it is not in production systems that are actually used.
The problem is the exposure to resource unavailable due to locks needed and or held for CATMAINT. Not surprising the catalog processing for CATMAINT is massive and there is no practical way to get through it if you have other members in the SYSPLEX doing work. [...]
16138 139 29_Re: DB2 for Z/OS V8 Migration0_17_sjvagnier@AEP.COM30_Mon, 6 Feb 2006 14:29:38 -0500567_US-ASCII I need to make this migration as error free as possible, so I will take the outage.
Thanks for the response...
**************************************** 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
Avram Friedman To Sent by: DB2 Data DB2-L@WWW.IDUGDB2-L.ORG Base Discussion cc List Re: [DB2-L] DB2 for Z/OS V8 Migration [...]
16278 40 27_Re: Governor and Static SQL12_Myron Miller22_myronwmiller@YAHOO.COM30_Mon, 6 Feb 2006 12:12:10 -0800535_iso-8859-1 If your SP is a SQL sp, then you can use ASUTIME and it works just like the governor in that once the Statement hits that max time, the SP will be terminated.
Myron
--- Matt Steele wrote:
> I realize the db2 governor cannot be used for static sql but, what can be > used to ensure a static SQL thread does not take too long. We use Stored > Procedures on Z/OS and ran into an issue where a thread was running for over > 12 hours. Our area is charged back and we took a huge [...]
16319 55 21_DB2 V8 z/OS question.9_Joe Burns25_joseph.burns@HIGHMARK.COM30_Mon, 6 Feb 2006 14:17:57 -0600403_- I'm trying to use a new performance feature of DB2v8, but it doesn't seem to be working.
According to the "DB2v8 Everything you wanted to know redbook" (section 10.15).
A non-correllated subquery does not need to be completely materialized once you are in V8. (we are in V8 NFM). But my testing seems to indicate that it is still materializing. The query would look something like: [...]
16375 71 27_Re: Governor and Static SQL11_Suresh Sane21_data_arch@HOTMAIL.COM30_Mon, 6 Feb 2006 16:21:22 -0600650_- ASUTIME applies to all stored procedures, not just SQL SP's (including COBOL, C, Java etc).
Thanks, Suresh
>From: Myron Miller >Reply-To: DB2 Database Discussion list at IDUG >To: DB2-L@WWW.IDUGDB2-L.ORG >Subject: Re: [DB2-L] Governor and Static SQL >Date: Mon, 6 Feb 2006 12:12:10 -0800 > >If your SP is a SQL sp, then you can use ASUTIME and it works just like the >governor in that once the Statement hits that max time, the SP will be >terminated. > >Myron > >--- Matt Steele wrote: > > > I realize the db2 governor cannot be used for static [...]
16447 91 25_Re: DB2 V8 z/OS question.11_Suresh Sane21_data_arch@HOTMAIL.COM30_Mon, 6 Feb 2006 16:30:11 -0600694_- Joe,
Just curious if the EXISTS predicate (stand-alone) avoids materialization.
Can you try...
SELECT * FROM ( SELECT * FROM DB2VER8.TABX WHERE EXISTS (SELECT 1 FROM DB2VER8.TABY WHERE COL2 < 500000) ) AS NTE WHERE COL1 < '0010'
Thanks, Suresh
>From: Joe Burns >Reply-To: DB2 Database Discussion list at IDUG >To: DB2-L@WWW.IDUGDB2-L.ORG >Subject: [DB2-L] DB2 V8 z/OS question. >Date: Mon, 6 Feb 2006 14:17:57 -0600 > >I'm trying to use a new performance feature of DB2v8, but it doesn't seem >to be working. > >According to the "DB2v8 Everything you wanted to know redbook" (section >10.15). > >A [...]
16539 42 61_FW: Information Integrator/Datajoiner - DB2 aliases to Oracle22_Ramesh Balasubramanyan15_RAMBAL@SAIF.COM30_Mon, 6 Feb 2006 16:14:38 -0800625_US-ASCII > ______________________________________________ > From: Ramesh Balasubramanyan > Sent: Monday, February 06, 2006 4:14 PM > To: DB2 Data Base Discussion List > Subject: Information Integrator/Datajoiner - DB2 aliases to > Oracle > > Hello, Has anybody used DB2 aliases to Nicknames defined in > Information Integrator/Datajoiner in CICS/PowerBuilder ? > We are trying to grasp, the extent of modification that will be > involved in the existing applications that are currently talking to > DB2. > > There is a 'VISION' (!) to move the data to Oracle and let datajoiner > present a DB2 view of the Oracle table. [...]
16582 70 25_Re: DB2 V8 z/OS question.12_Isaac Yassin20_yassini@BEZEQINT.NET30_Tue, 7 Feb 2006 07:10:19 +0200399_US-ASCII Hi,
Any Index on COL2 ?
Isaac Yassin
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Joe Burns Sent: Monday, February 06, 2006 10:18 PM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: [DB2-L] DB2 V8 z/OS question.
I'm trying to use a new performance feature of DB2v8, but it doesn't seem to be working. [...]
16653 77 17_Impact of Set SQL28_Chandrasekaran, Balachandran35_balachandran.chandrasekaran@FMR.COM30_Tue, 7 Feb 2006 04:38:54 -0500340_us-ascii Hello List,
Is there any claim or lock overhead in any of the SET SQL to set current date , or timestamp for example ??? We are doing a COMMIT even after SELECT statements to release all CLAIMS over the table pages. I am just getting surprised when there is a commit after the host variable is set a current timestamp. [...]
16731 172 29_Re: DB2 for Z/OS V8 Migration11_Mike Turner29_michael_turner@COMPUSERVE.COM30_Tue, 7 Feb 2006 09:40:52 -0000626_US-ASCII Steve
One other thing. To activate the new level 2 locking protocol you will need to stop all members MODE(QUIESCE) at the same time after you are in NFM. There are significant performance advantages to using the new locking protocol.
Regards Mike Turner Email: michael_turner@compuserve.com Tel: +44 (0)1565 873702 web: www.michael-turner.ltd.uk
-----Original Message----- From: owner-db2-l@WWW.IDUGDB2-L.ORG [mailto:owner-db2-l@WWW.IDUGDB2-L.ORG]On Behalf Of sjvagnier@AEP.COM Sent: Monday, 06 February 2006 19:30 To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] DB2 for Z/OS V8 Migration [...]
16904 113 34_Antwort: [DB2-L] Impact of Set SQL11_Roy Boxwell16_R.Boxwell@SEG.DE30_Tue, 7 Feb 2006 12:14:18 +0100315_us-ascii my guess is that this is old code that used a dummy select against SYSDUMMY1 and has been rewritten to use the SET functions available in V5 and no-one bothered to remove the COMMIT. As far as I am aware the SET does not set any locks or claims anywhere as you do not have table in the SET statement. [...]
17018 305 51_Antwort: Re: [DB2-L] DB2 ZOS V8 CPU Usage verses V711_Roy Boxwell16_R.Boxwell@SEG.DE30_Tue, 7 Feb 2006 12:17:13 +0100317_iso-8859-1 A bit of a self referencing Vendor plug here as we sell a very nice V7 to V8 tool that allows you to do a EarlyPrecheck, among many other things. Works great, easy to install and everyone loves it (Of course I have to say that!!!) Check out our home page for more details about our ImpactExpert tool. [...]
17324 39 35_Re: Z/Os DB2 V7 : DSSIZE alteration19_Jardine, Lawrence J19_JardineLJ@AETNA.COM30_Tue, 7 Feb 2006 06:49:56 -0500307_iso-8859-1 This is what I do. I am a big fan of the cross-load faci Yes!
This is what I do. I am a big fan of the cross-load facility. With this method, the table is in RO for the duration of the cross load (several hours). The only full outage is during the rename and rebinds (a few minutes). [...]
17364 136 35_Re: Z/Os DB2 V7 : DSSIZE alteration13_Michael Ebert18_mebert@AMADEUS.COM30_Tue, 7 Feb 2006 12:59:20 +0100403_US-ASCII The disadvantage of using a "rename Table" approach is that you have to change to another TS (we're talking partitioned tables/TSs here, remember) and rename the indexes as well, which may require a lot of batch process changes downstream (e.g. for COPY, RUNSTATS after a LOAD).
Dr. Michael Ebert DB2 & Oracle Database Administrator aMaDEUS Data Processing Erding / Munich, Germany [...]
17501 84 35_Re: Z/Os DB2 V7 : DSSIZE alteration21_Sriramulu, Vijayababu28_vijayababu.sriramulu@EDS.COM30_Tue, 7 Feb 2006 17:33:53 +0530599_iso-8859-1 Hi Larry,
After the RENAME's, will AUTO REBIND not take effect. Since there are no changes/alterations to the table structure, the auto rebinds are expected to go fine. Is that right ? Am I missing something here ?
I am planning to go with this approach.
Please advice.
Thanks & Regards Vijay
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Jardine, Lawrence J Sent: Tuesday, February 07, 2006 5:20 PM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] Z/Os DB2 V7 : DSSIZE alteration [...]
17586 229 29_Re: DB2 for Z/OS V8 Migration0_17_sjvagnier@AEP.COM30_Tue, 7 Feb 2006 08:05:46 -0500535_US-ASCII Thanks Mike for the additional V8 Migration information.
Regards,
**************************************** 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
Mike Turner To Sent by: DB2 Data DB2-L@WWW.IDUGDB2-L.ORG Base Discussion cc List Re: [DB2-L] DB2 for Z/OS V8 Migration [...]
17816 110 35_Re: Z/Os DB2 V7 : DSSIZE alteration19_Jardine, Lawrence J19_JardineLJ@AETNA.COM30_Tue, 7 Feb 2006 08:10:47 -0500386_iso-8859-1 Technically, I don't do renames. In my shop, all access to tables is through views, so instead of renaming the table, I drop the view and recreate it (referring to the new table). AutoBind would be in affect for this, but I like to do them manually so that I can verify access paths and avoid problems. Finally, a good reason to have views that are SELECT * from table. [...]
17927 42 18_Load utility query21_Deepak Gujaba Gaikwad24_deepakggaikwad@YAHOO.COM30_Tue, 7 Feb 2006 05:11:32 -0800376_iso-8859-1 Hello All,
We are facing a recovery problem with one of huge table (in test). A job was run to insert 20+ millions records in 2 tables reading from a file. The job ran for several hours and abended with errors. DB2 could not handle the backout and terminated abnormally. DB2 restarts attempt failed due to backout processing. DB2 was then cold started. [...]
17970 134 22_Re: Load utility query13_Richard Fazio21_rfazio@TRANSUNION.COM30_Tue, 7 Feb 2006 07:21:36 -0600329_US-ASCII First thing I would do is check the INDEX(es). Potentially, you will have to rebuilt.
Then, change up to LOAD SHRLEVEL CHANGE. You'll LOG the rows, but it will kick out the dups as they happen, rather than during a discard phase. It's very fast, quite efficient and I believe handle your concern about dups. [...]
18105 180 22_Re: Load utility query13_Michael Ebert18_mebert@AMADEUS.COM30_Tue, 7 Feb 2006 14:26:53 +0100423_US-ASCII Well, it's the DBAs responsibility to make sure a table (TS) is recoverable, not the programmers. At least that's my opinion.
How was the data loaded - with INSERTs or with LOAD REPLACE or LOAD RESUME? What were the job abend errors? What do you mean with "DB2 could not handle the backout" - error messages, return codes, reason codes? Ditto for "restart attempts failed due to backout processing"... [...]
18286 169 66_Re: FW: Information Integrator/Datajoiner - DB2 aliases t o Oracle14_Neff Stephen R22_Stephen.R.Neff@IRS.GOV30_Tue, 7 Feb 2006 08:29:29 -0500313_- DataJoiner is no longer supported. Information Integrator is. You need II and an ORACLE wrapper. The DB2 communication tables on the mainframe point to the II server which has nicknames defined that point to the ORACLE tables and database. See DB2 for z/OS to Oracle for Unix connectivity posted messages. [...]
18456 26 53_Re: Db2 V8 stored procedure calling cobol sub program12_Roger Miller19_millerrl@US.IBM.COM30_Tue, 7 Feb 2006 07:22:41 -0600350_- The stored procedure will need to detect, handle or pass on the error. Adding a parameter to the list of the stored procedure seems to be the usual technique, unless the stored procedure handles the problem found in the called routine.
Roger Miller
On Wed, 1 Feb 2006 10:51:58 -0600, Sherry Simmerl wrote: [...]
18483 41 34_Re: Z/Os DB2 V7 : CLOB column size12_Roger Miller19_millerrl@US.IBM.COM30_Tue, 7 Feb 2006 07:29:16 -0600376_- Is that the maximum size or the average size of your CLOB? There is a redbook on LOBs, SG24-6571, with a chapter on performance. I think you should start there. There are some substantial improvements for LOBs in V8 and in the release after that.
Roger Miller
On Mon, 6 Feb 2006 18:56:32 +0530, Sriramulu, Vijayababu wrote: [...]
18525 299 55_Re: Antwort: Re: [DB2-L] DB2 ZOS V8 CPU Usage verses V712_Gerald Hodge26_ghodge@HLSTECHNOLOGIES.COM30_Mon, 6 Feb 2006 19:49:32 -0800461_us-ascii Since we are going this way, please see the HLS Technologies, Inc. ad on page 5 of the current IDUG Journal. HLS offers the customer a rental program that takes the funds from your operational budget and not the capital budget. We have a set of tools for the set of problems that the V8 conversion will present. We also offer a free two day on site seminar on finding the Access Path changes before they hurt you. (This has limited availability.) [...]
18825 105 35_Re: Z/Os DB2 V7 : DSSIZE alteration33_Walter Jani=?ISO-8859-1?Q?=DFen?=27_walter.janissen@VICTORIA.DE30_Tue, 7 Feb 2006 07:43:29 -0600436_ISO-8859-1 On Tue, 7 Feb 2006 17:33:53 +0530, Sriramulu, Vijayababu wrote:
As soon as you rename the table, all packages dependent on that object will be flagged invalid. If such a package will be executed just in between the rename process, it gets inoperative and a manual rebind is necessary. Also you have better control over rebinds, access path an so on, if you do the rebinds manually. [...]
18931 99 34_Re: Z/Os DB2 V7 : CLOB column size13_Robert Tilkes22_TILKESR@NATIONWIDE.COM30_Tue, 7 Feb 2006 07:51:16 -0600580_US-ASCII Vijay,
I would suggest looking at the IBM Redbook regarding large objects. The redbook was a huge help. I have post several questions about LOBs in this forum and receive little results. I think performance depends on how the application is designed. Does the application need to return the LOB everytime a row is requested from the table. We currently have a 3M LOB in production. The LOBs we have are insert, select, and delete. We do not do any updates. Our LOBs are inserted through the WEB to the Mainframe. We have had no performance issues of late. It [...]
19031 86 22_Re: Load utility query21_Deepak Gujaba Gaikwad24_deepakggaikwad@YAHOO.COM30_Tue, 7 Feb 2006 06:12:22 -0800493_iso-8859-1 Thanks Ebert,
The data was loaded using INSERT via COBOL program. After 10+ millions uncommited inserts, the tablespace reached max extents causing job to fail. The job was run on weekend in the test db2. Throughout the execution DB2 continuously reached log full conditions. During the rollbacks, DB2 hit S80A.
So we are thinking if perform LOAD RESUME and then rebuild index, we should be able to recover from this problem. The data file is about 10GB. [...]
19118 135 35_Re: Z/Os DB2 V7 : DSSIZE alteration12_Isaac Yassin20_yassini@BEZEQINT.NET30_Tue, 7 Feb 2006 17:40:56 +0200649_iso-8859-1 Use ALIAS
Isaac Yassin
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Jardine, Lawrence J Sent: Tuesday, February 07, 2006 3:11 PM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] Z/Os DB2 V7 : DSSIZE alteration
Technically, I don't do renames. In my shop, all access to tables is through views, so instead of renaming the table, I drop the view and recreate it (referring to the new table). AutoBind would be in affect for this, but I like to do them manually so that I can verify access paths and avoid problems. Finally, a good reason to have [...]
19254 16 25_Re: DB2 V8 z/OS question.15_Patrick Bossman18_bossman@US.IBM.COM30_Tue, 7 Feb 2006 12:49:03 -0600619_- Hello, Joe's access path shows matching index access, so the index has to exist.
Joe - I suggest opening a pmr on this.
Regards, Pat
--------------------------------------------------------------------------------- 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
19271 22 17_Re: NPGTHRSH = -115_Patrick Bossman18_bossman@US.IBM.COM30_Tue, 7 Feb 2006 12:59:01 -0600373_- Hi Mark, I don't know of anyone that has set this parameter to -1. I also would not suggest that anyone do it.
If you've got "not small" tables that are being non-matching index scanned, or table space scanned - what's the deal with the candidate indexes? Are they not selective, poorly clustered, predicates S2, no index that satisfies available predicates? [...]
19294 37 25_Re: DB2 V8 z/OS question.12_Isaac Yassin20_yassini@BEZEQINT.NET30_Tue, 7 Feb 2006 21:00:59 +0200408_US-ASCII Hi, As ACCESSTYPE is D this looks strange. That's why I asked on the index on COL2 as that info. Is not in the original mail.
Isaac Yassin
-----Original Message----- From: DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] On Behalf Of Patrick Bossman Sent: Tuesday, February 07, 2006 8:49 PM To: DB2-L@WWW.IDUGDB2-L.ORG Subject: Re: [DB2-L] DB2 V8 z/OS question. [...]
19332 120 25_Re: DB2 V8 z/OS question.9_Joe Burns25_joseph.burns@HIGHMARK.COM30_Tue, 7 Feb 2006 16:12:45 -0500558_us-ascii Hello,
Apologize for the late reply, I ran into something else unrelated to this.
There is an index on COL2 which is on TABY. This is the index that Pat referred to (XPRC120C is the index name in the explain).
The PREFETCH of "D" apparently means that DB2 anticipates using DYNAMIC prefetch (first I've ever seen that). But not what I was hoping for in this case since it really only needs to find a single instance on TABY. As soon as it finds that first instance, it can stop. The EXISTS predicate is true at that point. [...]
19453 161 62_Heart of America DB2 User Group (Kansas City) and Bonnie Baker8_Dan Tuck18_dan.tuck@GMAIL.COM30_Tue, 7 Feb 2006 16:01:27 -0600584_WINDOWS-1252 Heart of America DB2 User Group Meeting and Training Meeting: March 13, 2006 Training: March 14 and 15, 2006 Holiday Inn Overland Park 8787 Reeder Road Overland Park, KS 66204 Check in: 8:00 am, Meeting: 8:30 am BONNIE BAKER -- TRAINING NOTIFICATION FOR MARCH 2006 Heart of America DB2 User Group Members: We have arranged for Bonnie to present at our March 13th meeting and to provide two days of training for those of you who are interested. The topics below have emphasis on DB2 Version 8. Bonnie's presentation on March 13, 2006 is part of the quarterly HOA DB2 [...]
19615 99 65_Re: FW: Information Integrator/Datajoiner - DB2 aliases to Oracle12_Kirk Hampton28_khampto1@CAPGEMINIENERGY.COM30_Tue, 7 Feb 2006 16:47:43 -0600414_us-ascii Hello Ramesh,
As someone already commented, Datajoiner is out-of-date. We are
using its follow-on, Relational Connect, which is also out-of-date. The
latest iteration is Information Integrator. But to answer your question,
YES, you can point a DB2 Alias to an Information Integrator Nickname
which in turn points to an Oracle table. We did it without ever changing [...]
19715 79 29_Re: DB2 for Z/OS V8 Migration15_Philip Sevetson22_db2.dba.guru@GMAIL.COM30_Tue, 7 Feb 2006 17:49:26 -0500464_ISO-8859-1 Mike --
Do you know; would STOPping the whole subsystem work equally well to engage the new protocol?
--Phil S.
On 2/7/06, Mike Turner wrote: > > Steve > > One other thing. To activate the new level 2 locking protocol you will > need > to stop all members MODE(QUIESCE) at the same time after you are in NFM. > There are significant performance advantages to using the new locking > protocol. > > m >
19795 54 31_SYSDBASE error while doing bind5_tsp l19_tspdba001@GMAIL.COM30_Tue, 7 Feb 2006 21:52:05 -0500779_ISO-8859-1 Hi experts,
Could anyone help me resolving this. I am getting the following error while doing the bind.
DSNT500I -DTI2 DSNXOTL RESOURCE UNAVAILABLE REASON 00C90088 TYPE 00000302 NAME DSNDB06 .SYSDBASE.X'012B3E' DSNT233I -DTI2 UNSUCCESSFUL BIND FOR
Thanks in advance.
Tp
--------------------------------------------------------------------------------- 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
19850 299 35_Re: SYSDBASE error while doing bind16_Rasmussen, Steen22_Steen.Rasmussen@CA.COM30_Tue, 7 Feb 2006 22:08:12 -0500343_us-ascii 00C90088 tells you that you are involved in a deadlock. Your BIND must have locks on some page in the catalog while trying to acquire another lock which is on a page already being hold by another application. Are you running simultaneous BIND, DDL or so ?
Kind regards Steen Rasmussen CA Senior Consultant [...]
20150 207 35_Re: SYSDBASE error while doing bind5_tsp l19_tspdba001@GMAIL.COM30_Tue, 7 Feb 2006 22:26:58 -0500616_ISO-8859-1 Yes, Thats correct i was running simultaneous BIND. Thanks for your response.
On 2/7/06, Rasmussen, Steen wrote: > > 00C90088 tells you that you are involved in a deadlock. Your BIND must > have locks on some page in the catalog while trying to acquire another lock > which is on a page already being hold by another application. Are you > running simultaneous BIND, DDL or so ? > > > > Kind regards > Steen Rasmussen > CA > Senior Consultant > ------------------------------ > > *From:* DB2 Data Base Discussion List [mailto:DB2-L@IDUGDB2-L.ORG] *On > Behalf Of *tsp [...]