1 WWW.IDUGDB2-L.ORG /home/listserv/home/db2-l March 2000, week 1
2 30 45_DB2 Personal Edition & Java Stored Procedures9_SrinivasG17_SRINIVASG@INF.COM30_Wed, 1 Mar 2000 13:18:01 +0530507_iso-8859-1 Hi,
I have DB2 For Win 95/98 & Win NT V6.1 Personal edition as well as Enterprise Edition. When I install this product on Win95/98 , for the Client Configuration Assistant, after I choose TCPIP as Protocol , it does not give the option to connect to OS/390 , AS/400 or VM etc. When I install the same product on Win NT, I get these options. Which means I get to connect to the Mainframe database directly when I use the Product on Win NT and not on Win 95/98 . Why is this happening? [...]
33 16 7_CLOTECB11_Roland Chua32_rolandchua@SINGAPOREEXCHANGE.COM30_Wed, 1 Mar 2000 16:11:49 +0800390_us-ascii Hi, Can anyone tell me what is db2 resource - CLOTECB and will it affect the CICS or DB2 performance. Thanks in advance.
Roland Chua
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
50 41 30_Archiving Data - Datawarehouse0_23_J.P.Slot@RF.RABOBANK.NL30_Wed, 1 Mar 2000 09:47:50 +0100643_us-ascii Hello all,
This is a re-post of a previous question.
I am wondering about the ARCHIVE for DB2 product as mentioned earlier (Princeton Softtech's Archive for DB2). Is it possible to ARCHIVE DB2 data (automated) that isn't used(touched) for a while (a pre-selected period) and RESTORE's it if it's needed (touched). Such a feature would be perfect in datawarehouse surroundings. A collegue of mine suggested to use OAM with an in-house build cobol-program to ARCHIVE and RESTORE data in such a manner. Does the above mentioned product provide this feature? Maybe you can provide me with information on other products [...]
92 159 19_Re: Machine Upgrade0_18_mebert@AMADEUS.NET30_Wed, 1 Mar 2000 09:50:05 +0100629_us-ascii Not true. More parallel tasks (jobs, processes, threads...) means more administrative overhead for the operating system, that's less CPU for your problem program. Even if you assume that you need NO overhead (that also means the tasks are independent; otherwise you'd get communication overhead), you lose: because if you at some time have less tasks than processors, part of your multi-CPU engine sits idle, and you irretrievably lose processing time (that's queuing theory for you). So from a theoretical point of view, ALWAYS go for the fewer, faster CPUs engine (you might also want to look up "Amdahl's Law"). [...]
252 49 21_Re: Why Not DB2??????14_Massimo Scarpa16_mscarpa@CESVE.IT30_Wed, 1 Mar 2000 10:16:31 +0100408_us-ascii In previous shop I worked we migrated from 2 hosts to 103 servers, with two UNIX central
machine (host-like) for data replication. A nice architecture to employ 300+ persons....
I'm sure that in DB2 you can recover ALL your data in less than two moths.......(a real episode in
my life). I'm sure that you'll never lost a part of your data (data may be non accessible for a [...]
302 43 28_Re: DB2 V4 or V5 requirement41_=?iso-8859-1?Q?Cl=E1udio_Meluzzi_Mendes?=18_P000002@BEG.COM.BR30_Wed, 1 Mar 2000 08:31:51 -0300473_iso-8859-1 Dear Anthony,
We have DB2 V4.1 for OS/390 with ODBC access enabled through DB2 Connect. An IIS (Microsoft Internet Information Server) server installed on Windows NT enables ASP web pages to gather information from data stored in DB2. The ODBC connection is established only between the web server and the database. Therefore workstations do not need to have any ODBC driver installed on it. Should you want more information feel free to contact us. [...]
346 97 19_Re: Machine Upgrade14_Joel Goldstein29_joel_goldstein@COMPUSERVE.COM30_Wed, 1 Mar 2000 08:15:48 -0500560_ISO-8859-1 Yes it is true, and there have been many many real world examples over the years to verify this. Again, it all depends on your workload. I was also in a large installation early last year, where part of JES was in a loop, consuming > 90% of one engine - for an entire week. People don't always understand they have a problem - even when it smacks them in the face. Amdahl's law - all processors wait at the same speed. Certainly there is more overhead with each additional processor. This has been know, evaluated, and benchmarked for decades. [...]
444 15 36_DB2 Connect password change from ADO20_Laurence M Crouch Jr24_lmcrouch@DUKE-ENERGY.COM30_Wed, 1 Mar 2000 08:43:36 -0500509_us-ascii DB2 Connect 5.2 has the capability to change your mainframe password. Page 6 of the DB2 Connect User's Guide indicates that a password can be changed from within an ODBC login dialog. Does anyone know if (and how) a password can be changed using ADO? Thanks!
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
460 54 27_Re: DB2 connect w/ MS Query12_Myron Miller22_myronwmiller@YAHOO.COM30_Wed, 1 Mar 2000 05:53:13 -0800557_us-ascii I've been using various versions of MS Query with Windows 95, 98, NT with no problems for several years. Having the parm options of DB2 Connect setup correctly has been the key for me.
What abend are you getting?
--- John Reid wrote: > Hi list, > > Has any one had any problems with using MS QUERY > retrieving DB2 OS/390 data > using DB2 Connect? > > We are trying to create MS EXCEL spreadsheets using > some data from DB2 tables > and continually get a fatal exception in app > msqry32.dbg( MS EXCEL is using [...]
515 37 15_Re: DB2 Experts12_Kalman, Jeff20_JKalman@UNCH.UNC.EDU30_Wed, 1 Mar 2000 09:05:30 -0500535_iso-8859-1 Good morning people, I'd like to know if anyone has ever experience the following problem and if so, what was done to correct it. The job that I am running takes historical data off of tape GDG and inserts the data to the DB2 table.
For two days I received the same abend when running this job using 2 different tape GDG files as input. After the first abend I ran a job to restore the indexes. The abend that I am getting is a System 04E, Reason Code 00C9021E. I looked up the Reason Code in Bookmanager and it [...]
553 16 26_Disaster recovery software17_Ben Ollo NONLILLY27_OLLO_BEN_NONLILLY@LILLY.COM30_Wed, 1 Mar 2000 09:11:45 -0500402_us-ascii Does anyone know of other vendors for DB2 disaster recovery software besides CA/Platinum Recovery Analyzer and the BMC product?
Thanks. Ben Ollo
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
570 30 27_DB2 Recover and Tape Mounts11_Jeff Schade15_JSCHADE@PCH.COM30_Wed, 1 Mar 2000 09:17:38 -0500575_- We have run into an interesting situation involving a DB2 recovery for multiple tablespaces and tape mounts. The imagecopies being used for input are stacked on a 3490 tape cartridge and the recover statements are coded in the exact same order as the imagecopy statements. The problem is that for every tablespace or partition being recovered DB2 requests a new tape drive. This is forcing the operator to mount the tape multiple times, sometimes on the exact same drive. I do not want to have to code DD cards in the recover JCL for each tablespace but I cannot find [...]
601 56 31_Re: DB2 Recover and Tape Mounts15_Michael Bancale21_mbancale@TXFB-INS.COM30_Wed, 1 Mar 2000 08:24:05 -0600287_iso-8859-1 We faced the same problem several months ago. For the times we had to recover we had the operator 'Hard Mount' the tape on a drive. If memory serves me right this worked. It isn't exactly the best solution but for the time we had to recover it sure did shorten the time. [...]
658 45 31_Re: DB2 Recover and Tape Mounts24_Marchese, Frank D, CSCIO18_fdmarchese@ATT.COM30_Wed, 1 Mar 2000 09:24:15 -0500673_iso-8859-1 Try looking at the RETAIN subparameter of the VOL parameter in the JCL manual.
-----Original Message----- From: Jeff Schade [mailto:JSCHADE@PCH.COM] Sent: Wednesday, March 01, 2000 9:18 AM To: DB2-L@RYCI.COM Subject: DB2 Recover and Tape Mounts
We have run into an interesting situation involving a DB2 recovery for multiple tablespaces and tape mounts. The imagecopies being used for input are stacked on a 3490 tape cartridge and the recover statements are coded in the exact same order as the imagecopy statements. The problem is that for every tablespace or partition being recovered DB2 requests a new tape drive. This is forcing [...]
704 52 31_Re: DB2 Recover and Tape Mounts13_Philip Nelson21_teamdba@ATTGLOBAL.NET30_Wed, 1 Mar 2000 14:19:51 +0000307_us-ascii We had this problem as well. What we did was coded a WTO (write to operator) step immediately after the first recover step and asked the operators to mount the tape with retain, with a similar step at the end to get them to dismount it. This only works if all the copies are on one cartridge. [...]
757 5134 21_Re: Why Not DB2??????5_Buz W14_buzw@YAHOO.COM30_Wed, 1 Mar 2000 06:39:56 -0800366_us-ascii Joel,
In your post you stated, "If you look at the real benchmarks, and TPC and TPD data (and can avoid the apple to football comparisons) DB2 blows away everything else in the marketplace".
Surely you meant to specify RELATIONAL databases. There's not a relational database on the planet that can compete with IDMS for pure performance. [...]
5892 57 31_Re: DB2 Recover and Tape Mounts15_Murley, Michael22_Michael_Murley@BMC.COM30_Wed, 1 Mar 2000 08:35:39 -0600479_ISO-8859-1 I think there are two ways to handle this with the DB2 Recover utility:
1. Code multiple DD statements in the JCL with appropriate volume referbacks (and hope you don't have more DD's than the JES limit!)
2. Issue an MVS Mount command for the volume.
Neither seems very attractive operationally.
BMC RECOVER PLUS for DB2 optimizes stacked tape usage and will only issue one tape mount per volume without coding multiple DD statements. [...]
5950 53 31_Re: DB2 Recover and Tape Mounts20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM30_Wed, 1 Mar 2000 09:38:13 -0500658_us-ascii Jeffrey
The recover also need the DD to be coded using referbacks and "VOL=(,RETAIN,,10,REF=*.SYSCPY01)". Either coding the DD or console command to keep the tape mounted is necessary.
I guess (just guess) V7 will keep the tapes mounted.
Regards -Venkat Pillay
> -----Original Message----- > From: Jeff Schade [SMTP:JSCHADE@PCH.COM] > Sent: Wednesday, March 01, 2000 9:18 AM > To: DB2-L@RYCI.COM > Subject: DB2 Recover and Tape Mounts > > We have run into an interesting situation involving a DB2 > recovery for multiple tablespaces and tape mounts. The > imagecopies being used for input are stacked on a 3490 tape [...]
6004 76 15_Re: DB2 Experts19_Briggs, N. - Neil -25_N.Briggs@CANON-EUROPA.COM30_Wed, 1 Mar 2000 15:42:46 +0100340_iso-8859-1 I have had a similar problem when a number of tablespaces were copied to tape. Unfortunately the datasets caused a second tape to be used. Indicated by the + next to the tape volume (=3.4). The tablespace dataset was catalogued on the first tape but not on the second, due to naming standards in our tape management system. [...]
6081 87 31_Re: DB2 Recover and Tape Mounts22_Leslie Pendlebury-Bowe43_Leslie.Pendlebury-Bowe@CCMAIL.EU.SONY.CO.JP30_Wed, 1 Mar 2000 13:47:39 +0000518_ISO-8859-1 Guys firstly .. you are not (that I know of) going to get away with avoiding coding DD statements for the pagesets that are in the recovery set defined .. you have to code them and code them with the RETAIN option ....
We use Platinum Recovery Analyser here and that takes care of our DR jobs for us everynight .. I cannot praise the product enough .. it works, works well and we are left to work on other things safe in the knowledge that our DR is offsite and ready to go at the drop of a hat [...]
6169 73 31_Re: DB2 Recover and Tape Mounts8_Long, Ed15_Ed.Long@FMR.COM30_Wed, 1 Mar 2000 09:53:16 -0500563_- I use the feature Michael alludes to in REcover+ and it works as advertised even on V6.1
> -----Original Message----- > From: Murley, Michael [SMTP:Michael_Murley@BMC.COM] > Sent: Wednesday, March 01, 2000 9:36 AM > To: DB2-L@RYCI.COM > Subject: Re: DB2 Recover and Tape Mounts > > I think there are two ways to handle this with the DB2 Recover utility: > > 1. Code multiple DD statements in the JCL with appropriate volume > referbacks > (and hope you don't have more DD's than the JES limit!) > > 2. Issue an MVS Mount command for the volume. > > [...]
6243 20 36_DB2 Version 6 and CICS Version 2.1.213_donald tipton29_donald.tipton@DAS.STATE.OH.US30_Wed, 1 Mar 2000 09:57:24 -0500544_us-ascii Is anyone running DB2 version 6 on OS/390 along with CICS version 2.1.2? IBM's official line is that CICS version 4 is required for compatibility, but I was wondering if anyone has tried to make DB2 V6 and CICS 2.1.2 work together. Any informaion would be appreciated.
Thanks Don Tipton
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
6264 99 31_Re: DB2 Recover and Tape Mounts22_Leslie Pendlebury-Bowe43_Leslie.Pendlebury-Bowe@CCMAIL.EU.SONY.CO.JP30_Wed, 1 Mar 2000 14:22:15 +0000336_ISO-8859-1 what if you only have IBM utilities....
If we use Platinum(CA) Fast Recover then the DD statements are not coded .. but with IBM there is no way around it .. either way it works for us .. and we build both types of jobs every night .. sorry PRA builds them for us .. and we get on with more interesting work :-). [...]
6364 92 15_Re: DB2 Experts12_Kalman, Jeff20_JKalman@UNCH.UNC.EDU30_Wed, 1 Mar 2000 10:22:14 -0500572_iso-8859-1 Thank you for your reply.
Jeff
-----Original Message----- From: Briggs, N. - Neil - [mailto:N.Briggs@CANON-EUROPA.COM] Sent: Wednesday, March 01, 2000 9:43 AM To: DB2-L@RYCI.COM Subject: Re: DB2 Experts
I have had a similar problem when a number of tablespaces were copied to tape. Unfortunately the datasets caused a second tape to be used. Indicated by the + next to the tape volume (=3.4). The tablespace dataset was catalogued on the first tape but not on the second, due to naming standards in our tape management system. [...]
6457 142 31_Re: DB2 Recover and Tape Mounts11_Jerry, Jeff24_Jeff.D.Jerry@NORWEST.COM30_Wed, 1 Mar 2000 09:34:25 -0600540_iso-8859-1 We have dealt with this problem in the past as well. We do not have any third party tools to generate the jcl that you seek.
Suggestions:
* Through the magic of DB2 some tables, and SQL you can do this. We run this process on a daily basis and during a disaster recovery test we run it to set up our application recoveries after the catalog and directory have been recovered. We have developed ours in such a way that we can specify if we wish to do "TO CURRENT", "TO COPY" or "TO RBA" recoveries. * You could [...]
6600 115 31_Re: DB2 Recover and Tape Mounts15_Murley, Michael22_Michael_Murley@BMC.COM30_Wed, 1 Mar 2000 09:49:33 -0600666_ISO-8859-1 BMC RECOVERY MANAGER for DB2 also builds the DD statements for the IBM recover utility.
-----Original Message----- From: Leslie Pendlebury-Bowe [mailto:Leslie.Pendlebury-Bowe@CCMAIL.EU.SONY.CO.JP] Sent: Wednesday, March 01, 2000 8:22 AM To: DB2-L@RYCI.COM Subject: Re: DB2 Recover and Tape Mounts
what if you only have IBM utilities....
If we use Platinum(CA) Fast Recover then the DD statements are not coded .. but with IBM there is no way around it .. either way it works for us .. and we build both types of jobs every night .. sorry PRA builds them for us .. and we get on with more interesting work [...]
6716 130 31_Re: DB2 Recover and Tape Mounts8_Long, Ed15_Ed.Long@FMR.COM30_Wed, 1 Mar 2000 10:48:50 -0500553_iso-8859-1 Hi Leslie.
With IBM recover, and assuming no third party software assistance, all I have ever seen are the zillions of DD statements usually hand coded one time and one at a time by a DBA just prior to being shipped to the nut house.
That smart dynamic allocation feature was one of the reasons we bought BMC R+ in the first place lo these many years ago . I have seen folks build REXX execs of SAS programs to generate the jcl with all the affs, referbacks, and label parms coded correctly and sometimes they even worked. [...]
6847 40 31_Re: DB2 Recover and Tape Mounts12_BOB JEANDRON21_BOB.JEANDRON@USDA.GOV30_Wed, 1 Mar 2000 10:17:00 -0500633_- We use the DD statement method, just make sure you use the DD method described in the V5.1 Utilities guide EXACTLY as it appears on P.2-185. We have had problems with recovering tablespaces spanning multiple tape volumes when using a referback in the DD statement.
Usual Disclaimers. >>> DB2-L@RYCI.COM@inter2 03/01/00 08:17AM >>> We have run into an interesting situation involving a DB2 recovery for multiple tablespaces and tape mounts. The imagecopies being used for input are stacked on a 3490 tape cartridge and the recover statements are coded in the exact same order as the imagecopy statements. The problem is [...]
6888 38 25_PLAN TABLE USING AN ALIAS15_Backes, Michael26_BackeM@MAIL.OA.STATE.MO.US30_Wed, 1 Mar 2000 10:20:31 -0600575_iso-8859-1 I am just wanting to verify something and maybe get an explanation as to why, but in the course of my duties several developers have wanted to start doing explains on their queries by themselves. i have told them it is kind of pointless as the test database they would be running them on does not have current statistics in the catalog, and even if it did the data is not production like nor of a production volume. nevertheless they still want to proceed with this exercise. to that end, there already is a plan_table with the dbadm ownerid(a.plan_table) in [...]
6927 46 56_JOBS: DB2 Software Developers- F/T-Perm- Austin, TX, USA11_DeeDee Dial25_dddial@PEDLEY-RICHARD.COM30_Wed, 1 Mar 2000 10:19:21 +0000643_us-ascii Leading developer of Database and Systems Management Software Tools has immediate openings for individuals with 5+ years experience in commercial software development using C and/or Assembler/ALC for DB2, MVS, and/or OS390 systems environments. Canddiates should understand DB2 structures, OS/390 and/or MVS internals, and be able to work at the system level. Knowledge of multi-programming/multi-tasking techniques preferred. Knowledge of distributed systems environments a plus. Individuals must be enjoy working as part of a team and able to interface with cross functional software engineering, QA, and support team members. [...]
6974 31 22_Cancelling a Reorg Job13_Scott Lindsey29_sflindsey@HIGHLIGHTS-CORP.COM30_Wed, 1 Mar 2000 11:43:29 -0500285_us-ascii I have a job running right now that is reorging partitions of a table where the clusterratio feel below 90%. If I wait for the step of the job that is currently running to finish and cancel it the job as soon as the next step starts, how many problems should I run into. [...]
7006 44 31_Re: DB2 Recover and Tape Mounts16_Duane Lee - ATCX22_DLee@MAIL.MARICOPA.GOV30_Wed, 1 Mar 2000 09:47:27 -0700593_- Use the MOUNT console command to mount your tape and then UNLOAD when you are finished.
> -----Original Message----- > From: Jeff Schade [SMTP:JSCHADE@PCH.COM] > Sent: Wednesday, March 01, 2000 7:18 AM > To: DB2-L@RYCI.COM > Subject: DB2 Recover and Tape Mounts > > We have run into an interesting situation involving a DB2 > recovery for multiple tablespaces and tape mounts. The > imagecopies being used for input are stacked on a 3490 tape > cartridge and the recover statements are coded in the exact > same order as the imagecopy statements. The problem is > that for every [...]
7051 69 29_Re: PLAN TABLE USING AN ALIAS20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM30_Wed, 1 Mar 2000 11:48:54 -0500475_us-ascii Mike, My 2 cents worth.
I copy the production catalog statistics on to the test catalogs. This way, although the test tables are not like production in terms of volume but the access path is almost (most of the time) same as that of production. I also prefer creation of PLAN_TABLE for each creator or qualifier. This way develops are aware right during development that they might go for non-matching or tablespace scan rather than discovering it late. [...]
7121 19 14_Helpdesk help!14_Penrod, Dale E19_dale.penrod@EDS.COM30_Wed, 1 Mar 2000 10:47:13 -0600342_- I have been charged with the development of a helpdesk script to determine if a reported problem is related to DB2 or a DB2 support software. I am sure, this type of script must exist somewhere in the DB2 IS world. Can anyone share with me a script, templates, ideas, experiences, or thoughts, on how to develop this type of document? [...]
7141 110 26_Re: Cancelling a Reorg Job0_18_mebert@AMADEUS.NET30_Wed, 1 Mar 2000 18:00:19 +0100381_us-ascii TERMinating a Reorg job in the Unload phase is no problem because the original data is not changed yet (and still there). It does not leave any restrictive states on either TS or indexes. I assume your TS partitions are large, so you shouldn't have a problem hitting the Unload phase. TERMinating an OLR job, by the way, is never a problem, it just may take a while. [...]
7252 5236 21_Re: Why Not DB2??????12_Greg P Mucha24_Greg_P_Mucha@CARADON.COM30_Wed, 1 Mar 2000 17:04:16 +0000628_us-ascii I think I can come up with a reasonable argument AGAINST an RDBMS putting less burden on programmers than VSAM or Sequential processing, which pretty much trashes the arguments in FAVOR of using relational Databases.
When it comes to testing, having an RDBMS environment that can be used by a STAFF of programmers in a maintenance and development shop quickly becomes absurd. And with DBA's who have a right to want to restrict the authority of developers to create their own databases (even temporarily), you're often stuck with data situations that don't even come close to providing an environment where [...]
12489 142 31_Re: DB2 Recover and Tape Mounts22_Leslie Pendlebury-Bowe43_Leslie.Pendlebury-Bowe@CCMAIL.EU.SONY.CO.JP30_Wed, 1 Mar 2000 16:08:25 +0000506_ISO-8859-1 indeed it does ..
infact I think RECOVER+ & recovery manager are great products .. I have only used recover+ ..
what I am saying is you still need the DD statements in there .. I do not care how you get them there .. for IBM that is
;-)
Les
______________________________ Reply Separator _________________________________ Subject: Re: DB2 Recover and Tape Mounts Author: "Murley; Michael" at Internet Date: 3/1/00 9:49 AM [...]
12632 165 31_Re: DB2 Recover and Tape Mounts22_Leslie Pendlebury-Bowe43_Leslie.Pendlebury-Bowe@CCMAIL.EU.SONY.CO.JP30_Wed, 1 Mar 2000 16:06:34 +0000458_ISO-8859-1 Hey - all I am suggesting is that PRA or BMC make life easy .. no support costs etc .. I love rexx .. as a lot of people on here will tell you ..
there are many ways to skin this cat ..
Leslie Pendlebury-Bowe
______________________________ Reply Separator _________________________________ Subject: Re: DB2 Recover and Tape Mounts Author: "Jerry; Jeff" at Internet Date: 3/1/00 9:34 AM [...]
12798 156 31_Re: DB2 Recover and Tape Mounts22_Leslie Pendlebury-Bowe43_Leslie.Pendlebury-Bowe@CCMAIL.EU.SONY.CO.JP30_Wed, 1 Mar 2000 16:09:52 +0000440_ISO-8859-1 Ed Yep - sport on boring but important - that is why I am presenting to customers on Friday about PRA .. and how we use it in a dynamic environment.
let me know how you geton ..
all the best
Les
______________________________ Reply Separator _________________________________ Subject: Re: DB2 Recover and Tape Mounts Author: "Long; Ed" at Internet Date: 3/1/00 10:48 AM [...]
12955 67 31_Re: DB2 Recover and Tape Mounts22_Leslie Pendlebury-Bowe43_Leslie.Pendlebury-Bowe@CCMAIL.EU.SONY.CO.JP30_Wed, 1 Mar 2000 16:10:36 +0000470_ISO-8859-1 yes but it will not stop the tape being rewound ...
it will just keep it permie mounted in the drive
????
Les
______________________________ Reply Separator _________________________________ Subject: Re: DB2 Recover and Tape Mounts Author: Duane Lee - ATCX at Internet Date: 3/1/00 9:47 AM
Use the MOUNT console command to mount your tape and then UNLOAD when you are finished. [...]
13023 27 76_UK only - transition training from OS/390 and DB2 5.1 to AIX and DB2 UDB 6.110_Robert Ord21_robertord@HOTMAIL.COM28_Wed, 1 Mar 2000 17:10:14 GMT593_- Can anyone recommend a transition training course for OS/390 DB2 DBA's who might need to get upto speed with DB2 UDB on AIX. Your help will be much appreciated.
Thanks
Robert Ord
______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
13051 40 11_Re: CLOTECB9_Fox, Dave16_Dave_Fox@BMC.COM30_Wed, 1 Mar 2000 11:24:10 -0600471_ISO-8859-1 Roland,
The resource name CLOTECB stands for CICS Life-of-task ECB (an MVS event control block).
I presume you see this resource named in a message or in a trace. Chances are that you have either hit THRDMAX for your CICS connections, or have set the dispatching priority too high for DB2 threads in your RCT (e.g. DPMODE=HIGH). If you are seeing transactions waiting for CLOTECB, try changing DPMODE to something lower (e.g. DPMODE=EQUAL). [...]
13092 32 21_DB2 and AIX Max Users0_23_sjtoeniskoetter@AEP.COM30_Wed, 1 Mar 2000 13:04:20 -0500401_us-ascii We just experienced a problem that I'd never heard of, and I'm wondering if anyone else is familiar with it. We started getting the following errors when trying to connect remotely (DB2 AIX v6):
SQL30082N Attempt to establish connection failed with security reason "21" ("RESOURCE TEMPORARILY UNAVAILABLE"). SQLSTATE=08001
New local connection requests were still working. [...]
13125 48 26_Re: Cancelling a Reorg Job20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM30_Wed, 1 Mar 2000 11:51:46 -0500578_us-ascii Scott
Stop the next partition right now or go to spufi and do a select WITH RR but do not commit. This way your next step is going to fail before it can start and then you could terminate the utility and do whatever is needed
> -----Original Message----- > From: Scott Lindsey [SMTP:sflindsey@HIGHLIGHTS-CORP.COM] > Sent: Wednesday, March 01, 2000 11:43 AM > To: DB2-L@RYCI.COM > Subject: Cancelling a Reorg Job > > I have a job running right now that is reorging partitions of a table > where the clusterratio feel below 90%. If I wait for the step [...]
13174 41 41_Fwd: DB2 Version 6 and CICS Version 2.1.216_Stephanie Jordan28_jordan_stephanie@HOTMAIL.COM28_Wed, 1 Mar 2000 10:27:25 PST631_- By copying the DSNC* modules from the DB2 V5 loadlib to the DB2 V6 loadlib, we were able to successfully connect and process to a CICS 2.1.2. I know IBM says not a good idea and not supported, but we were able to make it work...
Stephanie Jordan
>From: donald tipton >Reply-To: DB2 Data Base Discussion List >To: DB2-L@RYCI.COM >Subject: DB2 Version 6 and CICS Version 2.1.2 >Date: Wed, 1 Mar 2000 09:57:24 -0500 > >Is anyone running DB2 version 6 on OS/390 along with CICS version >2.1.2? IBM's official line is that CICS version 4 is required [...]
13216 50 31_Re: DB2 Recover and Tape Mounts12_Bing Alabata25_bing.alabata@MCMASTER.COM30_Wed, 1 Mar 2000 12:41:38 -0600599_us-ascii Hi Leslie,
I recall discussions on this issue druing my days as DB2 Columnist for the Data Management Review magazine. Both approaches were covered, i.e. an exit-triggered permanent tape mount and the VOL=(,RETAIN) parameter in the DD statement. More readers favored the JCL-based solution.
I will second Bob's recommendation. Read p. 2-185 of the DB2 V5 Utility Guide and Reference manual, particularly the "Retaining Tape Mounts" section. Note that it outlines the handling of both noncataloged and cataloged image copies in the case of a multi-dataset imagecopy tape. [...]
13267 62 30_Re: Disaster recovery software10_Paul Wendt14_PaulW@ENET.COM30_Wed, 1 Mar 2000 10:48:27 -0800500_us-ascii E-Net provides two disaster recovery products for DB2, RRDF and Log Apply. RRDF runs on the primary site to intercept all log records, and transmits to a recovery site.
At the recovery site, log records are archived to tape (or disk) in their original format (RRDF also supports IMS, ADABAS, IDMS, etc). For DB2, RRDF creates BSDS control cards to update the recovery site BSDS with the transmitted logs, to allow recovery up to the point of failure when a disaster is declared. [...]
13330 114 52_Re: Platinum RCM problem - anyone else experiencing?0_54_Shan_Leatherman/MO/americancentury@AMERICANCENTURY.COM30_Wed, 1 Mar 2000 13:07:50 -0600270_us-ascii Hi Gary, Good to hear from you. Your suggestion sounded good. We gave it a try but we get the same error. Oddly enough we have found,we don't get the error if we try the change in RC/Update , but when we use RC/Migrator we get the error. Truly mysterious. [...]
13445 216 43_Re: Equivalent of "Describe" command in DB216_Milligan, Andrea25_Andrea_Milligan@BCBST.COM30_Wed, 1 Mar 2000 14:20:26 -0500646_- I would be interested in the PROC, FORM and QUERY.
Andrea Milligan DataBase Administrator BlueCross BlueShield of Tennessee
> -----Original Message----- > From: Mallett, Steven [SMTP:Steven.Mallett@CORPMAIL.TELSTRA.COM.AU] > Sent: Tuesday, February 29, 2000 5:44 PM > To: DB2-L@RYCI.COM > Subject: Re: Equivalent of "Describe" command in DB2 > > We've got a query that runs on DB2 V5.1 using QMF 3.3 which gives > for each table in a DB2 Database > column name, description (format) and nullability/defaults > #pages and #rows in table > index columns, type, order > #pages and #nodes on each index > foreign key columns and [...]
13662 146 31_Re: DB2 Recover and Tape Mounts16_Milligan, Andrea25_Andrea_Milligan@BCBST.COM30_Wed, 1 Mar 2000 15:00:17 -0500629_- Do you have VOL=(,,,RETAIN) in your JCL on each of the IC DD's?
Andrea Milligan DataBase Administrator BlueCross BlueShield of Tennessee
> -----Original Message----- > From: Jeff Schade [SMTP:JSCHADE@PCH.COM] > Sent: Wednesday, March 01, 2000 9:18 AM > To: DB2-L@RYCI.COM > Subject: DB2 Recover and Tape Mounts > > We have run into an interesting situation involving a DB2 > recovery for multiple tablespaces and tape mounts. The > imagecopies being used for input are stacked on a 3490 tape > cartridge and the recover statements are coded in the exact > same order as the imagecopy statements. The problem [...]
13809 53 80_Re: UK only - transition training from OS/390 and DB2 5.1 to AIX and DB2 UDB 6.124_The Fillmore Group, Inc.18_fillmore@CLARK.NET30_Wed, 1 Mar 2000 15:05:44 -0800738_us-ascii Robert, Go to http://www.ibm.com/services/learning/uk and search on course code CF21U.
Frank
On Wed, 1 Mar 2000 17:10:14 GMT, Robert Ord wrote:
>Can anyone recommend a transition training course for OS/390 DB2 DBA's who >might need to get upto speed with DB2 UDB on AIX. Your help will be much >appreciated. > >Thanks > >Robert Ord > > > > > > >______________________________________________________ >Get Your Private, Free Email at http://www.hotmail.com > >================================================ >To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM. [...]
13863 57 35_Re: ERwin support for V6 for OS390?11_p.forestall19_forestall@SPRINT.CA30_Wed, 1 Mar 2000 15:46:59 -0500623_- My Canadian CA rep says middle to late June GA for ERwin 4.0. No word on UDB 6.0 but ya gotta figure...
"Lisa Westcott-Dryer" wrote in message news:OF21EDF750.86C3A4E3-ON85256894.004FB75D@mailrouter.net... > Through my rep and product coordinator Cindy Gillis 248 263 9700 #12825 at > CA, they say they have no immediate release date for the support of version > 6 UDB (MVS or whatever) with the Erwin product. Let's hassle 'em and get > 'em going! I have not checked the web yet. > > Lisa > > > > > Joe Luthman @RYCI.COM> on 02/29/2000 09:19:20 AM > > [...]
13921 27 91_Re: Loading data into DB2/UNIX via load command from DB2 OS/390 DSNTIAUL unload output file12_Raymond Bell20_raymondb@NZ1.IBM.COM30_Thu, 2 Mar 2000 09:43:23 +1300545_us-ascii "It has been a long time since I have done this, but if you have DB2 Connect, can't you export from DB2/OS/390, then import into DB2/Unix? Gary"
Yes you can, but it gets a bit time consuming if you're trying to export several million rows. Thanks, Rob (Lake), for your earlier post. I'm also trying to unload an EBCDIC table on DB2 for OS/390 V5 that has both character and numeric fields, FTP/send it to my PC and load it into a DB2 UDB for NT 5.2 table. We'd prefer not to have to externalise the numeric fields but might [...]
13949 52 11_DB2 & Indus18_Whittaker, Stephen26_stephen.whittaker@CPLC.COM30_Wed, 1 Mar 2000 15:49:30 -0500516_iso-8859-1 We are currently at DB2 V5 and have just installed INDUS Version 7. I have asked this question before of this group if anyone was using INDUS with the new DB2 Version 6. The answers I got back was 'no', no-one is currently running Indus with DB2 V6. I've found out that Indus V7 will 'not' support going to DB2 V6 and they have told me that they do not know if Indus V7 is even tolerant of DB2 V6. Basically if we upgrade to DB2 V6 we are on our own if something within Indus breaks. (at least thats [...]
14002 33 21_Re: Why Not DB2??????13_Seibert, Dave26_Dave_Seibert@COMPUWARE.COM30_Wed, 1 Mar 2000 16:00:40 -0500613_iso-8859-1 Hi Buz, I can't question your assertion that IDMS out-performs any other database. Maybe someone else can.
But I have to comment that all the developers, DBAs, and other IDMS professionals I've worked with over the past dozen years or so have removed IDMS from their resumes. I assume that's more from a perception(or mis-perception) of marketability. Sadly, I also have to admit that this is the same argument I used to get from folks back in my OS/2 bigot days. (Just because IBM couldn't market OS/2 as successfully as Microsoft doesn't mean it isn't(wasn't) a vastly superior product). [...]
14036 24 21_Alter Tablespace Help13_Joe Abbagnaro25_Joe_Abbagnaro@PRAXAIR.COM30_Wed, 1 Mar 2000 16:06:32 -0500358_us-ascii I am trying to do an alter on SYSCOPY keep getting invalid verb tried many combinations but can not get it to work
What keyword is missing ???
READY DSN SYSTEM(DSNQ) DSN -ALTER TABLESPACE DSNDB06.SYSCOPY PCTFREE 10 DSN9021I ? VERB '-ALTER' REQUIRED KEYWORD IS MISSING DSN9023I ? DSN9SCND '-ALTER ' ABNORMAL COMPLETION DSN [...]
14061 46 25_Re: Alter Tablespace Help15_Backes, Michael26_BackeM@MAIL.OA.STATE.MO.US30_Wed, 1 Mar 2000 15:12:47 -0600468_iso-8859-1 i may not understand what you are trying to do, however, any alter tablespace command i have ever done is just simple ' alter ts xxxx whatever' not '-alter ts whatever' if im wrong, sorry for wasting your time.
Mike Backes e-mail: backem@mail.oa.state.mo.us
-----Original Message----- From: Joe Abbagnaro [mailto:Joe_Abbagnaro@PRAXAIR.COM] Sent: Wednesday, March 01, 2000 3:07 PM To: DB2-L@RYCI.COM Subject: Alter Tablespace Help [...]
14108 59 25_Re: Alter Tablespace Help0_22_BILL_GALLAGHER@PHL.COM30_Wed, 1 Mar 2000 16:13:03 -0500413_us-ascii Joe,
You are trying to run an SQL ALTER statement under the DSN command processor. It needs to run under some sort of SQL processor, such as SPUFI, QMF, batch, etc.
The DSN processor can only execute DB2 commands (as opposed to SQL statements). The only "ALTER" DB2 commands that would be able to run under DSN would be "ALTER BUFFERPOOL", "ALTER GROUPBUFFERPOOL", or "ALTER UTILITY". [...]
14168 20 23_global temporary tables10_Tom Taylor17_ttaylor@CHUBB.COM30_Wed, 1 Mar 2000 16:30:56 -0500287_us-ascii Hi all
When I create a Global temp table, I noticed I get an EXCLUSIVE lock on a datapage in sysdbase until I commit. Is this correct and is there any way to avoid it. My app (dsntep2) is running longer than the timeout interval and other apps are getting timeouts. [...]
14189 16 28_Alter Tablespace Help - Done13_Joe Abbagnaro25_Joe_Abbagnaro@PRAXAIR.COM30_Wed, 1 Mar 2000 16:34:37 -0500380_us-ascii You guys are correct I ran it under SPUFI and all is well again.
It has been a while since I have done this.
Thanks
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
14206 74 38_Re: DB2 Recover and Tape Mounts -Reply11_Jeff Schade15_JSCHADE@PCH.COM30_Wed, 1 Mar 2000 16:40:20 -0500560_- I would like to thank everyone for their responses to this obviously widespread issue. We were not including ANY DD statements for the recovery jobs since it is not required, except as I am finding out to handle this particular tape mount issue. I agree that we have taken the time to code the required DD statements for the imagecopies and I plan on pursuing a REXX exec that will take these DD cards and create the required recover DD statements with the RETAIN parameter included. I believe this approach is the simplest given that I do not have any [...]
14281 29 27_Re: global temporary tables14_Wayne Driscoll19_wdriscoll@QUEST.COM30_Wed, 1 Mar 2000 13:55:14 -0800589_iso-8859-1 Since Creating a Global Temporary is a DDL statement, it inserts rows into the System Catalog. Because of this, there will be locks on the catalog. What is your dsntep2 app doing? Wayne Driscoll Product Developer Quest Software Inc. wdriscoll@quest.com Note: All opinions are strictly my own.
Hi all
When I create a Global temp table, I noticed I get an EXCLUSIVE lock on a datapage in sysdbase until I commit. Is this correct and is there any way to avoid it. My app (dsntep2) is running longer than the timeout interval and other apps are getting timeouts. [...]
14311 18 18_RACF & DB2 Connect11_Jeff Faughn22_Jeff_Faughn@MAY-CO.COM30_Wed, 1 Mar 2000 15:59:20 -0600327_us-ascii Hello DB2 Land!
We are OS/390 2.6, DB2 6.1 running DB2 Connect and RACF. We don't issue TSO id's to all the PC users (they have select access only) but, want them to access the Mainframe data through their PC (Lotus Approach) and be able to change their own password! Has anyone done anything like this??? [...]
14330 28 40_'Reply with History' - please be careful12_Raymond Bell20_raymondb@NZ1.IBM.COM30_Thu, 2 Mar 2000 10:55:59 +1300568_us-ascii Esteemed colleagues,
Sorry, but maybe I just got up on the wrong side of the bed this morning. I've often been amused by the number of people that click on 'reply with history' when sending to the list without really seeing how much 'history' there is. I admit it's nice to see just exactly what you're replying to, but I do tend to draw the line at replying with history twice when the history in question is a 6000-line digest. I nearly fainted at the thought of sifting through a 12,000-line digest this morning, but was (kinda) pleased it was [...]
14359 74 27_REASON 00C90101 ERQUAL 53CB0_22_Rohn.Solecki@MTS.MB.CA30_Wed, 1 Mar 2000 16:07:14 -0600335_us-ascii I am trying to recover a production backup to development for the last 3 days! But I keep getting "REASON 00C90101 ERQUAL 53CB" type errors on 3 of my restored tables.
Here are the steps I've been following:
- Create a devel version of the table using prod ddl changed to point to devel db. [...]
14434 133 21_Re: Why Not DB2??????0_30_Milorad_Cerovac@TAC.VIC.GOV.AU30_Thu, 2 Mar 2000 09:09:07 +1000546_us-ascii Joel, have you ever worked at a site that uses anything other than DB2 or Oracle? You may be surprised to note that DB2 may not be the fastest. The issue of performance benchmarks can easily spark a wider and greater debate - companies consistently refer to benchmarks that make their product look great, but are they really? At the end of the day the client has to be happy - a number of sites that have converted to DB2 from IDMS had major performance problems that resulted in the IT dept unable to meet its SLAs. The client was [...]
14568 61 91_Re: Loading data into DB2/UNIX via load command from DB2 OS/390 DSNTIAUL unload output file14_Manas Dasgupta25_manas.dasgupta@PFSFHQ.COM30_Wed, 1 Mar 2000 17:14:42 -0500545_iso-8859-1 Raymond:
When you say its time consuming, where do you see most of the time consumed? Is it during EXPORT or during IMPORT. If its during IMPORT you may want to use LOAD instead.
About 6 months back I had to setup AIX tables with data from fairly large os390 db2 tables(about 8 tables half of which had abot 1-20 million rows). I used EXPORT (via db2 connect) to get the data to AIX and then used LOAD to load the data to the AIX tables. I don't remember the exact timings but we were pretty happy with the elapsed [...]
14630 34 25_Re: Alter Tablespace Help10_David Cohn19_dcohn@ATTGLOBAL.NET30_Wed, 1 Mar 2000 14:35:53 -0800560_us-ascii Hi Joe, You are trying to use alter as a DB2 command. Alter is not a command. Execute the alter using SPUFI, QMF, or some like mechanism for executing an SQL statement. Hope this helps, Dave
Joe Abbagnaro wrote:
> I am trying to do an alter on SYSCOPY > keep getting invalid verb tried many combinations but can not get it to work > > What keyword is missing ??? > > READY > DSN SYSTEM(DSNQ) > DSN > -ALTER TABLESPACE DSNDB06.SYSCOPY PCTFREE 10 > DSN9021I ? VERB '-ALTER' REQUIRED KEYWORD IS MISSING > DSN9023I ? DSN9SCND '-ALTER ' [...]
14665 24 31_Re: DB2 Recover and Tape Mounts0_19_Tim.Lowe@STPAUL.COM30_Wed, 1 Mar 2000 16:37:54 -0600581_us-ascii Due to the enhancements in DB2 V6 (parallel backup and recovery when the datasets are NOT on tape), I was thinking that most people would be looking into changing their backup jobs to go to disk instead of tape (for all small to medium sized tablespaces), and then sweep the datasets to tape. With that in mind, I do not think that too much time should be spent on purchased products or "home grown processes" to avoid tape mounts during recoveries. By moving to backups to disk, the result should be faster, cheaper backup and recovery, and better tape utilization. [...]
14690 76 31_Re: DB2 Recover and Tape Mounts14_James Campbell29_James.Campbell@HANCORP.COM.AU30_Thu, 2 Mar 2000 09:36:33 +1000393_iso-8859-1 Should be only one comma: VOL=(,RETAIN)
There is a section in the Utilities Guide for RECOVER titled "Retaining tape mounts". It suggests you also need LABEL=1, LABEL=2 etc. I've never not used LABEL, so I don't know if its really necessary.
/* standard disclaimer */ James Campbell DBA Hansen Corporation, Doncaster +61 3 9840 3864 James.Campbell@Hancorp.com.au [...]
14767 96 31_Re: REASON 00C90101 ERQUAL 53CB15_Glenn Johnstone30_glenn.johnstone@TRICON-YUM.COM30_Wed, 1 Mar 2000 17:46:59 -0500471_US-ASCII I think this is occurring because the source table has had an alter done to it at some point. This causes the DBD to recognize that short or variable length rows are allowed in this table, until such time as a reorg is done. So when you dsn1copy, the data being copied contains short rows but the target DBD does not realize that short rows are to be allowed. You should be able to fix this by first reorging your source tables and then doing the dsn1copy. [...]
14864 67 43_Re: Equivalent of "Describe" command in DB20_19_Tim.Lowe@STPAUL.COM30_Wed, 1 Mar 2000 17:35:34 -0600288_us-ascii David, I do not know how the Oracle "describe" command works. But, take a look at DB2 Apar PQ24584 / DB2 V5 ptf UQ31275 / DB2 V6 ptf UQ31276. This provides support for the ODBC "SQL DESCRIBE INPUT" statement. (Note: DB2 connect V6 is required) Does this solve your problem? [...]
14932 26 33_Re: DB2 to RACF: project planning10_Peter Boot22_Peter.Boot@MAIL.ING.NL30_Wed, 1 Mar 2000 13:38:07 +0000466_US-ASCII Hi all,
I want to thank those of you who responded to my query about DB2 external security in RACF. Many valuable things have been said which will help us in doing this conversion. However, none of you really addressed the issue of project planning. So, I'll take the liberty of asking another time: does anyone have figures on the number of hours/dollars/euros spent in the conversion (possibly with the number of GRANT's which you converted)? [...]
14959 44 37_Re: DB2 Connect & Password Encryption12_MILLS,Robert26_robert.mills@DEWRSB.GOV.AU30_Thu, 2 Mar 2000 12:13:06 +1100584_iso-8859-1 Ben,
I have successfully used the Password Encryption Support with DB2 Connect. This was using Connect V6.1 with fixpack 2 and OS/390 V5.1.2. I had a couple false starts with it and here are some things to watch out for:
There is a prerequisite PTF required on DB2 OS/390 - see APAR PQ21252. If you have the PTF applied check that the module DSNLTBSF is in your DB2 load library, this module is loaded by DDF to provide the cryptography services. Do a List DB directory on the database entry and make sure AUTHENTICATION is set to DCS_ENCRPYT. I believe [...]
15004 25 36_DB2 Connect Enterprise Edititon v6.111_Ellen Chiem25_ellenchiem@SYNERGY.COM.AU30_Thu, 2 Mar 2000 12:04:46 +1000440_us-ascii Hi all,
I have just try to install DB2 UDB Connect Enterprise Edition version 6.1. I pick everything ( Administrative client, Software Developer's kit, and of course Connect Enterprise edition). I thought by choosing Administrative client and Software Developer's kit, it should have install the necessary OBDC drivers ( i.e db2.o). However, that isn't the case?? Have I done something wrong in the installation part?? [...]
15030 71 21_Re: Why Not DB2??????14_Joel Goldstein29_joel_goldstein@COMPUSERVE.COM30_Wed, 1 Mar 2000 20:45:38 -0500577_ISO-8859-1 Milorad, Yes I have. I cut my dbms teeth on IMS starting in 1969, even before it had a TP front end. I have also worked at sites that had IDMS. Any dbms can run like #%@* (including IDMS) when the design is wrong. DB2 is simply more overhead than IMS or IDMS - because you get more set processing power and functionality; however, when applications are moved to DB2, with the proper re-design for a relational system, they run quite well, and within projected performance and response time criteria (and yes it's often higher than originally). However, when an [...]
15102 205 21_Re: Why Not DB2??????14_Richard Yevich21_ryevich@ATTGLOBAL.NET30_Wed, 1 Mar 2000 20:14:51 -0500443_iso-8859-1 During the period from 1985 to 1990, the number of licenses of IDMS started to decline rapidly, not due to relational, not due to CA, but due to it's inability to grow functionally at that time. There was a new release coming, for a long long time they said, which became as I remember IDMS-10 (and some will remember the promises in the mid-80's of IDMS-R). By then the total number of licenses was already declining rapidly. [...]
15308 82 29_Reply to DB2 Recover and Tape16_JOHN G. MATTHEWS26_HC#JGM@MVS.MEDCTR.UCLA.EDU30_Wed, 1 Mar 2000 18:38:00 -0800744_- *** Original Author: JSCHADE@PCH - Schade; 03/01/00 06:22
>Received: from mshieldsrv.medctr.ucla.edu (149.142.194.75) by > MVS.MEDCTR.UCLA.EDU > (IBM MVS SMTP V3R2) with TCP; Wed, 01 Mar 00 06:22:40 PDT >Return-Path: >Received: from ryci.com ( 208.225.214.81 ) by mshieldsrv.medctr.ucla.edu with > SMTP (MailShield v1.5); Wed, 01 Mar 2000 06:24:30 -0000 >Received: from jupiter (jupiter.ryci.com) by ryci.com (LSMTP for Windows NT > v1.1b) with SMTP id <0.000A2165@ryci.com>; Wed, 1 Mar 2000 8:10:59 -0600 >Received: from RYCI.COM by RYCI.COM (LISTSERV-TCP/IP release 1.8d) with spool > id 13633 for DB2-L@RYCI.COM; Wed, 1 Mar 2000 08:10:59 -0600 >Received: from mail.pch.com by ryci.com (LSMTP for Windows [...]
15391 74 11_Re: CLOTECB11_Roland Chua32_rolandchua@SINGAPOREEXCHANGE.COM30_Thu, 2 Mar 2000 10:42:35 +0800333_us-ascii Hi Fox,
Thanks for the response. This resource name CLOTECB appeared in the Omegamon CICS screen. I would like to know will it affect our DB2 or CICS performance. If it does, could you explain in more detail? I had checked and our system did not hit THRDMAX. We do have certain threads defined with DPMODE=HIGH. [...]
15466 89 31_Re: DB2 Recover and Tape Mounts13_Viswanathan N33_Viswanathan.N@POWAIMAIL.LTITL.COM30_Thu, 2 Mar 2000 08:35:20 +0530412_us-ascii hi !!
I just want to ask if you are specifying the same cartridge in multiple DD statements in the same step of the JCL ?? If thats true then I have a solution .
regards Vishy
BOB JEANDRON on 03/01/2000 08:47:00 PM
Please respond to DB2 Data Base Discussion List
To: DB2-L@RYCI.COM [...]
15556 48 38_Accessing data on OS390 from Window NT11_Shirley Fan25_shirley.fan@ZURICH.COM.AU30_Thu, 2 Mar 2000 14:58:26 +1000495_us-ascii Hi all, I have installed IBM UDB Enterprise Edition v6 on a window NT and tried to connect to os390 db2 v5 via tcp/ip drda.
When I right click the tables in the control centre, the only option appeared on the object tree was 'Define As Replication Source'. But I can select the mainframe tables from the Command Centre. (so I assume the connection is ok). I tried two different ways to set up the remote database (using CCA and Control Centre), but the result was the same. [...]
15605 72 29_Re: PLAN TABLE USING AN ALIAS14_James Campbell29_James.Campbell@HANCORP.COM.AU30_Thu, 2 Mar 2000 15:06:58 +1000600_- You've got some unusual developers - actually concerned about the performance of their SQL! I'ld give them all the encouragement I could.
The only way I know to avoid having a plan_table per developer is to have an 'EXPLAIN program'. It takes an SQL statement, explains it, prints a report from the plan_table. Bind it with dynamicrules(bind) and have a package for each environment. Anyone can use it for any SQL - even against the production data. BUT, since all it does is a "delete old info from plan_table", explain, and "extract info from plan_table" there is no security breach. [...]
15678 290 21_Re: Why Not DB2??????0_30_Milorad_Cerovac@TAC.VIC.GOV.AU30_Thu, 2 Mar 2000 16:03:09 +1000611_us-ascii Richard,
One can argue that IDMS was stifled under the ownership of Cullinet, but one cannot then assume that IDMS has not moved forward under the control of CA. Rel 10 may have been late in coming, but it came and it went, and then R12 came (offering significant improvements to 24 hour processing, separating physical/logical database definitions etc) and now R14 has been around for a few years - though scalability was a problem it no longer is; with R14 capable of exploiting IBM's parallel SYSPLEX environment; and R15 is just around the corner with additional functionality. And ... [...]
15969 87 31_Re: DB2 Recover and Tape Mounts22_Leslie Pendlebury-Bowe43_Leslie.Pendlebury-Bowe@CCMAIL.EU.SONY.CO.JP30_Thu, 2 Mar 2000 05:34:08 +0000340_ISO-8859-1 Bing My one and only point with all this is the RETAIN .. if you do not get that little boy to work - and as you rightly point out there are several ways this all works, then you end up with repeated rewinds or remounts or whatever based on how you coded your JCL , and it can BITE your backside off at DR time wise :-) ... [...]
16057 15 43_Kurt Rompot/Schneider is out of the office.11_Kurt Rompot21_RompotK@SCHNEIDER.COM30_Thu, 2 Mar 2000 01:02:28 -0600389_us-ascii I will be out of the office starting 03/02/2000 and will not return until 03/06/2000.
I will respond to your message when I return.
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
16073 98 95_=?iso-8859-1?B?UulmLiA6IEFjY2Vzc2luZyBkYXRhIG9uIE9TMzkwIGZyb20gV2luZG93?= =?iso-8859-1?B?IE5U?=0_26_denzil.coulter@UNICIBLE.CH30_Thu, 2 Mar 2000 08:24:28 +0100454_iso-8859-1 Look at the 'no cost' DB2 Management Tools Package feature (FMID JDB551D). This activates OS390 enablement for the Control Center by installing the necessary stored procedures on the host DB2.
Shirley Fan @RYCI.COM> le 02.03.2000 05:58:26
Veuillez répondre à DB2 Data Base Discussion List
Envoyé par : DB2 Data Base Discussion List [...]
16172 50 31_Re: DB2 Recover and Tape Mounts14_Massimo Scarpa16_mscarpa@CESVE.IT30_Thu, 2 Mar 2000 08:57:36 +0100296_us-ascii Why I have to pay a lot of $$$ for a 3rd party product that make an implicit allocate of a
dataset read from catalog tables and execute stupid -SET NOCOPYPEND to reset some TS
statuses (and to pay every time I upgrade my machine, every time a new company is bought by [...]
16223 98 31_Re: DB2 Recover and Tape Mounts4_IBM116_IBM1@UBIL.COM.TR30_Thu, 2 Mar 2000 10:14:01 +0200272_- altough I totally agree with the general concept of massimo's idea, still my life as a systems person would be a sad and sorry one if I had to spend my entire working life looking at 15 different monitors when a "third party" tool has been created to do JUST THAT. [...]
16322 97 31_Re: DB2 Recover and Tape Mounts14_Tom Decabooter27_Tom.Decabooter@HONDA-EU.COM30_Thu, 2 Mar 2000 09:11:11 +0100627_us-ascii I fully agree. I automated our complete DB2 disaster recovery using REXX/ISPF/OPC, starting with the directory/catalog up to the last tablespace. No third party tools required.
Tom Decabooter System Administrator Honda Europe N.V.
Massimo Scarpa on 02/03/2000 08:57:36
Please respond to DB2 Data Base Discussion List
To: DB2-L@RYCI.COM
cc: (bcc: Tom Decabooter/HE/EU/HONDA)
Subject: Re: DB2 Recover and Tape Mounts [...]
16420 65 31_Re: DB2 Recover and Tape Mounts14_Massimo Scarpa16_mscarpa@CESVE.IT30_Thu, 2 Mar 2000 10:52:29 +0100552_us-ascii May be out-of theme (have you never heard CEO's version ?), but in many & many years of experience with 3rd part product sometimes:
1) It was a justification for a new price of a tool. Really , it was a sales rapresentative's justification for a BIG increase of price decided by a CEO (and ALL products of that company were fired). It's only a joke to say that some increase in prices haven't any justification and someone agree with me. If you agr(rrrrr)e or not does't matter (it seems to me that this is the real reason of your [...]
16486 137 31_Re: DB2 Recover and Tape Mounts22_Leslie Pendlebury-Bowe43_Leslie.Pendlebury-Bowe@CCMAIL.EU.SONY.CO.JP30_Thu, 2 Mar 2000 09:22:08 +0000485_ISO-8859-1 ok guys anybody can write REXX until it comes out their ear holes .. and I am sure most people at my site here are fed up with me saying "we could do this with rexx and save pounds".
But when I arrived at this shop I was given PRA .. as we are unable to get rid of this product for N years I decided it was better to work with it and see what it could do rather than make my boss feel he had dropped a clanger by buying a product he didn't need. As a consultant [...]
16624 50 33_Re: DB2 to RACF: project planning0_23_J.P.Slot@RF.RABOBANK.NL30_Thu, 2 Mar 2000 08:44:38 +0100690_us-ascii Peter,
If you get any figures i'll be interested too..
Regards,
Jaap Slot
----------------------[Reply - Original Message]----------------------
Sent by:Peter Boot Hi all,
I want to thank those of you who responded to my query about DB2 external security in RACF. Many valuable things have been said which will help us in doing this conversion. However, none of you really addressed the issue of project planning. So, I'll take the liberty of asking another time: does anyone have figures on the number of hours/dollars/euros spent in the conversion (possibly with the number of GRANT's which you converted)? [...]
16675 64 31_Re: DB2 Recover and Tape Mounts14_Massimo Scarpa16_mscarpa@CESVE.IT30_Thu, 2 Mar 2000 12:00:10 +0100366_us-ascii I disagree.
An example of real life: 2 (two) consultants made many REXX to create CONTROL-M chains in 2
(two) months. In the new installation where I work now we made the SAME thing in 2 (two) days
(and I think in a more clever way).
For this reason I recommend to do your own (small & stupid) REXX 'coz you know your needs [...]
16740 55 31_Re: DB2 Recover and Tape Mounts12_Allen, Colin31_Colin.Allen@ABBEYNATIONAL.CO.UK30_Thu, 2 Mar 2000 11:06:00 -0000499_- I agree entirely with the sentiments below. Some of the 3rd party tools I have used in the past have taken weeks of effort to try to get them to do what I wanted and didn't necessarily succeed in the end. Where my opinion differs is that many DBA's CAN'T write REXX, they are never trained in it and don't find the time to learn it. I feel that REXX, ISPF panels, skeletons, JCL, etc are essential to DBA's but Management don't seem to realise that a structured training program is required. [...]
16796 26 31_Re: DB2 Recover and Tape Mounts14_Massimo Scarpa16_mscarpa@CESVE.IT30_Thu, 2 Mar 2000 12:17:57 +0100418_us-ascii I think that this is the answer: a lot of people CAN'T write REXX execs or CAN'T write simple
REXX or doesn't knows hot to use REXX to communicate with DB2, for instance.
Or many consultant CAN write REXX but they don't have any experiences in managing a
production/DR environment.
So they buy (or suggests to buy) products that use REXX/CLIST/Panels.....and this costs $$$$. [...]
16823 77 31_Re: DB2 Recover and Tape Mounts22_Leslie Pendlebury-Bowe43_Leslie.Pendlebury-Bowe@CCMAIL.EU.SONY.CO.JP30_Thu, 2 Mar 2000 10:38:04 +0000674_ISO-8859-1 Colin .. some very good points. I was a programmer for a decade before going into DBA .. it helped.
Les
______________________________ Reply Separator _________________________________ Subject: Re: DB2 Recover and Tape Mounts Author: "Allen; Colin" at Internet Date: 3/2/00 11:06 AM
I agree entirely with the sentiments below. Some of the 3rd party tools I have used in the past have taken weeks of effort to try to get them to do what I wanted and didn't necessarily succeed in the end. Where my opinion differs is that many DBA's CAN'T write REXX, they are never trained in it [...]
16901 50 44_Re: 'Reply with History' - please be careful12_Allen, Colin31_Colin.Allen@ABBEYNATIONAL.CO.UK30_Thu, 2 Mar 2000 11:48:00 -0000321_- I actually like to see some history although I agree that some selective editing would be nice. While we are on the subject, summarising replys is ok, but I'm sure no one will be offended if 'thank you's are taken as read or sent to just the author, no the whole list, similarly 'me too' requests for sample code. [...]
16952 17 31_Re: DB2 Recover and Tape Mounts22_Leslie Pendlebury-Bowe43_Leslie.Pendlebury-Bowe@CCMAIL.EU.SONY.CO.JP30_Thu, 2 Mar 2000 10:54:19 +0000292_ISO-8859-1 ...i'm finished with this :)
Les
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
16970 16 43_Re: Equivalent of "Describe" command in DB211_John Rosser24_jhrosser@DUKE-ENERGY.COM30_Thu, 2 Mar 2000 06:34:11 -0600358_- If you have QMF, try this command: DRAW tablename (TYPE=INSERT
Regards, John Rosser
my opinions only
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
16987 99 29_Re: PLAN TABLE USING AN ALIAS13_Lynne Flatley17_LFlatley@NEFN.COM30_Thu, 2 Mar 2000 08:10:53 -0500550_us-ascii James is right on when he says...encourage them! This is great. Create a plan table with the same creator ID as the RACF group that the developers are in. If they are trying to explain a dynamic SQL statement, then they need to set their current SQLid to the RACF group (this also assumes that the query being explained has its tables using the same creator ID, if not, aliases WILL work for those tables). If they are trying to use explain(yes) on a bind, the qualifier needs to be the creator ID of both the plan_table and the tables [...]
17087 63 21_Re: Why Not DB2??????14_Richard Yevich21_ryevich@ATTGLOBAL.NET30_Thu, 2 Mar 2000 07:16:17 -0500601_iso-8859-1 Going nowhere -- time for EOT!
Richard
> -----Original Message----- > From: DB2 Data Base Discussion List [mailto:DB2-L@RYCI.COM]On Behalf Of > Milorad_Cerovac@TAC.VIC.GOV.AU > Sent: Thursday, March 02, 2000 1:03 AM > To: DB2-L@RYCI.COM > Subject: Re: Why Not DB2?????? > > > Richard, > > One can argue that IDMS was stifled under the ownership of Cullinet, but > one cannot then assume that IDMS has not moved forward under the > control of > CA. Rel 10 may have been late in coming, but it came and it went, and > then R12 came (offering significant improvements to 24 [...]
17151 19 31_Re: DB2 Recover and Tape Mounts13_Aratos Sicyon25_aratos_sicyon@HOTMAIL.COM28_Thu, 2 Mar 2000 15:19:35 EET422_- Please, please, STOP!
A CEO's husband.
______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
17171 16 45_Re: DB2 Recover,Tape Mounts and CEO's husband14_Massimo Scarpa16_mscarpa@CESVE.IT30_Thu, 2 Mar 2000 14:34:46 +0100361_us-ascii Stop what ? Why ? I missed something ?
"If you don't like what you read, don't read."
Max Scarpa
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
17188 40 22_Re: RACF & DB2 Connect9_Chu, Pius14_ChuP@CONED.COM30_Thu, 2 Mar 2000 08:55:01 -0500389_- As long you have an valid RACF id, it should be fine. In our environment, I used our unique employee number as the RACF ID and connected to a RACF group. The RACF group is granted the table authority in DB2. The unique employee number RACF id has no other access. The unique RACF id enable our support group to track down the individuals and also less questions from auditing dept. [...]
17229 73 27_Re: global temporary tables10_Tom Taylor17_ttaylor@CHUBB.COM30_Thu, 2 Mar 2000 09:00:06 -0500400_us-ascii WAYNE
I'm generating object statistics and it takes a long time, maybe 15 min. during this time there are other apps that need the same data page and can't get it, then timeout. I would create the table permemently, but were short on dasd. any suggestions. Tom
Wayne Driscoll on 03/01/2000 04:55:14 PM [...]
17303 85 29_Re: PLAN TABLE USING AN ALIAS16_Ruane, Paul (P.)16_pruane1@FORD.COM30_Thu, 2 Mar 2000 08:55:56 -0500700_iso-8859-1 PLAN_TABLE must be defined as type T in SYSIBM.SYSTABLES. Global Temporary Tables aren't allowed either.
-----Original Message----- From: Pillay, Venkat (PCA) [mailto:venkat_pillay@ML.COM] Sent: 01 March 2000 16:49 To: DB2-L@RYCI.COM Subject: Re: PLAN TABLE USING AN ALIAS
Mike, My 2 cents worth.
I copy the production catalog statistics on to the test catalogs. This way, although the test tables are not like production in terms of volume but the access path is almost (most of the time) same as that of production. I also prefer creation of PLAN_TABLE for each creator or qualifier. This way develops are aware right during development [...]
17389 88 21_Re: Why Not DB2??????0_22_BILL_GALLAGHER@PHL.COM30_Thu, 2 Mar 2000 09:31:15 -0500518_us-ascii This has turned out to be a very entertaining thread! Maybe the IDMS vs DB2 issue can be settled once and for all in a "steel-cage" match at Wrestlemania 2000.
Bill
Richard Yevich on 03/02/2000 07:16:17 AM
Please respond to ryevich@attglobal.net
To: DB2-L@RYCI.COM cc: (bcc: BILL GALLAGHER/Phoenix Home Life Mutual Insurance) bcc: BILL GALLAGHER/Phoenix Home Life Mutual Insurance Subject: Re: Why Not DB2?????? [...]
17478 13 21_Re: Why Not DB2??????10_Bill Brown25_bill_brown@SPRINGMAIL.COM30_Thu, 2 Mar 2000 09:37:33 -0500548_- Boy has this thread strayed from its original point. As I was the originator of the thread it has been sort of humorous to see the IDMS thing jump in here. Talk about left field. Somebody needs to go take a valium.
Back to the original point. I was trying to instigate a debate over why projects the size that the big three auto makers and Sears are undertaking seem to be ignoring the technical superiority of DB2 and to some extent the mainframe. I realize that Oracle has "mind share" in the market, but not all of these decisions [...]
17492 51 21_Re: Why Not DB2??????20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM30_Thu, 2 Mar 2000 09:44:06 -0500298_us-ascii I know ORACLE list-server would be more appropriate to ask this question.
I am long been wondering what is the biggest ORACLE database (in terms of GB, in terms of # of users and in terms of # of transaction a day) ? and How do people feel about performance in that big shop ? [...]
17544 45 35_Re: ERwin support for V6 for OS390?16_Rasmussen, Steen22_Steen.Rasmussen@CA.COM30_Thu, 2 Mar 2000 14:43:25 -0000662_- (to satisfy certain requests - history deleted on purpose)
Hi .....
The upcoming version of ERwin (version 4.0) is focused on model management, traceability of design object between design layers, reuse of physical design across heterogeneous database environments, enhanced managemet of very large models, enhanced support for corporate design standard and improved GUI. The "big thing" is the ability to have independent models for different purposes in different design layers. As an example you can have the same problem modelled in a logical model, a generic physical model, a UDB model and a Oracle model. All of them independent, but [...]
17590 82 21_EDMPOOL Fragmentation9_Andy Hunt33_Andy.Hunt@SCOTTISH-SOUTHERN.CO.UK30_Thu, 2 Mar 2000 14:55:21 +0000440_us-ascii Hi, I believe we may be suffering from EDMPOOL fragmentation problems (DB2 V5.1 @ 9906 ). Typical Omegamon shows, Pool Usage Pages Pct Max Pages Entry Count of Entries ----------- ------- ---- --------------------- -------------------- In Use 4438 24% DBDs 677 4% 122.0 38 CTs 65 0% 8.0 74 PTs 3696 20% 49.0 980 Available 14312 76% SKCTs 672 4% 33.7 602 SKPTs 9840 52% 81.0 1676 Free 3800 20% 14.0 1240 Total Pages 18750 100% [...]
17673 81 49_Re: DB2 Personal Edition & Java Stored Procedures14_Julian Stuhler27_julian.stuhler@TRITON.CO.UK30_Thu, 2 Mar 2000 07:00:29 -0800573_- Hi Srinivas.
I believe the first issue is due to the packaging of DB2 Connect with the personal and enterprise editions of DB2 UDB. I believe that There were some postings to the newsgroup a few weeks ago on this subject, so it may be worth checking the archives.
As to your second question, the DB2 stored procedure builder does not yet offer direct support for deploying Java stored procedures directly to the S/390 platform. However, you can certainly use it to create and test the SP on the NT or windows platform, then simply move the code to the [...]
17755 84 64_A query tuning challenge from the world of DB2 system monitoring0_34_L-IS.Kirkpatrick@MUTUALOFOMAHA.COM30_Thu, 2 Mar 2000 09:03:26 -0600365_us-ascii We have a query that lists the most recent image copy for each table space. It also lists the tablespaces that do not have an image copy. We use this query to ensure that all DB2 tablespaces in the shop have a valid image copy. We also compare the results of this list with the MVS catalog to ensure that MVS is still tracking each of these datasets. [...]
17840 31 31_Re: REASON 00C90101 ERQUAL 53CB14_Wayne Driscoll19_wdriscoll@QUEST.COM30_Thu, 2 Mar 2000 07:17:04 -0800606_iso-8859-1 If you find that the problem is indeed due to these "short rows" you can get around the problem by executing an ALTER TABLE xxx ADD COLUMN yyy CHAR(1) NOT NULL WITH DEFAULT. This will cause the DBD to be flagged to allow variable length rows. This problem will only occur if NONE of the rows in the table currently are defined as VARCHAR. Another method to get variable length columns supported would be to alter the tablespace to allow compression (COMPRESS YES). Hope this helps. Wayne Driscoll Product Developer Quest Software Inc. wdriscoll@quest.com Note: All opinions are strictly my [...]
17872 74 31_Re: DB2 Recover and Tape Mounts15_Robert Lawrence21_rlawrence@BOSCOVS.COM30_Thu, 2 Mar 2000 10:17:21 -0500404_us-ascii Bing, I really enjoyed and learned a lot from your column. It became the first (and a lot of times the only) article I read in that magazine. Thanks again
Bob Lawrence DBA Boscov's Dept Stores Inc
-----Original Message----- From: Bing Alabata [SMTP:bing.alabata@MCMASTER.COM] Sent: Wednesday, March 01, 2000 1:42 PM To: DB2-L@RYCI.COM Subject: Re: DB2 Recover and Tape Mounts [...]
17947 40 27_Re: global temporary tables14_Wayne Driscoll19_wdriscoll@QUEST.COM30_Thu, 2 Mar 2000 07:33:28 -0800549_iso-8859-1 Tom, Global Temporary Tables can (and should) be defined (CREATED) once, then used repeatedly. You could have a job that defines the table, run it at a time when there is little contention. Then, when you need to use the table, the job would just instanciate the table by inserting one or more rows into it. The contents of the table will only be known to the UOW that performed the instanciation, but the tables catalog definition is availble to be used by other users, at the same or any other time. If your job is, in one unit of [...]
17988 40 21_Re: Why Not DB2??????13_Dan S Brogdon22_Dan.S.Brogdon@AEXP.COM30_Thu, 2 Mar 2000 08:52:01 -0700500_- For those who have designed or maintained a network database vs. relational vs hierarchical there are industry/processing needs that fit one better than the other.
What I'd like to note is that DB2/IBM is finally adopting some of the features in DB2 that have been available in IDMS for several decades: 1. The ability to run in local vs cv mode (running batch without DBMS over head). 2. Providing next , prior and owner pointers.(fetch prior). 3. Optimizer hints (not needed in IDMS) [...]
18029 120 21_Re: Why Not DB2??????9_Don Alden26_Don.Alden@PKSIS.LEVEL3.COM30_Thu, 2 Mar 2000 08:52:58 -0700467_iso-8859-1 Ladies and Gentlemen of DB2 LSERV,
Seems like this has been the most widely discussed issue since the DB2 recovery tape handling. Adding my 2cents about 20+ years old technology, IMS is still the NUMBER ONE DBMS that handles LARGE corporations and has not and will not go away anytime soon. And, by the way, still "blows the doors off DB2". Although, as was mentioned earlier, it has "nowhere near" the flexibility and capability that DB2 has. [...]
18150 84 40_Re: DB2 Connect password change from ADO0_15_leon@CA.IBM.COM30_Thu, 2 Mar 2000 11:10:04 -0500435_us-ascii You can change password from an application. This function is supported from all programming APIs (ADO, ODBC, JDBC, SQLJ, Embedded SQL, DB2 CLI). For ADO or ODBC or JDBC all you need to do is add NEWPW=xxxx to your connection string (the string you provide as an argument to the connection object OPEN method). Is is trivial to do. Here is a sniplet of VBScript code that uses ADO to change DB2 password from a web page: [...]
18235 19 21_Re: Why Not DB2??????0_19_Tim.Lowe@STPAUL.COM30_Thu, 2 Mar 2000 10:55:11 -0600298_us-ascii To add my 2 cents, I have heard many times at several sites that a UNIX or NT server "must be rebooted" due to problems it is having. It is extremely rare that anything like this ever occurs on OS/390 systems. That alone explains why large sites use OS/390 for critical applications. [...]
18255 42 21_Re: Why Not DB2??????16_David Sanfilippo23_david.sanfilippo@DB.COM30_Thu, 2 Mar 2000 10:57:33 -0600541_us-ascii Maybe a Bachmann vs Codd rematch on the undercard?
---------------------------------------- Message History ----------------------------------------
From: BILL_GALLAGHER@phl.com on 03/02/2000 02:31 PM GMT
Please respond to DB2-L@ryci.com
To: DB2-L@ryci.com cc: Subject: Re: Why Not DB2??????
This has turned out to be a very entertaining thread! Maybe the IDMS vs DB2 issue can be settled once and for all in a "steel-cage" match at Wrestlemania 2000. [...]
18298 88 28_DB2 5.2 AIX - Delete Failure11_John Pilbin25_John.Pilbin@GRATTAN.CO.UK30_Thu, 2 Mar 2000 17:22:58 -0000378_iso-8859-1 We are trying to delete data from a database supporting Net.Commerce, with Text Extenders on the Product Table. The delete function works fine on test, but when we run it on live, we get :- SQL0901N The SQL statement failed because of a non-severe system error. Subsequent SQL statements can be processed. (Reason "unexpected column usage type".) SQLSTATE=58004 [...]
18387 91 21_Re: Why Not DB2??????13_Lynne Flatley17_LFlatley@NEFN.COM30_Thu, 2 Mar 2000 12:44:08 -0500352_us-ascii I wasn't going to jump in but...
I've worked with many DBMSs including (but not limited to) IMS, IDMS and DB2 for many years as both a DBA and a programmer. I even worked at Cullinane/Cullinet for a time.
Which one do I love working with now from the perspective of ease of use as both a developer and a DBA? DB2, DB2, DB2. [...]
18479 43 4_REXX13_Crowley, Paul21_CrowleP@DIVINVEST.COM30_Thu, 2 Mar 2000 11:50:13 -0600379_iso-8859-1 Can anyone recommend a good book to learn about coding REXX execs? Thanks
-----Original Message----- From: Massimo Scarpa [mailto:mscarpa@CESVE.IT] Sent: Thursday, March 02, 2000 6:18 AM To: DB2-L@RYCI.COM Subject: Re: DB2 Recover and Tape Mounts
I think that this is the answer: a lot of people CAN'T write REXX execs or CAN'T write simple [...]
18523 26 8_Re: REXX11_Daryl Wells18_dwwell1@US.IBM.COM30_Thu, 2 Mar 2000 12:59:01 -0500493_us-ascii A good book to learn how to use REXX is REXX in the TSO Environment, Gabriel F Gargiulo, Wiley-QED IBM Mainframe series, ISBN 0-471-56188-6.
Also, about a month ago, Mark Zelden posted a link to a REXX class that he assembled. Here it is:
http://www.flash.net/~mzelden/mvsfiles/rexxintro.zip
------------------------------------------------------------------ From: Daryl Wells DB/DC Support IBM Global Services Tie Line 427-3242 Phone (813) 878-3242 [...]
18550 16 21_Re: Why Not DB2??????10_John Klock18_john.klock@BBH.COM30_Thu, 2 Mar 2000 13:01:25 -0600495_- First, thanks List for all the information over the last several months. Secondly, I also can not resist, in IDMS 12.01 we experienced frequent system crashes, and a shortage of programmers willing to work on IDMS projects. DB2 V5 has been extremely stable and there seems to be a larger pool of programmers. I must note however, that the feature ROWID in V6 sounds an improved version of DBKEY. Both DBMS have their own strengths but DB2 in my opinion has the greater strengths overall. [...]
18567 24 21_Re: Why Not DB2??????12_Jim Harrison12_jimh@QIS.NET30_Thu, 2 Mar 2000 14:30:08 -0500588_us-ascii Interesting. Our IDMS regions were always extremely stable (R10 days) . I don't think I ever saw one crash - unless *I* caused it, and I was known for doing that frequently!
At 01:01 PM 3/2/00 -0600, you wrote: >First, thanks List for all the information over the last several months. >Secondly, I also can not resist, in IDMS 12.01 we experienced frequent >system crashes, and a shortage of programmers willing to work on IDMS >projects. DB2 V5 has been extremely stable and there seems to be a larger >pool of programmers. I must note however, that the feature [...]
18592 1432 20_stogroup vol err msg14_Marg Gozdowski27_marg.gozdowski@LA-Z-BOY.COM30_Thu, 2 Mar 2000 14:34:25 -0500323_iso-8859-1 Under db2 510, I tried to create a stogroup with a volume of all #s. received err msg:
my other stogroups have vols of xxx###
According to the example in the manuals, I should be able to use volume of all #s.
Do you know of any bug in db2 that would cause the err listed above? [...]
20025 269 21_Re: Why not DB2??????12_Steve Fuller18_vantech@RADIKS.NET30_Thu, 2 Mar 2000 13:49:34 -0600552_iso-8859-1 Bill,
I realize this string has gotten away from your original thought, but if you would bare with me, I have a non-technical thought about the IDMS vs DB2 arena.
Quite a few of you in the United States know me, as we have talked in the past, and will probably talk in the future. I was an IBM Programmer Analyst for about 3 plus years in IDMS and ADS/O, from 1982-1985, found that this was not my calling, and became a recruiter. I specialized in IDMS DBA's from coast to coast from mid-1985 to mid 1990, when I switched [...]
20295 1459 24_Re: stogroup vol err msg9_John Reid14_jreid@FBFS.COM30_Thu, 2 Mar 2000 14:09:19 -0600549_us-ascii
Marg, do you mean asterisks? Your E-mail came through with pound signs? If it was asterisks then here is what happened. It was an error in 4.0 that was fixed in 5.1 or so. You now can only have one asterisk in stogroup volume. We had the same issue and to fix it we took the DB2 subsystem down, reset the DSNZPARM to allow update's to the system catalog, used SQL update to set all the volids to one asterisk, took the DB2 subsystem down again and then reset the DSNZPARM to not allow update on the system catalog. IBM [...]
21755 21 40_Logging Insert Records with -803 SQLCODE12_Mehdi Fakhar21_MFakhar@GUARANTEC.COM30_Thu, 2 Mar 2000 14:14:43 -0600557_- Following is the scenario we have encounterd. Table has 2 unique index. Index 1 is primary, made up of timestamp manipulation. Index 2 is a 4 column composite key. Program was looping during the insert. It gets the timestamp, makes up the key(Index 1) calls up an I/O program to insert, it gets an -803 due to index 2, returns to calling program which makes up another key (Index 1), calls up the I/O routine again, .... The problem is when the transaction was aborted, it tried to Rollback about 8 million rows that had tried to insert. My question [...]
21777 70 21_Re: Why Not DB2??????0_22_Rohn.Solecki@MTS.MB.CA30_Thu, 2 Mar 2000 14:42:08 -0600550_us-ascii Q: How many news group subscribers does it take to change a light bulb? A: 1,343, broken down as follows . . . - 1 to change the light bulb and to post to the newsgroup that the light bulb has been changed; - 14 to share similar experiences of changing light bulbs and how the light bulb could have been changed differently; - 7 to caution about the dangers of changing light bulbs; - 27 to point out spelling/grammar errors in posts about changing light bulbs; - 53 to flame the spell checkers; - 1 to correct spelling/grammar flames; [...]
21848 111 64_A query tuning challenge from the world of DB2 system monitoring14_Michael Hannan24_mhannan@C031.AONE.NET.AU30_Fri, 3 Mar 2000 08:16:08 +1000341_us-ascii Larry,
You haven't given the access path, but by far the most likely problem is that the Left Join is using nested loop join to join to a materialised right hand table in a work file (which has no index).
To force Merge Scan Join, change LEFT JOIN to FULL JOIN and add a WHERE predicate at the end to compensate: [...]
21960 37 26_Stored Procedure Debugging8_LAM, Leo17_l.lam@IOOF.COM.AU30_Fri, 3 Mar 2000 08:16:37 +1000672_iso-8859-1 I've some questions regarding debugging Stored Procedures.
Is anyone using the IBM DEBUG Tool to debug DB2 Stored Procedures ?
My reading of the IBM web page would require an OS/2 workstation to drive the tool ?
How are other people developing and debugging COBOL stored procedures on OS/390 ?
Thanks
Leo
****************** IMPORTANT - PLEASE READ **************************************** Please confirm receipt of this email. This email is confidential - if you receive this email by mistake you must not use it or the information in any way. If there is a problem with this email or you have received it [...]
21998 49 87_Loading data into DB2/UNIX via load command from DB2 OS/390 DSNTIAUL unload output file12_Raymond Bell20_raymondb@NZ1.IBM.COM30_Fri, 3 Mar 2000 10:27:03 +1300296_us-ascii Manas,
I was talking about the time it takes to transmit the data from OS/390 to AIX/NT. Admittedly it's not much to do with the export command itself, more the limited bandwidth we have to play with. And yes, when the data finally arrives, I will be using the Load utility. [...]
22048 13 18_Re: Why IDMS??????0_17_JTonchick@AOL.COM28_Thu, 2 Mar 2000 16:50:26 EST307_US-ASCII Please take this IDMS discussion to the IDMS-L listserv.
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
22062 28 24_DB2 Patch (PTF) question0_22_Rohn.Solecki@MTS.MB.CA30_Thu, 2 Mar 2000 17:00:28 -0600367_us-ascii A couple of questions about PTF's (our service provider handles them) . . .
- are all/most of them applied to next version up. i.e. are all PTF's for V5 applied to V6 when it comes out (I realize there has to be a cutoff date for the new version) - if yes, is there a list somewhere where we can check to see which ones make it into next version [...]
22091 50 44_Re: Logging Insert Records with -803 SQLCODE15_Di Carlo, Donna23_Donna_Bermender@BMC.COM30_Thu, 2 Mar 2000 17:08:59 -0600331_ISO-8859-1 I think I can answer the easy question. DB2 will first insert the row into the table, then the primary index. It doesn't know it has a duplicate until it inserts into the second index. After receiving the error, it must then delete the row from the primary index, then the table, creating compensation log records. [...]
22142 91 31_FW: TDAN.com March Announcement13_Morrill, John12_JohnM@VP.NET30_Thu, 2 Mar 2000 16:15:56 -0700585_iso-8859-1 -----Original Message----- From: Robert S. Seiner (CIBER/TDAN) [mailto:rseiner@tdan.com] Sent: Tuesday, February 29, 2000 9:56 PM To: TDAN Reader List Subject: TDAN.com March Announcement
Re: The Data Administration Newsletter -- http://www.tdan.com
The next issue of TDAN.com will be announced on April 1. The latest issue includes articles on web mining, warehouse CSF, stewardship, XML and EDI, knowledge management, repository implementation, and more? These articles will be archive-bound when the next issue is announced in a month. [...]
22234 63 28_Re: DB2 Patch (PTF) question14_James Campbell29_James.Campbell@HANCORP.COM.AU30_Fri, 3 Mar 2000 10:22:25 +1000371_- Based on my reading of various posts on IBM-MAIN: - yes, if they are applicable. - you have to use IBMLink to check. As the fix might involve different coding in different releases, fixes might not became available for all affected releases at the same time.
PQ28894 has UQ33487 for the V5 fix and UQ33488 for the V6 fix. Both became available on 99/09/02. [...]
22298 1045 43_CSP template for DB2 CURSOR WITH HOLD, CICS14_Linda Hagedorn29_LLHAGEDORN@GWMAIL.STATE.NV.US30_Thu, 2 Mar 2000 11:30:09 -0800317_US-ASCII Hello,
I'm need a copy of a CSP (Cross System Product) template using CURSOR WITH HOLD, fetching through a cursor result set, and paging forwards and backwards (if CSP has tricks or methods to do this). Of most importance is CURSOR WITH HOLD to page through a result set across multiple screens. [...]
23344 24 27_REASON 00C90101 ERQUAL 53CB28_Dan Courter (1-408-463-3870)27_daniel_courter@VNET.IBM.COM28_Thu, 2 Mar 2000 15:51:41 PST508_- Rohn,
The 53CB is a return code 4 from an SPROC, due to finding a row that should be fixed length (obdrelvf=no) but the actual data row is missing a column. Possible explanations:
- damaged tablespace - a bad sproc (generated assembler code to extract data, specific to a fetch) - down level/inconsistent plans/packages (sprocs are stored in plans sometimes) - inconsistent recoveries and/or modify scenarios + wrong data copied in with DSN1COPY + bad PIT recovery (to when obdrelvf=no) [...]
23369 20 17_Stored Procedures30_GroupIT - Jenny Kuan Kong Lang19_jennykkl@MAS.COM.MY30_Fri, 3 Mar 2000 10:15:46 -0800477_us-ascii Hi,
Is there any difference in stored procedures written for the distributed platforms e.g. NT, AIX versus stored procedures written for UDB on OS/390. If so, are the differences manageable or drastic?
jenny kuan
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
23390 42 20_2GB/4GB PRIQTY LIMIT15_Backes, Michael26_BackeM@MAIL.OA.STATE.MO.US30_Thu, 2 Mar 2000 20:43:21 -0600475_iso-8859-1 I apologize in advance because i am pretty sure this was recently talked about in this forum, but i cannot seem to find it in the archives. so i will ask and beg forgiveness.
a 3390 device only has roughly 3000 or so cylinders. effectively making any dataset on it limited to 2 gig? i think i have read somewhere that the limit for a partitition is 4 gig?? etc. basically i am confused as to what the best approach is to handling these large datasets. [...]
23433 49 26_Links about DB2 for OS/3906_DB2usa18_db2usa@HOTMAIL.COM30_Thu, 2 Mar 2000 21:28:41 -0600520_ISO-8859-1 Hi,
Here is a non-profit website about DB2 for OS/390 (IBM mainframes):
- English : http://www.multimania.com/db2usa/eindex.htm - Français : http://www.multimania.com/db2usa/index.htm
Last updated on Sunday, February 20th 2000 ===========================================
- Links about DB2 for OS/390: http://www.multimania.com/db2usa/eliendb2.htm Joe Celko wrote the SQL for Smarties column in DBMS magazine.
INRIA's Rodin project was canceled and replaced by Caravel. [...]
23483 19 18_Re: Why IDMS??????12_Jim Harrison12_jimh@QIS.NET30_Thu, 2 Mar 2000 22:50:14 -0500490_us-ascii Well, it might be most helpful if you would kindly provide an address at which we could subscribe, would it not?
At 04:50 PM 3/2/00 EST, JTonchick@AOL.COM you wrote:
>Please take this IDMS discussion to the IDMS-L listserv. >
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
23503 52 34_UDB Backup/Restore strange problem0_29_Greg.Palgrave@BANKWEST.COM.AU30_Fri, 3 Mar 2000 12:00:14 +0800357_us-ascii I have a bizarre problem with NT UDB backup/restores, and I am hoping someone on the list can shed some light on the cause and/or cure.
Here's the scenario: If I run a backup of a remote database and direct the files to d:\backup (on the server) it runs (no problem). If I try to restore on the server from that backup it is also fine. [...]
23556 77 38_Re: UDB Backup/Restore strange problem11_Roland Chua32_rolandchua@SINGAPOREEXCHANGE.COM30_Fri, 3 Mar 2000 12:58:02 +0800588_us-ascii Hi Greg, Do you copy the whole backup directory or just the backup file itself. For it to work you must copy the whole directory.
Greg.Palgrave@BANKWEST.COM.AU on 03/03/2000 12:00:14 PM
Please respond to DB2 Data Base Discussion List
To: DB2-L@RYCI.COM cc: (bcc: ROLAND CHUA CHONG KEE/DATACEN/IT/SGX) Subject: UDB Backup/Restore strange problem
I have a bizarre problem with NT UDB backup/restores, and I am hoping someone on the list can shed some light on the cause and/or cure. [...]
23634 99 38_Re: UDB Backup/Restore strange problem0_29_Greg.Palgrave@BANKWEST.COM.AU30_Fri, 3 Mar 2000 13:17:09 +0800666_us-ascii Roland,
Yes, I copy the file AND entire directory structure.
Thanks, Greg
---------------------------------------- Message History ----------------------------------------
From: Roland Chua on 03/03/2000 12:58
Please respond to DB2 Data Base Discussion List
To: DB2-L@RYCI.COM cc: (bcc: Greg Palgrave/SDG/SS/BankWest)
Subject: Re: UDB Backup/Restore strange problem
Hi Greg, Do you copy the whole backup directory or just the backup file itself. For it to work you must copy the whole directory. [...]
23734 61 8_Re: REXX12_Nurmi, Seppo23_Seppo.Nurmi@RIKSBANK.SE30_Fri, 3 Mar 2000 08:38:06 +0100655_- Hi, More of REXX can be found on the mailing list: TSO-REXX@VM.MARIST.EDU archives: http://www.marist.edu/htbin/wlvindex?TSO-REXX /Seppo
> -----Original Message----- > From: Crowley, Paul [SMTP:CrowleP@DIVINVEST.COM] > Sent: Thursday, March 02, 2000 6:50 PM > To: DB2-L@RYCI.COM > Subject: REXX > > Can anyone recommend a good book to learn about coding REXX execs? > Thanks > > -----Original Message----- > From: Massimo Scarpa [mailto:mscarpa@CESVE.IT] > Sent: Thursday, March 02, 2000 6:18 AM > To: DB2-L@RYCI.COM > Subject: Re: DB2 Recover and Tape Mounts > > > I think that this is the answer: a lot of people CAN'T write REXX execs or [...]
23796 71 24_Re: 2GB/4GB PRIQTY LIMIT12_Nurmi, Seppo23_Seppo.Nurmi@RIKSBANK.SE30_Fri, 3 Mar 2000 09:14:23 +0100531_- I have been thinking to ask a similar question, the manuals are not very clear in this point (maybe I don't find the right spot). We have a tablespace that goes just over 2 gig, only one dataset, though, but the VSAM dataset is allocated on two volumes. In which situation does this happend? And in which situation does DB2 allocate a new VSAM dataset. Of which size then, primary or secundary quantity? Something to do with DB2 versions? We have now V5 but the tablespace have not been changed since V2 or V3 maybe. /Seppo [...]
23868 85 24_Re: 2GB/4GB PRIQTY LIMIT12_Nurmi, Seppo23_Seppo.Nurmi@RIKSBANK.SE30_Fri, 3 Mar 2000 09:23:00 +0100529_- > Hi, something went wrong and i got a rejection message. I wrote: > > I have been thinking to ask a similar question, the manuals are not very > clear in this point (maybe I don't find the right spot). We have a > tablespace that goes just over 2 gig, only one dataset, though, but the > VSAM dataset is allocated on two volumes. > In which situation does this happend? And in which situation does DB2 > allocate a new VSAM dataset. Of which size then, primary or secundary > quantity? Something to do with DB2 versions? [...]
23954 83 24_Re: 2GB/4GB PRIQTY LIMIT10_Fritz Rapp21_FritzRapp@T-ONLINE.DE30_Thu, 2 Mar 2000 18:10:28 +0000432_us-ascii Hi Seppo, if a situation comes up so that DB2 discovers there isnt enough space on the volume where your VSAM-Cluster resides and you have several volumes defined in the stogroup you use - then DB2 allocates another cluster on a different volume with the same PRIQTY-Space as the original dataset. If you dont have enough space you could minimize the PRIQTY and enlarge the SECQTY and run another REORG. regards Fritz [...]
24038 68 30_FYI.... DB2 V6 DATA CORRUPTION11_Mohamed Beg19_mohamedb@ABSA.CO.ZA30_Fri, 3 Mar 2000 10:28:15 +0200631_iso-8859-1 db2 v6 data corruption. >>This flash is intended for the immediate notification of all customers >>using or planning to use: >> >> DB2 UDB for OS/390 V6 in Data Sharing >> Coupling Facility Control Code (CFCC) Release Level 7 or Level 8 >> APAR PQ23043/PTF UQ35798 or APAR PQ32199/UQ38517 >> >>Depending on the CFCC Release and Service Level, this combination may cause >>severe data corruption and data loss in DB2 data sharing environments. >> >>To use the above combination safely, Customers must ensure that they have >>the correct CFCC service level on their coupling facilities, see below, as >>documented in [...]
24107 19 18_Re: Why IDMS??????14_Massimo Scarpa16_mscarpa@CESVE.IT30_Fri, 3 Mar 2000 09:44:29 +0100443_us-ascii Why not if it's useful to understand advantages (& disadvantages) of DB2 & others DBMS ?
I've learned a lot of new things from this 'flamed' discussion.
Regards
MAx Scarpa
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
24127 69 8_Re: REXX4_IBM116_IBM1@UBIL.COM.TR30_Fri, 3 Mar 2000 11:09:46 +0200690_- REXX in the TSO Environment,
an IBM book (code ISBN 0-471-56188-6).
ciao
Aurora Dell'Anno Systems Engineer IBM Software Support e-mail ibm1@ubil.com.tr aurora.dellanno@sistinf.it
> -----Original Message----- > From: Crowley, Paul [SMTP:CrowleP@DIVINVEST.COM] > Sent: Thursday, March 02, 2000 7:50 PM > To: DB2-L@RYCI.COM > Subject: REXX > > Can anyone recommend a good book to learn about coding REXX execs? > Thanks > > -----Original Message----- > From: Massimo Scarpa [mailto:mscarpa@CESVE.IT] > Sent: Thursday, March 02, 2000 6:18 AM > To: DB2-L@RYCI.COM > Subject: Re: DB2 Recover and Tape Mounts > > > I think that this is the answer: a lot [...]
24197 1494 24_Re: stogroup vol err msg13_Adrian Savory24_adrian.savory@ZURICH.COM30_Fri, 3 Mar 2000 10:57:37 +0000347_us-ascii
Marg,
It just so happens I tried something very similar yesterday. We're trying to find a way of seperating SAP tablespaces and indexes, where the VCAT is the same for both - due to the SAP r.4.5 naming standards you cannot tell from the dataset name whether you are looking at an index or tablespace dataset. [...]
25692 99 33_Re: Time consuming Load Processes12_Isaac Yassin23_yassin@NETVISION.NET.IL30_Fri, 3 Mar 2000 13:15:56 +0200605_us-ascii Hi, I did it many times with varchar - no problems. You get as reorged as in prod (or the image copy you use for input into the dsn1copy) - no more, no less.
Isaac Yassin
I Singh wrote: > > DSN1COPY might not be a feasible solution if you have varchar fields for > then you will have to reorg your tables on test. > ---------------------- Forwarded by Inderpal Singh/IRM/DST/US on 02/29/2000 > 01:49 PM --------------------------- > > Isaac Yassin on 02/28/2000 12:11:17 PM > > Please respond to DB2 Data Base Discussion List > > To: [...]
25792 59 18_Re: Why IDMS??????16_Michael McMorrow25_michael.g.mcmorrow@AIB.IE30_Fri, 3 Mar 2000 11:50:52 -0000623_ISO-8859-1 I agree. I'm not sure which of Rohn's categories this falls into but I've worked mainly as DBA for the IMS, DB2 and NCR/Teradata lightbulbs and am increasingly being asked by business sponsors about other lightbulb brands and technologies. The ORACLE lightbulb is rightly or wrongly gaining a lot of industry credit at the moment and as either a neutral DBA or DB2-biased DBA I want to contribute to my companies lightbulb architecture from a position of open-minded knowledge rather than from a blinkered 'DB2 is great and the rest are necessarily and eternally rubbish' stance. The IDMS lightbulb tangent [...]
25852 57 8_Re: REXX11_Hayden, Lee18_Lee_Hayden@CSX.COM30_Fri, 3 Mar 2000 08:06:45 -0500566_iso-8859-1 Check out the REXX Language Home Page at:
http://www2.hursley.ibm.com/rexx/rexx.html
There is LOTS of information here.
-----Original Message----- From: Crowley, Paul [mailto:CrowleP@DIVINVEST.COM] Sent: Thursday, March 02, 2000 12:50 PM Subject: REXX
Can anyone recommend a good book to learn about coding REXX execs? Thanks
-----Original Message----- From: Massimo Scarpa [mailto:mscarpa@CESVE.IT] Sent: Thursday, March 02, 2000 6:18 AM To: DB2-L@RYCI.COM Subject: Re: DB2 Recover and Tape Mounts [...]
25910 60 20_FW: the *NEW* IDMS-L13_Jackson, Jeff20_Jeff.Jackson@FMR.COM30_Fri, 3 Mar 2000 08:15:02 -0500750_iso-8859-1 For those of you who want to know about the IDMS mailing list....
Jeff Jackson FPI&BG Database Services Fidelity Investments 82 Devonshire St., MZ V12C Boston, MA, 02109 * - 617-392-0908 *mailto:Jeff.Jackson@fmr.com
-----Original Message----- From: Chris Hoelscher [mailto:hoelsche@US.IBM.COM] Sent: Friday, June 18, 1999 11:38 AM To: IDMS-L@UGA.CC.UGA.EDU Subject: the *NEW* IDMS-L
IUA - the CA-IDMS Database & Applications User Association is pleased to announce the inauguration of the *new* IDMS-L. This list will carry forward the tradition started by James Bradshaw in February 1989 and carried on through today. We thank James Bradshaw for his work and hope that he [...]
25971 70 24_Re: 2GB/4GB PRIQTY LIMIT0_23_Mike_Levine@TEKHELP.NET30_Fri, 3 Mar 2000 08:22:59 -0500549_us-ascii Hi,
A non "large" DB2 tablespace can be 64 GB max. Each partition of a partitioned tablespace can be 4GB. The VSAM size limit for a single dataset is 2GB. If you cannot partition the table it may be best to allocate a smaller priqty and secqty and let each data set go into extents. We have a large segmented tablespace and it is allocated in DB2 as PRIQTY 262000 SECQTY 131000. This gives a primary allocation of 364 cyls and a secondary of 182 cyls. What happens is that each 2 GB dataset (A001, A002, etc) uses 43,695 tracks [...]
26042 82 27_Re: global temporary tables10_Tom Taylor17_ttaylor@CHUBB.COM30_Fri, 3 Mar 2000 08:38:44 -0500867_us-ascii Wayne I created the global temp table outside of my batch job. Everything is running smooth.... Thanks Tom
Wayne Driscoll on 03/02/2000 10:33:28 AM
Please respond to DB2 Data Base Discussion List
To: DB2-L@RYCI.COM
cc: (bcc: Thomas G Taylor/ChubbMail)
Subject: Re: global temporary tables
Tom, Global Temporary Tables can (and should) be defined (CREATED) once, then used repeatedly. You could have a job that defines the table, run it at a time when there is little contention. Then, when you need to use the table, the job would just instanciate the table by inserting one or more rows [...]
26125 41 10_IXF Format11_David Price22_dprice@UBIQUITY.COM.AU30_Sat, 4 Mar 2000 00:40:22 +1100458_us-ascii PC/IXF format is an Unload format supported by non-OS/390 DB2 UDB utilities EXPORT, IMPORT, LOAD and DB2MOVE. As well as the data, IXF has information about the tables and columns so that IMPORT, for example, can redefine the tables without any DDL.
Its format is documented in the (non-mainframe) DB2 UDB 6.1 "Data Movement Utilities Guide & Reference" manual, Appendix C. (The DB2 UDB 5.2 format has some different field lengths BTW.) [...]
26167 45 25_NPI's & load performance.14_Jessen Michael29_JessenMichel@JDCORP.DEERE.COM30_Fri, 3 Mar 2000 07:54:02 -0600467_- I searched the archives and found similar problems with NPI build times, but I didn't see any posted solutions (if there are any?)... This is for DB2 OS/390 V5.1 with a 300+ million row table, 72 partitions & 2 npis.
1. Why does the build phase of a load into... replace part ## take so long --> when I'm loading an empty dataset? Ok, It takes so long because it's rebuilding the entire NPI (right?) - Is there anyway to not rebuild the 'entire' index? [...]
26213 124 24_Re: 2GB/4GB PRIQTY LIMIT0_23_Mike_Levine@TEKHELP.NET30_Fri, 3 Mar 2000 09:13:53 -0500625_us-ascii Hi Adrian,
Yes I agree! Thanks for pointing this out.
Regards,
Michael Levine Premier Data Services, Inc.
> >Mike, > >This may be a trivial comment (hence not to DB2-L) but wouldn't it be better to >allocate PRIQTY 262080 SECQTY 131040? This would give you cylinder allocation >(e.g. PRIQTY = 364 * 720, SECQTY = 182 * 720). > >Just a thought. > >Adrian Savory >ACSIS Ltd >----- Forwarded by Adrian Savory/ZI/England/Zurich on 03/03/00 13:37 ----- >|--------+-----------------------> >| | Mike_Levine@T| >| | EKHELP.NET | >| | | >| | 03/03/00 | >| | 13:22 | >| | Please | >| [...]
26338 92 24_Re: 2GB/4GB PRIQTY LIMIT19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM30_Fri, 3 Mar 2000 08:14:32 -0600604_iso-8859-1 Hi Michael, Mike, The last time I was involved with this issue, I observed that a non-partitioned DB2 tablespace (segmented or simple) will extend when it reaches its 2 or 4GB limit as DS "vcat.dbname.tsname.I0001.A002", then A003, etc. As of DB2 for OS/390 V5, when the 2 or 4GB limit is reached, I've seen DB2 take the PRIQTY on a candidate volume (SMS-managed, STOGROUP, or user-defined), if available, else the SECQTY, if available. I have even seen DB2 take an amount less than the SECQTY on a candidate volume (V4). Another factor is how RAID logical volumes are defined size-wise. [...]
26431 145 29_Re: NPI's & load performance.0_18_mebert@AMADEUS.NET30_Fri, 3 Mar 2000 15:36:25 +0100477_us-ascii Well actually DB2 is not Rebuilding the NPI. Rebuild is a sort of low-level operation, where the index is constructed in one go from raw data. The BUILD phase of a NPI is different if you do partition-level processing (that's why it's called BUILD2 in OLR), this is more a SQL type of operation, and thus much slower. If you REPLACE PART # from DUMMY, DB2 has to scan the entire NPIs and delete the RID entries corresponding to the deleted rows in the partition. [...]
26577 158 31_Re: DB2 Recover and Tape Mounts0_18_mebert@AMADEUS.NET30_Fri, 3 Mar 2000 15:44:32 +01001639_us-ascii Did you consider posting this disaster recovery REXX et al. to the DB2-L-DOCUMENTS Listserver?
Regards
Dr. Michael Ebert DB2 Database Administrator aMaDEUS Data Processing Erding / Munich, Germany
From: Tom Decabooter on 02/03/2000 08:11 GMT
Please respond to DB2 Data Base Discussion List
|---------> | | |---------> >---------------------------------------------------------------------------> | | >---------------------------------------------------------------------------> >-------------------------------------------| | | >-------------------------------------------| |---------> |To: | |---------> >---------------------------------------------------------------------------> |DB2-L@RYCI.COM | >---------------------------------------------------------------------------> >-------------------------------------------| | | | | >-------------------------------------------| |---------> |cc: | |---------> >---------------------------------------------------------------------------> | (bcc: Michael Ebert/MUC/AMADEUS) | >---------------------------------------------------------------------------> >-------------------------------------------| | | | | >-------------------------------------------| |---------> | | |---------> >---------------------------------------------------------------------------> | | >---------------------------------------------------------------------------> >-------------------------------------------| | | >-------------------------------------------| [...]
26736 148 68_Re: A query tuning challenge from the world of DB2 system monitoring0_34_L-IS.Kirkpatrick@MUTUALOFOMAHA.COM30_Fri, 3 Mar 2000 08:46:19 -0600312_us-ascii This did the trick. I changed the query to a full join and the access path changed from a nested loop to a merge scan. The result is that the query now runs in 5 sec of CPU instead of 4+ minutes.
By the way, we are on version 5 of DB2. Would this have been optimized correctly in version 6? [...]
26885 69 18_Re: db2-l@ryci.com12_Greg Popwell27_greg.popwell@MOTION-IND.COM30_Fri, 3 Mar 2000 08:54:37 -0600508_iso-8859-1 Those of you who told me to get more current on maintenance to solve this problem were right. As info, APAR PQ25401 exactly described my problem. PTF's UQ29219 (V5) and UQ29220 (V6) have been available since 4/27/99. (pause to hide my face in embarrassment)
Thanks to those who responded!
Greg Popwell
-----Original Message----- From: Ray H Janes [mailto:ray_h_janes@EMAIL.MOBIL.COM] Sent: Monday, February 28, 2000 9:03 AM To: DB2-L@RYCI.COM Subject: Re: db2-l@ryci.com [...]
26955 69 44_Re: Logging Insert Records with -803 SQLCODE15_Di Carlo, Donna23_Donna_Bermender@BMC.COM30_Fri, 3 Mar 2000 09:03:38 -0600576_ISO-8859-1 Tim Lowe and I had an offline conversation regarding this topic. He asked "Wouldn't (DB2) have to go to the clustering index first to determine the optimum page to put the row in the tablespace?". That got me curious, so I set up a simple test where the clustering index was the unique index. It looks like DB2 will detect the duplicate while determining row placement and avoid any unnecessary updates to the space. But, if the unique index was a secondary index, the updates to the table space and index occur before DB2 realizes this is a duplicate. Thanks [...]
27025 66 14_Re: IXF Format17_Alessandro Brezzi28_brezzi.isac@INTERBUSINESS.IT30_Fri, 3 Mar 2000 16:27:19 +0100635_us-ascii Hi David, if you own QMF OS/390 you can find in the Appendix C of Developing QMF Applications SC26-4722-04 the IXF host format.
HTH
Alessandro
>PC/IXF format is an Unload format supported by non-OS/390 DB2 UDB >utilities EXPORT, IMPORT, LOAD and DB2MOVE. As well as the data, >IXF has information about the tables and columns so that IMPORT, >for example, can redefine the tables without any DDL. > >Its format is documented in the (non-mainframe) DB2 UDB 6.1 >"Data Movement Utilities Guide & Reference" manual, Appendix C. >(The DB2 UDB 5.2 format has some different field lengths BTW.) > >Because [...]
27092 36 30_Re: Stored Procedure Debugging12_Steve Grimes30_Steve_Grimes@AISMAIL.WUSTL.EDU30_Fri, 3 Mar 2000 09:21:33 -0600363_us-ascii Leo,
We are just breaking ground on Stored Procedure debugging under OS/390, V6.
We don't have WLM so we can't take advantage of GUI Stored Proc. builder/debugger. At this point were writing Cobol Stored Procedures and using the DISPLAY verb. We had to add the following DD to the stored procedure address space to allow this to work: [...]
27129 23 36_DB2 Contoken Conversion to Timestamp15_Michael Bancale21_mbancale@TXFB-INS.COM30_Fri, 3 Mar 2000 09:27:35 -0600512_iso-8859-1 Does anyone have a REXX or any other type program or method which will convert the DB2 Contoken in the DB2 Catalog to a timestamp?
TIA
Michael Bancale mbancale@txfb-ins.com Texas Farm Bureau Insurance Company www.txfb-ins.com Waco, Texas
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
27153 47 40_Re: DB2 Contoken Conversion to Timestamp20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM30_Fri, 3 Mar 2000 10:58:01 -0500385_us-ascii Below is algorithm - I am sure you could write a REXX to do this.
Take the conttoken and split it into two 4 bytes halves.
second half needs (Shift Left Single) by 3 bits. So convert it to decimal and multiply with 8 --->Result = PARTB again Shift Left Double by 3 bits ---> decimal(first half + hex(PARTB) ) * 8 ---> result = decimal value for STORE CLOCK [...]
27201 24 14_Re: IXF Format10_Peter Boot22_Peter.Boot@MAIL.ING.NL30_Fri, 3 Mar 2000 16:39:57 +0000405_US-ASCII David,
>And does anyone know anything about "Host IXF" format? >It's mentioned in the non-OS/390 manuals but I haven't >come across it in the DB2 OS/390 manuals.
The host IXF format is described, partially, in appendix C of the manual 'Developing QMF Applications'. There's a reference there to a 'Data Extract' product, also from IBM, which is said to define the full format. [...]
27226 160 60_Re: JOBS: DB2 Software Developers- F/T-Perm- Austin, TX, USA12_Kapil Mathur25_Kapil.Mathur@WACHOVIA.COM30_Fri, 3 Mar 2000 11:11:55 -0500381_us-ascii
hi, I am not qualified for this job (no knowledge of C) but I was wondering if the company that you mentioned is BMC in Texas/Houston - because I am a mainframe DB2 for OS/390 Database Administrator (6.5 years) with a COBOL/CICS background and I was wondering if you knew if there was any suitable openings at BMC (if that is the client that you support) ... [...]
27387 64 14_Re: IXF Format17_Rabindra Senapati21_rsenapati@HOTMAIL.COM28_Fri, 3 Mar 2000 08:30:06 PST363_- I can not answer your questions.But i experience some problem using other two formats on nt platform which i sloved using ixf format. It's varchar field which puts some carriage ruturn ascii code in the varchar fields(if that is empty) using other formats which case my load fails to load my backup.But ixf does perfectly for me.May help to get some idea. [...]
27452 68 40_Re: DB2 Contoken Conversion to Timestamp9_Rob Crane22_racrane@CONCENTRIC.NET30_Fri, 3 Mar 2000 09:41:39 -0700781_iso-8859-1 I assume by "convert to timestamp" what you want is the ability to compare the contoken in the package, DBRM library member, and the loadlib library member. Here is a simple query that can be used to get the contoken in the proper format for that type of search.
SELECT A.COLLID AS COLLECTION, A.NAME AS PACKAGE, HEX(A.CONTOKEN) AS DBRM_TOKEN, SUBSTR(HEX(A.CONTOKEN),9,8) || SUBSTR(HEX(A.CONTOKEN),1,8) AS LOADLIB_TOKEN, A.VERSION, A.PDSNAME FROM SYSIBM.SYSPACKAGE A WHERE A.NAME = ‘????????’ ORDER BY A.NAME, A.COLLID, A.VERSION WITH UR; -------------------------------------------OUTPUT------------------- COLLECTION PACKAGE DBRM_TOKEN LOADLIB_TOKEN AUTO_OLTP P6980002 15D1906116E27DD8 16E27DD815D19061 AUTO_OLTP P6980002 15D1906116EF8EC8 16EF8EC815D19061 [...]
27521 66 14_FW: XDB Vendor14_Richard Bourke25_Richard.Bourke@MERANT.COM30_Fri, 3 Mar 2000 17:35:56 +0100638_- Bill, congragulations on finding it. We unfortunately hide the provenance of the various component products quite successfully! Machadi, if the general context of the "SQL Option for DB2" within Mainframe Express seems surprising, bear in mind what it is being used for within Mainframe Express, to allow off-loading of host development by providing compatible components on the developer desktop - like XDB for DB2 - for os/390 system components. There is no actual requirement that you have to use it that way however. XDB Software in the past positioned the server for different uses. Anyway, Merant India's contact details are [...]
27588 96 40_Re: DB2 Contoken Conversion to Timestamp20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM30_Fri, 3 Mar 2000 11:43:04 -0500526_iso-8859-1 Rob
There is small glitch in this approach. You said >>search using hex representation.
>>DBRM code search -> FIND x'15D1906116E27DD8' >>Loadlib code search -> FIND x'16E27DD815D19061'
The Conttoken in the LOAD module may not be in 8 byte format. It could split into two 4 bytes fields being swapped or if have C program load module, it would be split into four 2 bytes fields and scattered across load module in any fashion. One should search the load module keeping this factor in mind. [...]
27685 51 37_Cursor name on DECLARE/OPEN and FETCH0_20_vpacheco@AMADEUS.NET30_Fri, 3 Mar 2000 17:56:53 +0100446_us-ascii Dear DB2 Community
I would like to ask you with more Programming experience a solution to a problem we are facing in our site:
Following the syntax definition for the declaration/open/fetch of a cursor, the cursor name needs to be a constant: >>--DECLARE--cursor-name--CURSOR->
We have tried to use a host var on the position of the cursor name, but we had always errors: EXEC SQL DECLARE :cur FOR SELECT ... ; [...]
27737 16 21_Re: Why Not DB2??????10_Ken Koleto23_kkoleto@PAINEWEBBER.COM30_Fri, 3 Mar 2000 12:00:45 -0500292_iso-8859-1 Rohn:
Right on. ROFL!
Ken
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
27754 21 23_Force Commit on Delete?9_Tim Towle22_tim.towle.db2@JUNO.COM30_Fri, 3 Mar 2000 11:58:59 -0500364_us-ascii All,
I am deleting 300,000 rows from a table containing 1,000,000 rows with a single delete statement. This is causing contention problems due to locking with other jobs accessing this table. To alleviate this problem is there a way to force commit after a certain number of rows or after a time interval when executing this delete statement? [...]
27776 25 20_stogroup vol err msg14_Marg Gozdowski27_marg.gozdowski@LA-Z-BOY.COM30_Fri, 3 Mar 2000 12:17:59 -0500376_iso-8859-1 I'm sorry that the err msg did not display correctly from my last e-mail.
I have stogroups that I have built with 3 char. at beginning & 3 numbers at the end.
I tried to create a stogroup with 6 numbers & I received an err msg (under 510) of -104 with error illegal symbol(123456). some symbols that might be legal are: [...]
27802 45 27_Re: Force Commit on Delete?9_Don Alden26_Don.Alden@PKSIS.LEVEL3.COM30_Fri, 3 Mar 2000 10:26:32 -0700676_iso-8859-1 Howdy Tim,
You didn't mention how you are doing this (DB2BATCH,SPUFI etc.). All you need to do is issue the COMMIT statement after N number of rows.
Good Luck, Don Alden
-----Original Message----- From: Tim Towle [mailto:tim.towle.db2@JUNO.COM] Sent: Friday, March 03, 2000 9:59 AM To: DB2-L@RYCI.COM Subject: Force Commit on Delete?
All,
I am deleting 300,000 rows from a table containing 1,000,000 rows with a single delete statement. This is causing contention problems due to locking with other jobs accessing this table. To alleviate this problem is there a way to force commit after a certain number of rows or [...]
27848 41 27_Re: Force Commit on Delete?16_Duane Lee - ATCX22_DLee@MAIL.MARICOPA.GOV30_Fri, 3 Mar 2000 10:37:41 -0700597_- You might be better off doing an unload of the data you want to keep and then LOAD REPLACE back into the table with, possibly, LOG NO followed by imagecopy.
Duane
> -----Original Message----- > From: Tim Towle [SMTP:tim.towle.db2@JUNO.COM] > Sent: Friday, March 03, 2000 9:59 AM > To: DB2-L@RYCI.COM > Subject: Force Commit on Delete? > > All, > > I am deleting 300,000 rows from a table containing 1,000,000 rows with a > single delete statement. This is causing contention problems due to > locking with other jobs accessing this table. To alleviate this problem > is there [...]
27890 63 27_Re: Force Commit on Delete?8_Kurian B23_Bejoy_Kurian@MAY-CO.COM30_Fri, 3 Mar 2000 11:45:32 -0600539_us-ascii You can use REORG with DISCARD option if you are version 6. Bejoy
From: Duane Lee - ATCX on 03/03/2000 11:37 AM
Please respond to DB2 Data Base Discussion List
To: DB2-L@RYCI.COM cc: (bcc: Bejoy Kurian/MDSC/MAYCO)
Subject: Re: Force Commit on Delete?
You might be better off doing an unload of the data you want to keep and then LOAD REPLACE back into the table with, possibly, LOG NO followed by imagecopy. [...]
27954 31 27_Re: Force Commit on Delete?15_Matos, Isalinda22_Isalinda_Matos@BMC.COM30_Fri, 3 Mar 2000 11:42:15 -0600598_ISO-8859-1 Hi, Tim
If you can use several DELETE statements with a WHERE clause to reduce the 300.000 rows into smaller amounts, then you can issue a COMMIT, after every DELETE execution.
What I used to do, whenever possible.. 1. Unload the 300.000 rows to a dataset, if I need them archived on cart 2. Unload the 700.000 rows to another dataset, to reload afterwards 3. Load the 700.000 rows back into the table, with options REPLACE and LOG NO 4. Full Image Copy and runstats It's faster, the tablespace will be reorganized afterwards and although I had the table unavailable, [...]
27986 244 19_Re: Machine Upgrade11_Jon Nolting18_nolting@AMDAHL.COM30_Fri, 3 Mar 2000 09:38:46 -0800423_us-ascii Just a clarification on one of Amdahl's Laws referenced in this thread. In 1967, Dr. Gene Amdahl presented "Validity of the single processor approach to achieving large scale computing capabilities" at the AFIPS conference. This presentation and paper described a relationship that defines a limit on the benefit a given workload can receive from additional processors. This law is defined mathematically by: [...]
28231 69 24_Re: stogroup vol err msg10_Ron Graham23_rgraham@CDBSOFTWARE.COM30_Fri, 3 Mar 2000 11:47:51 -0600553_ISO-8859-1 Hi,
Put the quotes around the volume name:
ALTER STOGROUP MYGROUP ADD VOLUMES('123456') ;
Ron
-----Original Message----- From: marg.gozdowski@LA-Z-BOY.COM Sent: Friday, March 03, 2000 11:21 AM To: DB2-L@RYCI.COM Subject: stogroup vol err msg
-------------------------------------------------------------------------- -- I'm sorry that the err msg did not display correctly from my last e-mail.
I have stogroups that I have built with 3 char. at beginning & 3 numbers at the end. [...]
28301 15 8_LIMITKEY13_Toppins, Mike19_MToppins@UNIPAC.COM30_Fri, 3 Mar 2000 10:56:46 -0700483_windows-1252 Does anyone have any routines they would be willing to share to "translate" LIMITKEY values, more specifically for DATE, TIME, and TIMESTMP type fields. I've written a "DDL" generator and this is the last piece to complete. TIA.
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
28317 77 27_Re: Force Commit on Delete?0_22_Rohn.Solecki@MTS.MB.CA30_Fri, 3 Mar 2000 12:05:23 -0600563_us-ascii No automatic way of saying commit after x rows then continue the same SQLstatement. So if you can't afford the time for unload/reload here are a couple more options: 1 - write a short program using a cursor and looping counter, every X rows do a commit then continue with the cursor. 2 - use several DELETE statements. You'll have to add an additional where clause condition to limit the number of rows qualifying for each delete statement, something like WHERE ... (key > x AND key < y) , then COMMIT, then on the next DELETE, bump limiting values [...]
28395 39 12_Re: LIMITKEY17_Patricia Candella21_pcandella@METLIFE.COM30_Fri, 3 Mar 2000 13:09:22 -0500692_us-ascii Did you consider using the limitkey values in systablepart? Is there more that you need than just taking the data frm the limitkey column in that table?
"Toppins, Mike" on 03/03/2000 12:56:46 PM
Please respond to "DB2 Data Base Discussion List"
To: DB2-L@RYCI.COM cc: (bcc: Patricia Candella/Bsg/MetLife/US) Subject: LIMITKEY
Does anyone have any routines they would be willing to share to "translate" LIMITKEY values, more specifically for DATE, TIME, and TIMESTMP type fields. I've written a "DDL" generator and this is the last piece to complete. TIA. [...]
28435 57 12_Re: LIMITKEY13_Toppins, Mike19_MToppins@UNIPAC.COM30_Fri, 3 Mar 2000 11:13:46 -0700561_windows-1252 Thank you, thank you, thank you.
I didn't see the LIMITKEY column in SYSTABLEPART; I've been using the LIMITKEY column from the SYSINDEXPART table and it's in DB2 internal format instead of external.
-----Original Message----- From: Patricia Candella [mailto:pcandella@METLIFE.COM] Sent: Friday, March 03, 2000 11:09 AM To: DB2-L@RYCI.COM Subject: Re: LIMITKEY
Did you consider using the limitkey values in systablepart? Is there more that you need than just taking the data frm the limitkey column in that table? [...]
28493 26 27_Re: Force Commit on Delete?14_Wayne Driscoll19_wdriscoll@QUEST.COM30_Fri, 3 Mar 2000 10:35:29 -0800542_iso-8859-1 If you use option 1, remember that the cursor must be defined WITH HOLD or else the COMMIT will close the cursor. Wayne Driscoll Product Developer Quest Software Inc. wdriscoll@quest.com Note: All opinions are strictly my own.
No automatic way of saying commit after x rows then continue the same SQLstatement. So if you can't afford the time for unload/reload here are a couple more options: 1 - write a short program using a cursor and looping counter, every X rows do a commit then continue with the cursor. [...]
28520 104 27_Re: Force Commit on Delete?0_19_Tim.Lowe@STPAUL.COM30_Fri, 3 Mar 2000 12:29:34 -0600657_us-ascii Reorg with DISCARD has been available in DB2 V5 with apar PQ19897 since 11/24/98. There have also been some enhancements to this since then.
Kurian B on 03/03/2000 11:45:32 AM
Please respond to DB2 Data Base Discussion List
To: DB2-L@RYCI.COM
cc: (bcc: Tim Lowe/sfm/spc)
Subject: Re: Force Commit on Delete?
You can use REORG with DISCARD option if you are version 6. Bejoy [...]
28625 13 24_Re: stogroup vol err msg0_16_HesterRC@AOL.COM28_Fri, 3 Mar 2000 13:23:31 EST303_US-ASCII Try placing the volid in single quotes, ex: '123456'
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
28639 18 59_Selective Partition Locking in non-Data Sharing environment29_=?iso-8859-1?Q?Jos=E9_Moura?=25_jose.antonio.moura@CGD.PT30_Fri, 3 Mar 2000 18:55:28 -0000542_us-ascii I would like to know if anyone has took any advantage (at what cost ?) from ALTERing partitioned tablespaces to LOCKPART YES.
Or, the real benefits (better concurrency, less XES contention) will only come from it's exploitation under a Data Sharing environment ?
Jose Moura CGD
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
28658 74 27_Re: Force Commit on Delete?10_Ron Graham23_rgraham@CDBSOFTWARE.COM30_Fri, 3 Mar 2000 13:20:43 -0600338_ISO-8859-1 Hi Tim,
A better way would be to delete the rows during a REORG. Our products, CDB/SuperReorg and CDB/RW-Reorg, both allow rows to be deleted and archived during a REORG using a WHERE clause. IBM's has REORG (V6?) also allows to delete rows during a REORG and I'm sure the other REORG vendors have this capability. [...]
28733 41 38_RUNSTATS on large partition tablespace16_Alison Pelletier26_alison_pelletier@YAHOO.COM30_Fri, 3 Mar 2000 11:44:18 -0800447_us-ascii Hello everyone,
We are DB2 OS/390 V5. I need to run runstats for the 1st time on a huge tablespace. I see recommendations to run on tablespace 1st and then you can run on individual partitions (254)...my problem being runstats on the tablespace is very long running and I've exceeded our cpu limit of 120 mins. And when we go prod. with this monster, I'm going to have to allow about 6+ hours to run runstats. My statement is [...]
28775 19 44_Getting Userid from Access using DB2 Connect12_Myron Miller22_myronwmiller@YAHOO.COM30_Fri, 3 Mar 2000 12:04:45 -0800357_us-ascii One of our DBAs is modifying some Access code that connects to DB2 via DB2 Connect. He'd like to be able to retrieve the USERID that the person connected with. Can someone help us with getting that USERID? __________________________________________________ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com [...]
28795 67 36_Re: RUNSTATS on large partition tabl11_Robert Jans21_Jans.Robert@AMSTR.COM30_Fri, 3 Mar 2000 13:26:00 -0700323_- Alison, the obvious solutionwould be to break your statement into several jobs if you can. ie: RUNSTAT parts 1 through 50, 51-100, etc and do the indexes individually. This will allow you to run jobs conncurrently. BTW, the break on part 50 and 51 for a new job is totally arbitary and was meant as an example only. [...]
28863 93 36_Re: RUNSTATS on large partition tabl14_Bistricer, Avi20_BistricerA@CONED.COM30_Fri, 3 Mar 2000 15:28:11 -0500578_- Evaluate using SAMPLE nn Avi Bistricer, DBA 212.460.6695
-----Original Message----- From: Robert Jans [SMTP:Jans.Robert@AMSTR.COM] Sent: Friday, March 03, 2000 3:26 PM To: DB2-L@RYCI.COM Subject: Re: RUNSTATS on large partition tabl
Alison, the obvious solutionwould be to break your statement into several jobs if you can. ie: RUNSTAT parts 1 through 50, 51-100, etc and do the indexes individually. This will allow you to run jobs conncurrently. BTW, the break on part 50 and 51 for a new job is totally arbitary and was meant as an example only. [...]
28957 101 36_Re: RUNSTATS on large partition tabl20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM30_Fri, 3 Mar 2000 15:32:16 -0500482_us-ascii Normally, I would prefer RUNSTAT on entire tablespace with "TABLE ALL INDEX ALL" for the benefits of accesspath. However you could do the following.
1. Run the RUNSTAT on entire tablespace with SHRLEVEL CHANGE and let it take 6 hours. 2. Run the RUNSTAT part by part and over the period of time then update the catalog manually for the consolidated statistics by summing up or by intelligent guess looking from SYSTABLESTATS and SYSINDEXSTATS. 3. Use sampling. [...]
29059 130 42_Re: RUNSTATS on large partition tablespace0_19_Tim.Lowe@STPAUL.COM30_Fri, 3 Mar 2000 14:32:38 -0600469_us-ascii Alison, Since you have not included "TABLE", then you will only gather column statistics for the first column of the indexes. If you want to gather statistics on all columns of the table then you could specify: RUNSTATS dbname.tsname TABLE INDEX(ALL) In order to gather column statistics on all of the columns of the index, you could specify each column name in the table clause. But, I don't do that since it introduces too many possibilities for error. [...]
29190 43 36_Re: RUNSTATS on large partition tabl0_19_Tim.Lowe@STPAUL.COM30_Fri, 3 Mar 2000 15:38:58 -0600575_us-ascii I need to make a correction, about the runstats statement at the end of my last message: RUNSTATS dbname.tsname TABLE INDEX(ALL KEYCARD FREQVAL NUMCOLS 1 COUNT 10 FREQVAL NUMCOLS 2 COUNT 10 FREQVAL NUMCOLS 3 COUNT 10 FREQVAL NUMCOLS 4 COUNT 10 FREQVAL NUMCOLS 5 COUNT 10 FREQVAL NUMCOLS 6 COUNT 10 FREQVAL NUMCOLS 7 COUNT 10 FREQVAL NUMCOLS 8 COUNT 10 FREQVAL NUMCOLS 9 COUNT 10 FREQVAL NUMCOLS 10 COUNT 10) This gathers multi-key column cardinality, as well as the 10 most frequent values for the first column of all indexes, as well as the first 2 through 10 [...]
29234 66 29_Re: NPI's & load performance.14_Michael Hannan24_mhannan@C031.AONE.NET.AU30_Sat, 4 Mar 2000 08:50:45 +1000670_us-ascii Michael,
Make sure NPIs are Reorged with suitable freespace BEFORE the Load Replace Partition. NPI is updated rather than fully rebuilt. Reorg makes the update more sequential.
Performance would be massively better without the NPIs.
From: Michael Hannan
>From: Jessen Michael >Subject: NPI's & load performance. >To: DB2-L@RYCI.COM > >I searched the archives and found similar problems with NPI build times, but >I didn't see any posted solutions (if there are any?)... >This is for DB2 OS/390 V5.1 with a 300+ million row table, 72 partitions & 2 >npis. > >1. Why does the build phase of a load [...]
29301 187 68_Re: A query tuning challenge from the world of DB2 system monitoring14_Michael Hannan24_mhannan@C031.AONE.NET.AU30_Sat, 4 Mar 2000 08:50:51 +1000569_us-ascii Larry,
I'm not absolutely certain if DB2 V6 will fix your access path. Most problem Materializations are avoided. However I am a bit concerned by the subquery against the right hand table. I will have to check my V6 notes since I don't have V6 yet.
DB2 V6 is expected to automatically convert the Full Join back to a Left Join automatically behind the scenes (where it is logically a Left Join) so if the right hand materialization is not avoided, then V6 could actually make it worse. I will come with a way to disable the V6 conversion of [...]
29489 27 31_Re: REASON 00C90101 ERQUAL 53CB0_22_Rohn.Solecki@MTS.MB.CA30_Fri, 3 Mar 2000 15:47:47 -0600421_us-ascii Thanks all for suggestions. Here is what worked in end.
Final solution: CREATE TABLESPACE (with all production values except STOGROUP) CREATE TABLE, omitting last column in the definition ALTER TB to add the missing column WITH DEFAULT NULL create other objects ie views and synonyms, grant to all objects
I could then DSN1COPY OBIDXLAT the Imagecopy from production into the new devel table. [...]
29517 119 30_Re: Disaster recovery software10_Wang.James24_Wang.James@AAA-CALIF.COM30_Fri, 3 Mar 2000 14:10:19 -0800573_us-ascii Try DB2I2 by JRH GoldenState Software, Inc. www.Db2i2.com
James Wang Sr. Systems Programmer Automobile Club of Southern California (714) 850-2851 Wang.James@aaa-calif.com
-----Original Message----- From: Ben Ollo NONLILLY [SMTP:OLLO_BEN_NONLILLY@LILLY.COM] Sent: Wednesday, March 01, 2000 06:12 AM To: DB2-L@RYCI.COM Subject: Disaster recovery software
Does anyone know of other vendors for DB2 disaster recovery software besides CA/Platinum Recovery Analyzer and the BMC product? [...]
29637 22 24_Re: 2GB/4GB PRIQTY LIMIT12_MacLane, Don29_Donald.MacLane@PSS.BOEING.COM30_Fri, 3 Mar 2000 14:54:24 -0800350_- In DB2 UDB for OS/390 V6 From SQL Reference For PRIQTY and SECQTY: For any page size, if integer is greater than 4194304, n is 4194304. That's about 5,825 cylinders! ---
Don MacLane Database Administrator, The Boeing Company Phone: (425) 234-8968 Mail stop: 6C-HW Email: don.maclane@boeing.com Pager: (206) 416-0267 Fax: (425) 234-4470 [...]
29660 42 48_Re: Getting Userid from Access using DB2 Connect0_15_leon@CA.IBM.COM30_Fri, 3 Mar 2000 18:57:42 -0500787_us-ascii How about simply examining the value of CURRENT SQLID and/or USER special registers?
Leon Katsnelson, DB2 Connect Development Manager mailto:leon@ca.ibm.com
Myron Miller on 03/03/2000 03:04:45 PM
Please respond to DB2 Data Base Discussion List
To: DB2-L@RYCI.COM cc: Subject: Getting Userid from Access using DB2 Connect
One of our DBAs is modifying some Access code that connects to DB2 via DB2 Connect. He'd like to be able to retrieve the USERID that the person connected with. Can someone help us with getting that USERID? __________________________________________________ Do You Yahoo!? Talk to your friends online with Yahoo! Messenger. http://im.yahoo.com [...]
29703 74 42_Re: RUNSTATS on large partition tablespace14_Richard Yevich21_ryevich@ATTGLOBAL.NET30_Sat, 4 Mar 2000 09:56:55 -0500344_iso-8859-1 Allison,
I think all your questions were answered except one. Yes, the first time you need to run it on the entire tablespace. After that, you can run on individual partitions. Helps to get it run on the entire tablespace before you load up all the data. Then you can run it on the individual partitions whenever needed. [...]
29778 15 25_Reason for plan valid "N"12_Ranjay Sinha24_rsinha@LOT.TATASTEEL.COM28_Sat, 4 Mar 2000 17:40:55 GMT428_us-ascii Greetings While going through the catalog table sysibm.sysplan I found the valid "N" status of the some plan. Where I can find the reason for which the plan status is valid "N".
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
29794 62 29_Re: NPI's & load performance.12_Isaac Yassin23_yassin@NETVISION.NET.IL30_Sat, 4 Mar 2000 20:35:58 +0200404_us-ascii Hi, Having NPI on big tables can really kill you on LOAD and REORG, try to avoid it. If you want a heretic solution: Instead of having NPI build it as another table where the NPI key+PK is the key, have an index on that. You can update it any time after the load, but of course it will not be updated automatically upon regular insert/update/delete (unless you are on V6 and use triggers). [...]
29857 220 68_Re: A query tuning challenge from the world of DB2 system monitoring14_Michael Hannan24_mhannan@C031.AONE.NET.AU30_Sun, 5 Mar 2000 09:30:05 +1000463_us-ascii Larry,
I can't see anything in DB2 V6 notes that suggests DB2 will need to materialize the right hand table (Nested Table Expression) of the Left Join. So given this, the access path should be O.K. in V6 even if DB2 used a LEFT JOIN rather than a FULL JOIN. However we want to be absolutely safe so I would recommend the following predicate in WHERE clause at the end to confuse DB2 into not doing internal rewrite of FULL JOIN to LEFT JOIN. [...]
30078 41 29_Re: Reason for plan valid "N"17_Linda F. Claussen18_lindafc@NETINS.NET30_Sat, 4 Mar 2000 18:06:46 -0600391_ISO-8859-1 When you DROP an object that a plan is dependent on then the plan is marked a invalid.
The next execution of the plan will attempt to do an auto rebind.
Check your DB2 Admin. Guide for implications for Dropping Objects.
Linda F. Claussen Claussen & Associates, Inc DB2 Consulting and Training lindafc@netins.net http://www.netins.net/showcase/lclaussen/ [...]
30120 48 15_Problem In BIND10_ajay kumar19_db2v5r1@HOTMAIL.COM28_Sun, 5 Mar 2000 17:36:53 IST445_- Hi
We are using BD2 V5R1
The problem is During BIND, the job is simply idle for more than 25 mts, it is not coming out? We are binding the package to 'xxx' collection sometimes the bind will be idle for more than 1 hrs, but the thread status is '*'.
We displayed all threads, we found the corresponding package - plan is running in the CICS, Even we stoped the transaction, the Bind JOB never completed or timed out. [...]
30169 37 16_regarding volume17_ravi kumar hassan18_ravibh@HOTMAIL.COM28_Sun, 5 Mar 2000 08:09:03 PST443_- Hey All,
i have a question on space for the segmented tablespace.
what is max size of the segmented tablespace. i think each VSAM dataset can be max of 2gb that is apporximately 1 volume. we allocated like one stogroup has one volume. and i allocated priqty 400 cyls and 100 cyls secqty. suppose if the dataset(A0001) exceeds 2gb what will happen.....? does it says volume is full and fails or does go to the next volume. [...]
30207 24 15_Dates in Keys ?22_Lindsay, William (GES)24_WLindsay@EXCHANGE.ML.COM30_Sun, 5 Mar 2000 11:17:35 -0500344_iso-8859-1 Does anyone have any comments good, or bad as to using dates in primary keys ? We are looking at keeping data historically and we are considering using from to dates to distinguish the rows.
Thanks
Bill Lindsay Group Employee Services 401K Distribution 732 764-3464 (voice) 732 764-5902 (fax) 888 MERRIL-0 (pager) [...]
30232 31 68_Re: A query tuning challenge from the world of DB2 system monitoring16_Daniel Boulanger19_dboulang@CA.IBM.COM30_Sun, 5 Mar 2000 13:00:32 -0500306_us-ascii Good day,
We're running DB2 V5 in our shop and by curiousity I tried the query that was listed in the previous message.
It does not seem to run as fast as you mentioned. Perhaps I did not apply the modification properly, so I would appreciate if you could post the final query. [...]
30264 66 14_Re: IXF Format12_MILLS,Robert26_robert.mills@DEWRSB.GOV.AU30_Mon, 6 Mar 2000 14:34:27 +1100470_iso-8859-1 David,
The OS/390 IXF format is described in the QMF Application Development Guide.
There was a program called IXFUTIL, written by IBM Denmark I believe, which we used to use to convert between mainframe and PC IXF formats. We used it several years ago with DB2 on OS/2 IXF files, I haven't tried it on a recent UDB IXF file (and we don't use OS/2 anymore!). These days we use Import/Export through DB2 Connect if we have to transfer data. [...]
30331 38 32_DB2 UDB v6.1 Certification tests14_Romeo B Titong26_rbtitong@ACISYSTEMS.COM.PH30_Mon, 6 Mar 2000 14:13:14 +0800380_us-ascii Hi!
i'll be taking certification tests for DB2 UDB 6.1. beside the training manuals, i am also reading the DB2 UDB Certification Guide V5.2 book.
will there be a big difference between the certification tests for 6.1 and 5.2? do i need to buy the certification guide book for 6.1? or do i wait until the certification tests for v7.0 will be available? [...]
30370 23 21_To Know about DB2-UDB8_Narahari34_mulbagal.narahari@TATAINFOTECH.COM30_Mon, 6 Mar 2000 12:17:51 +0530542_us-ascii Hi I have joined newly to the group to benefit/upgrade my knowledge about DB2
.If U have any material/presentations on the Latest DB2-Universal Database pls share with me.Also I am interested to know about the WEB Enabling concepts with DB2 as Database server.
Regards Narahari
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
30394 146 36_Re: RUNSTATS on large partition tabl0_18_mebert@AMADEUS.NET30_Mon, 6 Mar 2000 09:34:16 +0100575_us-ascii I have just 2 days ago run this type of statement on all our production tablespaces:
RUNSTATS dbname.tsname TABLE ALL SAMPLE nn INDEX(ALL KEYCARD FREQVAL NUMCOLS 1 COUNT 10 FREQVAL NUMCOLS 2 COUNT 10 FREQVAL NUMCOLS 3 COUNT 10) SHRLEVEL CHANGE
Our largest TSs are about 35 GB; the statement above took a maximum of 330 mins CPU. The SAMPLE was adjusted such that I would sample 50 MB+5% of the TS. Some experiments I did showed that sampling does not reduce the CPU time required as much as one might hope for. In one case, no sampling took 17.36 [...]
30541 127 38_Re: UDB Backup/Restore strange problem7_revadee18_revadeev@SCB.CO.TH30_Mon, 6 Mar 2000 19:20:00 +0700343_iso-8859-1 Dear, Greg Have you try to map network drive from the server that keep backup image file?. (Map on d:/backup to your workstation in case you have one image timestamp.) And you try to restore db from this path. And another case is your version of db2 in your workstation. If it has lower than the backup image, it doesn't work. [...]
30669 19 23_DRDA - Three part names10_Robert Ord21_robertord@HOTMAIL.COM28_Mon, 6 Mar 2000 13:26:57 GMT482_- What release of DB2 OS/390 supports three part names in DRDA ? Can an APAR be applied to version 5.1 ?
Regards
Rob ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
30689 63 68_Re: A query tuning challenge from the world of DB2 system monitoring14_Michael Hannan24_mhannan@C031.AONE.NET.AU30_Tue, 7 Mar 2000 00:48:39 +1000420_us-ascii Daniel,
I will leave it to Larry to answer your question since I never analysed what the original query was doing, and I never tested it nor had any knowledge of indexes or access paths.
Its worth mentioning however, as my college Mohan Chandekar pointed out to me, the query had 'WITH UR;' at the end which can only work efficiently if your shop has has type 2 indexes on the DB2 Catalogue. [...]
30753 174 68_Re: A query tuning challenge from the world of DB2 system monitoring0_34_L-IS.Kirkpatrick@MUTUALOFOMAHA.COM30_Mon, 6 Mar 2000 08:01:32 -0600540_us-ascii I am posting the revised query below. There are some shop specific items that you should be aware of:
1) There is a predicate in the query to look at objects with a VCAT that begins with "P". The systems people in our shop manage the backup of the system objects (the catalog and directory). All of the application tablespaces use a DSN naming convention that begin with a "P". Since I am not a systems person but I am a DBA manageing applications, I am interested in the "P" type datasets here (Your milage may vary). [...]
30928 41 27_Re: DRDA - Three part names12_craig patton21_prgpatton@HOTMAIL.COM28_Mon, 6 Mar 2000 06:03:45 PST715_- Rob, this support goes back all the way to version 3 of DB2. You do NOT need an apar for V5.1. Are you having a particular problem?
HTH, Craig Patton Independent DB2 DBA
>From: Robert Ord >Reply-To: DB2 Data Base Discussion List >To: DB2-L@RYCI.COM >Subject: DRDA - Three part names >Date: Mon, 6 Mar 2000 13:26:57 GMT > >What release of DB2 OS/390 supports three part names in DRDA ? Can an APAR >be applied to version 5.1 ? > >Regards > >Rob >______________________________________________________ >Get Your Private, Free Email at http://www.hotmail.com > >================================================ >To change your subscription options [...]
30970 51 27_Re: DRDA - Three part names15_Toine Michielse18_vndobtm@US.IBM.COM30_Mon, 6 Mar 2000 15:02:41 +0000479_us-ascii Hello Rob,
It's DB2 v6. There's no APAR for DB2 v5.
Regards,
Toine Michielse DB2 S/390 Advocate Santa Teresa Laboratory Mobile: +31 6 537 23 256 Email: vndobtm@us.ibm.com Lotus: Toine Michielse/Santa Teresa/Contr/IBM@IBMUS
Robert Ord @RYCI.COM> on 03/06/2000 01:26:57 PM
Please respond to DB2 Data Base Discussion List
Sent by: DB2 Data Base Discussion List [...]
31022 73 27_Re: DRDA - Three part names12_craig patton21_prgpatton@HOTMAIL.COM28_Mon, 6 Mar 2000 06:08:51 PST697_- OK, I'll be stupid here, WHAT IS BEING SUPPORTED in V6 that is NOT already supported with 3 part names since V3????? Am I missing the point of this question??
Craig Patton
>From: Toine Michielse >Reply-To: DB2 Data Base Discussion List >To: DB2-L@RYCI.COM >Subject: Re: DRDA - Three part names >Date: Mon, 6 Mar 2000 15:02:41 +0000 > >Hello Rob, > >It's DB2 v6. There's no APAR for DB2 v5. > >Regards, > >Toine Michielse >DB2 S/390 Advocate >Santa Teresa Laboratory >Mobile: +31 6 537 23 256 >Email: vndobtm@us.ibm.com >Lotus: Toine Michielse/Santa Teresa/Contr/IBM@IBMUS > > >Robert Ord @RYCI.COM> [...]
31096 109 27_Re: DRDA - Three part names15_Toine Michielse18_vndobtm@US.IBM.COM30_Mon, 6 Mar 2000 15:17:26 +0000440_us-ascii Craig,
DRDA for three part names... Previously, using three part names (location.creator.tabname) would always result in using DB2's private protocol. With V6, you can specify which protocol to use DRDA/private through a bind option.
Regards,
Toine Michielse DB2 S/390 Advocate Santa Teresa Laboratory Mobile: +31 6 537 23 256 Email: vndobtm@us.ibm.com Lotus: Toine Michielse/Santa Teresa/Contr/IBM@IBMUS [...]
31206 32 52_DB2 UDB DBA LONGTERM Contract Opportunity - Illinois0_18_jtulman@MAAINC.COM30_Mon, 6 Mar 2000 09:21:17 -0500543_us-ascii Ladies and Gentlemen,
A client of mine located in the Central Illinois area is looking for a highly qualified UDB DB2 DBA for a minimum of 1.5 YEAR on a consulting basis. The client is looking for this person to install UDB ver 6.1 as well as Administer version 5.2 of UDB on the NT platforms. Any experience with AIX/RS6000 or UDB EE or EEE as well as IBM Certifications would be a HUGE plus but not required. Client is looking for this person to start no later than 3 weeks from interview. We can work with both Corp to [...]
31239 48 30_Strange Problem - DB2 Stogroup9_Chu, Pius14_ChuP@CONED.COM30_Mon, 6 Mar 2000 09:26:21 -0500576_- Our DB2 application objects are STOGROUP defined. We have five 3380 packs (XX15, XX16, XX17, XX01, XX02) that defined under DB2 STOGROUP "XYZ". The packs were defined original with three packs XX15, XX16, and XX17. Later it was alter to add packs XX01 and XX02. Packs XX15, XX16 and XX17 were 100 percent full while packs XX01 and XX02 were 50 percent full when we're getting the following error messages for any objects belonging to this STOGROUP "XYZ" that need to extent. Finally we removed the five packs from STOGROUP "XYZ" and add the same to the order of (XX01, [...]
31288 92 34_Re: Strange Problem - DB2 Stogroup13_Karen Broskey20_kbroskey@METLIFE.COM30_Mon, 6 Mar 2000 09:51:14 -0500515_us-ascii Hi! I ran into this same problem in my old shop. We were always told that the order of the packs in the stogroup did not matter. Yet we encountered the exact same situation as you. When we moved the packs with space to the beginning of the volumes list, our datasets were able to extend. Another thing to keep in mind is that supposedly DB2 will only try a certain number of times to get an extent (I'm not sure how many). So, if the 3 packs that are 100% full were tried first, DB2 may have given up [...]
31381 25 68_Re: A query tuning challenge from the world of DB2 system monitoring16_Daniel Boulanger19_dboulang@CA.IBM.COM30_Mon, 6 Mar 2000 09:50:12 -0500375_us-ascii Thanks for the prompt response.
It was "obvious", with some reading, that the VCATNAME specification was proper to your shop. On the other hand, the two user indexes requirement, was less evident, at least for people like me who understand what SQL is, but does not have the necessary expertise to determine what is precisely happening behind the scene. [...]
31407 24 34_Re: Strange Problem - DB2 Stogroup9_Stan Hoey24_stan_hoey@CIRCLE-DL2.COM30_Tue, 7 Mar 2000 14:50:42 -0800338_iso-8859-1 Pius, I don't think that there is any way that you can tell from SYSIBM.SYSSTOGROUP or SYSIBM.SYSVOLUMES the order in which DB2 will use the candidate volumes in a Storage Group. You have to remember the order in which they were added.
I don't think that this is "strange behaviour". This is "working as designed". [...]
31432 81 34_Re: Strange Problem - DB2 Stogroup0_22_BILL_GALLAGHER@PHL.COM30_Mon, 6 Mar 2000 09:59:23 -0500483_us-ascii We experienced similar problems a few years ago. I opened up a PMR with IBM, asking why the extents were not going to the packs in the stogroup that had available space. They never really were able to give me a satisfactory answer . . . instead, they strongly suggested that we move to SMS and define the DB2 stogroups as VOLUMES('*'). We ended up doing this, and have not had any problems since. We've been running with SMS managing our DB2 storage for 2-3 years now. [...]
31514 30 28_Hardware compression (again)15_Jim Lewandowski22_jlewand@STARNETINC.COM31_Tue, 29 Feb 2000 17:20:17 -0600424_us-ascii I hate to bring up the dreaded subject but I found out some more information via a DASD hardware vendor's benchmarks.
On a CMOS machine (9672-R75? - 7 CPs), the X'B263' CMPSC hardware instruction (used by DB2 compression) is NOT a full hardware implementation (i.e. no true "coprocessor"). The 33% CPU overhead (33% more CPU per row retrieval) that I had quoted from my own analysis IS in the ballpark. [...]
31545 55 34_Re: Strange Problem - DB2 Stogroup0_22_BILL_GALLAGHER@PHL.COM30_Mon, 6 Mar 2000 10:05:20 -0500439_us-ascii I remember about 6 or 7 years ago, somebody posted a message to this listserv about how do do a DSN1PRNT of tablespace DSNDB06.SYSGROUP, and actually trace through the pointers in SYSIBM.SYSVOLUMES to see the actual order in which the volumes have been defined for a stogroup, and the order in which DB2 will consider the volumes. Unfortunately, I don't have this info anymore, but does this ring any bells for anybody else? [...]
31601 24 26_Start procedure automation12_Fertaki Gina20_GFertaki@EUROBANK.GR30_Mon, 6 Mar 2000 17:14:21 +0200573_ISO-8859-7 Hello , we have a DRDA connection between our AS/400 and mainframe (DB2 Version 4). We have AS/400 programs that run an SQL CALL to a stored procedure on DB2. If this procedure, for some reason, has abended the AS/400 program receives a -471 (SQL Call for stored procedure failed ). We then have to manually issue a -START PROCEDURE command and rerun the application program. The question is what we can do to automate this process in production, so that when the program on the AS/400 receives an SQL code of -471,it will trigger a -START PROCEDURE in DB2 [...]
31626 35 36_Re: RUNSTATS on large partition tabl0_19_Tim.Lowe@STPAUL.COM30_Mon, 6 Mar 2000 09:16:03 -0600661_us-ascii Richard, I don't agree that with runstats, "the first time you need to run it on the entire tablespace. After that, you can run on individual partitions.".
If Allison's intent is to get statistics on an entire tablespace, she can gather these statistics one partition at a time. O course, until statistics have been gathered on all partitions, warning messages like this will be issued: DSNU623I -DB2T DSNUSTBL - SYSIBM.SYSTABLES CATALOG NOT UPDATED WITH AGGREGATE STATISTICS FOR ...tablename... BECAUSE SOME PARTITIONS HAVE NO VALID STATISTICS When runstats have been gathered on all partitions, then runstats on the "last partition" will [...]
31662 32 20_LOG YES on Utilities16_Goldston, Leland25_LGoldston@SSS.SUNGARD.COM30_Mon, 6 Mar 2000 10:19:49 -0500431_iso-8859-1 We have a client who has hired a DBA to set up the DB2 infrastructure. The DBA has completely rejected all utilities which have LOG NO and instead wants LOG YES. This decision is solely based on the disaster recovery process this person has set up. The DR process involves restoring DFDSS volume dumps and log forward apply. We agree that for strictly DR purposes this is the best and most pristine way to proceed. [...]
31695 62 24_Re: LOG YES on Utilities20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM30_Mon, 6 Mar 2000 10:35:42 -0500532_us-ascii If imagecopies are not taken, even I would be scared but LOG YES is not the solution. If you are on V5 and UP, you could run inline imagecopies during LOAD or REORG. If you have big tablespaces then LOG YES is big NO.
I was working in a shop, which was doing no imagecopies but solely relying on DFDSS volume dumps. In such situations probably concurrent copy is an alternative except DSN1COPY is not allowed using concurrent copy back-up. Ideal thing would be reconsider your DR process and setup imagecopies. [...]
31758 29 28_user-defined structured type11_Alfred Moos21_moos@FH-HEIDELBERG.DE30_Mon, 6 Mar 2000 16:44:43 +0100316_iso-8859-1 Hello all,
I am looking for a good description of the DB2-concept of "user-defined structured types". Very appreciated is also a cookbook-like description from the idea to the declaration of types and the corresponding tables to loading and using the tables.
Best regards Alfred Moos
31788 21 12_Online Reorg0_26_mmetcalf@NOTES.STATE.NE.US30_Mon, 6 Mar 2000 10:13:59 -0600522_us-ascii I have a 28 partition table with 22,582,640 totals rows. There are 2 non partitioned indexes and 1 npi. I ran an online reorg for the npi and caused about 14 -904's during the BUILD2 phase which ran approx 3.53 minutes. It was my understanding that the -911's and -904's would most likely occur during the switch phase which only ran for 6 seconds. Is this because during the BUILD2 phase the index dataset doesn't get deleted and redefined, rather just updated/corrected? Thanks in advance for any insight. [...]
31810 68 24_Re: LOG YES on Utilities12_Allen, Colin31_Colin.Allen@ABBEYNATIONAL.CO.UK30_Mon, 6 Mar 2000 16:37:00 -0000547_- We have a similar recovery strategy on one system (without the log apply), but would not consider using LOG YES. There are several options depending on when you do the reorgs and what down time you have, eg. 1. VOL DUMP - REORG - VOLDUMP while the system is down 2. Use a mixed strategy where you roll the catalog forward then look for IC's more recent than your last VOLDUMP and do 'normal' recoveries on these, followed by LOG ONLY on the rest. We do IC's before/after reorgs to facilitate recovery if the reorgs fail and to speed things [...]
31879 35 59_Selective Partition Locking in non-Data Sharing environment0_18_cmunson@US.IBM.COM30_Mon, 6 Mar 2000 09:39:38 -0700635_us-ascii Jose, Selective Partition Locking (SPL) was designed for a data sharing environment to reduce parent L-lock contention and improve data sharing efficiency, however, there is a valid use in a non-data sharing environment. For non-data sharing environments if you experience lock escalation then SPL can help. SPL uses the partition as the parent in the parent/child locking hierarchy so when you escalate with SPL the gross lock is obtained on the partition, not the tablespace, thus allowing access to other parts. If you don't have an issue with escalation then SPL shouldn't be considered for non-data sharing systems. [...]
31915 15 26_DataPropagator environment12_Martin, Paul22_Paul.Martin@ECOLAB.COM30_Mon, 6 Mar 2000 10:45:44 -0600467_iso-8859-1 Anyone have OS/390 V6 Capture running on OS/390 DB2 V5 with AIX V5 Apply?
In our DB2 V6 migration plan (from V5) we would like to make it simple by migrating Capture to V6 before we migrate DB2 to version 6.
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
31931 39 16_Re: Online Reorg12_Greg Popwell27_greg.popwell@MOTION-IND.COM30_Mon, 6 Mar 2000 10:56:27 -0600697_iso-8859-1 In my experience and according to the Utility Guide, you are correct. The BUILD2 phase "corrects" NPI's if REORG TABLESPACE or INDEX PART SHRELEVEL CHANGE or REFERENCE is specified.
Greg Popwell
-----Original Message----- From: mmetcalf@NOTES.STATE.NE.US [mailto:mmetcalf@NOTES.STATE.NE.US] Sent: Monday, March 06, 2000 10:14 AM To: DB2-L@RYCI.COM Subject: Online Reorg
I have a 28 partition table with 22,582,640 totals rows. There are 2 non partitioned indexes and 1 npi. I ran an online reorg for the npi and caused about 14 -904's during the BUILD2 phase which ran approx 3.53 minutes. It was my understanding that the -911's and -904's would most [...]
31971 55 24_Re: LOG YES on Utilities19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM30_Mon, 6 Mar 2000 10:59:01 -0600363_iso-8859-1 Leland, If you've got a fairly large table on test, benchmark reorg with log no and log yes and then project those figures to the 7+M row table on prod. Time to recover can be very important. Would it be possible to show how a full forward recovery from a recent image copy would be much faster than from an older DFDSS volume dump and log apply. [...]
32027 93 24_Re: LOG YES on Utilities16_Goldston, Leland25_LGoldston@SSS.SUNGARD.COM30_Mon, 6 Mar 2000 12:09:36 -0500295_iso-8859-1 We have a "conversion" system with similar volume. I am in the process of benchmarking both, LOG YES and NO. As for time to recover, I don't believe I can benchmark. All I can go by on this is past experience in recovering a similarly sized system and extrapolating to the 7+ M. [...]
32121 110 16_Re: Online Reorg0_18_mebert@AMADEUS.NET30_Mon, 6 Mar 2000 18:06:15 +0100582_us-ascii I assume what you did was run OLR for a single partition of a partitioned TS which had at least one non-partitioned index (NPI). During the BUILD2 phase, the Logical partition of the NPI corresponding to the partition of the TS you are reorging is locked (the logical partition displays as Lnn, i.e. with leading Ell, when you do a -DIS DB(db) TS(ts)). This can cause timeouts in SQL trying to use this logical partition of the NPI. The logical partition consists of all RIDs (not Keys!) in the index addressing rows in the partition you reorged. Because you probably [...]
32232 104 24_Re: LOG YES on Utilities16_Goldston, Leland25_LGoldston@SSS.SUNGARD.COM30_Mon, 6 Mar 2000 12:14:48 -0500557_iso-8859-1 We are on V5 PUT9902.
All other clients use what has been recommended. Basically, allow volume dumps to be restored, followed by BSDS recovery from archive log at quiesce point, catalog and directory recovery from image copy plus forward log apply, then application recoveries via image copy and forward log apply. We are in the process of expanding on this with placing archive log on disk and volume dumps of the catalog, directory and BSDS just after system weekly image copies. However, we have not put this in production as yet. [...]
32337 77 36_Re: RUNSTATS on large partition tabl14_Richard Yevich21_ryevich@ATTGLOBAL.NET30_Mon, 6 Mar 2000 11:14:12 -0500480_iso-8859-1 Tim,
Perhaps, I have not tested it recently. Was told that there is additional information updated when you run the whole that will not be updated if you just do partition by partition. Yes, once you do partition by partition for all partitions, the aggregate statistics needed for access path selection will be complete. When I get a chance in a couple weeks, I will verify what each methods does and if all stats get covered by just doing all partitions. [...]
32415 55 32_Re: Hardware compression (again)14_Richard Yevich21_ryevich@ATTGLOBAL.NET30_Mon, 6 Mar 2000 11:14:15 -0500645_iso-8859-1 Jim,
Are you saying that there is a 25% to 33% increase in CPU in DB2 for a single compressed row to be decompressed? Please clarify.
Regards, Richard +===+===+===+===+====+ ryevich@attglobal.net
> -----Original Message----- > From: DB2 Data Base Discussion List [mailto:DB2-L@RYCI.COM]On Behalf Of > Jim Lewandowski > Sent: Tuesday, February 29, 2000 6:20 PM > To: DB2-L@RYCI.COM > Subject: Hardware compression (again) > > > I hate to bring up the dreaded subject but I found out some more > information via a DASD hardware vendor's benchmarks. > > On a CMOS machine (9672-R75? - 7 CPs), the X'B263' CMPSC [...]
32471 66 27_Re: DRDA - Three part names13_Lynne Flatley17_LFlatley@NEFN.COM30_Mon, 6 Mar 2000 12:16:51 -0500636_us-ascii How does this affect the CONNECT statement, then? Will this 'eliminate' the need for the CONNECT statement?
> -----Original Message----- > From: Toine Michielse [SMTP:vndobtm@US.IBM.COM] > Sent: Monday, March 06, 2000 10:03 AM > To: DB2-L@RYCI.COM > Subject: Re: DRDA - Three part names > > Hello Rob, > > It's DB2 v6. There's no APAR for DB2 v5. > > Regards, > > Toine Michielse > DB2 S/390 Advocate > Santa Teresa Laboratory > Mobile: +31 6 537 23 256 > Email: vndobtm@us.ibm.com > Lotus: Toine Michielse/Santa Teresa/Contr/IBM@IBMUS > > > Robert Ord @RYCI.COM> on 03/06/2000 01:26:57 PM > [...]
32538 61 24_Re: LOG YES on Utilities0_23_Mike_Levine@TEKHELP.NET30_Mon, 6 Mar 2000 12:21:22 -0500562_us-ascii Hi,
This is not the reply you were looking for but...we run nine concurrent reorg jobs of very large tables with LOG YES that must complete in a three hour monthly window. We did notice a small increase in run time from LOG NO but since we did not have to take after reorg image copies (shrlevel reference) the overall down time was actually reduced! Also, since the reorgs are run during times of low system activity there were no logging problems. Sure we use additional tapes for archives but we would have had to use the tapes anyway to [...]
32600 41 16_Package versions0_19_mike.holmans@BT.COM30_Mon, 6 Mar 2000 16:53:09 -0000584_- I'm trying to convert our production system to using packages rather than old-style plans, and we've hit a problem with versioning.
The precompiler options which are used are VERSION(AUTO) and LEVEL(1).
For the purposes of this question, you should assume that LEVEL(1) cannot be changed.
When we get an updated program, we try to bind the package ACTION(REPLACE), and it refuses on the grounds that something already exists with the same CONTOKEN (of 1). Presumably this is because the existing package has a different VERSION to the new one being bound. [...]
32642 104 27_Re: DRDA - Three part names15_Toine Michielse18_vndobtm@US.IBM.COM30_Mon, 6 Mar 2000 18:59:18 +0000477_us-ascii Hello Lynne,
Using DRDA for three part names does not change the difference between system directed access (three part names) and application directed access (using CONNECT statements).
Regards,
Toine Michielse DB2 S/390 Advocate Santa Teresa Laboratory Mobile: +31 6 537 23 256 Email: vndobtm@us.ibm.com Lotus: Toine Michielse/Santa Teresa/Contr/IBM@IBMUS
Lynne Flatley @RYCI.COM> on 03/06/2000 05:16:51 PM [...]
32747 119 24_Re: LOG YES on Utilities12_craig patton21_prgpatton@HOTMAIL.COM28_Mon, 6 Mar 2000 10:20:54 PST516_- I would think that LOG YES would be the exception, rather than the rule. In cases like Mike's it can make sense to use LOG YES and skip the IC. My concern is with any recoveries that need to be done and the extra time it will take to read the LOGs. This can add significantly to the time it takes to recover data (and we all know that users don't care about recovery, until it's needed, then they want it done sooner than: 'right now!'). Mike did not mention whether he starts with an IC, but no matter what, [...]
32867 126 27_Re: DRDA - Three part names12_craig patton21_prgpatton@HOTMAIL.COM28_Mon, 6 Mar 2000 10:25:07 PST715_- Toine,
Does this mean that for DRDA, you would CONNECT, and STILL use the 3 part name?
Craig Patton
>From: Toine Michielse >Reply-To: DB2 Data Base Discussion List >To: DB2-L@RYCI.COM >Subject: Re: DRDA - Three part names >Date: Mon, 6 Mar 2000 18:59:18 +0000 > >Hello Lynne, > >Using DRDA for three part names does not change the difference between >system directed access >(three part names) and application directed access (using CONNECT >statements). > >Regards, > >Toine Michielse >DB2 S/390 Advocate >Santa Teresa Laboratory >Mobile: +31 6 537 23 256 >Email: vndobtm@us.ibm.com >Lotus: Toine Michielse/Santa Teresa/Contr/IBM@IBMUS > > [...]
32994 63 20_Re: Package versions9_Rob Crane22_racrane@CONCENTRIC.NET30_Mon, 6 Mar 2000 11:46:32 -0700591_us-ascii The version has nothing to do with what DB2 is executing. The version is not even passed in the parm list, the consistency token tells DB2 which package to execute. The Version parameter allows different versions of a package to coexist within the same collection.
Using the level keyword is a bad idea. Instead of letting DB2 generate the contoken, you are supplying the value for contoken through the level precompiler option. If the SQL did in fact change, then your level should have been changed as well. If the program change was not tied to SQL modules, then you [...]
33058 163 27_Re: DRDA - Three part names15_Toine Michielse18_vndobtm@US.IBM.COM30_Mon, 6 Mar 2000 19:39:51 +0000411_us-ascii Craig,
I'm currently at a customer and do not have time for an extensive answer. I'd have no problem with answering you question as soon as I have a little more time. Please send me an email to remind me.
Regards,
Toine Michielse DB2 S/390 Advocate Santa Teresa Laboratory Mobile: +31 6 537 23 256 Email: vndobtm@us.ibm.com Lotus: Toine Michielse/Santa Teresa/Contr/IBM@IBMUS [...]
33222 30 31_DB2 Recovery TIMEOUTs - PROBLEM15_Jeffery A Price17_jprice@IPALCO.COM30_Mon, 6 Mar 2000 13:45:25 -0500605_us-ascii Hello List,
I have a question for you. We experience a great deal of TIMEOUTs on SYSDBASE when attempting to run as little as four (4) concurrent recovery jobstreams. There is an APAR concerning DEADLOCKs, but it's not us (PQ25543). The workaround for our shop was to increase the value of ZPARM UTIMOUT until we could successfully run concurrent recovery jobstreams w/o TIMEOUTs. I was wondering if, in V6, if there was any improvement in DB2 concurrency - especially in the RECOVER arena? Increasing UTIMOUT allows the jobs to run successfully, but they sure do take a long time to [...]
33253 95 23_[Fwd: Package versions]9_Rob Crane22_racrane@CONCENTRIC.NET30_Mon, 6 Mar 2000 11:59:53 -0700405_us-ascii Forgot to answer your original question. Yes you are correct, you can use the VERSION(user supplied) and LEVEL parms together to get around the issue of VERSION(AUTO) using the level value instead of the timestamp that would normally be picked up.
If you specify version(auto) then the current timestamp will be used for the version identifier unless you also specify the LEVEL parm, which
33349 145 24_Re: LOG YES on Utilities0_23_Mike_Levine@TEKHELP.NET30_Mon, 6 Mar 2000 13:50:27 -0500611_us-ascii Hi Craig
Good points. If the data changes they really should consider taking full image copies. Daily shrlevel change image copies followed by an application wide QUIESCE would be the least disruptive from an availability standpoint and provide good recovery. Also, going back to our reorg procedure. Before the reorgs, we do take incremental, shrlevel change image copies followed by QUIESCE. If you really wanted to, after the LOG YES reorgs, you could take a full shrlevel change image copy followed by QUIECE but we choose to let the full image copies occur later that day as part of a [...]
33495 40 54_IDUG Headquarters Update: Asia Pacific 2000 Conference4_IDUG12_IDUG@SBA.COM30_Mon, 6 Mar 2000 13:00:17 -0600408_iso-8859-1 Dear IDUG® Member -
Save the Dates!
IDUG 6th Annual Asia Pacific Conference November 15-17, 2000 Melbourne, Australia
Join the professional DB2 community in Melbourne this November. Attending IDUG Asia Pacific will help you and your organisation stay ahead of the learning curve, build your network of DB2 contacts, and provide solutions to all your database challenges! [...]
33536 99 45_Furthering a topic (aka beating a dead horse)14_William Gannon41_wgannon@INET-SERVICES.PALMBEACH.K12.FL.US30_Mon, 6 Mar 2000 14:00:17 -0500317_us-ascii This 'situation' - (Alter TB but never doing Reorg ... resulting in VB length) has been mentioned in the past and I am curious, Has anyone written a Catalog query to isolate objects in this quasi state? Is it even possible to identify them or is it an internal mechanism not available to the user / DBA [...]
33636 49 32_Re: Hardware compression (again)12_Isaac Yassin23_yassin@NETVISION.NET.IL30_Mon, 6 Mar 2000 21:39:02 +0200576_x-user-defined Hi Jim, Absolutely yes. Isaac Yassin
Jim Lewandowski wrote: > > I hate to bring up the dreaded subject but I found out some more > information via a DASD hardware vendor's benchmarks. > > On a CMOS machine (9672-R75? - 7 CPs), the X'B263' CMPSC hardware > instruction (used by DB2 compression) is NOT a full hardware > implementation (i.e. no true "coprocessor"). The 33% CPU overhead (33% > more CPU per row retrieval) that I had quoted from my own analysis IS in > the ballpark. > > I will be getting more formal numbers from the hardware [...]
33686 26 32_Re: Hardware compression (again)12_Vernon, John27_John.Vernon@TWC.STATE.TX.US30_Mon, 6 Mar 2000 14:13:14 -0600601_- Hi Jim, Absolutely yes.
John Vernon DB2 System Administrator Texas WorkForce Commission 101 East 15th Street Austin, Texas 78778 Vice-President, Heart of Texas DB2 Users Group Email: john.vernon@twc.state.tx.us jvernonrr@aol.com Phone: 512-936-3170 Fax: 512-936-3190 "Ability may get you to the top, but it takes character to keep you there."
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
33713 21 21_DB2 Hangs at Shutdown16_Dempsey, Michael20_MDempse@UNCH.UNC.EDU30_Mon, 6 Mar 2000 15:14:16 -0500351_iso-8859-1 Hey Folks... Pretty nebulous question, but we tried to bring down DB2 with the normal shutdown (@stop DB2). When it took over 30 minutes and nothing, we issued the stop with a force. Then waited another 30 minutes and finally cancelled. It generally takes about 10 minutes to come down under normal circumstances. Any ideas? Thanks... [...]
33735 39 49_Re: Furthering a topic (aka beating a dead horse)14_Wayne Driscoll19_wdriscoll@QUEST.COM30_Mon, 6 Mar 2000 12:33:10 -0800569_iso-8859-1 Bill, While this would be a great thing to have, it isn't an easy thing to provide. The information required isn't stored in the catalog, but rather it resides in the DBD for the database that the table resides in. In a table DBD entry there are 2 bits that identify the allowability of variable length records. One of them (I don't have a diag guide in front of me) identifies that the table will ALLOW variable records, and the second identifies that the table EXPECTS variable records. If there are variable length columns in the table, or the table [...]
33775 50 25_Re: DB2 Hangs at Shutdown12_Hall, Robert25_Robert.Hall@OCC.TREAS.GOV30_Mon, 6 Mar 2000 15:41:14 -0500584_iso-8859-1 Hi Michael,
Did you check to see if there were any active threads? We have seen problems at shutdown when there are active threads that cannot be purged. In some cases, the threads were from QMF under CICS, but CICS was down and DB2 seemed to be waiting for a response from a non-existent CICS region. On another occasion, I logged on to TSO early one Saturday morning and DB2 promptly shut down. Apparently, DB2 was waiting for a SPUFI thread from my userid to terminate before DB2 would shutdown. We have also had cases where distributed threads were causing [...]
33826 32 30_DBA Job Opportunity in Atlanta13_penny andrews19_pzandrews@YAHOO.COM30_Mon, 6 Mar 2000 12:49:51 -0800319_us-ascii Dear List....
My company has an opening for a UDB DBA in Atlanta to be my counterpart there. The position is in support of a large data warehouse, and the company is a large East Coast financial institution. The offices are in the Buckhead area, and are quite posh. UDB DBA experience is required. [...]
33859 185 25_AW: DB2 Hangs at Shutdown16_Roland Schiradin28_SchiradinR@ALTE-LEIPZIGER.DE30_Mon, 6 Mar 2000 21:44:13 +0100524_iso-8859-1 Michael,
I had the same question months ago. Here is what I got ---------------------------------------------------------------------------- -----
Long append warning.
You did not indicate your current release of DB2. There are major improvements in stop and restart times in Version 4 and in Version 5.
The most severe problems are where there are large numbers of allocations. We have some customers running with more than 15,000 concurrent today, but without official support. [...]
34045 19 49_Re: Furthering a topic (aka beating a dead horse)25_Edward(Ed) J. Finnell,III23_efinnell@SEEBECK.UA.EDU30_Mon, 6 Mar 2000 15:03:21 -0600418_us-ascii Seems like it would be easier? Had one application that was down for almost a day because they had dropped a table. I tried to restore and restart their application. Forget the abend, but it did everytime. Called support center and they guy was really sharp. "You're trying to reload a table that's been extended." So, went back and recreated the table and had the apps do the extend then do the reload. [...]
34065 32 57_Dynamic File Allocation from a COBOL DB2 Stored Procedure17_ROBERT HARRINGTON23_RHARRIN3@EMAIL.USPS.GOV30_Mon, 6 Mar 2000 16:10:34 -0600365_iso-8859-1 Subject: Dynamic Allocation of a flat file in a DB2 Stored Procedure.
I was wondering if anyone had a way to dynamically allocate a file in a DB2 Stored Procedure. We have archived data from DB2 to flat files and have stored enough information in a DB2 table to determine the name of the file that contains the data that we want to retrieve. [...]
34098 51 16_Re: Online Reorg11_Kwan, James18_James_Kwan@BMC.COM30_Mon, 6 Mar 2000 15:37:57 -0600617_ISO-8859-1 Michelle,
It is most likely that you will experience timeouts during build2 phase than in switching phase. Switching phase normally last few seconds if there are only few datasets involved. However build2 phase can last quite a long time. I have seen many cases that the reorg phases of one partition of partitioned tablespace took only few minutes, but the build2 phase lasted for almost 1/2 hour. What happened during OLR is the logical partition of the NPI is copied to a temporarily dataset. After the switching phase, these logical NPI datasets have to be updated to the original datasets [...]
34150 33 61_Re: Dynamic File Allocation from a COBOL DB2 Stored Procedure12_Jim Harrison12_jimh@QIS.NET30_Mon, 6 Mar 2000 17:31:55 -0500393_us-ascii Wow - talk about timing. I just put together a SP to do this in our shop as a debugging aid. It calls an assembler routine I wrote years ago to do dynamic allocation w/ SVC 99. I can send you source for both if you are interested, or, better yet, if I can find the directions, I'll post them to the other list because the last time I offered this up I had close to 100 requests. [...]
34184 29 38_Re: UDB Backup/Restore strange problem15_Lynette Roberts21_llroberts@KSCABLE.COM30_Mon, 6 Mar 2000 16:30:35 -0600508_- We have also received this error and have found it to be due to errors in the command parameters used for "source-database-alias" and/or "target-database-alias".
In our environment, we are usually attempting to restore the database to a new name, which causes our confusion. Verify that in your restore command, the first parameter is the "source-database-alias" -- the alias of the source database from which the backup was taken, and that the second parameter is the "target-database-alias". [...]
34214 38 61_Re: Dynamic File Allocation from a COBOL DB2 Stored Procedure14_James Campbell29_James.Campbell@HANCORP.COM.AU30_Tue, 7 Mar 2000 10:10:06 +1000587_- There is a program on the CBT tape (www.cbttape.org) in file 352 named R062A10/S062A10 to do dynamic allocations from high level languages.
/* standard disclaimer */ James Campbell DBA Hansen Corporation, Doncaster +61 3 9840 3864 James.Campbell@Hancorp.com.au > > Subject: Dynamic Allocation of a flat file in a DB2 Stored Procedure. > > I was wondering if anyone had a way to dynamically allocate a file in > a DB2 Stored Procedure. We have archived data from DB2 to flat files > and have stored enough information in a DB2 table to determine the > name of the file that [...]
34253 92 34_Re: Strange Problem - DB2 Stogroup14_James Campbell29_James.Campbell@HANCORP.COM.AU30_Tue, 7 Mar 2000 11:21:38 +1000474_- Quite simple:
In the catalog, the actual format of some rows is:
6 bytes header; 1 or more pointers to child, sibling and parent rows. These are standard 4 byte RID values; selectable data.
SYSSTOGROUP has a pointer to the first SYSVOLUMES child row SYSVOLUMES has two pointers: the first points to the next sibling row, the second points to the prior sibling row. Both point back to the parent SYSSTOGROUP when there isn't a next/prior sibling. [...]
34346 111 34_Re: Strange Problem - DB2 Stogroup0_29_Greg.Palgrave@BANKWEST.COM.AU30_Tue, 7 Mar 2000 09:19:17 +0800321_us-ascii Bill,
The DB2 Diagnosis Guide and Reference has all the internal structure/pointer definitions for Catalog etc. I believe the order of the volumes is also hidden in the DBD, but I've never bothered hunting for it. Following pointer chains does not rank highly on my list of "fun things to do".... :) [...]
34458 152 20_Re: Package versions27_Thakral, Deepak Kumar (CTS)24_TDeepak@CHN.CTS-CORP.COM30_Tue, 7 Mar 2000 01:06:04 -0500322_iso-8859-1 Rob,
1.If you say that version has nothing to do with what DB2 is executing, then how do we find out what is the consistency token for the package we are executing. Does have package have a consistency token by default. If I query SYSIBM.SYSPACKAGES, the contoken column looks something like this : [...]
34611 51 34_Re: DBA Job Opportunity in Atlanta11_Bill Pettit18_billp@ORMUTUAL.COM30_Mon, 6 Mar 2000 22:07:36 -0800560_iso-8859-1 send me an address to mail one to.
> -----Original Message----- > From: DB2 Data Base Discussion List [mailto:DB2-L@RYCI.COM]On Behalf Of > penny andrews > Sent: Monday, March 06, 2000 12:50 PM > To: DB2-L@RYCI.COM > Subject: DBA Job Opportunity in Atlanta > > > Dear List.... > > My company has an opening for a UDB DBA in > Atlanta to be my counterpart there. The position > is in support of a large data warehouse, and > the company is a large East Coast financial > institution. The offices are in the Buckhead > area, and are quite [...]
34663 77 37_Cursor name on DECLARE/OPEN and FETCH0_20_vpacheco@AMADEUS.NET30_Tue, 7 Mar 2000 08:21:18 +0100432_us-ascii Dear all
This is a repost of my note of last Friday. As I did not get any reply I decided to repost it. Can someone bring some light on this issue?
Much Obliged Vitor Pacheco ---------------------- Forwarded by Vitor Pacheco/MUC/AMADEUS on 07/03/2000 08:20 ---------------------------
From: Vitor Pacheco (ADP/DS/SO) on 03/03/2000 17:56
To: DB2 Data Base Discussion List - [...]
34741 42 25_Re: DB2 Hangs at Shutdown12_Nurmi, Seppo23_Seppo.Nurmi@RIKSBANK.SE30_Tue, 7 Mar 2000 09:02:29 +0100557_- We had similar problems before. The reason was distributed threads from workstations leaved hanging on. DB2 tries to wait out them and that can sometimes take a very long time. There are installation parameters that can be consired to be changed (can't rememer just now which ones) that automatically time out idle threads. We now let operators check that all workstations are logged off, and force them log off whenever necessary. We take down all CICS systems first before we even try to shut down DB2. I hope these experiences are helpful. /Seppo [...]
34784 26 36_TAPE Usage with SYSUT1 in Utilities.32_Cuneyt Goksu (Garanti Teknoloji)22_CuneytG@GARANTI.COM.TR30_Tue, 7 Mar 2000 10:11:39 +0200309_- Hello List,
I want to use TAPE instead of DASD in SYSUT1 for Utilities for example in RECOVER INDEX. I know it's optional to use SYSUT1 from the restartablity and I/O perspective.
Has anybody experinced to use TAPE for SYSUT1 DDs ? Is it possible to use ? I want to hear some experiences [...]
34811 62 40_Re: TAPE Usage with SYSUT1 in Utilities.16_Rasmussen, Steen22_Steen.Rasmussen@CA.COM30_Tue, 7 Mar 2000 08:30:36 -0000334_iso-8859-1 Hi Goksu -
I always have used TAPE for SYSUT1 when the objects are quite big (how big, BIG is - depends upon your own definition). Sometimes this is the ONLY way to get the job done. Expecially for the NPI's on partitioned tablespaces, using TAPE instead of DASD can save some reruns due to lack of DASD-space. [...]
34874 188 20_Re: Package versions19_Nagaraj Pudukotai S33_nagaraj.pudukotai@HPSBLR.SOFT.NET30_Tue, 7 Mar 2000 14:05:17 +0530512_iso-8859-1 Hi Deepak
1. Use HEX scalar function on CONTOKEN to get the hex value of it. 2. Use LEVEL precompiler option if you do not change the SQL stmts in your program. The value you supply in this will be used by DB2 to generate the consistency token. You can do away with your bind. 3. It is the consistency token which is used to pick the correct package by DB2. You can the find the first 4 or the last 4 bytes of the consistency token scattered in the load module(a COBOL pgm's load module). [...]
35063 60 25_Re: DB2 Hangs at Shutdown9_A. Bohacz22_abohacz@OZEMAIL.COM.AU30_Tue, 7 Mar 2000 19:55:41 +1100386_iso-8859-1 Michael, A few areas you might like to look at:
(a) Is DB2 really hanging? If you look at the DB2 started tasks, you may see one of them using quite a bit of CPU while DB2 shuts down. If DBM1 is using CPU you might like to look at whether DB2 is performing an abnormal number of writes to SMF. Review whether the means by which DB2 writes to SMF is satisfactory. [...]
35124 39 32_Connecting Two DB2 V5 Via TCP/IP13_Dieter Scharf18_D.Scharf@SIGNAL.DE30_Tue, 7 Mar 2000 10:58:48 +0100314_us-ascii I have a problem in connecting two OS390 DB2 V5.1 via TCP/IP. During installation, we made exactly every step of chapter 3.3 in the 'Installation Guide'. When i try to connect from DB2 (DSNX) to DB2(DSNO) in Spufi, i receive the following error message. Does anybody know the reason for this error ? [...]
35164 115 41_Re: Cursor name on DECLARE/OPEN and FETCH10_ajay kumar19_db2v5r1@HOTMAIL.COM28_Tue, 7 Mar 2000 15:47:11 IST681_- HI
I think, You have to use Dynamic SQL to accomplish this. Use PREPARE and EXECUTE statement to open the cursor.
regards karhtik
>From: vpacheco@AMADEUS.NET >Reply-To: DB2 Data Base Discussion List >To: DB2-L@RYCI.COM >Subject: Cursor name on DECLARE/OPEN and FETCH >Date: Tue, 7 Mar 2000 08:21:18 +0100 > >Dear all > >This is a repost of my note of last Friday. As I did not get any reply I >decided >to repost it. >Can someone bring some light on this issue? > >Much Obliged >Vitor Pacheco >---------------------- Forwarded by Vitor Pacheco/MUC/AMADEUS on 07/03/2000 >08:20 --------------------------- > >From: Vitor Pacheco [...]
35280 81 40_Re: TAPE Usage with SYSUT1 in Utilities.13_McDonald, Ian32_Ian.McDonald@ABBEYNATIONAL.CO.UK30_Tue, 7 Mar 2000 11:35:00 -0000524_iso-8859-1 Cuneyt
We use it all the time here. The DASD bods are very stingy with supplying loads of spare DASD for utility/work type functions so sod 'em we get round it by utilising cartridges. They tell us there is enough spare DASD out there but being a cynical old DBA I don't believe them and I don't want to be woken up a 2 a.m. every time a REORG job goes bang! We occasionally get moaned at for using 3 cart drives but if the REORGs are planned then it's no problem and if it isn't planned then we need [...]
35362 52 61_Re: Dynamic File Allocation from a COBOL DB2 Stored Procedure19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM30_Tue, 7 Mar 2000 06:24:19 -0600611_iso-8859-1 Jim, I'd be interested. Might as well post them to the document site at DB2-L. Thank You, Rick Davis "This e-mail and any files transmitted with it are the property of SBC, are confidential, and are intended solely for the use of the individual or entity to whom this e-mail is addressed. If you are not one of the named recipient(s) or otherwise have reason to believe that you have received this message in error, please notify the sender at 314-235-6854 and delete this message immediately from your computer. Any other use, retention, dissemination, forwarding, printing, or copying of this [...]
35415 64 36_AW: Connecting Two DB2 V5 Via TCP/IP16_Roland Schiradin28_SchiradinR@ALTE-LEIPZIGER.DE30_Tue, 7 Mar 2000 13:26:25 +0100643_iso-8859-1 1128 X'468' ECONNREFUSED The attempt to connect was rejected.
Check the port with the Netstat command
Roland
> -----Ursprüngliche Nachricht----- > Von: Dieter Scharf [SMTP:D.Scharf@SIGNAL.DE] > Gesendet am: Dienstag, 7. März 2000 10:59 > An: DB2-L@RYCI.COM > Betreff: Connecting Two DB2 V5 Via TCP/IP > > I have a problem in connecting two OS390 DB2 V5.1 via TCP/IP. > During installation, we made exactly every step of chapter 3.3 in the > 'Installation Guide'. > When i try to connect from DB2 (DSNX) to DB2(DSNO) in Spufi, > i receive the following error message. > Does anybody know the reason [...]
35480 93 34_Re: Strange Problem - DB2 Stogroup19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM30_Tue, 7 Mar 2000 07:02:38 -0600592_iso-8859-1 James, DB2 for OS/390 -- huh? Regards, Rick Davis "This e-mail and any files transmitted with it are the property of SBC, are confidential, and are intended solely for the use of the individual or entity to whom this e-mail is addressed. If you are not one of the named recipient(s) or otherwise have reason to believe that you have received this message in error, please notify the sender at 314-235-6854 and delete this message immediately from your computer. Any other use, retention, dissemination, forwarding, printing, or copying of this e-mail is strictly prohibited." [...]
35574 122 41_Re: Cursor name on DECLARE/OPEN and FETCH13_Lockwood Lyon23_Lockwood_Lyon@AMWAY.COM30_Tue, 7 Mar 2000 08:05:20 -0500433_us-ascii Did I miss something obvious in this one? Your statement is
EXEC SQL DECLARE :cur FOR SELECT ... ;
Shouldn't it be
EXEC SQL DECLARE :cur CURSOR FOR SELECT ... ;
-- Lockwood Lyon Amway Corp 616.787.1068
Please respond to DB2 Data Base Discussion List
Sent by: DB2 Data Base Discussion List
To: DB2-L@RYCI.COM cc: [...]
35697 60 25_Re: DB2 Hangs at Shutdown19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM30_Tue, 7 Mar 2000 07:08:23 -0600590_iso-8859-1 Michael, All great replies so far. One other possibility is that DB2 was in some long backout processing for a failed thread. Also, for -STOP DB2 (implied MODE(QUIESCE)) to have a chance to work, all address spaces that use DB2 should be shutdown first.
HTH, Rick Davis "This e-mail and any files transmitted with it are the property of SBC, are confidential, and are intended solely for the use of the individual or entity to whom this e-mail is addressed. If you are not one of the named recipient(s) or otherwise have reason to believe that you have received this [...]
35758 92 25_Re: DB2 Hangs at Shutdown13_Lockwood Lyon23_Lockwood_Lyon@AMWAY.COM30_Tue, 7 Mar 2000 08:13:51 -0500690_us-ascii Rick:
Interesting. So you'd recommend doing a DSNC STOP from any CICS connected to the DB2 first, prior to the shutdown?
-- Lockwood Lyon Amway Corp 616.787.1068
Please respond to DB2 Data Base Discussion List
Sent by: DB2 Data Base Discussion List
To: DB2-L@RYCI.COM cc:
Subject: Re: DB2 Hangs at Shutdown
Michael, All great replies so far. One other possibility is that DB2 was in some long backout processing for a failed thread. Also, for -STOP DB2 (implied MODE(QUIESCE)) to have a chance to work, all address spaces that use DB2 should be shutdown first. [...]
35851 102 24_Re: LOG YES on Utilities0_23_Mike_Levine@TEKHELP.NET30_Tue, 7 Mar 2000 08:17:04 -0500571_us-ascii Hi Rick,
Yes, you would be able to recover the tablespace (to current - end of log) using the image copy and the log. However, if you wanted a consistent point in time (RECOVER TORBA/TOLOGPOINT) you would also have to take a QUIECE after the SHRLEVEL CHANGE image copy. You could also take just a QUIESCE after the REORG LOG YES and in the (hopefully) unlikely event of a DASD failure, your only exposure is that recovery would take longer until your next full image copies are taken (i.e., DB2 would start with the before REORG image copy and apply [...]
35954 90 25_Re: DB2 Hangs at Shutdown19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM30_Tue, 7 Mar 2000 07:45:15 -0600608_iso-8859-1 Lockwood, My reply about when DB2 should/could be STOPped was a practical answer. Obviously, if no address spaces were calling DB2, and DB2 was in a position to be STOPped, a -STOP DB2 (implied MODE(QUIESCE)) would work regardless of other address spaces that were up. Regret any confusion.
Regards, Rick Davis "This e-mail and any files transmitted with it are the property of SBC, are confidential, and are intended solely for the use of the individual or entity to whom this e-mail is addressed. If you are not one of the named recipient(s) or otherwise have reason to believe that [...]
36045 70 34_Re: DBA Job Opportunity in Atlanta13_penny andrews19_pzandrews@YAHOO.COM30_Tue, 7 Mar 2000 05:49:17 -0800554_us-ascii Bill...
Send it to me: pzandrews@yahoo.com.
Thanks!
--- Bill Pettit wrote: > send me an address to mail one to. > > > -----Original Message----- > > From: DB2 Data Base Discussion List > [mailto:DB2-L@RYCI.COM]On Behalf Of > > penny andrews > > Sent: Monday, March 06, 2000 12:50 PM > > To: DB2-L@RYCI.COM > > Subject: DBA Job Opportunity in Atlanta > > > > > > Dear List.... > > > > My company has an opening for a UDB DBA in > > Atlanta to be my counterpart there. The position > > is in support [...]
36116 124 24_Re: LOG YES on Utilities19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM30_Tue, 7 Mar 2000 07:52:50 -0600365_iso-8859-1 Michael, Thanks for reply. One more question; did you observe the elapsed time you saved by comparing the cost of logging to the LOG NO/COPY SHRLEVEL REFERENCE? Finally, <>, I once had so many Image Copies running at one time that a couple timed out trying to write SYSCOPY, geez. [...]
36241 48 40_Re: TAPE Usage with SYSUT1 in Utilities.0_23_Mike_Levine@TEKHELP.NET30_Tue, 7 Mar 2000 08:59:03 -0500650_iso-8859-1 Hi,
Works great if you have a VTS or similar tape system! Otherwise slow but still works.
Regards,
Michael Levine Premier Data Services, Inc.
>Hello List, > >I want to use TAPE instead of DASD in SYSUT1 for Utilities for example in >RECOVER INDEX. >I know it's optional to use SYSUT1 from the restartablity and I/O >perspective. > >Has anybody experinced to use TAPE for SYSUT1 DDs ? Is it possible to use ? >I want to hear some experiences > >Thank you so much. > >Cüneyt GÖKSU >GARANTI TEKNOLOJI >Tel : 0090-212-657 0404 Ext: 1218 >mailto:Cuneytg@garanti.com.tr >Visit : http://www.kultur.gov.tr [...]
36290 47 25_Re: DB2 Hangs at Shutdown14_Massimo Scarpa16_mscarpa@CESVE.IT30_Tue, 7 Mar 2000 15:03:56 +0100342_us-ascii Hi all folks.
Michael,
is this hang up real or apparent ? In our shop when we stop DB2(V 5) , we can see that the DB2
normal termination message appears SDSF system log (i.e. DB2 is stopped and the shutdown
was correct; no thread waiting, no job running, etc etc) but DB2 is still 'live' in SDSF (DA [...]
36338 72 25_Re: DB2 Hangs at Shutdown9_Ted Pesta18_tpesta@AMQUEST.COM30_Tue, 7 Mar 2000 09:15:51 -0500379_iso-8859-1 This is real ! You have described our situation exactly. I've asked our MVS person to look at it and searched IBMLINK. No answers so far.
Ted Pesta AmQUEST, Inc.
-----Original Message----- From: Massimo Scarpa [mailto:mscarpa@CESVE.IT] Sent: Tuesday, March 07, 2000 9:04 AM To: DB2-L@RYCI.COM Subject: Re: DB2 Hangs at Shutdown
Hi all folks. [...]
36411 18 25_AW: DB2 Hangs at Shutdown16_Roland Schiradin28_SchiradinR@ALTE-LEIPZIGER.DE30_Tue, 7 Mar 2000 15:17:02 +0100445_- Do *NOT* cancel DB2, this can cause a lot of errors. It can happen what you need an IPL to restart DB2
If DB2 needs a long time to shutdown, check your SMF-dataset-usage ("D SMF").
Roland
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
36430 23 25_Re: DB2 Hangs at Shutdown12_Allen, Colin31_Colin.Allen@ABBEYNATIONAL.CO.UK30_Tue, 7 Mar 2000 14:33:00 -0000370_- Perhaps we are lucky, but cancelling the IRLM brings DB2 down and has never caused any problems on the odd occasion we needed to do it.
Colin.
----------
Do *NOT* cancel DB2, this can cause a lot of errors. It can happen what you need an IPL to restart DB2
If DB2 needs a long time to shutdown, check your SMF-dataset-usage ("D SMF"). [...]
36454 53 29_Re: AW: DB2 Hangs at Shutdown14_Massimo Scarpa16_mscarpa@CESVE.IT30_Tue, 7 Mar 2000 15:38:26 +0100321_us-ascii I don't know if Roland's post is a suggestion following my post, but I agree, in general it's not a
good idea to cancel DB2. I experimented some (small) problems during restart.
What we do is to get rid of 'ghost' address spaces ('DB2 garbage'). We are pretty sure that
DB2 is stopped. [...]
36508 17 35_Need a screen-width plan_table view0_22_kmulcahy@CHECKFREE.COM30_Tue, 7 Mar 2000 09:43:41 -0500259_us-ascii I need an SQL view or statement that can be used from SPUFI to compress a plan_table row to about the width of the screen. I lost mine changing jobs and it was very useful. With V5 I could no longer fit the row on the screen, but, it was close! [...]
36526 82 25_Re: DB2 Hangs at Shutdown13_Dan S Brogdon22_Dan.S.Brogdon@AEXP.COM30_Tue, 7 Mar 2000 07:51:22 -0700420_- We use the following: stop ddf mode force stop proc action reject - wait 1 min stop db2 mode force - wait 2 min modify Irlm
We haven't had to modify IRLM yet- we have 48 DB2 systems. DB2 comes down within 2 min - usually around 1 min.
From: RD8246%MOMAIL.SBC.COM@Internet on 03/07/2000 06:11 AM To: DB2-L%RYCI.COM@Internet cc: (bcc: Dan S Brogdon) Subject: Re: DB2 Hangs at Shutdown [...]
36609 130 39_Re: Need a screen-width plan_table view16_Robert Blackwell31_robert.blackwell@PIEDMONTNG.COM30_Tue, 7 Mar 2000 09:58:52 -0500757_us-ascii SELECT PROGNAME CONCAT ' ' CONCAT SUBSTR(COLLID,01,06) CONCAT ' ' CONCAT SUBSTR(VERSION,01,04) AS PROGNAME_COLL_VERS ,SUBSTR(DIGITS(QUERYNO),06,05) CONCAT '-' CONCAT SUBSTR(DIGITS(QBLOCKNO),05,01) CONCAT '-' CONCAT SUBSTR(DIGITS(PLANNO),05,01) AS QURY#_B_P ,CASE METHOD WHEN 0 THEN ' ' WHEN 1 THEN 'N' WHEN 2 THEN 'M' WHEN 3 THEN 'S' WHEN 4 THEN 'H' END AS M ,CASE SUBSTR(CREATOR,03,04) WHEN '0 ' THEN ' ' CONCAT SUBSTR(CREATOR,01,03) CONCAT '.' CONCAT SUBSTR(TNAME,01,08) WHEN ' ' THEN ' ' ELSE SUBSTR(CREATOR,01,06) CONCAT '.' CONCAT SUBSTR(TNAME,01,08) END AS TABLE_NAME ,ACCESSTYPE CONCAT ' ' CONCAT SUBSTR(ACCESSNAME,01,08) CONCAT ' ' CONCAT SUBSTR(DIGITS(MATCHCOLS),05,01) AS AT_INDXNME_MC ,INDEXONLY CONCAT ' ' CONCAT SORTN_UNIQ CONCAT [...]
36740 45 25_Re: DB2 Hangs at Shutdown19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM30_Tue, 7 Mar 2000 09:06:55 -0600395_iso-8859-1 Colin, I concur; never yet had a problem with "C"ancelling xxxxIRLM. Never had fragmented MVS storage problems that can be related. There might be an issue related to doing it via SDSF though. Also, "C"ancelling xxxxIRLM has, so far, brought down all other DB2 ASIDs regardless of what they were doing. Finally, it provides a way to restart DB2 with BACKOUT=NO when appropriate. [...]
36786 214 16_FW: iehlist REXX12_Allen, Colin31_Colin.Allen@ABBEYNATIONAL.CO.UK30_Tue, 7 Mar 2000 15:23:00 -0000475_iso-8859-1 I'm not sure this made it the first time as our external mail is a bit erratic. ---------- Heres the REXX below, we have REXXTOOLS which extends standard REXX, so you will have to write your own sort routine to replace STEMSORT.
Incidentally, I run a prior batch job which generates the JCL for me by running a QUICKREF space report in batch and stripping out the volume names so I don't have to remember to change the job each time a volume is added. [...]
37001 134 16_Re: Online Reorg0_14_dcreed@CSC.COM30_Tue, 7 Mar 2000 09:23:59 -0600601_us-ascii I was just putting on some old maintenance and noticed some documentation that may apply to this thread. If not.. oops.
It appeared that UQ25008 provided a new parm to the online reorg
Online Reorg results in SQL -911 code failures if there are updaters at the time OLR attempts to drain writers near the end of the log phase until the end of the reorganization. Analysis has shown that in certain cases these -911 failures represent deadlocks between transactions and OLR because of the type of claim (or drain) requested along with the order that objects are claimed (or [...]
37136 95 40_Re: TAPE Usage with SYSUT1 in Utilities.0_19_Tim.Lowe@STPAUL.COM30_Tue, 7 Mar 2000 09:26:14 -0600574_us-ascii You specified "for utilities", but I think that it might be better to break this down by utility.
Tape use for SYSUT1 in recover/rebuild index is not a good idea in my opinion since this dataset can be eliminated instead to speed up the elapsed time a great deal by writing directly to the sortwork datasets. This does mean that restart is from the beginning, but the majority of times that I have needed restartability for this utility are due to space abends on this dataset anyway. So, eliminating the use of this dataset is really more of a benefit [...]
37232 48 25_Re: DB2 Hangs at Shutdown14_Massimo Scarpa16_mscarpa@CESVE.IT30_Tue, 7 Mar 2000 16:31:54 +0100441_us-ascii I'm wondering if a CANCEL DB2 (or IRLM) may became the 'standard' DB2 shutdown
procedure. In my opinion : absolutely NO.
Ok, during time and with production consolidation (and needs) this may became the 'standard'
DB2 shutdown procedure.....until something goes wrong (remeber first Murphy's law).
And I heard terrifying histories about multiple IPLs, CRC restarts, internal inconsistency errors at [...]
37281 46 25_AW: DB2 Hangs at Shutdown16_Roland Schiradin28_SchiradinR@ALTE-LEIPZIGER.DE30_Tue, 7 Mar 2000 16:38:15 +0100585_iso-8859-1 Cancel DB2xxxx is not the solution, look for the real problem
Roland
> -----Ursprüngliche Nachricht----- > Von: Allen, Colin [SMTP:Colin.Allen@ABBEYNATIONAL.CO.UK] > Gesendet am: Dienstag, 7. März 2000 15:33 > An: DB2-L@RYCI.COM > Betreff: Re: DB2 Hangs at Shutdown > > Perhaps we are lucky, but cancelling the IRLM brings DB2 down and has > never > caused any problems on the odd occasion we needed to do it. > > Colin. > > ---------- > > Do *NOT* cancel DB2, this can cause a lot of errors. > It can happen what you need an IPL to restart DB2 > [...]
37328 25 26_batch program connectivity4_IBM116_IBM1@UBIL.COM.TR30_Tue, 7 Mar 2000 17:43:40 +0200525_- help help.
does anybody know if is it possible for a batch program to connect to more than one DB in DB2 Server for VSE?
thanks for a speedy help.
Aurora Dell'Anno Systems Engineer IBM Software Support e-mail ibm1@ubil.com.tr aurora.dellanno@sistinf.it
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
37354 66 25_Re: DB2 Hangs at Shutdown24_Marchese, Frank D, CSCIO18_fdmarchese@ATT.COM30_Tue, 7 Mar 2000 10:15:31 -0500549_iso-8859-1 We had to do this at another place that I worked at and using an earlier release of DB2 (maybe 2.3 or 3). It works because IRLM is not an optional addr space for DB2. However, we still viewed it as a 'last resort'. I don't believe we felt any more or less confident than if we had tried to cancel DB2MSTR...we were still nervous on that first START after the cancel. When time allowed, we would do that START using ACCESS(MAINT) so we could look around at the general health of everything before the general population could have at [...]
37421 145 42_Re: LOG YES on Utilities - what is logged?0_23_Mike_Levine@TEKHELP.NET30_Tue, 7 Mar 2000 11:22:31 -0500339_us-ascii Hi Rick,
No I don't recall the actual numbers. What I did notice was that when LOG YES was specified, the REORGs did not run much longer than with LOG NO. I don't believe that before and after images of the rows are logged during reorg like other logging. Anyone know offhand exactly what is logged during REORG? Tks. [...]
37567 211 42_Re: LOG YES on Utilities - what is logged?16_Goldston, Leland25_LGoldston@SSS.SUNGARD.COM30_Tue, 7 Mar 2000 11:37:58 -0500473_iso-8859-1 I ran several REORGs on a 2M+ row segmented tablespace. There was no significant difference between CPU, CWU, EXCP, etc. The only difference I saw, of course, was the log activity. The amount of log CI's written was nearly 100% of the active pages of the tablespace. We have DATA CAPTURE CHANGES for the table (and many others). I would like to know what is logged (as below) and does DATA CAPTURE CHANGES have an effect on the record written to the log ?? [...]
37779 61 29_Re: AW: DB2 Hangs at Shutdown0_23_Mike_Levine@TEKHELP.NET30_Tue, 7 Mar 2000 11:37:35 -0500664_iso-8859-1 Hi,
I have seen performance monitors like OMEGAMON hang up a normal shutdown until the started task is stopped.
Regards,
Michael Levine Premier Data Services, Inc.
>Cancel DB2xxxx is not the solution, look for the real problem > >Roland > > >> -----Ursprüngliche Nachricht----- >> Von: Allen, Colin [SMTP:Colin.Allen@ABBEYNATIONAL.CO.UK] >> Gesendet am: Dienstag, 7. März 2000 15:33 >> An: DB2-L@RYCI.COM >> Betreff: Re: DB2 Hangs at Shutdown >> >> Perhaps we are lucky, but cancelling the IRLM brings DB2 down and has >> never >> caused any problems on the odd occasion we needed to do it. >> >> Colin. [...]
37841 138 20_ODBC and Access 200023_Donald W. Woodruff, Jr.30_dwoodruf@CI.NEWPORT-NEWS.VA.US30_Tue, 7 Mar 2000 11:48:56 -0500393_iso-8859-1 Hello all.
I am having a problem linking to a DB2 table (DB2 v5.1 for MVS) using the ODBC provided with DB2 Connect.
Description: Using Access 2000--LINK a table using a ODBC data source (configured via CAE). When I try to view the table in Access, I get the following: "Microsoft Access can't open the table in Datasheet view." I can IMPORT the table just fine. [...]
37980 150 20_Re: Package versions9_Rob Crane22_racrane@CONCENTRIC.NET30_Tue, 7 Mar 2000 10:10:28 -0700455_iso-8859-1 Deepak,
1) Try the following query, customize to your needs (this assumes source code in COBOL). Enter your package name where you see the ????????.
SELECT A.COLLID AS COLLECTION, A.NAME AS PACKAGE, HEX(A.CONTOKEN) AS DBRM_TOKEN, SUBSTR(HEX(A.CONTOKEN),9,8) || SUBSTR(HEX(A.CONTOKEN),1,8) AS LOADLIB_TOKEN, A.VERSION, A.PDSNAME FROM SYSIBM.SYSPACKAGE A WHERE A.NAME = ‘????????’ ORDER BY A.NAME, A.COLLID, A.VERSION WITH UR; [...]
38131 46 23_DB2PM - Am I going mad?22_Leslie Pendlebury-Bowe43_Leslie.Pendlebury-Bowe@CCMAIL.EU.SONY.CO.JP30_Tue, 7 Mar 2000 16:13:55 +0000636_ISO-8859-1 Guys needs some help here.
I issue the following request to DB2PM :
STATISTICS REDUCE INTERVAL(30) BOUNDARY(60) INCLUDE (SUBSYSTEMID(ssid1,ssid2)) REPORT FROM(03/07/00,09:59:59.00) TO(03/07/00,10:59:59.00) LAYOUT(LONG) DDNAME(STRPTDD) SCOPE(GROUP) ORDER(INTERVAL) EXEC
.....and assumed that I would end up with group information for those group bufferpools that have been connected to during the period requested.
The thing is I only get information for GBP0 .. and it is driving me mad as I know that there have been at least three other Group Buffer Pools connected to by these members ... [...]
38178 248 42_Re: LOG YES on Utilities - what is logged?14_Massimo Scarpa16_mscarpa@CESVE.IT30_Tue, 7 Mar 2000 18:17:03 +0100411_us-ascii Have you never experimented a rollback following an abend of utility ? Just a curiosity.
I made some test (many & many years ago) and with small number or rows there's little
difference in elapsed (LOG NO/YES). The difference was greater when I had some million rows
and if utility abended. The rollback was very long. But it was DB2 v 3, however. Rollback may be faster now. [...]
38427 286 24_Re: ODBC and Access 200015_Craig Wahlmeier27_cwahlmeier@DATA-TRONICS.COM30_Tue, 7 Mar 2000 11:31:45 -0600356_iso-8859-1 Donnie,
I have had various problems with Access 2000 and DB2 for OS/390. All were fixed when I applied MDAC 2.5, (Microsoft Data Access Components), to the client machine. I assume you are not running Windows 2000, (it ships with MDAC 2.5). You can download it for free from Microsoft at - http://www.microsoft.com/data/download.htm [...]
38714 248 42_Re: LOG YES on Utilities - what is logged?20_Pillay, Venkat (PCA)20_venkat_pillay@ML.COM30_Tue, 7 Mar 2000 12:33:05 -0500428_iso-8859-1 With DATA CAPTURE CHANGES the log changes are big. The entire row is logged. The logical logging is charged to User TCB but physical goes to SSAS as SRB (refer: "system performance matrix" - Joel Goldstein). The log buffers are flooded, your job may not see the problems but others on the system might be overwhelmed. The logs will be archived faster than they should. You won't see the EXCP on your job output. [...]
38963 25 22_UDB 5.0 Crash Recovery14_Robin Wooyhand22_woolyhr@NATIONWIDE.COM30_Tue, 7 Mar 2000 12:46:24 -0600495_- I have UDB 5.0 NT Personal Edition installed on a laptop.
The database seems to go through a Crash Recovery at least once a day.
The function code indicates that it is having trouble with data protection and that it could have something to do with the locking or the logging.
My first hunch is that the user is not exiting from the application properly and then shutting down the machine which would cause an automatic Crash Recovery the next day when they power up. [...]
38989 19 39_DB2 for MVS and Command Level Interface17_Giorgio De Nunzio25_giorgio.denunzio@UNILE.IT30_Tue, 7 Mar 2000 20:27:10 +0100454_US-ASCII Hi all. I need to know if DB2 for MVS, in any version, has Call Level Interface (CLI), which is found in other operating systems DB2 versions.
Thank you very much!
Giorgio De Nunzio
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
39009 73 25_Re: DB2 Hangs at Shutdown19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM30_Tue, 7 Mar 2000 13:09:32 -0600595_iso-8859-1 Hi Max, You've got a great idea; there are about 2300? people on db2-l, let's see how many can stop DB2 with -STOP DB2 (defaulted MODE(QUIESCE)). You're #1. Best Regards, Rick Davis "This e-mail and any files transmitted with it are the property of SBC, are confidential, and are intended solely for the use of the individual or entity to whom this e-mail is addressed. If you are not one of the named recipient(s) or otherwise have reason to believe that you have received this message in error, please notify the sender at 314-235-6854 and delete this message immediately from [...]
39083 145 42_Re: LOG YES on Utilities - what is logged?19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM30_Tue, 7 Mar 2000 13:15:51 -0600624_iso-8859-1 Michael, Thanks again. You might what to rephrase this and sent it out to DB2-L; maybe someone knows. Regards, Rick Davis "This e-mail and any files transmitted with it are the property of SBC, are confidential, and are intended solely for the use of the individual or entity to whom this e-mail is addressed. If you are not one of the named recipient(s) or otherwise have reason to believe that you have received this message in error, please notify the sender at 314-235-6854 and delete this message immediately from your computer. Any other use, retention, dissemination, forwarding, printing, or copying [...]
39229 100 25_Re: DB2 Hangs at Shutdown8_Kurian B23_Bejoy_Kurian@MAY-CO.COM30_Tue, 7 Mar 2000 13:23:32 -0600454_us-ascii Hi all, In my previous shop, we have 10+ DB2 subsystems and most of them connected with various attachments like Omegamon, CICS, CSP, DDF's etc. We have a scheduled IPL every week on all LPARS. Our operators used to check and disconnect all the attachments and 95% time the shutdown will be normal. Some case DB2 never stops and normally i used to issue the following command to terminate IRLM. MODIFY irlmproc,ABEND,NODUMP. DB2 used stop. [...]
39330 35 25_Re: DB2 Hangs at Shutdown12_Kirk Hampton16_khampto1@TXU.COM30_Tue, 7 Mar 2000 13:48:38 -0600548_us-ascii Okay, I'll add my $0.02 here. We have 5 production and 4 test DB2 subsystems across 4 LPARS running OS/390 v2.6 and DB2 for OS/390 v5.1, and we stop them all every Sunday for an ABARS backup, with just a -STOP DB2 (implicit MODE(QUIESCE)) command. We do preface the -STOP with a Rexx program which runs a -DIS THD(*) and then cancels any active DDF threads it finds, and also issues an MVS C(ancel) on any TSO or BATCH threads that it finds. Next we issue a DSNC STOP to all attached CICS's via an MVS (modi)F(y) console command, and [...]
39366 22 43_Re: DB2 for MVS and Command Level Interface14_Wayne Driscoll19_wdriscoll@QUEST.COM30_Tue, 7 Mar 2000 11:56:12 -0800333_iso-8859-1 Giorgio, DB2 MVS introduced CLI support in Version 5. Wayne Driscoll Product Developer Quest Software Inc. wdriscoll@quest.com Note: All opinions are strictly my own.
Hi all. I need to know if DB2 for MVS, in any version, has Call Level Interface (CLI), which is found in other operating systems DB2 versions. [...]
39389 30 25_Re: DB2 Hangs at Shutdown15_David A. Zelmer25_david.zelmer@CITICORP.COM30_Tue, 7 Mar 2000 13:52:13 -0600320_us-ascii I have 23 DB2 subsystems; test, QA and prod, and they all shutdown with a -STOP DB2 (defaulted MODE(QUIESCE)).
DAVIS, RICK (SBCSI) wrote: > > Hi Max, > You've got a great idea; there are about 2300? people on db2-l, > let's see how many can stop DB2 with -STOP DB2 (defaulted MODE(QUIESCE)). > You're #1.
39420 39 43_Re: DB2 for MVS and Command Level Interface17_Alessandro Brezzi28_brezzi.isac@INTERBUSINESS.IT30_Tue, 7 Mar 2000 21:14:21 +0100390_us-ascii Hi Giorgio, the answer is: DB2 5.0 and up: all came with CLI, but you can access CLI function only from C/C++ for OS/390 compiler.
Hope this help
Ciao
Alessandro
>Hi all. >I need to know if DB2 for MVS, in any version, has Call Level >Interface (CLI), which is found in other operating systems DB2 versions. > >Thank you very much! > > Giorgio De Nunzio
39460 54 25_Re: DB2 Hangs at Shutdown19_DAVIS, RICK (SBCSI)21_RD8246@MOMAIL.SBC.COM30_Tue, 7 Mar 2000 14:34:33 -0600569_iso-8859-1 That's 3
Rick Davis "This e-mail and any files transmitted with it are the property of SBC, are confidential, and are intended solely for the use of the individual or entity to whom this e-mail is addressed. If you are not one of the named recipient(s) or otherwise have reason to believe that you have received this message in error, please notify the sender at 314-235-6854 and delete this message immediately from your computer. Any other use, retention, dissemination, forwarding, printing, or copying of this e-mail is strictly prohibited." [...]
39515 18 21_WLM Stored procedures11_Jeff Faughn22_Jeff_Faughn@MAY-CO.COM30_Tue, 7 Mar 2000 14:45:04 -0600452_us-ascii Hello DB2LAND!
We are OS/390 2.6 DB2 6.1 and trying to turn on WLM Stored Procedures (in Compatibility mode). We are trying to execute "DSNUTILS" from a COBOL program. Has anyone ever done this?
================================================ To change your subscription options or to cancel your subscription visit the DB2-L webpage at http://www.ryci.com/db2-l. The owners of the list can be reached at DB2-L-REQUEST@RYCI.COM.
39534 73 25_Re: DB2 Hangs at Shutdown0_23_Mike_Levine@TEKHELP.NET30_Tue, 7 Mar 2000 16:01:35 -0500658_us-ascii Hi,
I think DB2 will eventually come down with MODE(QUIESCE)...it's just a matter of when!
Regards,
Michael Levine Premier Data Services, Inc.
>That's 3 > >Rick Davis >"This e-mail and any files transmitted with it are the property of SBC, >are confidential, and are intended solely for the use of the individual >or entity to whom this e-mail is addressed. If you are not one of the >named recipient(s) or otherwise have reason to believe that you have >received this message in error, please notify the sender at 314-235-6854 >and delete this message immediately from your computer. Any other use, [...]
39608 47 25_Re: WLM Stored procedures0_22_BILL_GALLAGHER@PHL.COM30_Tue, 7 Mar 2000 16:08:33 -0500497_us-ascii Jeff,
I've been able to do this . . . what do you need to know?
Bill Gallagher, DBA Phoenix Home Life Enfield, CT
Jeff Faughn on 03/07/2000 03:45:04 PM
Please respond to DB2 Data Base Discussion List
To: DB2-L@RYCI.COM cc: (bcc: BILL GALLAGHER/Phoenix Home Life Mutual Insurance) bcc: BILL GALLAGHER/Phoenix Home Life Mutual Insurance Subject: WLM Stored procedures [...]
39656 101 25_Re: DB2 Hangs at Shutdown15_David A. Zelmer25_david.zelmer@CITICORP.COM30_Tue, 7 Mar 2000 16:08:13 -0600356_us-ascii As in all situations, it depends. However, for what it is worth, 4 of our production DB2 systems had been active from Nov 20/21, 1999 and were not shut down until Feb 26/27, 2000. These four DB2 systems all shut down within 1 minute of the -STOP DB2 command. This was prior to a scheduled IPL, so CICS and Omegamon were also being shut down. [...]
39758 82 49_Re: Furthering a topic (aka beating a dead horse)10_Leo Flores23_leoflores@EARTHLINK.NET30_Tue, 7 Mar 2000 14:10:41 -0800449_us-ascii Wayne,
I'm not proficient in SQL, but
To determine this couldn't one just perform the following?
1) From SYSIBM.SYSTABLES TA Select DBNAME, TSNAME, ALTEREDTS Where ALTEREDTS > CREATEDTS
2) From SYSIBM.SYSCOPY CO Select the MAX(TIMESTAMP) Where CO.DBNAME = TA.DBNAME AND CO.TSNAME = TA.TSNAME AND ICTYPE in ('R'(Load Replace Log(Yes), 'S'(Load Replace Log(No), 'W'(Reorg Log(No), 'X'(Reorg Log(Yes)) [...]
39841 94 32_Before Insert Trigger Limitation12_Steve Grimes30_Steve_Grimes@AISMAIL.WUSTL.EDU30_Tue, 7 Mar 2000 16:45:40 -0600483_us-ascii Hello all.
This note is forwarding IBM's reply to my problem of being unable to modify the contents of a row being inserted, in a before trigger, with information from another table. It seems I'm out-of-luck, dead-in-the-water, nowhere-to-turn on this one!
For example, for an inserted transaction row, I had envisioned the before trigger would obtain the "control date" from the control date table and load it into the "posting" date of the transaction. [...]
39936 34 13_db2/cs V5 AIX11_Shirley Fan25_shirley.fan@ZURICH.COM.AU30_Wed, 8 Mar 2000 09:52:08 +1000278_us-ascii Hello all, Were using AIX, DB2 UDB V5. We have a 2Gb tablespace with hundreds of table, and have deleted half of the data. How can we reclaim the space and release it to the O/S? There is no alter drop container command. We want to redefine the tablespace to 1Gb. [...]
39971 53 25_Re: DB2 Hangs at Shutdown24_Gendron, Dave (Exchange)28_Dave.Gendron@COASTALCORP.COM30_Tue, 7 Mar 2000 18:23:10 -0500638_- Kirk,
Can you share your REXX program that kills DDF threads. We currently do it manually and it would be nice to automate this step. Thx.
Dave Gendron Colorado Springs, CO dave.gendron@coastalcorp.com
-----Original Message----- From: Kirk Hampton Sent: Tuesday, March 07, 2000 12:49 PM To: DB2-L@RYCI.COM Subject: Re: DB2 Hangs at Shutdown
Okay, I'll add my $0.02 here. We have 5 production and 4 test DB2 subsystems across 4 LPARS running OS/390 v2.6 and DB2 for OS/390 v5.1, and we stop them all every Sunday for an ABARS backup, with just a -STOP DB2 (implicit MODE(QUIESCE)) command. We do preface [...]
40025 26 10_BMC unload9_Jagan Rao17_Jagan_Rao@GAP.COM30_Tue, 7 Mar 2000 16:42:42 -0800316_us-ascii Hi all,
We need to download 5M records from a DB2 table to flat file using BMC unload.During unload ,the DB2 table will also subjected to continuous inserts by CICS back ground tasks.
I know BMC SHR level change unlioad will allow this. But it will download the uncommitted data as well. [...]
40052 46 14_Re: BMC unload14_Maxfield, John21_John_Maxfield@BMC.COM30_Tue, 7 Mar 2000 19:16:29 -0600454_ISO-8859-1 SHRLEVEL CHANGE CONSISTENT YES can be used to unload a consistent, point-in-time, snapshot while allowing full read/write access to your data. It does this using BMC's XBM SnapShot Upgrade Feature. So... if you this feature, you shouldn't have any problems doing what you need to do.
-----Original Message----- From: Jagan Rao [mailto:Jagan_Rao@GAP.COM] Sent: Tuesday, March 07, 2000 6:43 PM To: DB2-L@RYCI.COM Subject: BMC unload [...]
40099 71 25_Re: DB2 Hangs at Shutdown0_29_Greg.Palgrave@BANKWEST.COM.AU30_Wed, 8 Mar 2000 10:27:57 +0800284_us-ascii A CANCEL of the IRLM is OK on the rare occasion DB2 misbehaves, the real danger is the gung-ho FORCE command if a region doesn't respond to the cancel. It's a last resort that may leave you having to IPL to clear the wreckage. Been there, done that, don't recommend it. [...]