Partitioning Esscmd Questions

I have a couple of questions about ApplyOTLChanges and PurgeOTLChangeFile Esscmds.With ApplyOTLchangeFile you have to name the chg files that you want to apply. The only thing is that the name of this file seems to change. I have two sources, 1 target (no overlap in partitions) and the chg files in the target are named ess00001.chg and ess00001.chg one time and then can be ess00002.chg and ess00004.chg the next. What is the logic for the numbering of these chg files in the target?On the purgeOTLchangeFile command, I always seem to get the error "no matching region defined for the specified server,app,db: [servername,appname,dbname]" error 1023001.Can anyone shed any light?Thanks

The number is the partition number in the app, (look in Partition Manager). The same partition can have diff numbers when viewed from diff apps.

Similar Messages

FORTE M2 - NT partition compilation question

Microsoft (R) 32-Bit Incremental Linker Version 5.00.7022
Copyright (C) Microsoft Corp 1992-1997. All rights reserved.
c:\forte/install/lib/qqDB.lib : fatal error LNK1106: invalid file or disk
cannot seek to 0x38dff853
make[2]: *** [all] Error 1
Any suggestion ? (knowing I got 300 Mo left on drive)
1) Can that offset be translated in bytes ?
2) Does this rely on Oracle Client installation (DB) ?
Thanks,
j-paul gabrielli
sema DTS

Thanks for the info - installed.
Btw, another point:
can I tell Microsoft that I really like their
installation paths ?
rem
rem Root of Visual Developer Studio Common files.
set VSCommonDir=C:\PROGRA~1\MI017E~1\Common
rem
rem Root of Visual Developer Studio installed files.
rem
set MSDevDir=C:\PROGRA~1\MI017E~1\Common\msdev98
I fear that I cannot consider this as something portable :-/
j-paul
-----Message d'origine-----
De: Adamek, Zenon [mailto:ZAdamekpurolator.com]
Date: mardi 30 mai 2000 15:39
À: 'Jean-Paul.Gabriellisema.fr'
Cc: kamranaminyahoo.com
Objet: RE: (forte-users) FORTE M2 - NT partition compilation question
Hi
You are using the wrong version (5.0) of MSVC++. Forte version M requires
the MSVC++ 6.0 compiler (Linker Version 6.00). See technote 12083.
Zenon Adamek
Purolator
-----Original Message-----
From: Jean-Paul Gabrielli [SMTP:Jean-Paul.Gabriellisema.fr]
Sent: Tuesday, May 30, 2000 9:34 AM
To: Forte User Group
Subject: (forte-users) FORTE M2 - NT partition compilation question
Microsoft (R) 32-Bit Incremental Linker Version 5.00.7022
Copyright (C) Microsoft Corp 1992-1997. All rights reserved.
c:\forte/install/lib/qqDB.lib : fatal error LNK1106: invalidfile or disk
cannot seek to 0x38dff853
make[2]: *** [all] Error 1
Any suggestion ? (knowing I got 300 Mo left on drive)
1) Can that offset be translated in bytes ?
2) Does this rely on Oracle Client installation (DB) ?
Thanks,
j-paul gabrielli
sema DTS
For the archives, go to: http://lists.xpedior.com/forte-users and use
the login: forte and the password: archive. To unsubscribe,send in a new
email the word: 'Unsubscribe' to: forte-users-requestlists.xpedior.com--
For the archives, go to: http://lists.xpedior.com/forte-users and use
the login: forte and the password: archive. To unsubscribe, send in a new
email the word: 'Unsubscribe' to: forte-users-requestlists.xpedior.com

Embarrassing ESSCMD question (re: error logs)

Hi all you kind folks,This is one that is probably incredibly obvious, but I just can't find the answer. I use various ESSCMD scripts to load oodles of data into Essbase cubes. The data files are very large. My scripts are set up basically like this (simplified):OUTPUT 1 "dataload.log";IMPORT 3 "datafile1" 4 "Y" 3 "loadrules.rul" "N" "dataload.log";IMPORT 3 "datafile2" 4 "Y" 3 "loadrules.rul" "N" "dataload.log";...and so on. The problem is that after each data file loads to the cube, the error messages in 'dataload.log' are overwritten by the next data file's errors.When the script is finished, only the errors for the last data file are in theerror log. I need to be able to capture all the errors, for all the data files.My question: is there an Essbase setting somewhere that would let mespecify that the errors for successive data file loads be appended to theerror log, instead of overwriting it each time? Or do I just have to send theerrors for each individual data file to a separate log file? Thanks very much! gregp.s. I jacked up the DATAERRORLIMIT setting to 100 million, thinking thatthis was the problem - no luck.

I think you made a confusion between the general log of your script (OUTPUT...) and the error log of each loading (for each loading line).As the error log is erased each time, I am affraid that you should mention a different error file for each loading.At the end, if you want, you can concatenate these files.Hope this helps.Denis ZubaTHESYS [email protected]://www.Thesys-Solutions.comT?l.: +41 21 653 56 12Mobile: +41 79 688 80 12

Partitioning HD Question

Hey, I have a question. How can I partition my HD so that I can have 2 partitions (one for Mac and one for Linux) without having to reformat and delete my Mac Disk?
TIA

Hi!, There are a couple of programs that purport to do that. One is DiskStudio from Micromat and iPartition
Tom

Import Manager Partitioning/Pivot Question

My source data has two fields (Part Number, Page Number). There are multiple rows for each part number with a single value for page number. I need to combine the page number values for all instances of each part number and import into MDM as a comma delimited value.
For example:
Part Number | Page Number
A1 | 1
A1 | 2
I want to store in destination field "Page Numbers" as "1, 2" for single record where part number is "A1".
I do not understand how to accomplish this with Import Manager.

Hi ,
I tried in one more case also but there was limitaion .In that case you will end up with more no. of delimiters or nulls
You can try with same procedure but just define a clone field for partnumber and create pivot table with the same constraints
just swaaping with newly introduced field.
After this just partiton all the new fields and then combine .
the newly partition field treat as page number and map to destination field and also map source part number with destination field.
This is also an option with limitation of trailng or preceeding null .I think we can ignore that also for that we have to do some setting of ignoring nulls.
Please try this alos but if you are able to add new field than that is the best one.
please revert back with your results.
Thanks,
Sudhanshu

Partitioning design question

I'm investigating partitioning one or more tables in our schema to see if that will speed performance for some of our queries. I'm wondering if the following structure is viable, however.
Table structure - this is a snippet of relevant info:
CREATE TABLE ASSET (
asset NUMBER, -- primary key
assetType NUMBER,
company NUMBER,
created SYSTIMESTAMP,
modified SYSTIMESTAMP
lobData CLOB
...)The current table has ~ 60 million rows. All queries are filtered at least on the company column, and possibly by other criteria (never/rarely by date). The number of rows a company can have in this table can vary greatly - the largest company has about 2.4 million, and the smallest about 1000. This table is joined by several other tables via the primary key, but rarely queried itself by the primary key (no range pkey queries exist).
I'm thinking of partitioning by company (range) - however, I'm not sure if the uneven distribution of company data makes that an effective partition. The number of companies is relatively small (~6000 ) and does not grow significantly (perhaps 1-2 new companies a day). The data in this table is pretty active - ~200k deletes/inserts a day.
Does it make sense to range partition by company? I was thinking of partitioning per company (1 partition per company) - but the partitions would be quite different in size. Is there a limit to the number of partitions a table can have (is it 64k?). Does partitioning even make sense for this table structure?
Any thoughts or insights would be most helpful - thank you.

kellypw wrote:
I'm investigating partitioning one or more tables in our schema to see if that will speed performance for some of our queries. I'm wondering if the following structure is viable, however.
Table structure - this is a snippet of relevant info:
CREATE TABLE ASSET (
asset NUMBER, -- primary key
assetType NUMBER,
company NUMBER,
created SYSTIMESTAMP,
modified SYSTIMESTAMP
lobData CLOB
...)The current table has ~ 60 million rows. All queries are filtered at least on the company column, and possibly by other criteria (never/rarely by date). The number of rows a company can have in this table can vary greatly - the largest company has about 2.4 million, and the smallest about 1000. This table is joined by several other tables via the primary key, but rarely queried itself by the primary key (no range pkey queries exist).
I'm thinking of partitioning by company (range) - however, I'm not sure if the uneven distribution of company data makes that an effective partition. The number of companies is relatively small (~6000 ) and does not grow significantly (perhaps 1-2 new companies a day). The data in this table is pretty active - ~200k deletes/inserts a day.
The version of Oracle is very important.
Partitioning on company looks like a sensible option since you ALWAYS filter on company - but list partitioning makes more sense than range partitioning because it is more "truthful"
Unfortunately it looks, at first sight, as if you have a logical error in the design - I'm wondering if the company should be part of the primary key of the asset. If you partition by company you won't be able to do partition-wise joins to the other tables when joining on primary key (I've interpreted your statement to mean that the asset is the foreign key in other tables) unless you happen to be running 11g and use "ref partitioning".
It's hard to predict the impact of 6,000 partitions, especially with such extreme variations in size. With list partitioning it's worth thinking about putting each large company into its own partition, but using a small number of partitions (or even just the default partition) for all the rest.
Regards
Jonathan Lewis

Partition Size Question

Hi all!
New here, and I just need some clarification on something. I'm thinking of reformatting my Macbook Pro and installing Windows 7 on it once I get a copy. Ideally I would like my entire hard drive to be Windows. Is this possible or do I need to allocate some space for OSX? If it's not possible through Boot Camp is there some other software I could use or something?
Thanks in advance

I have not tried it, but I don't think you can do it at least with Apple software alone. Having a small OS X partition shouldn't be too bothersome. But, [see this thread|http://discussions.apple.com/thread.jspa?threadID=2273104&tstart=0].

Partition pruning question

Oracle Version: 11.2.0.2
OS platform : Solaris 10
From EXPLAIN PLAN output (or any other method) of an SQL how can we tell that Partition pruning is not happening?
What can be done so that optimizer will look at only the relevant partition(not the entire table) for records it is looking for ?

What can be done if the optimizer seem to scan full tabe or lots of partitions instead of the relevant partition ?On partitioned tables there are two statistics: GLOBAL and LOCAL, if when we take the statistics we provide one partition (using the PARTNAME parameter) the global statistics will be also collected.
The GRANULARITY parameter will allow you to decide the level of stats to be taken, GLOBAL as a whole object, PARTITION or SUBPARTITION.
Options for the GRANULARITY parameter:
'ALL' – gathers all (subpartition, partition, and global) statistics
'AUTO'- determines the granularity based on the partitioning type. This is the default value.
'DEFAULT' – gathers global and partition-level statistics. This option is obsolete, and while currently supported, it is included in the documentation for legacy reasons only. You should use the ‘GLOBAL AND PARTITION‘ for this functionality. Note that the default value is now ‘AUTO‘.
'GLOBAL' – gathers global statistics
‘GLOBAL AND PARTITION‘ – gathers the global and partition level statistics. No subpartition level statistics are gathered even if it is a composite partitioned object.
'PARTITION ‘- gathers partition-level statistics
'SUBPARTITION' – gathers subpartition-level statistics.
NOTE: Global statistics will not collect individual statistics for partitions and subpartitions.
If we choose PARTITION or SUBPARTITION the stats will be taken in a fraction of the time, but there is a price for this speed, the GLOBAL stats for the whole table as an object will be missing, therefore only the queries that are using the partition key will have the stats.
Conclusion : If you can afford to take GLOBAL stats then do it, you will have complete statistics for all your database, just set the GRANULARITY parameter to AUTO, this will also avoid stats problems when queries do not specify the partition key.
Aggregated stats have worked out very well for me, but that might not be the case on all databases.
So, just keep statistics updated and believe upon Optimizer, because when we are providing the correct and updated statistics, now optimizer will do best for us for sure.
Regards
Girish Sharma

Partition Exchange questions

I’m on 10gR2.
1) Is it possible to exchange the partition between two LIST partitioned tables? I know it’s possible to exchange partition between a partitioned table and a non-partitioned table but not sure on two partitioned table. Any one has done this?
2) IF I perform following, what would happen to SELECT that already in progress on the Partitioned table P1? Will it continue or will it be interrupted? Will my result be based on old data or exchanged data? i.e. Read consistency be applied in this case?
ALTER TABLE T1 EXCHANGE PARTITION P1 with table T2
Thanks in advance for your help.

max71 wrote:
I’m on 10gR2.
1) Is it possible to exchange the partition between two LIST partitioned tables? I know it’s possible to exchange partition between a partitioned table and a non-partitioned table but not sure on two partitioned table. Any one has done this?No,
You can exchange a simple table with a partition of a partitioned table.
You can exchange an entire (simple) partitioned table with a single partition of a composite partitioned table. To move a partition from one list-partitioned table to another you will have to do a "double-exchange" (or worse).
e.g. To swap partition pX of table pt1 with partition pY of table pt2:
Create empty simple table T of the right shape
alter table PT1 exchange partition pX with table T
alter table pT2 exchange partition pY with table T
alter table pT1 exchange partition pX with table T
Drop table T
2) IF I perform following, what would happen to SELECT that already in progress on the Partitioned table P1? Will it continue or will it be interrupted? Will my result be based on old data or exchanged data? i.e. Read consistency be applied in this case?
ALTER TABLE T1 EXCHANGE PARTITION P1 with table T2
In simple cases, the result will be correct. Oracle introduced a form of "cross-DDL" read-consistency in Oracle 8.0 for exactly this reason. It is possible (though a little unlikely after so many years) that you might find some bugs in it in complicated cases. I know that in the very early days I had some cases where queries would silently lose partitions after an exchange and give the wrong results.
The thing to watch out for is that the data segments you renamed (and still need current queries to see) do not get over-written. For example, a query against T1 that started just before the exchange you've quote above will want to see the data in the thing that is now table T2 - and that's possible. Even if you drop table T2, the query can still complete safely. But if you drop T3, and then create a new object that overwrites the hole left by T2, the query will crash when it starts to read that space with an error like ORA-01410 invaliid rowid, or something similar.
Regards
Jonathan Lewis
http://jonathanlewis.wordpress.com
http://www.jlcomp.demon.co.uk
"The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge." Stephen Hawking.

AE with 3 external drives: Partition privacy question

I have an AE at work today with two 1Tb drives on it, shared via "disk password". I want to add a third drive, partitioned into 3 partitions (management, internal, and general). I've already partitioned it from my MBPro (Mac OS Extended-Journaled) with GUID Partition Table (I'm pretty sure that's what I needed to do).
When I connect it to the AE, I need to restrict access to the management partition to managers, internal partition to a list of specific users, and then make the remaining partition "open" to anybody who can see the AE (disk password, like the other two whole drives).
I'd do the experimentation myself, but I can't bring it down to play and experiment right now. I can make any changes needed as long as I schedule the outage.
It it as simple as
1) changing the AE setting for Disks --> File Sharing setting for "Secure shared disks" from "Use Disk Password" to "With accounts",
2) Adding all users in my network to the accounts list
3) Add them appropriately to the two existing drives, and the third 3-partition drive?
TIA for any insight/tips,
Steve

<BUMP>

Partitions Related Questionn...

Hi All,
I want to move a partition from one table to another temporary table, by retaining the actual copy of the data in the initial table as well... Is it possible.... Howw,, Any documentationnn??
What happens when we move a partition?? Does the partition goes into read-mode only automaticallyy... As when this movement is happening, is it possible to insert the data in the partition simoultaneouslyy...
Are there anyy online documentss available through whichh i can grill more into Partitioning Conceptss... I have gone through the basic concepts of the partitioningg...
Thanks in Advance...

rahul12289471 wrote:
What happens when we move a partition?? The partition contents do not move - merely the ownership. With a partition exchange, the owner of the table data is swapped with the owner of the partitioned data. So the table now owns the partition data (and indexes) and the partition owns the table's data (and indexes).
The actual data does not move. Which is why exchanging a 1TB partition is as efficient (fast) as exchanging a 0 byte partition. And why this makes a very effective means of managing data in partition sets.
As there is no copying/moving of data involved, you cannot "retain" a copy of the data. To make a copy means requiring I/O to read the entire partitioned data set (all 1TB if that is the case) and then duplicating that (writing 1TB of data if that is the case).
Also keep in mind that duplicating data sets in a RDBMS is usually a silly and wrong thing to do. Relational design and relational databases are all about eliminating the need to duplicate data and then programatically maintain these duplicate data sets.
I/O is also the slowest and thus most expensive operation on a database. Why would you want to contribute to that, slowing down the database and server, by doing totally redundant I/O by reading and writing the exact same data multiple times?

PARTITION column question

I am assuming that it is not a good design practice to have a column, that a table is partitioned on, to be nullable.
Any ideas?
Thank u.

Hi,
When we partition a table with values of certain
attributes and when the value of the incoming data is
null then we need to handle it as it may not fall
under any specific category.Since we handle them
seperately it may take more time to load.I think you've said it all right there, you need to do special things when you find NULL values based on your companies business requirements. That is not to say that ALL data loading will be impacted by a NULL value on a partitioned column, just that yours "may" be.
Again, you'd have to test it out (in your specific circumstance) before you can say one way or the other.
Message was edited by:
Tubby

Partition areas question

I have two cubes, one of which is connected to the accounts system and contains the "actual" financial data, the other is a forecasting cube and contains, well, forecasts.I want to link the two via a replicated partition so that actuals can be directly accessed in the Forecasts cube without having to do data export/loads, etc.All the members from all the dimensions in the Actuals cube exist in the Forecasts cube, but they are not neatly divided up. We are a magazine publisher, and each issue is a separate activity, so the activities dimension in the Actuals cube looks like this:Magazine X...........Issue 1...........Issue 2while the same bit of the Forecasts cube looks like this:Magazine X...........Issue 1...........Issue 2...........Issue 3...........Issue 4issues 3 and 4 not having been published yet, no have no actuals data so don't exist in the Actuals cube.Defining the area I want to replicate is easy in the Actuals cube, but I can't see how to define the matching area in the Forecasts cube because of the extra members?Any ideas for a way forward gratefully accepted![email protected]

In my experience, mismatched replicated partitions work fine. It just writes a message to the log file if a match isn't found, (so don't have too many) but the ones that do match tranfer their data fine. Take for example, one cube has one year and the other has seven years. mapping y2003 to @children(years), will not validate, but the transfer of data will always work.

Cloning an OS to an external drive - partition order question

Hi
when you clone your internal OS to an external hd which is for example, has partitions - does it matter which of the three partitons you should clone your OS to (to make it bootable) ? Does it have to be in a particular positioned partition on that disk - eg first or last ?
Also is it okay to clone two bootable OS to that example external HD
for example - one OS being a bootable clone of my current internal and the other OS being a bootable clone of my previous OS (panther)
thanks for time
d

it doesn't matter what partition you clone it to-the system will just view it as multiple disks attached to the Mac.
Good luck!

Windows Partition Installed - Questions ??

OK, can't believe how easy that was. I been dreading that for two days. Boot Camp is a lovely tool and the Apple software was smoooooooth.
OK, here's some questions:
1. How do I get right click on the touch pad? Two fingers aren't getting it or is there a touchpad icon somewhere? I tried the mouse under Control Panel and not luck.
2. How do I get tap click enabled? Again wasn't an option under my mouse options under Control Panel.
So far, that's it. I am sooooooooo pleased and looking forward to working the Apple side for my audio and video work and for now my other editing suite on the Windows side. I'm sooooooooo glad I took those two PC's back and got the Mac. Now I can retire my old dying Toshiba that I'm typing this one while Windows Update is busy updating :o)

drdimento,
Hey again! You were in the right place to change those preferences (for 2-finger clicking and tap-clicking). You must have been using Windows for too long, man :-D You want to change a preference for the trackpad, right? Well, look in the trackpad tab! (right next to the mouse tab)
Scott

Partitioning Esscmd Questions

Similar Messages

Maybe you are looking for