How are discs managed under a RAC data server

I am considering using Oracle 9i for a very large database and I am being offered RAC as a scalable architecture. I am trying to find out how the 'shared disc' part of the architecture maps into physical reality. All storage devices in my experience can only attach an individual logical unit of storage to one server at any one time. There is no mechanism to request use of a shared resource. (The logical unit can however be passed around between servers in a cluster transition).
There are two obvious mechanisms:-
1)Oracle has a request/lock/release mechanism to pass logical units around between nodes.
2)Cache Fusion abstracts the physical layer so that the location of the storage is invisible. The storage is shared between the nodes and each piece of physical storage attached to one database node.
Or it could be something I have not thought of.
Any information gratefully received.

Here you can see all the process flows : http://scn.sap.com/docs/DOC-8292
Input file server is used to retrieve/save report, output is used to store report instances.
If you have multiple FRS servers - they all have to point to the same shared location (one for IFRS and one for OFRS)
Report processing servers utilize local cache directories. In some cases it makes sense to share them in some not.
There are several KB articles on webi cache and CR cache configs in addition to default practices described in Admin guide.

Similar Messages

How are valid-from and valid-to dates set in TRPROD table in APO during CIF

First question:
How are the valid-from and valid-to dates set in the APO TRPROD table?
For example, if I activate a transportaion lane model on March 1, 2008, will the valid-from and valid-to in TRPROD be 20080301 and 99991231?
Or are these dates defined in the model itself? Or in APO config somewhere?
Next question:
Suppose I rerun the model on May 1 2008 to add some records that didn't get added in the March 1 run. But suppose I want to back-date the valid-from of these records to March 1, rather than use May 1 as the valid-from.
Can this be done?
Or is the valid-from always the date of the CIF that creates the TRPROD records?
Please advise.
Thanks

Hi Srinivas -
Thanks very much for replying so quickly, but your response didn't answer the questions.
Let me try again.
1)
If I run a CIF on March 1 2008 based on a model that will generate APO TLane data, will the TRPROD valid-from and valid-to get automatically set to 20080301 and 99991231?
2)
If the answer to this question is yes, then is it also true that if I run a new set of records in on May 1, 2008 (no duplications with the original set), then the valid-from and valid-to dates will be 20080501 and 99991231.
3)
But if (1) and (2) are both true (if you answer them both "yes", then suppose I want the May 1 record set to have valid-from dates of March 1 (for example, if the May 1 record set includes records that didn't get included in the March 1 set due to an error in selection criteria). How do I get the valid-from to be March 1 when I run the CIF on May 1?
On the other hand, if (1) and (2) are NOT true, then how DO the two dates get set during a CIF ????
Thanks
djh

How to revert managed client to unmanaged without server

The scenario is that a group of 10.6 client macs were managed by an OS X server, and that server no longer exists. How can the clients be reverted back to an unmanaged state without using a server.
I've tried manually erasing the '/Library/Managed Preferences' and '/Library/Preferences/DirectoryService' contents but they are restored from somewhere, I assume some hidden system cache (because I can't find copies of these files anywhere, and a hidden locker makes sense). I've also tried changing some of these preferences using the 'defaults' command, hoping that'd propagate back to the cache, but they still get restored. I've also cleaned out the global com.apple.loginwindow MCX preferences, to no avail.
I have perused the local directory with 'dscl' and found nothing obvious, but it wouldn't surprise me if there was binary data in there supporting the automatic recovery.
From observation, I'd guess there might be a process run from 'launchd' periodically that restores the values, because I've seen them come back while I was working at the command line.
There are only a few clients so it's not a problem if I have to do it manually or even in single-user mode, but I'd really, really like to avoid having to wipe and reinstall because there's a great deal of user state on them. Everything's backed up with good TM backups going back a long way, but I'm afraid of restoring the server binding if I have to reinstall and recover, and I'd really like to avoid that hassle, too.
I'm a very seasoned Unix geek and have no fear of working in system directories from the command line as root, but so far I haven't been able to find whatever it takes.
So, who knows how to actually do this?
Thanks.

Antonio Rocco wrote:
Hi
sudo rm -R -v /Library/Preferences/DirectoryService
sudo rm -R -v /Library/Preferences/edu.mit.kerberos
sudo rm -R -v /Library/Managed\ Preferences
sudo rm -R -v /Library/Caches/*
Restart the mac.
This is the approach I have used as well. However if the client Mac is on the same network as a server managing that client then the server will re-apply the managed settings after a reboot.
I know the original poster said the server was no longer present, but is there another server that might be applying settings? Perhaps an Open Directory replica?
It would not apply in this case because the original poster said the clients are running Mac OS X 10.6, but much older versions of Mac OS X client also kept a copy of managed preferences in NetInfo and you needed to use NetInfo Manager to delete them. Mac OS X 10.6 does not use NetInfo.

Associated Arrays...How are they managed?

Hi I have a question about how the associated arrays are managed by oracle. Does Oracle keep the data in memory and start paging it to disk once they reach a certain size or does it keep certain amount in memory and use some temporary table to which it insert and retrieve data as and when needed.
This is important to me as I feel that I may end up putting a lot of data in the associated arrays and I know for certain that paging is really slow.
Thanks for your insight..
Regards,
Sumit

Associative Array are important for what they represent, not for how their memory is managed. Associative Arrays are equivalent to Java/C# Hash tables/collections. Previously, Oracle only allowed collections/arrays to be indexed by standard datatypes like binary_integer.
Jason

How to export app +underlying schema(and data)

Hi
I am trying to export my application and its underlying tables to a new system .
I want to know if it can be done together and not seperately..(export app and then export each table)
Is there away to package the app and its underlying repository?
Thanks

No....
You will have to do them seperately as there is no logical connection between the application and the tables.
Your application just has the access to use those tables and does not "own" any of them.
a) Oracle database tables/ application tables
You can export the entire schema and all the objects and data in that schema using the export and import utilities.
All these objects are owned by the schema (eg. emp under scott) and hence are related logically.
b) Oracle apex metadata tables.
All the information about apex objects is stored as metadata in database tables (FLOWS_0300 is a schema).
So, if you install Oracle apex on a new machine, all these repository tables will be created.
When you create applications / objects, these tables will have metadata for those applications.
Hope this helps,
Rajesh.

How are YOU managing files?

I know this post doesn't necessarily pertain to just FCP but I'm looking for a way whether it's automated or manual to back up media files.
I'm working in a SAN situation and I have an external drive that I use to backup my files in case of a SAN crash or whatever.
Since I'm working with more than just FCP (i.e. After Effects, Photoshop, Illustrator, DVD Studio Pro) I have a lot of folders within folders. Pics, audio files, project files, and all sorts of elements to backup. I have a good file structure in place to keep all of this organized but when backing this stuff up, I have a lot of logistical issues:
Since only some of my folder's contents change, how can I possibly keep track of which ones to replace without deleting and duplicating the new ones? Do I have to cherry pick through all my data so I don't have to delete all my media on the backup and then replace it with updated backup files?
Some have suggested to me using Automator with iCal. I know nothing about scripting or how to go about this.
So finally my question.. How do you go about backing up your files in this kind of situation? Are you using some sort of software? Do you back up a root folder onto the drive every week and once that's loaded you delete the old backups? Do you meticulously drag and drop each sub folder and replace any duplicate files? There has to be a simple solution to this mess!
thanks for reading my long post, it's my longest yet!!!

You may want to look into dedicated backup software. I use Deja Vu but there are others. It should have the ability to designate which folders are backed up, when it happens and how duplicates are handled among a host of other issues.
As for structure. While the system drive gets cloned before any software update it is not regularly backed up as no FCP material AT ALL goes on the system drive.
A second drive is used for scratch capture and renders. It is not backed up as that material can be recaptured/regenerated if lost.
All non-TC material exists on third/fourth/fifth/etc drives. This is the stuff that is backed up.
Good luck.
x

How are Podcasts managed?

I subscribe to about 10 different Podcasts. Whenever I fire up iTunes, any new episodes of these podcasts are downloaded.
However, on my iPod, which I dock often, the podcasts never seem to be updated. I have iTunes set to keep the last 3 episodes.
What I would like, is for the 3 most recent podcasts waiting for me on the iPod after I dock it.
Am I missing something?
Thanks,
Rob

I did a search to locate this old post, but I am having the same issue. I have been through all of the iPod option settings, and it is set to update every hour and sync the last 5 episodes (and I have toggled these settings various ways). The updates are not consistent in iTunes; for instance, I can show a particular podcast as having the last update one week ago, and then I manually select update, and it finds a newer one. This is after the iPod is docked and it says it is updating. Then, the iPod updates are not consistent with the iTunes updates. I have podcasts that are shown in iTunes, but not updated on the iPod.
The podcast updates worked flawlessly a few months ago, but sometime in the last 3-4 months in a newer iTunes version, I am seeing inconsistency updating.
Any ideas?

How are you checking quality in the upgrade process?

When upgrading your EBS, how are you managing quality?
Are you using solutions such as Quality Center and Quick Test Pro (from Mercury)? Are you using other tools?
Which test sets are you using to ensure that the upgrade had completed successfully?

Hi,
I am not very good at testing and nor I have used any testing tools.
But I think there are various post upgrade task which measures the quality of upgrade.
In my view upgrade quality can be decided on two factors
1.) Performance of new instance
2.) Is there any issue with Data ( Check for all tests and reports)
Pravin

Connect to a data server with topology manager

Hi! I am new to ODI...
I've created a master and work repository. Now I'm in topology manager to create a data server from Physical Architecture. I succeed in connecting to this data server with "Oracle Toad" using a service name...so I select JDBC driver and then I insert jdbc:oracle:oci8:@<tns_alias> (with the right service name).
Then it give me the message that it can connect to the data server, but when i click ok, it says me:
{color:#ff0000}java.lang.UnsatisfiedLinkError: no ocijdbc10 in java.library.path
at java.lang.ClassLoader.loadLibrary(Unknown Source)
at java.lang.Runtime.loadLibrary0(Unknown Source)
at java.lang.System.loadLibrary(Unknown Source)
at oracle.jdbc.driver.T2CConnection.loadNativeLibrary(T2CConnection.java:3013)
at oracle.jdbc.driver.T2CConnection.logon(T2CConnection.java:228)
at oracle.jdbc.driver.PhysicalConnection.<init>(PhysicalConnection.java:361)
at oracle.jdbc.driver.T2CConnection.<init>(T2CConnection.java:142)
at oracle.jdbc.driver.T2CDriverExtension.getConnection(T2CDriverExtension.java:79)
at oracle.jdbc.driver.OracleDriver.connect(OracleDriver.java:595)
at com.sunopsis.sql.SnpsConnection.u(SnpsConnection.java)
at com.sunopsis.sql.SnpsConnection.a(SnpsConnection.java)
at com.sunopsis.sql.SnpsConnection.testConnection(SnpsConnection.java)
at com.sunopsis.sql.SnpsConnection.testConnection(SnpsConnection.java)
at com.sunopsis.graphical.frame.b.jz.dP(jz.java)
at com.sunopsis.graphical.frame.b.jz.bD(jz.java)
at com.sunopsis.graphical.frame.bo.bz(bo.java)
at com.sunopsis.graphical.frame.b.jz.em(jz.java)
at com.sunopsis.graphical.frame.b.jz.<init>(jz.java)
at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
at sun.reflect.NativeConstructorAccessorImpl.newInstance(Unknown Source)
at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(Unknown Source)
at java.lang.reflect.Constructor.newInstance(Unknown Source)
at com.sunopsis.graphical.frame.bb.b(bb.java)
at com.sunopsis.graphical.tools.utils.swingworker.v.call(v.java)
at edu.emory.mathcs.backport.java.util.concurrent.FutureTask.run(FutureTask.java:176)
at com.sunopsis.graphical.tools.utils.swingworker.l.run(l.java)
at edu.emory.mathcs.backport.java.util.concurrent.ThreadPoolExecutor$Worker.runTask(ThreadPoolExecutor.java:665)
at edu.emory.mathcs.backport.java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:690)
at java.lang.Thread.run(Unknown Source)
{color}
My environment variable for Oracle are:
C:\oracle\ora92\bin;C:\Programmi\Oracle\jre\1.3.1\bin;
I succeed in connecting with OCI using Toad...why can't I connect with ODI? Need I to install another component? (and which?)
Sorry for my english...please help me! thanks....
Alessia
Edited by: user1762560 on 26-feb-2009 9.06

jdbc:oracle:oci8:@tns_alias
it doesn't contain the sign '< ' or '>'

How to create a nuw field in data model

Hi,
I am using report builder data model, i have 2 query (Q_1, Q_2). Q_1 has a field F_1, and Q_2 has a field F_2, is it possible I can create a new field for both query: (Q3=Q2-Q1).
thanks

Question: how are you making sure that the data frequency is same? If you can and you are certain then why do you need to have two different groups?
Accepting your design... in the following situation it should work
1) if Q2 group is a child of Q1 then you create a formula column (where you define the equation in the pl/sql editor) in Q1.
2) there is a data link from Q1 to Q2 then create the formula column (mentioned in #1) in Q2. If you create formula column in Q1 that will not work.
- Saumen.

How do people manage QoS Policies in large network without using QPM

We are using QPM to manage QoS polices however we are looking at decommissioning CiscoWorks. How are people managing with their QoS settings in large environments?

I have no idea about the modem and bridge mode (I don't do networking -- hopefully Bob Timmons, Tesserax, or one of the other networking gurus will drop in and address that).
But . . . you should be able to back up to the TC as long as it's on your network and recognized by your Macs. I think being in bridge mode means it will be rather slow, but it should work. Until/unless we hear otherwise, you might want to see #Q1 in Using Time Machine with a Time Capsule.

Windows Server 2008 - How to remotely manage account passwords?

I'm hoping to get direction on how to remotely manage 'user accounts' on Windows Server 2008 machines. The remote management would be from an automated perspective. For example, I have a custom application (maybe C#, Java, etc) that runs on one
server that makes connections to various 'Windows Server 2008' servers; and I want it to be able to do the following.
Query Windows Server 2008 box for full list of user accounts
Update password for user account
Lock account
Unlock account
Check policy information to find out when password may expire
I'm trying to find out what may already be available and running on Windows Server 2008 that I can take advantage to get this done programmatically.
-Wes

I agree with Mekac, the gallery should provide a lot of script of what you want to do.
Like ;
https://gallery.technet.microsoft.com/Password-Expiry-Email-177c3e27 - Password Expiry Email Notification.
Regards, Philippe
Don't forget to mark as answer or vote as helpful to help identify good information. ( linkedin endorsement never hurt too :o) )
Answer an interesting question ? Create a
wiki article about it!

LDAP Data Server

I need to pull xMII user information out of my LDAP database. Ultimately I'm wanting to create a 'last login' report listing xMII users and when they last logged in. As part of that I need to know who my xMII users are.
Is there an LDAP data server available for me to make the connection to the database?

David
Since u want to see the xMII User List who have recently Logged on, u can directly see the Active Sessions option under System Management in the xMII Menu.
Here the Active Session displays the list of recent logged xMII users along with his/her login time and expiration time.
You can access to the folder
\ServletExec AS\se-xMII\webapps\default\Lighthammer\WEB-INF\classes\com\lighthammer\Illuminator\services\admin where the SessionList.Class is present.
Also u can look into the folder \ServletExec AS\se-xMII\webapps\default\Lighthammer\Admin where the JSP of Active Session is present.
All those will help u to show the list of xMII User List with their last logged time using JSP
Regards
Som

XML Data Server and model help

All,
I have the following files
1) travel.dtd
2) travel.xml.
I've set up a XML data server with JDBC driver com.sunopsis.jdbc.driver.xml.SnpsXmlDriver and URL = jdbc:snps:xml?d=c:/XML/travel.dtd. I have created a model in designer and have successfully reversed the datastores generated from the dtd.
The travel.xml file has data populated within it, but when I right click on the datastores, there is nothing displayed. This makes sense because the xml server is configured to point at the .dtd. Do I need to change my topology connection to connect to the xml document as opposed to the dtd. Can someone tell me what i'm missing?
thanks for your help.

Since its not displaying anything as you are saying you just change the property to "f" and check.
Example
jdbc:snps:xml?f=/xml/myxml.xml&d=/xml/myxcd.xsd&s=MYSCHEMANAME
("s=MYSCHEMANAME": This schema will be selected when creating the physical schema under the XML data server.)
Hope it helps.
Thanks

On iPhone 5c iOS 8.2, I go to settings, cellular, use cellular data for: Under "use cell data for are my apps but they are greyed out how can I ungrey them??

iOS 8.2 on iPhone 5c, goto settings, cellular, use cellular data for: all my apps under "use cellular data for" are greyed out. I've restart and hard reset. Looked thru all settings. Can someone help??

Did you try to reset the phone by holding the sleep and home button for about 10sec, until the Appel logo comes back again? You will not lose data by resetting.
If this does not work, did you try to switch off Cellular Data, restart the phone and switch it back on again?

How are discs managed under a RAC data server

Similar Messages

Maybe you are looking for