The personal view on the IT world of Johan Louwers, specially focusing on Oracle technology, Linux and UNIX technology, programming languages and all kinds of nice and cool things happening in the IT world.
Some time ago when I was working at a previous employee I was part of a team who was evaluating some business intelligence solutions. One of the systems we where evaluating was Microsoft reporting services, to get a good understanding of the system I did some trials, installed the system on a test server and played around with it. Also I have printed a lot of documentation to get a good read about things.
Now cleaning up my office, I am working on that project already for a long time, I (re)found those prints and to make sure I can find them back and give you the change to also find them online in case you are looking at Microsoft reporting services I have decided to post a little explanation about them and some links.
"With 40,000 employees servicing markets in more than 160 countries around the world, PMI is continually monitoring a complex array of business data to drive its ongoing success. To continue its leadership in a rapidly changing global market, PMI’s corporate executives and worldwide market managers needed a solution to speed and improve access to consistent sales data across the global organization. They wanted to more quickly access business performance data at the corporate level, and to implement a consistent method for reporting and analyzing information across many different local markets and product brands."
The second paper is about "Business Intelligence and Data Warehousing in SQL server 2005" and is discussing the possibilities of sql server 2005 and BI solutions. They also provide some coding examples and it is generally a good introduction paper to this subject.
The third paper is from Progressive Strategies and is named "Comparing Business Intelligence Platfroms". A short introduction to the paper:
"This white paper compares the features of Microsoft® SQL Server™ 2000 Analysis Services, IBM DB2 OLAP Server 8.1, and Hyperion Essbase 6.5. As you'll learn, Analysis Services has functionality that compares favorably against other offerings such as IBM DB2 OLAP Server 8.1 and Hyperion Essbase 6.5. SQL Server 2000 Analysis Services and Hyperion Essbase 6.5 lead the industry in online analytical processing (OLAP), according to The OLAP Report (www.olapreport.com), and Microsoft has taken a clear lead in the marketplace over Hyperion and the other vendors (Figure 1 OLAP Market Share Page 2). According to The OLAP Report, this lead is sure to increase in 2003. Customers1 such as MGM Mirage, Starbucks Coffee, and Comp USA have adopted Analysis Services to gain competitive advantages in their respective businesses. Numerous independent software vendors (ISVs), including Ascential Software, Brio Software, Cognos, ProClarity Corporation, Sagent Technology, and Business Objects, have implemented solutions on Analysis Services. It is a more modern, flexible, scalable, and usable OLAP database than Essbase or DB2."
From my short experience with reporting services I have to say that it is a real good product from Microsoft and that I really like working with it. I have plans to start gaining some more knowledge on this platform an will report about it on this weblog even do it is not Oracle or UNIX/Linux related :-)
One full year of blogging...... Started in 2006 I now have finished a complete year of blogging,... the year 2007. For those who are intereseted i have placed the visitor results here (2006 and 2007):
Google Gears is beta software offered by Google to enable off-line access to services that normally only work on-line. It installs a database engine, based on SQLite, on the client system to locally cache the data. Google Gears-enabled pages use data from this local cache rather than from the online service. Using Google Gears, a web application may periodically synchronize the data in the local cache with the online service. If a network connection is not available, the synchronization is deferred until a network connection is established. Thus Google Gears enables web applications to work even though access to the network service is not present.
As can be expected from Google there are already some example API's you can take a look at on the google code page.
There is also a movie about the new release here below:
Oracle just released a paper about “Managing Complex XML Data in Oracle XML DB 11g”. As XML is becoming more and more a standard in all kind of applications Oracle likes to stretch out more of the capabilities of XML within Oracle XML DB.
The paper will go into detail about the pros and cons of XML and using Oracle to store and maintain your XML structured data within the XML DB.
When trying to receive some goods you can come across the error message as shown below in text or in the screenshot above here. There are a couple of possible reasons for this error message to appear. You have entered a date that is not in a open GL (General ledger) period, also known as a accounting period. It can be that you have made a mistake by entering the date, the second is that the date is correct however the period is not yet opened.
APP-PO-14376: Please enter a GL Date within an open purchasing period.
Cause: You provided a GL date that is not within an open purchasing period.
Action: Enter a date that is within an open period.
Or use the Control Purchasing Periods window to open the period corresponding to the date you entered. If you're using encumbrance/budgetary control, also open the GL period in the Open and Close Periods window.
If the date you entered is a valid date and the period is still closed you will have to open Purchasing GL period. You can do so via the Purchasing Super User responsibility. From here go to Setup - Financials - Accounting - Open and Close periods. You will be able to open or close a GL periode. After this is done you should be able to receive goods without any error message.
When tuning Oracle jServ you can encounter the security.maxConnections directive in the jserv.properties settings file. security.maxConnections, indeed, sets the max number of connections that jServ can concurrently handle. It can be very tempting to set this to a very high level, this is however not a good idea. First, you have to consider if your server would be able to handle the load of a high number of concurrent connections. Secondly the effect of this setting is related to the number of file descriptors your operating system can handle.
A file descriptor is a handle created by a process when a file is opened. There is a limit to the amount of file descriptors per process. On how to set / reset the max number of file handlers is something I already described in my post file-max and semmni parameters. So when trying to upgrade the max number of connections your system can handle this is something to remember and check before you tune the settings in your jserv.properties file.
When trying to close a inventory period you might end up having some problems due to “Uncosted Transactions”. Uncosted Transactions are an indication you still have records pending in the mtl_material_transactions table. Normally the “Cost Manager” should clear this interface table and take action upon the records in the table. It could however that this concurrent process is somehow stopped or encountered an error during runtime.
To check this you can query the concurrent requests to see if a concurrent request named “Material cost transaction worker” is running. If this is not the case you most likely have a number of records in the mtl_material_transactions table. You can use the query below to check this: select count(1) from mtl_material_transactions where costed_flag in ('N', 'E') order by creation_date;
You will have to do some cleanup work before you can restart the process. First make sure the Cost Manager is NOT running. Now you have to check if there are error records. A error record is indicated by a “e” in costed_flag, you can use the query below to check the error records. You first have to solve the root cause of the error and than restart the Costing Manager.
select request_id, error_explanation, error_code, costed_flag from mtl_material_transactions where costed_flag in ('E') order by creation_date;
After you have solved all the root causes of the error records you also have to take some action on the non processed records which are identified by the letter “N” in costed_flag. You will have to run the update statement below to set all those ready to be processed again.
Update MTL_MATERIAL_TRANSACTIONS set COSTED_FLAG = ‘N’, set TRANSACTION_GROUP_ID = NULL where COSTED_FLAG = ‘E’ or COSTED_FLAG = ‘N’;
After you have taken those steps you can restart the Cost Manager again. Under the Inventory responsibility navigate to Setup – Transactions – Interface Managers. You have to set the frequency of the cost manager and start the cost manager from the Tools menu with the Launch Manager option. This should solve your problems. For more information you can check on Metalink note304313.1 and 105647.1 for more information and background details.
All of them are a good read and it is a shame if you have not been there to view it in real life. However the download can give you a good impression of what you have missed and can bring you up to speed anyway.
In my previous post I reported on my first steps with Google Maps Mobile on my Nokia E61i. I reported that I had to turn of my 3G network. I think this is for a part true, however this morning my train was re-routed on a different track due to some problems and I was looking into the route I was traveling without having 3G turned off. It looks like that if your travel along a GPS tower it will take this as a location point and it will skip all those UMTS towers. UMTS has precedence above GPS on the cell network. So it will depend on your location and the towers surrounding you if you will get a location or not. I still do not know if google simply did not index UMTS towers or that the 3G signal is not working with the application. However, I will be able to get my approximate position anyway I only have to turn of 3G from time to time.
A new version of google maps mobile is released and has a new option. To locate you on the map. This could already be done by GPS but now google released a version which will use the GSM cell tower you are connected to to place you on the map.
I have installed it on my Nokia E61i and in the beginning this option was not working. However it turned out that if you are using a 3G network instead of a plain old GSM network. After setting it to GSM it worked like a charm.
I found out that I made a stupid mistake in a script I am running on a Linux box, the result of this was that there was send a e-mail to my mailbox every 5 minutes so after correcting the mistake I wanted to clean the junk in my Mailbox dir by simply deleting all the messages that where out there. However when issuing the rm command the system presented my with the following error message:
I found a great article on moundalexis.com (http://www.moundalexis.com/archives/000035.php). The solution to the problem was simply do a find and pipe it to xargs rm
Oracle released a paper on how to build your own Oracle RAC Cluster on Oracle Enterprise Linux and ISCSI. The guide will help you to build your own RAC cluster for under US$2600. I however states that this is only for a testing environment and not for a production environment. The paper will go over most basic points of your hardware environment and storage solutions.
On the storage part there are some notes that go “against” the use of a standard NAS/NFS and tend to go more for a Direct NFS Client approach, Direct NFS Clients will integrate NFS directly to Oracle and by doing this Oracle will be able to control and fine-tune the I/O path between the Oracle software and the NFS server resulting in significant performance gains. In the example however will be worked with a ISCSI solution.
For the rest it will guide you step by step installing all the nodes and setting up networking and other environment parts. A great guide when you like to start with RAC in a test environment.
When receiving goods in Oracle InventoryINV the lines will be stored in the RCV_TRANSACTIONS_INTERFACE table so a concurrent process can act upon the data in the table and make the needed changes to for example the current availability. In some cases it might happen that the transaction status stays on “Pending” when the Processing Mode is “Immediate”.
In those cases you can “forward” those records by doing the following steps:
Update rcv_transactions_interface rti set processing_mode_code ='BATCH' , processing_status_code = 'PENDING' , transaction_status_code = 'PENDING', processing_request_id = NULL, validation_flag = NULL where rti.interface_transaction_id in (17583, 17584);
After you committed this update statement you have to run the “Receiving Transaction Processor” concurrent request. The reason for the lines to be stuck can be for example that rvctp was not invoked because the concurrent manager was down.
User experience, more and more software companies are experiencing with new ways of showing various parts of your system.... I found a cool link to a new way of dealing with desktops and more on the smashingmagazine.com website. The Bumptop company has a great way of dealing with the desktop of the future.
BumpTop is a fresh and engaging new way to interact with your computer desktop. You can pile and toss documents like on a real desk. Break free from the rigid and mechanical style of standard point-and-click desktops. Interact by pushing, pulling and piling documents with elegant, self revealing gestures. BumpTop's stunning interface makes clever use of 3D presentation and smooth physics-based animations for an engaging, vivid user experience.
You can view a talk about Bumptop at TED.com and/or view the demo I posted below.
Some time ago when I was working at a previous employee I was part of a team who was evaluating some business intelligence solutions. One of the systems we where evaluating was Microsoft reporting services, to get a good understanding of the system I did some trials, installed the system on a test server and played around with it. Also I have printed a lot of documentation to get a good read about things.
