Feed aggregator

As enterprises look to improve business strategies, big data analytics that draw on information from all mobile channels will be essential to better, real-time decision-making.

According to a study by New Vantage, 85 percent of companies with plans to mine big data cited improved analytics and smarter decisions as primary reasons for these initiatives. Harvard Business Review contributors Paul Barth and Randy Bean explained that speed is a key factor in the effectiveness of decisions, so realizing higher ROI from big data investments depends on analytics that are highly immediate. In order to meet big data goals, CIOs are integrating analytic systems with existing operations for more dynamic data discovery that draws on established algorithms while generating new patterns. With the support of database administration services, enterprises can build an infrastructure that enables a more intelligent model for decision-making. 

New uses for big data BI
Leveraging BI analytics for better decisions has the potential to benefit a range of industries, including healthcare organizations. Healthcare IT News contributor Allen Portela reported that technologies capable of mining big data that has accumulated on a multitude of mobile devices can enable healthcare providers to better measure patient care quality for strategic improvements. Real-time patient data will revolutionize the treatment process because doctors will be able to address medical problems more quickly and informatively. However, these benefits are not possible if data sources remain separated. Portela explained that integrating systems for a unified database infrastructure is imperative to realizing analytic advantages.

XConomy agreed that connectedness is the key to big data success. The source asserted that merging big data analytics with mobile initiatives will result in more contextualized insight, and will therefore offer more actionable value. As mobile machines become increasingly connected through WiFi and Near Field Communication (NFC) technologies, the barrier between people and devices is dissipating, giving enterprises an opportunity to gain a deeper understanding of what user behavior means for the business’ future. XConomy further argues that mobile and big data initiatives will eventually result in more personalized client interactions that will have a more immediate positive impact on business growth.

Firms from retail to healthcare can gain a competitive advantage from big data analysis. By leveraging analytic technologies and remote database support, these organizations can make quicker, more accurate decisions driven by a continual influx of information and new algorithms.

RDX’s business intelligence and big data experts assist customers in leveraging data contained in large data stores. For more information, please visit our Business Intelligence and Predictive Analytics pages or contact us.

In last two postings I wrote about how you might plan a project where you migrate your PeopleSoft application from one database platform to another, and how you might approach the Design phase.  I wanted to share my thoughts about the Build phase in this article.  I'll share my thoughts about the Test and Cutover phases in my next posting(s).

The Build Phase

The Build Phase is always my favorite part of any PeopleSoft project, probably because I come from a Development background.  The Build phase of a replatforming project is in some ways very straightforward, and in some ways it is more difficult.  The problem isn’t in the coding changes – it’s not too difficult to make a piece of SQL work on a different database platform -- the challenge is in Unit Testing.  Every SQL that is touched must be unit tested, and that will be the biggest part of the effort.

Most developers are used to unit testing their own work.  But it's a good idea to use a code and testing review where developers document each object change and unit test and another developer reviews the results.  Since there will be many small changes, the documentation requirement should be light, but it should include a trace file that proves that each App Engine step, PeopleCode SQL, and SQR Function was executed was tested.  How structured your process is will depend on the size and location of your team.  Insuring quality with process and documentation might not be as important in a small shop, but is critical to your success if you have a large development team located off shore.

Unit testing is the only opportunity you’ll have to actually test each piece of modified code.  Subsequent phases will test the system overall, but you will probably not achieve 100% code coverage.  Fortunately, almost all of your defects can actually be caught in unit testing of a replatforming project so you should use this to your advantage.  Defects that get missed will haunt you in later testing phases where they’ll be more visible and more expensive to fix.

Also as part of this phase, your DBA team should  execute another mock cutover using the tools and steps you decided you will use for the real cutover.  The resulting database (plus the code generated in the Build phase) will be the starting point for your first test database.

And the testing team should start building the test scripts for the subsequent test phases here.  Since we’re not changing application functionality, they should be able to leverage existing scripts from prior upgrades or implementation and enhance them for functionality that was added since the original scripts were created.

The good news is that I've successfully got my picture being flagged up with my blog articles in Google Search. 
The bad news is that sometimes (mostly ?) it has the picture from my deprecated 'domain' GPlus account rather than my primary one. I suspect I need to take more drastic steps to eliminate the old profile. 

Akhilesh pinged me today and told me that autoconfig was not working on a newly upgraded R12 instance.  It was failing with these errors:

AutoConfig could not successfully execute the following scripts:   Directory: /r12ascp/erpapp/10.1.3/perl/bin/perl -I /r12ascp/erpapp/10.1.3/perl/lib/5.8.3 -I /r12ascp/erpapp/10.1.3/perl/lib/site_perl/5.8.3 -I /usnapss1/erpapp/appl/au/12.0.0/perl -I /r12ascp/erpapp/10.1.3/Apache/Apache/mod_perl/lib/site_perl/5.8.3/sun4-solaris-thread-multi /r12ascp/erpapp/inst/apps/usnapss1_r12ascp/admin/scripts/adexecsql.pl sqlfile=/r12ascp/erpapp/inst/apps/r12ascp_r12ascp/admin/install     afwebprf.sql            INSTE8_PRF         1     amscmprf.sql            INSTE8_PRF         1     amswebprf.sql           INSTE8_PRF         1     cncmprf.sql             INSTE8_PRF         1     csfadmprf.sql           INSTE8_PRF         1     oksfrmprf.sql           INSTE8_PRF         1
   Directory: /r12ascp/erpapp/inst/apps/r12ascp_r12ascp/admin/install     ibywebprf.sh            INSTE8_PRF         1
If we tried running any of the above scripts, they failed with this error:

SQLPLUS Executable : /r12ascp/erpapp/10.1.2/bin/sqlplus


SQL*Plus: Release 10.1.0.5.0 - Production on Mon Feb 4 12:04:27 2013

Copyright (c) 1982, 2005, Oracle.  All rights reserved.

Enter value for 1: Enter value for 2: Enter value for 3: Connected.
declare
*
ERROR at line 1:
ORA-06502: PL/SQL: numeric or value error: hex to raw conversion error
ORA-06512: at "APPS.WF_EVENT", line 3600
ORA-06512: at "APPS.FND_WF_ENGINE", line 1025
ORA-06512: at "APPS.FND_PROFILE", line 3064
ORA-06512: at "APPS.FND_PROFILE", line 3514
ORA-06512: at "APPS.ADX_PRF_PKG", line 203
ORA-06512: at line 167
ORA-06512: at line 217

So I checked the code on line 3600 for wf_event package:
select line,textfrom dba_sourcewhere name='WF_EVENT'and line between 3550 and 3610;
 3600
 wf_event.local_system_guid := hextoraw(wf_core.translate('WF_SYSTEM_GUID'));I did another code search:
select line,textfrom dba_sourcewhere text like '%WF_SYSTEM_GUID%';This was the result of the query:   select text into l_result from wf_resources where name='WF_SYSTEM_GUID' andlanguage='US';
SQL> select text from wf_resources where name='WF_SYSTEM_GUID' and 2  language='US';
TEXT--------------------------------------------------------------------------------CAFECAFE-0013-0001-0029-ABCDEFABCDEF
That was weird as that is the GUID used by Oracle JRE plugin 1.3.1.29.
I checked the actual GUID:SQL> select guid from wf_systems;
GUID--------------------------------D37180AE23A4479AE04400212846C6CE
Updated the correct GUID:
SQL> update wf_resources 2  set text='D37180AE23A4479AE04400212846C6CE' 3  where name='WF_SYSTEM_GUID' and 4  language='US';
1 row updated.
SQL> commit;

Re-ran $INST_TOP/admin/install/adwebprf.sql apps appspassword ORACLE_SID

New error this time:

SQL> @afwebprf.sql apps appspassword r12ascp
Connected.
declare
*
ERROR at line 1:
ORA-06502: PL/SQL: numeric or value error: character string buffer too small
ORA-06512: at "APPS.WF_EVENT", line 3602
ORA-06512: at "APPS.FND_WF_ENGINE", line 1025
ORA-06512: at "APPS.FND_PROFILE", line 3064
ORA-06512: at "APPS.FND_PROFILE", line 3514
ORA-06512: at "APPS.ADX_PRF_PKG", line 203
ORA-06512: at line 167
ORA-06512: at line 217

Checked the code again:


    3602
 wf_event.local_system_status := wf_core.translate('WF_SYSTEM_STATUS');

SQL> select text from wf_resources where name='WF_SYSTEM_STATUS' and 2  language='US';
TEXT--------------------------------------------------------------------------------CAFECAFE-0013-0001-0029-ABCDEFABCDEF
So I checked with query:
select count(*) from wf_resources where text='CAFECAFE-0013-0001-0029-ABCDEFABCDEF';1207 rows
select count(*) from wf_resources;1775 rows


So I inquired about an existing good instance.  No R12 instance was available, so I copied from an 11i instance with sqlplus COPY command:

First put the tnsnames.ora entry for the good instance in the DB $TNS_ADMIN/tnsnames.ora file
Then used the sqlplus copy command, which is the fastest way to copy data between two oracle databases:

SQL> copy from apps/****@11iascp to apps/****@r12ascp create wf_resources_d1 using select * from wf_resources

Array fetch/bind size is 15. (arraysize is 15)
Will commit when done. (copycommit is 0)
Maximum long size is 80. (long is 80)
Table WF_RESOURCES_11i created.

  1717 rows selected from apps@usnapsd1.
  1717 rows inserted into WF_RESOURCES_11i.
  1717 rows committed into WF_RESOURCES_11i at apps@r12ascp.


Since there was a difference of 58 rows, I first copied those rows in a different table:

create table wf_resources_more as
( select * from wf_resources
where (name,type) in
(
select name,type from wf_resources
minus
select name,type from wf_resources_11i
);

Which created table wf_resources_more with 58 rows.

Then I delete those 58 rows from the wf_resources table:
delete wf_resources

where (name,type) in
(
select name,type from wf_resources
minus
select name,type from wf_resources_11i
);

Then I updated the table with 11i values, through a correlated update statement:
update wf_resources aset a.text=(select text from wf_resources_11i bwhere a.type||a.name = b.type||b.name)
That updated the 1717 common rows with correct values.
Then I inserted the extra 58 rows new in R12
insert into wf_resources (select * from wf_resources_more);commit;
I ran adwebprf.sql again and it gave a new error:
SQL> @afwebprf.sql apps appspass r12ascpConnected.declare*ERROR at line 1:ORA-01403: no data foundORA-06512: at "APPS.WF_EVENT", line 3604ORA-06512: at "APPS.FND_WF_ENGINE", line 1025ORA-06512: at "APPS.FND_PROFILE", line 3064ORA-06512: at "APPS.FND_PROFILE", line 3514ORA-06512: at "APPS.ADX_PRF_PKG", line 203ORA-06512: at line 167ORA-06512: at line 217
Again I checked code on line 3604 of WF_EVENT:
 select name into wf_event.local_system_name
     3605 from wf_systems
     3606 where guid = wf_event.local_system_guid;Then I realized that after the correlated update, the wf_resources table now had the GUID of 11i system:
SQL> select name,text 2  from wf_resources 3  where name='WF_SYSTEM_GUID';
NAME------------------------------TEXT--------------------------------------------------------------------------------WF_SYSTEM_GUIDCA43363B52162863E04400212846C6CE
So I updated it again:
I checked the actual GUID:SQL> select guid from wf_systems;
GUID--------------------------------D37180AE23A4479AE04400212846C6CE
Updated the correct GUID:
SQL> update wf_resources 2  set text='D37180AE23A4479AE04400212846C6CE' 3  where name='WF_SYSTEM_GUID' and 4  language='US';
1 row updated.
SQL> commit;

Re-ran $INST_TOP/admin/install/adwebprf.sql apps appspassword ORACLE_SID
SQL> @afwebprf.sql apps appsapss1 usnapss1Connected.[ APPS_WEB_AGENT ]Application Id : 0Profile Value  : http://usnapss1.tsg.ge.com:8004/pls/usnapss1Level Name: SITEINFO           : Updated/created profile option value..Deleted : 0[ APPS_SERVLET_AGENT ]Application Id : 0Profile Value  : http://usnapss1.tsg.ge.com:8004/OA_HTMLLevel Name: SITEINFO           : Updated/created profile option value..Deleted : 0[ APPS_JSP_AGENT ]Application Id : 0Profile Value  : http://usnapss1.tsg.ge.com:8004Level Name: SITEINFO           : Updated/created profile option value..Deleted : 0[ APPS_FRAMEWORK_AGENT ]Application Id : 0Profile Value  : http://usnapss1.tsg.ge.com:8004Level Name: SITEINFO           : Updated/created profile option value..Deleted : 0[ WF_MAIL_WEB_AGENT ]Application Id : 0Profile Value  :Level Name: SITEINFO           : Updated/created profile option value....
It succeeded.
I told Akhilesh about the result and asked him to run autoconfig.  We tailed the log at
$INST_TOP/admin/log/02042015/adconfig.log
Adconfig succeeded without errors this time:
AutoConfig is exiting with status 0
AutoConfig execution completed on Mon Feb  4 20:17:58 2013
Time taken for AutoConfig execution to complete : 2 mins  38 secs
I was overjoyed, and so was Akhilesh.  We still had 53 rows which had the value of text as:CAFECAFE-0013-0001-0029-ABCDEFABCDEF
select count(*) from wf_resourceswhere text='CAFECAFE-0013-0001-0029-ABCDEFABCDEF';53 rows.
We would investigate further, how this had happened.  I suspect human error.  Human error makes troubleshooting more interesting, just like human error makes chess more interesting.  If it was a computer playing chess, it would not make any mistakes and win the game.  It is the human errors which make chess and troubleshooting, an interesting pursuit.

There is a new Oracle book available covering data management and upgrade for PeopleSoft systems, and you can listen to a podcast with the authors, Jim Marion and Paula Dean to learn more about this book.  In the interview they discuss the value of the book, why they wrote it, and who its intended audience is.  They also talk about some of the new PeopleSoft technology that supports data management and upgrade activities, and explain how the book can help maximize data integrity, maintain peak application performance, and keep your PeopleSoft environment up to date. The book covers the latest techniques and provides best practices, case studies, and programming examples. You can learn how to develop and deploy data mover scripts, audit database health, apply patches, and generate project comparison reports.

You can listen to the podcast here.

You can acquire the book or learn more about it here.

Oracle's Author Podcasts are posted here.

What has six conferences in six cities across India and features presentations by thirteen experts in Oracle technologies? If your answer was OTN Yathra 2013, give yourself a round of applause.

OTN Yathra 2013, organized by Oracle ACE Director Murali Vallath, begins in Delhi on Saturday, 16 February 2013, and then makes its way to Mumbai (Monday,18 Feb), Pune (Wednesday, 20 Feb), Bangalore (Friday, 22 Feb), Hyderabad (Monday, 25 Feb), and concludes in Chennai on Wednesday 27 February. Among the thirteen speakers for the event are eight members of the Oracle ACE Program.

The goal of OTN Yathra 2013 is to reach out to the 400,000-member Oracle user community in India to "bring the Oracle community together, giving them awareness,improving the level of knowledge, and increasing the networking opportunities in the region."

Click here for registration and more information

Pythian now officially supports MongoDB both as On-Demand and Managed Services offerings. We’ve been dipping our toes into the MongoDB pool for some time now, as more and more of our customers adopt MongoDB into their data infrastructure, but now its official. Part of the decision to officially support MongoDB was driven by our desire...

I’m sure we’ve all dabbled with VMs a time or two (and if you haven’t, you should). However, I had a specific goal of creating a robust purely virtual environment for training, labs, and testing. While a plain old Linux VM loaded over Windows, Linux, or Mac would have been fine, it doesn’t have the flexibility to really give me all the power out of my hardware. At best I’ll have a couple VMs running at once, because at some point I’ll run out of RAM and my host OS won’t be able to handle it.

So, enter the vSphere Hypervisor. Using this software, I was able to create my own ‘whitebox’ out of commodity equipment from Amazon. The whole thing came to about $840 (though price increases make it about $900 now). Check out the stats:

Server Setup:

• AMD A10 Trinity 3.8GHz Quad Core Processor
• 32GB DDR3 1600 RAM
• 64GB SATA3 SSD (OS and NAS)
• 4TB 7200RPM SATA3 Disk
• VMWare vSphere Hypervisor Free 5.1

An exact hardware list is at the bottom of this post. The high level process is like this:

1. Put the server together. That was easy, right?
2. Burn the vSphere Hypervisor software to a USB thumb drive and boot up with it. Pause to enjoy the new UEFI (replacement to BIOS) with mouse support. Important: Make sure you use a USB 2.0 port for the thumbdrive. The UEFI doesn’t seem to play nice with booting from a USB 3 port.
3. Install ESXi to the 64GB SSD and use it as your boot volume (a decent guide can be found here).
4. Once ESXi is installed and loaded up (minus the boot drive), you can use another computer to go to the link displayed on the ESXi server. From there you can download the vSphere Client and connect to your ESXi server to create virtual machines.
5. Once you’re in the client and connected, the first step is to create a FreeNAS Virtual Machine in the default “datastore1″ datastore. Here is a good guide for that.
6. Using FreeNAS, you can create a ZFS volume out of the 2x2TB drives (I used RAID0) and configure them for iSCSI. Another guide!
7. Lastly, you need to setup ESXi for iSCSI and create a datastore out of your new 4TB volume. And here’s a guide for that.

Once you’re done with all these steps you basically have the ability to create as many virtual machines with any combination of four CPU cores (1×4, 2×2, 4×1) up to 32GB RAM using 4TB of disk space on your new datastore. On mine, I have Oracle 11gR2 on OEL6, a three node MySQL Cluster on OEL6, MarkLogic 6 on CentOS 6, OpenVPN, Postgres, and two NoSQL/Apache/PHP boxes. With that, I’m at about 30% capacity and barely a whimper on the CPU monitor.

There are a ton of extra tricks and things you can do with this setup. Since the motherboard/CPU support IOMMU (AMDs virtualization passthrough), you can enable it in the UEFI and it will auto-enable in ESXi. This gives you the ability to passthrough PCI, PCIe, and USB ports straight to a VM for a lot of awesome possibilities. You could also look at something like PLEX to make a Media Server, set up a Minecraft Server for the kids (sure, the kids), anything you want. You’ll have tons of room.

Random Notes:

• I’d recommend setting up a dedicated IP range in your home network for this. That way you can keep it all organized and assign static IPs to all the VMs.
• One really good VM is a VPN server. I used OpenVPN which is pretty decent and easy to set up. Forward a port on your router to it for an instant VPN into your home network.
• One downside of the FreeNAS setup of a VM inside your ESXi server is that when the server comes up, none of the VMs in your iSCSI datastore will be able to come on by default. You will need to start the FreeNAS VM and re-scan the iSCSI interface in ESXi each time after a reboot. Thankfully, you shouldn’t have to reboot the actual server very often. If you know a way around this, let me know!
• From what I understand, passthrough of a PCIe Video Card to a VM is possible with IOMMU enabled but is a pain in the butt. With Windows it’s nearly impossible, and a bug requires the Windows VM use no more than 2GB RAM. I think with Linux it is quite a bit easier. Also note, while the AMD A10 processor is an APU (CPU with a built in GPU), it can’t be passed through.

Here’s the final hardware list:

• $109.99 – Thermaltake ARMOR A30 VM70001W2Z Black SECC MicroATX Mini Tower Computer Case
• $58.24 – Corsair Builder Series CX 500 Watt ATX/EPS  80 PLUS (CX500)
• $4.40 – OKGear 18 inch Blue Premium SATA III Round Cable 6GB/s Straight to Right Angle w/latch
• $100.38 – Crucial m4 64GB 2.5-Inch Solid State Drive SATA 6Gb/s CT064M4SSD2
• 2 x $104.99 – WD Green 2 TB Desktop Hard Drive: 3.5 Inch, SATA III, 64 MB Cache – WD20EZRX
• $123.79 – AMD A10-5800K APU 3.8Ghz Processor AD580KWOHJBOX
• 2 x $89.99 – Corsair Vengeance 16GB (2x8GB) DDR3 1600 MHz (PC3 12800) Desktop Memory (CMZ16GX3M2A1600C10)
• $119.99 – ASUS AMD A85X SATA 6.0 Gb-s 240-Pin DDR3 2400 Motherboards F2A85-M PRO

TOTAL: $906.75

The post Make your own training/lab environment for $900 appeared first on Steve Karam :: The Oracle Alchemist.

Enkitec E4, the only exclusively Exadata-focused conference in the world, is going to be back this this year too! :-)

It will take place on 5-6 August 2013 in Four Seasons Hotel & Resorts in Irving, TX. It will be very hot in Texas then, but the hotel has a really cool pool there (which I intend to be able to use this year – last year I had to spend all my free time working in the hotelroom unfortunately). And they have beer inside too :)

There will be very good speakers showing up this year too (including the keynote speaker who’s awesome, but I can’t say the name yet) and I’ll hang out there too. I think I’ll deliver a live demo / internals hacking session this year (that way I don’t have to spend my night before the conference preparing slides and can focus more on the cool stuff).

So check out the Enkitec E4 page, it has a video from last year’s event and also a link for submitting your abstracts for the conference!

See you there! (in person or virtually :)

 

Related Posts

• ExaSnapper 0.7 beta download and the hacking session videos
• Do you like to watch TV? Watch Enkitec.TV from now on! :-)
• Enabling and Reading event 10046 / SQL Trace
• Even more Snapper – v4.03 now works in SQL Developer…
• Snapper v4.02 and the Snapper launch party video

I like design patterns.  Actually, I love design patterns.  I'd much rather leverage the experiences of others than burn my calories reinventing the wheel.

Oracle's UX Design patterns have really saved my bacon lately.  I've been working with building some very cool stuff with ADF Essentials.  Design patterns have not only made my work product better, they've also saved me time and prevented me from doing some really stupid things (all of which would have needed unwinding and rewinding prior to release).  It's really true:

Those who cannot remember the past are condemned to repeat it. - George Santayana, The Life of Reason, Vol. I - Reason In Common Sense.

So, if you're building your own apps...especially if you're building them with Oracle tools or on Oracle technology...save your own bacon here.

Good morning and welcome to Monday...

Here in the States, this is the day to recover from whatever Super Bowl festivities you might have had last night. While I personally wish my Patriots had been playing the Niners, there is always next year.  But that same approach can get you in trouble when it comes to how your organization stores and provides access to information.  The idea that you will "get to it next year" can end up costing you a lot of time, money and angst this year.

I recently had a little Q&A with Bryant Duhon from AIIM about managing information.  We are planning for the upcoming AIIM show in New Orleans (http://www.aiimconference.com) and Oracle is pleased to be a platinum sponsor of the event.  We hit on a topic that has been written about before but still remains a vexing problem for companies around the world.  Over time, information that has real value, and needs to be reused or accessed by other colleagues is scattered across different locations and different systems - all of which are usually not connected to one another.  These locations are often referred to as "information silos" because of how they isolate information within individual repositories.

Bryant asked me this question, "We've been talking about information silos in this industry for at least two decades; how do we move from talking to doing?"  It's a good question.  Clearly if it was so easy to do, it would not still be coming up as an issue! Personally, I've come to the conclusion that eliminating information silos is a worthwhile but not always achievable goal for most organizations.  In my response, I stated that eliminating information silos has been the ideal scenario for many years. But the reality is that over time, organizations take on new technologies for information management but are unable or unwilling to wean themselves from the earlier, now legacy, technologies. Eventually a company may find itself with several different content management systems, each still used by one or more departments for a vital business function. Software vendors periodically attempt to address this problem with new technologies that often end up adding additional layers of complexity. IBM Lotus Notes and Microsoft SharePoint are popular examples of collaboration and information sharing tools that instead of delivering on the promise of simplified content management, have over time exacerbated the information silo problem for most organizations.  The Cloud and its many variations, now offers a new take on this decades-old information management challenge, but that's a topic for another day.

So what is an information management professional like yourself supposed to do?  Will you just throw up your hands and allow new repositories to keep popping up like so many groundhogs (slightly apropos this week) across the IT landscape?  Well, that is not such a great idea either.  The more disconnected your approach to information management, the more disconnected your vital business systems become, the more fragmented your operations become and the more inefficient your employees and colleagues will be.  If anyone ever needs to find anything, they will first need to know where to start looking.  Good luck with that!  

The approach that often succeeds for many businesses is to take a step back, evaluate the information landscape and consider strategic improvements to how information is shared, utilized, archived and controlled.  Then take positive steps to minimizing the number of repositories in use with an eye toward integrating as many business processes as possible around one "primary" information store.

Since content migration is frequently difficult or prohibitively expensive, the best opportunity for making actual progress on this front is for organizations to standardize on one ideal repository moving forward. When this is not possible, the goal should be to minimize and reduce the number of content repositories for all business applications moving forward. Using application integration frameworks and connectors, this new go-forward repository must be leveraged within the context of as many business processes as possible – which means that finding the right information management system that aligns with your overall business strategy should be one of the leading selection criteria.

Stopping the creation of new information silos is job one. Then over time, business processes and applications can be updated to take advantage of this new repository and legacy systems can be put into maintenance mode where older information is still accessible but no new content is being added. This takes time, proper planning, and a firm hand at the wheel that helps minimize distractions created by the latest shiny new technology in order to drive information management costs down over time.  If you can spare 30 minutes or so, I recommend you take a look at a webcast we did here at Oracle about ways to think about consolidation and the rationale (cost savings!!) for doing so.  Click here to register and watch it immediately.

You can read the rest of our conversation on the AIIM blog, here: http://www.aiim.org/community/blogs/expert/ECM-One-Repository-and-the-Key-Role-of-Classification. As always, we welcome your comments, suggestions, ideas and feedback here on the WebCenter blog.  If I don't hear from you here, I hope to see you at the AIIM Conference in New Orleans next month!

A friend of mine the other day said an amazing thing I like coding in C++ I mean, seriously?  The land of friends, of people writing C code and debugging nightmares, had things got that much better, I mean I know there are some good threading libraries now but seriously, C++ is nice? All of the idiots code in Java, they don't know C++ And there we have the point.  Its not about what technology

A spike in SQL injection attacks has caused enterprises to re-consider security measures to prevent a devastating breach from poor database monitoring.

International Business Times reported that after Yahoo experienced one such attack at the end of last year, security professionals began to focus on vulnerabilities due to third-party code. Hackers are increasingly exploiting these loopholes to gain access to the database, steal information and exercise complete control of all systems.

FireHost’s recent report on the “Superfecta” of major cyber-attacks revealed that Cross-Site Scripting and SQL injection occurrences like Yahoo’s have been on the rise since the third quarter of last year. In fact, instances of cross-site scripting increased by approximately 160 percent in the final three months of 2012. Chris Hinkley, senior security engineer at FireHost, commented on the study’s implications.

“The change in frequency of the types of attack between quarters gives you an idea of how cyber-criminals are constantly working to identify the path of least resistance,” he explained. “This is why it is important to have an understanding of the kind of traffic that is accessing your hosted infrastructure, so that you can make sure that malicious traffic is diverted and that there is less risk to sensitive data.”

Security requires continual supervision
A major reason for SQL injection susceptibility is many enterprises with cloud-based servers still have no logging capabilities for file exchange. David Gibson, vice president of strategy at Varonis, told the source that this lack of visibility into the network keeps businesses blind to impending threats.

“We have found that, after a workload is virtualized, the actual details of managing file permissions and monitoring access is considered to be automatically ‘taken care of,’ ” he said. “It is also quite possible that the teams managing virtualization projects see file security and governance as outside their discipline. The security team may have no visibility of what is happening.”

While virtualizing systems undoubtedly gives enterprise employees flexibility in accessing business-critical data, without remote support, the database is vulnerable to continually diversifying SQL injection attacks. Enterprises can utilize database administration services to constantly monitor the SQL database, as well as set permission controls for file access to grant the advantages to cloud hosting without the risk of attack.

RDX’s highest priority is safeguarding customer information. To learn more about how RDX ensures data security, please visit our Focus on Security page or contact us.

Well my Oracle Cloud trial versions are active now and I have something to play around for a month.
The activation took a long time for me and there are lot of people out there who already had their hands on.

So what are the tools that are basically used on the Oracle Cloud offering

• Oracle Database
• version provided is 11g release 2 enterprise edition
• managed and secured on exadata
• data access and manupulation with sql and plsql standards
• apex - my favorite 
• data movement using sql developer
• it has a lot of productive applications
• it is mobile enabled
• Oracle Java
• it is all java out there
• enterprise manager for managing
• it runs on exalogic
• its associated database runs on exadata
• it runs on weblogic
• it uses identity managment for user/group management
• it uses access manager for single signon

Good to here from Oracle and I am going to explore it in the next up coming days. 

In march 2011, my colleague Gregory Steulet published an article about Simulating and testing I/O performances with Orion. In this article, he wrote about a bug concerning Orion on Oracle 11.2, reported in Oracle bug number 9104898: "ORION FAILS WITH ORA-27061: WAITING FOR ASYNC I/OS FAILED".

 

Trying to compare performances of storage devices on a virtual machine between a classical and a paravirtual storage adapter, I experienced the same error when using large IOs. I was using a Red Hat 5 based environment, which seems to be especially concerned by this problem: 

 

Error completing
IO(storax_aiowait)
ORA-27061: waiting for async I/Os failed
Linux-x86_64 Error: 14:
Bad addressAdditional information: -1
Additional information: 1048576
Test aborted due to errors.

 

If you have already faced the problem too, there is good news! Oracle finally published a patch in November 2012 to fix this problem, reported in January ... 2009!!! The patch is available for Oracle RDBMS 11.2.0.2 and 11.2.0.3.  You can download it at My Oracle Support website, searching for Patch 9104898 or Patch ID 15597001.

 

It works fine for me now.

 

Don't forget to download the last version of OPatch (6880880 in My Oracle Support) before applying this patch, as recommanded by Oracle.

Pour ceux qui se posent la question de la haute disponibilité de la solution Enterprise Manager 12c, vous trouverez des réponses dans le document "Deploying a Highly Available Enterprise Manager 12c Cloud Control" (complément à la documentation).

Once you have Migration Plans created the next step is assembling them into Migration Requests. The Migration Request is a collection of unrelated Migration Plans and associated filter criteria to decide the scope of the migration.

To create a Migration Request navigate to the Administration Menu and select the M --> Migration Request menu option and fill in the following:

• Migration Request - Name the Migration Request (it should be prefixed with CM to seperate it from the Migration Requests delivered with the product).
• Description - A short description to describe the Migration Request
• Detailed Descrption - A detailed description of the Migration Request
• Migration Plans - A list of Migration Plans to include in this Migration Request and the Selection Criteria to use to subset the requests. This means the following:
  • Migration Plan - The Migration Plan to include in the Migration Request. The Migration Plans in the Request are NOT related. Any relationships are documented in the Migration Plans.
  • Selection Type - The criteria to select the subset of records in the Primary object in the Migration Plan. The Configuration Migration Assistant supports SQL based, XPath based or Algorithm based criteria.
  • Key Selection - The selection criteria to use in the format as specified in the Selection Type. This is SQL WHERE clause, XPATH statement or algorithm to use for selection. You can yse the help icon to find examples.

   For example:

The above example uses the (1=1) SQL clause to indicate that ALL records are migrated.

Note: The product supplies the majority of the Migration Requests you would use in the implementation. This step is only necessary if you wanted to copy a base Migration Request and alter it or add custom Business Objects to Migration Plans/Requests.

This concludes the configuration of the Configuration Migration Assistant. The next blog entries on this subject will discuss the execution components of the feature.

For more information about this aspect of the Configuration Migration Assistant and other aspects refer to the Configuration Migration Assistant Overview (Doc Id: 1506830.1) whitepaper available from My Oracle Support.

A very informative Post by Rajesh Dhanapal from Infosys.

In the current era the customers are not interested to wait for months and years for BI project implementation. The customers are keen on quick wins rather than implementing the BI solution from scratch by following traditional BI approach. The Oracle BI Apps solution provides head start for the organisation to reach the to-be state quicker with reduced time and effort, and reduced risk.Read Details

One of the great things about working here at Rittman Mead is our corporate ethos of sharing our knowledge and experience. Of course, much of this is “paid for” work with our customers where we train, mentor, consult, develop, implement and support all manner of things Oracle BI. However, sharing with the community is also a major feature of our culture; there is, of course, the Rittman Mead Blog, but we are also keen supporters of user groups throughout the world.

I am delighted to say that Rittman Mead will be at the RMOUG conference being held in Denver, CO between February 12 and 13, 2013. I love this conference – it is large enough to allow several parallel streams and attracts many world class speakers, yet the conference also manages to remain a close gathering of friends.

This year we will have four members of our team presenting: Stewart Bryson, Jordan Meyer and Michael Rainey from Rittman Mead North America, I will be representing the European offices. Between us we will presenting 7 sessions over the event.

• Reporting against Transactional Schemas with OBIEE 11g – Stewart Bryson, Feb 12 8:30 am.
• Aggregation: The BI Server versus the Oracle Optimizer - Stewart Bryson, Feb 12 2:30 pm.
• GoldenGate and ODI – A Perfect Match for Real-Time Data Warehousing – Michael Rainey, Feb 12 5:15 pm.
• Social Network Analysis with Oracle Tools – Jordan Meyer, Feb 13 9:45 am.
• Data Science for OBI Professionals - Jordan Meyer, Feb 13 11:15 am.
• Extending Oracle’s Data Warehouse Reference Architecture for a Real Time and Big Data World – Peter Scott, Feb 13 1:30 pm.
• Tuning “Real Time” Data Warehouses – A Guide from the Field - Peter Scott, Feb 13 4:00 pm.

In addition Stewart will also be co-presenting with Kent Graziano on “Using OBIEE and Data Vault to Virtualize Your BI Environment: An Agile Approach”

After the meeting we will posting our presentations on the articles page of our website.

If you see any of us in Denver then come up and say hello – we would love to meet you.

After boring myself to tears watching The Hobbit, I lost my cinema mojo for a while. I’ve struggled to motivate myself to get involved. Tonight I decided to give Bullet to the Head a shot ( ouch ) at revitalizing my cinema spirit and you know something? It did it!

Stallone is a low grade hitman. His partner gets killed, so he accidentally teams up with a cop and goes on an ultra-violent killing spree of revenge.

The film plays to all Stallone’s strengths. He is doesn’t get involved in lengthy dialog. His lines are simple and direct, with a lot of attitude and a reasonable amount of humour. It’s a very basic action flick, but I think it delivers very well. If Stallone can keep delivering stuff like this, I’ll keep going to the cinema to see it.

On a mildly related note, I watched Dredd (2012) last night on cable at a mates house. It was pretty good. I should have seen it on the big screen.

Cheers

Tim…

Bullet to the Head… was first posted on February 2, 2013 at 11:13 pm.
©2012 "The ORACLE-BASE Blog". Use of this feed is for personal non-commercial use only. If you are not reading this article in your feed reader, then the site is guilty of copyright infringement.