Before we return to our series of blogs on 10G EM features and functionality, I think it is important for us to discuss the 10G Enterprise Manager Standalone Java Console. Previous blogs have contained information on troubleshooting 10G EM management server to agent communications. But what happens if the entire environment goes down and you still want to use a GUI version of the tool? In addition, the management server version of 10G EM currently does not provide the capability of administering some of Oracle’s advanced features such as streams, advanced queuing, etc..

This blog will show you how to install and configure the 10G Enterprise Manager Standalone Java Console. We’ll also compare and contrast the Java Console to its big brother that uses the management server and repository.

Before we return to our regularly scheduled 10G EM topics, I'd like to provide you with my opinions on Tom Kyte's latest book.

This is the final blog of a three part series on troubleshooting 10G EM agent to management server communication issues. I’ll start the discussion by summarizing a few key points from the previous two blogs.

I’ll also provide a few hints that will help you determine where to start the debugging process (management server or agent). I’ll complete this series by showing you how to activate detailed tracing on the agent and management server components. If you can’t identify the problem by analyzing error messages normally generated by the agents and management server, you’ll have to activate more detailed traces to gather additional diagnostic data.

Let's continue our discussion on debugging 10G Enterprise Manager agent to management server communications. My last blog covered some of Giant Eagle's experiences with agent administration and troubleshooting. But the management agents are only one piece of the puzzle; the problem could also be on the management server. As a result, when debugging communication failures we must also spend some time reviewing diagnostic information generated by the 10G EM management server environment.

As we continue down our path of 10G EM enlightenment, I thought it might be advantageous to deviate from our discussions on the advisors for yet another blog. Keeping the communications flowing between the agents and the management console can be somewhat tricky at times. I will admit that most of our issues were "self inflicted". If you are like us, you'll have to learn how to troubleshoot a problem or two until you gain experience.

We have been installing and administering the 10G agents for some time now and we think we have crested the top of the learning curve. Once you gain experience, the environment pretty works as advertised. I thought I would give you a few helpful facts on how to administer the agents and also provide you with some information on how we solved some of the problems we have encountered. In an upcoming blog I'll describe how to troubleshoot the management server.

We were in the process of discussing 10G EM’s Segment Analyzer before being interrupted by a one terabyte table that I discussed in the last blog. In this blog, I’ll get back on track by reviewing the different alternatives we can use to reclaim wasted space in Oracle10G.

Since the Segment Analyzer and online shrink command were introduced in Oracle10G, I’ll spend a few minutes discussing how you can identify and shrink objects in pre-10G releases.

Consider this to be yet another "Stop the Presses Blog". I originally intended to discuss the different toolsets that DBAs can use to reclaim unused space allocated to Oracle objects. This installment was to be the follow-up to my previous blog on the Segment Analyzer. But we have hit an important milestone here at Giant Eagle, so I thought I would deviate from my intended topic for one blog.

If you have been following this series, you'll know that my unit is responsible for administering several large data warehouses. The warehouse that we most recently converted to 10G has a single, partitoned table that has just grown to one terabyte in size. The entire database is multiple terabytes and rapidly growing at hundreds (and hundreds) of MEGs per month. We estimate that within 2 years, the database will be close to 10 terabytes in size. I thought it might be interesting to interview our "big data" DBA, Jim Dojonovic. Jim is responsible for administering our data warehouses.

The 10G segment advisor identifies segments that have become fragmented as a result of update and delete operations. Oracle describes these objects as being sparsely populated. Not only do sparsely populated objects waste space but they can also contribute to SQL performance problems.

This blog will show you how to use the 10G segment advisor to identify sparsely populated segments. In my next blog, we'll discuss the different alternatives we can use to reclaim the wasted space.

The memory advisors can only be activated when 10G's automatic shared memory management feature is turned off. Since the memory advisors are pretty simple, I thought I would include a discussion on 10G's automatic shared memory management feature. So consider this blog to be your first "two for one special" from your friendly ex-Oracle instructor.

SQL tuning can be a real pain. I have been doing it for 15 years and it is still one of the most challenging parts of my job. I have always described SQL tuning as more of an art than a science. Hopefully, as the SQL Tuning Advisor matures, it will help transform a once time-consuming and arduous burden into just one more line item on a DBA’s “TODO” list. This blog presents an overview of 10G EM’s SQL Tuning Advisor.