Sunday, June 3, 2012 - The Wit and Ramblings of David Giard
Demanding rigidly defined areas of doubt and uncertainty

If you are planning to attend Tech Ed North America or if you will be in the Orlando, FL area June 10, I recommend you check out the INETA Community Leadership Summit.

This free event will bring together leaders in the developer community to exchange ideas with one another. INETA is hosting the event at the Orlando City Hall on Sunday June 10 from noon to 5PM.

As part of the Summit, I will moderate a discussion titled "Getting Others Involved: Building a Leadership Team, Recruiting Volunteers & Growing Your Membership". This is based on my experience helping to grow the Great Lakes Area .NET User Group over the last couple years. Expect lots of interactions and a chance to listen and to share ideas. Other moderated discussion sessions will include “How to start a user group” and “Working with sponsors”.

In addition, this is a great chance to meet the INETA Board of Directors and other influencers in the development community.

The Summit is free and there is even an after-party hosted by Grape City, but you must register in advance at http://inetals2012.eventbrite.com/.

You can find more information at http://ineta.org/summit2012/.

Here is Steve Smith’s presentation on Common Design Patterns at the May 2012 Great Lakes Area .NET User Group (GANG).

Here is Kathleen Dollard’s presentation on .NET Framework Core Features at the April 2012 Great Lakes Area .NET User Group (GANG).

 

In .NET applications, Connection Strings are easily stored in and retrieved from a web.config file or an app.config file. Add a <connectionStrings> section to the config file, directly within the <configuration> section. Within the <connectionStrings> section, place an <add> tag for each connection string you wish to store and retrieve. This <add> tag contains two important attributes: name and connectionString. The name attribute will be used to look up an entry, so each <add> tag should have a unique name. The connectionString attribute contains the connection string you will use to connect to a database. If your application needs to connect to a number of different data types, it is worth setting the Provider property as well. This is the name of a database provider installed on the current machine, that will allow you to connect to a database.
An example is shown below:

<configuration>
  <connectionStrings>
    <add name="MyApp_Dev" 
        connectionString="Data Source=Server01;Initial Catalog=AwesomeDB_Dev;Integrated Security=True"/>
    <add name="MyApp_QA" 
        connectionString="Data Source=Server01;Initial Catalog=AwesomeDB_QA;Integrated Security=True"/>
    <add name="MyApp_Prod" 
        connectionString="Data Source=Server01;Initial Catalog=AwesomeDB;Integrated Security=True"/>
  </connectionStrings>
</configuration>

To retrieve a connection string by its associated name attribute, use the utilities found in the System.Configuration namespace of the System.Configuration assembly. First, set a reference to System.Assembly by selecting the menu option Project | Add Reference, selecting the System.Assembly in the Component Name column of the .NET tab of the Add Reference dialog; and, clicking the OK button. Then, add the following using statement to the top of your class file.

This namespace contains the static ConfigurationManager class, which exposes a number of methods for retrieving configuration information. ConfigurationManager.ConnectionStrings returns a collection containing all connection strings settings in the config file of the current assembly. Once we have this collection, we can retrieve a single item by specifying its name. Doing so returns a ConnectionStringSettings object that contains properties retrieved from a single <add> element of the config file. The important property is ConnectionString, which can be used with ADO.NET to connect to a database. Knowing the Provider property may also be useful if your application connects to different types of data.
Sample code is below.

ConnectionStringSettingsCollection cssCollection =
    ConfigurationManager.ConnectionStrings;
ConnectionStringSettings devCsSettings = cssCollection["MyApp_Dev"];
string devCs = devCsSettings.ConnectionString;

Episode 214

Elijah Manor on AmplifyJS

An App.config file or a Web.config file are great places to store configurable information – information that generally doesn’t change; but we want to be able to change easily (i.e., without rebuilding and redeploying the application.) Examples include connection strings (stored in the config file’s <connectionStrings> section) and application-wide name-value pairs (stored in the config file’s <appSettings> section).

We can add more flexibility by moving a section to an external file and linking to that file from the config file.

By splitting the file, we can manage and deploy only those settings separate from the rest of the configuration.

To do so, we create a new text file and copy that section into that file; then use the configSource attribute of the section tag in the original config file to point to the new file.

For example, the following app.config contains all the application’s connection strings and application settings

<?xml version="1.0" encoding="utf-8" ?>
<configuration>
  <connectionStrings>
    <add name="MyApp_Dev" connectionString="Data Source=SQL071;Initial Catalog=Awesome_Dev;Integrated Security=True"/>
    <add name="MyApp_QA" connectionString="Data Source=SQL071;Initial Catalog=Awesome_Dev;Integrated Security=True"/>
    <add name="MyApp_Prod" connectionString="Data Source=SQL071;Initial Catalog=Awesome_Dev;Integrated Security=True"/>
  </connectionStrings>
  <appSettings>
    <add key="CompanyName" value="The Awesome Company"/>
    <add key="PhoneNumber" value="(513) 555-4444"/>
  </appSettings>
</configuration>

We can accomplish the same functionality as the above app.config by creating 2 files: connections.config and appSettings.config and adding the following code to each file, respectively

connections.config:

<connectionStrings>
  <add name="MyApp_Dev" connectionString="Data Source=SQL071;Initial Catalog=Awesome_Dev;Integrated Security=True"/>
  <add name="MyApp_QA" connectionString="Data Source=SQL071;Initial Catalog=Awesome_Dev;Integrated Security=True"/>
  <add name="MyApp_Prod" connectionString="Data Source=SQL071;Initial Catalog=Awesome_Dev;Integrated Security=True"/>
</connectionStrings>

appSettings.config:

<appSettings>
  <add key="CompanyName" value="The Awesome Company"/>
  <add key="PhoneNumber" value="(513) 555-4444"/>
</appSettings>

Then, point to these files in the app.config, as shown below:

<?xml version="1.0" encoding="utf-8" ?>
<configuration>
  <connectionStrings configSource="connections.config" />
  <appSettings configSource="appSettings.config" />
</configuration>

One caveat to doing this: The configSource files (connections.config and appSettings.config in our example) must be in the same folder as the config file. We can accomplish this by selecting each configSource file in Solution Explorer and setting its Copy to Output directory property to either “Copy always” or “Copy if newer”.

Here is Phil Japikse’s presentation on Testing Code From The Pit of Despair at the March 2012 Great Lakes Area .NET User Group (GANG).

Episode 213

Max Trinidad on Powershell

In October, the Great Lakes Area .NET User Group (GANG) celebrated 10 years this past October with an all-day event. Here is Godfrey Nolan’s presentation on Executable Requirements or BDD in .NET.

The SQL Server master database contains many dynamic views that contain information about the current state of the SQL Server. One such view is dm_os_workers, which lists all active threads run by SQL Server and information about each thread. Of particular interest are the error columns:
One or more of the following bit columns will return 1 if there is anything is going wrong with a thread:

• is_sick
• is_in_cc_exception
• is_fatal_exception
• is_inside_catch

One limitation of this view is that it does not retain any history. If you want to keep a history of threads, you can create a Database to hold that history and the following SQL to copy the results of dm_os_workers to a table in that database. The following SQL copies the dm_os_workers view results to the dbo.ThreadsHistory table in the Instrumentation database.

IF  NOT EXISTS 
    (SELECT * 
    FROM Instrumentation.sys.objects 
    WHERE object_id = OBJECT_ID(N'[Instrumentation].[dbo].[ThreadsHistory]') 
    AND type IN (N'U'))
BEGIN
SELECT
        GETDATE() AS TimeLogged,
        worker_address, 
        status, 
        is_preemptive, 
        is_fiber, 
        is_sick, 
        is_in_cc_exception, 
        is_fatal_exception, 
        is_inside_catch, 
        is_in_polling_io_completion_routine, 
        context_switch_count, 
        pending_io_count, 
        pending_io_byte_count, 
        pending_io_byte_average, 
        wait_started_ms_ticks, 
        wait_resumed_ms_ticks, 
        task_bound_ms_ticks, 
        worker_created_ms_ticks, 
        exception_num, 
        exception_severity, 
        exception_address, 
        locale, 
        affinity, 
        state, 
        start_quantum, 
        end_quantum, 
        last_wait_type, 
        return_code, 
        quantum_used, 
        max_quantum, 
        boost_count, 
        tasks_processed_count, 
        fiber_address, 
        task_address, 
        memory_object_address, 
        thread_address, 
        signal_worker_address, 
        scheduler_address, 
        processor_group
    INTO [Instrumentation].[dbo].[ThreadsHistory]
    FROM sys.dm_os_workers 
    WHERE 1=0
END

DECLARE @TimeNow AS DATETIME
SELECT @TimeNow = GETDATE() 

INSERT INTO Instrumentation.dbo.ThreadsHistory
(
    TimeLogged,
    worker_address, 
    status, 
    is_preemptive, 
    is_fiber, 
    is_sick, 
    is_in_cc_exception, 
    is_fatal_exception, 
    is_inside_catch, 
    is_in_polling_io_completion_routine, 
    context_switch_count, 
    pending_io_count, 
    pending_io_byte_count, 
    pending_io_byte_average, 
    wait_started_ms_ticks, 
    wait_resumed_ms_ticks, 
    task_bound_ms_ticks, 
    worker_created_ms_ticks, 
    exception_num, 
    exception_severity, 
    exception_address, 
    locale, 
    affinity, 
    state, 
    start_quantum, 
    end_quantum, 
    last_wait_type, 
    return_code, 
    quantum_used, 
    max_quantum, 
    boost_count, 
    tasks_processed_count, 
    fiber_address, 
    task_address, 
    memory_object_address, 
    thread_address, 
    signal_worker_address, 
    scheduler_address, 
    processor_group
)
(
    SELECT
        @TimeNow,
        worker_address,
        status, 
        is_preemptive, 
        is_fiber, 
        is_sick, 
        is_in_cc_exception, 
        is_fatal_exception, 
        is_inside_catch, 
        is_in_polling_io_completion_routine, 
        context_switch_count, 
        pending_io_count, 
        pending_io_byte_count, 
        pending_io_byte_average, 
        wait_started_ms_ticks, 
        wait_resumed_ms_ticks, 
        task_bound_ms_ticks, 
        worker_created_ms_ticks, 
        exception_num, 
        exception_severity, 
        exception_address, 
        locale, 
        affinity, 
        state, 
        start_quantum, 
        end_quantum, 
        last_wait_type, 
        return_code, 
        quantum_used, 
        max_quantum, 
        boost_count, 
        tasks_processed_count, 
        fiber_address, 
        task_address, 
        memory_object_address, 
        thread_address, 
        signal_worker_address, 
        scheduler_address, 
        processor_group
    FROM sys.dm_os_workers 
)

You can use SQL Agent to schedule a job that runs this every 60 seconds (or however frequently you want) to keep a history of the threads being generated by SQL. This history can tell you if threads are generating exception and if thread counts are increasing.

---

Steve Latsch contributed to this article.