DbContext behave differently within different classes - c#

I have a local ASP.NET MVC 3 Windows Azure Project where I use a local instance of MSSQL Express for my emulator.
In my web.config I have the following connection string:
<add name="ExampleDb" connectionString="data source=.\SQLEXPRESS;Integrated Security=SSPI;Initial Catalog=ExampleDb;User Instance=true;MultipleActiveResultSets=True" providerName="System.Data.SqlClient" />
For debugging purposes I have the following in my WebRole.cs file:
public override bool OnStart()
{
ExampleDb context = new ExampleDb();
context.ExampleItemEntries.Add(new ExampleItem() { ExampleItemId = 1, Att1 = "1", Att2 = "2" });
context.SaveChanges();
return base.OnStart();
}
When I perform this action I can actually see the content in my code-first generated database (using Entity Framework). Database: ExampleDb, Table: ExampleItem.
However, doing the exact same thing in ExampleController does NOT update the database. Instead this data goes somewhere else (I have no idea where). When I visit my controller via the browser I can upload data and look at it but it is not stored in my ExampleDb database, only data sent from WebRole.cs is in the database.
Any ideas what's causing this behaviour?
ExampleDb.cs looks like this:
public class ExampleDb : DbContext
{
public ExampleDb() : base("ExampleDb") { }
public DbSet<ExampleItem> ExampleItemEntries { get; set; }
}

Your connection string contains 'User Instance=True'. This means that SQLEXPRESS uses a user instance to store your data. This is a completely different instance of SQL Server that contains a new set of databases.
I assume the code in WebRole.cs is run in a different user account than the code in ExampleController. When the ExampleController is run under a user account that is not an administrator, a user instance is used...
So when you remove 'User Instance=True' from your connectionstring and assign the proper access rights to your SQLEXPRESS database instance, that would solve the problem.

In CodeFirst Model you should define the connection string in the web.config first as below:
<connectionStrings>
<add name="ExampleDbConnection"
connectionString="data source=.\SQLEXPRESS;Integrated Security=SSPI;
database=ExampleDb;
AttachDBFilename=|YouDataDirectory|yourdb.mdf;
User Instance=true"
providerName="System.Data.SqlClient" />
<connectionStrings/>
The DbContext class constructor accepts a name-value pair specifiying connection-string name in web.config. So you can reference this connection string in your context:
class ExampleDb : DbContext
{
public ExampleDb() : base("name=ExampleDbConnection") { }
...
}
This code does work fine in a ASP.NET MVC Web Role so you can try as above.

Related

Correctly storing db connection strings in environment variables ASP.Net MVC Apps

I am trying to move my 2 database connection strings to environment variables for security reasons. Everything works fine when I include the 2 connection strings on web.config like so:
<connectionStrings>
<clear />
<add name="DefaultConnection" connectionString="Data Source=xxxxxx" providerName="System.Data.SqlClient" />
<add name="RDSContext" connectionString="Data Source=xxxxxx" providerName="System.Data.SqlClient" />
</connectionStrings>
I then removed the 2 connection strings from web.config and created 2 environment variables as follows:
setx CUSTOMCONNSTR_DefaultConnection "Data Source=xxxxx"
setx CUSTOMCONNSTR_RDSContext "Data Source=xxxxx"
Although I now get the following error when I startup IIS and visit the web app
Server Error in '/' Application.
Cannot attach the file 'C:\Users\xxx\xxx\App_Data\BookingSystem.Models.RDSContext.mdf' as database 'BookingSystem.Models.RDSContext'.
Can anyone tell me what I am doing incorrect?
Simply naming your environment variable in a certain way doesn't mean that MVC will pick them up. You may be assuming a little more magic is happening than there actually is.
When you do ConfigurationManager.ConnectionStrings["MyConnectionStringName"].ConnectionString you're only retrieving the connection string from your web.config file, or machine.config if you happened to have defined it there. There's nothing there to tell your app to grab it from anywhere else.
Instead, you might consider creating a class to represent your configuration:
public class MyApplicationDatabaseConfiguration
{
public string ConnectionString { get; set; }
}
Any class that needs to obtain this connection string can do so by depending on an instance of that MyApplicationDatabaseConfiguration.
public class MyDatabaseRepository
{
readonly MyApplicationDatabaseConfiguration _dbConfig;
public MyDatabaseRepository(MyApplicationDatabaseConfiguration dbConfig)
{
_dbConfig = dbConfig;
}
public void DoSomethingWithTheDatabase()
{
using(var connection = new SqlConnection(_dbConfig.Connectionstring))
{
//now you can use the connection
}
}
}
Then you can load your config however you like. Ex:
MyApplicationDatabaseConfiguration dbConfig = new MyApplicationDatabaseConfiguration();
dbConfig.ConnectionString = Environment.GetEnvironmentVariable("MyApplication_MyConnectionString");
Now the MyDatabaseRepository doesn't care how you load the database config, it simply says "I need the database config". This is extremely powerful, it allows you to change out the configuration by simply changing a line of code. Need to change back to using web.config? It's as simple as dbConfig.ConnectionString = ConfigurationManager.ConnectionStrings["MyConnectionStringName"].ConnectionString;
Microsoft has taken this one step further in Microsoft.Extensions.Configuration library, which is meant for .NET Core but targeted to .NET Standard, so you can use it in your .NET Framework library if you'd like.

Entity Framework - Invalid Object with custom connections string

Just for a brief overview this is how I added the database into my project:
I have added a datasource by adding an ADO.NET Data Model Entity and selecting EF Designer from Database.
Doing so has generated a connection string for me in my web.config. Integrated Security is set to true (if that matters).
Once connected I right clicked and selected 'Update model from database.
Since there is multiple environments I built a custom context with a parametized constructor. Code looks as such (condensed and censored):
public partial class DataEntities : DbContext
{
public DataEntities(string connectionString)
: base(connectionString)
{
}
}
With each environment there are different local sql accounts associated. Requiring me to generate a connection string that is associated with the correct account in the correct environment. Doing so my web config looks something like this:
<connectionStrings>
<add name="DataEntities" connectionString="metadata=res://*/Models.XXModel.XXModel.csdl|res://*/Models.XXModel.XXModel.ssdl|res://*/Models.XXModel.XXModel.msl;provider=System.Data.SqlClient;provider connection string="data source=MYSQLSERVER;initial catalog=Data;integrated security=True;MultipleActiveResultSets=True;application name=EntityFramework"" providerName="System.Data.EntityClient" />
</connectionStrings>
<secureConnectionStringsSection passwordPolicy="AllowLocalPasswordsForConnectionStrings">
<secureConnectionStrings>
<add name="DataEntities-Local" providerName="System.Data.SqlClient" connectionString="data source=MYSQLSERVER;user id=ACCNT; password=PASSWORD;" />
</secureConnectionStrings>
</secureConnectionStringsSection>
When I am instantiating my DataEntity object, I am calling a helper function I have written to get my custom connection string. I am doing so via the following code:
private DataEntities adDB = new DataEntities(XXX.Helpers.EFDBHelper.getDataConnectionString());
I can verify that my connection string helper does correctly pull the custom connection string that I have in my web.config. However once I actually try to make a call on the database I am left with the following error:
Invalid object name 'dbo.mycolumn'.
I know that my parametized constructor is being called with the correct connection string. I also know my connection string is valid, I have tested it in powershell:
$conn = New-Object System.Data.SqlClient.SqlConnection
$conn.ConnectionString = "data source=MYSQLSERVER;user id=ACCNT;password=PASSWORD"
$conn.Open()
$conn.Close()
I am certain that it is not an issue with the database itself as well. The moment I remove the parameter from my Entity initialization:
private DataEntities adDB = new DataEntities();
I am able to pull data from the database. I am assuming that it uses the auto-generated connection string. Which won't work since I cannot use integrated security once it goes past my local environment.
My apologies, I am new to the technology here. I am sure that it is just something small that I am missing.
In my connection string I did not specify a catalog.
Since I am newish to the technology, I tried to simply replicate some of the existing code that I had inherited. In the connection string I tried to replicate from, no catalog was expressed as well.
However... the account I am using to talk to the new datasource has access to all levels of the database so the catalog was required. The old inherited database had only one catalog the the sql account had access too.
You live and you learn!

Remove ConnectionString from Web.config using Entity Framework CodeFirst

Transform web.config won't work well for Entity Framework CodeFirst because the connection string is added dynamically after transform when publishing. We have a dynamic connection settings being set in code for CodeFirst when releasing to different environments.
I have tried the correct removal method but the published results ends up with the connection string in the web.config.
Are there any work arounds for this? Putting connection info in our web.configs isn't an option. Adding the connection info dynamically at runtime works well.
The problem is when publish adds the connection info in the web.config the app gets a 500 error because codefirst is attempting to use an invalid connection string for the environment that is not sandbox where it was created.
We are changing that at runtime here and that is working.
public MyAppDataContext()
{
this.Database.Connection.ConnectionString = OurDynamicConfigSettings.GetSetting("MyAppConnectionString");
}
The codefirst though is attempting to use what is in web.config before we set it dynamically. The fix is to remove the connection info from the web.config.
It all works except when building on build server or publishing, codefirst inserts the connection info back into the web.config every time we publish. Having to remember to remove this each time is not good practice and prone to errors if you forget.
Code for removing connection string in our transform file should work but doesn’t.
<connectionStrings>
<add xdt:Transform="Remove" xdt:Locator="XPath(configuration/connectionStrings[#name='MyAppConnectionString'])" />
</connectionStrings>
Your transform should be:
<connectionStrings>
<add xdt:Transform="Remove" xdt:Locator="Match(name)" name="MyAppConnectionString" />
</connectionStrings>
But, I have to ask.. are connection strings stored in your database or something? Why do you need to do them in code?
You could just do this:
public class MyDataContext : DbContext {
public MyDataContext(string connectionString) : base(connectionString)
...
}
Then when you create you contexts, you do this:
using(var context =
new MyDataContext(OurDynamicConfigSettings.GetSetting("MyAppConnectionString"))) {
...
}
public partial class MyContextEntities : DbContext
{
public MyContextEntities(string ConnectionString="[Your Entity Connection String Here]")
: base(ConnectionString)
{
}
protected override void OnModelCreating(DbModelBuilder modelBuilder)
{
throw new UnintentionalCodeFirstException();
}
}
Replace " with ' (single quote) in connection string

Dynamic MySQL database connection for Entity Framework 6

I wish to pass a dynamic connection string to the entity framework context. I have over 150 schemas which are identical (one per account) and I would like to select the connection as such:
ApplicationDbContext db = new ApplicationDbContext("dbName");
In theory this would be fairly easy, as I can create a connectionString and pass it as the argument for the constructor, for example:
public ApplicationDbContext(string dbName) : base(GetConnectionString(dbName))
{
}
public static string GetConnectionString(string dbName)
{
// The connectionString passed is something like:
// Server=localhost;Database={0};Uid=username;Pwd=password
var connString = ConfigurationManager
.ConnectionStrings["MyDatabase"]
.ConnectionString
.ToString();
return String.Format(connString, dbName);
}
I can connect successfully when I just pass the connection string name, but not when I generate it dynamically as below. I realize now that it's because the connection string in web.config has the providerName="MySql.Data.MySqlClient" attribute in it.
When I pass the actual connection string dynamically to the connection though, it assumes that it needs to connect to SQL Server rather than MySQL and fails due to the connection string being invalid.
The question is, how do I pass the provider name to the connection string if I am creating it dynamically?
Entity Framework 6 offers some handy subtle changes which aid in both getting MySQL working and also creating dynamic database connections.
Getting MySQL working with Entity Framework 6
First, at the date of my answering this question, the only .Net connector drivers compatible with EF6 is the MySQL .Net Connectior 6.8.1 (Beta development version) which can be found at the official MySQL website here.
After installing, reference the following files from your Visual Studio solution:
Mysql.Data.dll
Mysql.Data.Entity.EF6.dll
You will also need to copy these files somewhere where they will be accessible to the project during build time, such as the bin directory.
Next, you need to add some items to your Web.config (or App.config if on desktop based) file.
A connection string:
<connectionStrings>
<add name="mysqlCon"
connectionString="Server=localhost;Database=dbName;Uid=username;Pwd=password"
providerName="MySql.Data.MySqlClient" />
</connectionStrings>
Also add the provider, inside the <entityFramework /> and <providers /> nodes, optionally (this is an absolute must in the second part of my answer, when dealing with dynamically defined databases) you may change the <defaultConnectionFactory /> node:
<entityFramework>
<defaultConnectionFactory type="MySql.Data.Entity.MySqlConnectionFactory, MySql.Data.Entity.EF6" />
<providers>
<provider invariantName="MySql.Data.MySqlClient" type="MySql.Data.MySqlClient.MySqlProviderServices, MySql.Data.Entity.EF6" />
</providers>
</entityFramework>
If you change the defaultConnectionFactory from the default sql server connection, don't forget to remove the <parameter> nodes which are nested in the defaultConnectionFactory node. The MysqlConnectionFactory does not take any parameters for its constructor and will fail if the parameters are still there.
At this stage, it's quite easy to connect to MySQL with Entity, you can just refer to the connectionString above by name. Note that if connecting by name, this will work even if the defaultConnectionFactory node still points at SQL Server (which it does by default).
public class ApplicationDbContext: DbContext
{
public ApplicationDbContext() : base("mysqlCon")
{
}
}
The it is just a matter of connecting normally:
ApplicationDbContext db = ApplicationDbContext();
Connecting to a dynamically selected database name
At this point it's easy to connect to a database which we can pass as a parameter, but there's a few things we need to do.
Important Note
If you have not already, you MUST change the defaultConnectionFactory in Web.config if you wish to connect to MySQL
dynamically. Since we will be passing a connection string directly to
the context constructor, it will not know which provider to use and
will turn to its default connection factory unless specified in
web.config. See above on how to do that.
You could pass a connection string manually to the context like this:
public ApplicationDbContext() : base("Server:localhost;...")
{
}
But to make it a little bit easier, we can make a small change to the connection string we made above when setting up mySQL. Just add a placeholder as shown below:
<add name="mysqlCon" connectionString="Server=localhost;Database={0};Uid=username;Pwd=password" providerName="MySql.Data.MySqlClient" />
Now we can build a helper method and change the ApplicationDbContext class as shown below:
public class ApplicationDbContext: DbContext
{
public ApplicationDbContext(string dbName) : base(GetConnectionString(dbName))
{
}
public static string GetConnectionString(string dbName)
{
// Server=localhost;Database={0};Uid=username;Pwd=password
var connString =
ConfigurationManager.ConnectionStrings["mysqlCon"].ConnectionString.ToString();
return String.Format(connString, dbName);
}
}
If you are using database migrations, the following step is important
If you are using migrations, you will find that the ApplicationDbContext will be passed to your Seed method by the framework and it will fail because it will not be passing in the parameter we put in for the database name.
Add the following class to the bottom of your context class (or anywhere really) to solve that problem.
public class MigrationsContextFactory : IDbContextFactory<ApplicationDbContext>
{
public ApplicationDbContext Create()
{
return new ApplicationDbContext("developmentdb");
}
}
Your code-first migrations and seed methods will now target the developmentdb schema in your MySQL database.
Hope this helps someone :)
It's now 2019 of course things have changed a bit but Franciso's example really helped me on this. This is the simplest solution I could find and the only one that actually worked. I did change it a bit from what he has shown. Follow this to completion you should end up with a working solution.
I had to change a few things. I am going to be very explicit in what has to be done and I am going to use my actual file names etc so that you don't have to guess about substitutions. Many examples are also short on how to make it work at the end. This example has everything you need to know.
This was built on visual studio 2015 Entityframework 6 using MySql server 8.0.16.0.
Unfortunately the MySql connectors and libraries are a complete mess. The 8.0.xx.0 connector / net and MySql.Data.Entity.EF6 and MySql.Data are completely useless.
I have installed Connector Net 6.10.7.0, MySql.Data.Entity.EF6 6.10.7.0, and MySql.Data 6.10.7.0. That works for me and I will vigorously oppose changing this.
This is for MySql but I really don't know why it could not work for any db.
Scenario
I have a multi tenant situation where I have a common db and multiple tentant databases, one per customer The customer id is kept in the common db for login purposes and authorizaton and the customer id directs which database to use. The client db's are all called myclientdb_x where x is the client number. myclientdb_1, myclientdb_2, myclientdb_35 and so on.
I need to dynamically switch to whatever clientdb_x the code is currently serving. There is a initial database client called myclient_0 which is the template for all of the other myclient_x databases.
Step1
I created a specific connection string in my Web.config for this it looks like this. It allows connections to the clientdb_0
<add name="DefaultClientConnection" providerName="MySql.Data.MySqlClient"
connectionString="server=localhost;user id=xxx;
password=xxxx; persistsecurityinfo=True;database=clientdb_0" />
Step2
I created a new entity called ClientDbUserUpdater using the wizard. The data entity is called
ClientDbUserUpdater.edmx
I told it to use "DefaultClientConnection" as the DB connection
I told it to save this new connection string in the Web.config
This created new entity connection string in the Web.config file and it will look like
<add name="myclient_0Entities" connectionString="metadata=
res://*/Areas.Authorizations.Models.ClientDbUserUpdater.csdl|
res://*/Areas.Authorizations.Models.ClientDbUserUpdater.ssdl|
res://*/Areas.Authorizations.Models.ClientDbUserUpdater.msl;
provider=MySql.Data.MySqlClient;provider connection string="
server=localhost;user id=xxxx;password=yyyyy;
persistsecurityinfo=True;database=myclient_0"" providerName="System.Data.EntityClient" />
You might have to dig a bit because the wizard is not good about putting in \n in appropriate places.
Notice that this connection string is fundamentally the same as the initial connection string except for its name and the fact that it has
res://*/Areas.Authorizations.Models.ClientDbUserUpdater.csdl|
res://*/Areas.Authorizations.Models.ClientDbUserUpdater.ssdl|
res://*/Areas.Authorizations.Models.ClientDbUserUpdater.msl;
The res: strings are needed by the data entity and its why you can't just send a standard connection string into the data entity.
If you try to send in the initial connection string
<add name="DefaultClientConnection" providerName="MySql.Data.MySqlClient"
connectionString="server=localhost;user id=xxx;
password=xxxx; persistsecurityinfo=True;database=clientdb_0" />
you will get an exception from
protected override void OnModelCreating(DbModelBuilder modelBuilder)
{
throw new UnintentionalCodeFirstException();
}
Step3
This new connection string is the one you need to alter. I have not tested it but I am pretty sure if change the data entity model with the wizard you will need to make this change again.
Take string:
<add name="myclient_0Entities" connectionString="metadata=
res://*/Areas.Authorizations.Models.ClientDbUserUpdater.csdl|
res://*/Areas.Authorizations.Models.ClientDbUserUpdater.ssdl|
res://*/Areas.Authorizations.Models.ClientDbUserUpdater.msl;
provider=MySql.Data.MySqlClient;provider connection string="
server=localhost;user id=xxxx;password=yyyyy;
persistsecurityinfo=True;database=myclient_0"" providerName="System.Data.EntityClient" />
and change it to:
<add name="myclient_0Entities" connectionString="metadata=
res://*/Areas.Authorizations.Models.ClientDbUserUpdater.csdl|
res://*/Areas.Authorizations.Models.ClientDbUserUpdater.ssdl|
res://*/Areas.Authorizations.Models.ClientDbUserUpdater.msl;
provider=MySql.Data.MySqlClient;provider connection string="
server=localhost;user id=xxxx;password=yyyyy;
persistsecurityinfo=True;database={0}"" providerName="System.Data.EntityClient" />
Notice that the only part changed is database=myclient_0 to database={0}
Step 4
The data entity created some code behind ClientDbUserUpdater.edmx. The file is called ClientDbUserUpdater.Context.cs.
The code is ...
namespace what.ever.your.namespace.is
{
using System;
using System.Data.Entity;
using System.Data.Entity.Infrastructure;
public partial class client_0Entities : DbContext
{
public client_0Entities()
: base("name=client_0Entities")
{
}
protected override void OnModelCreating(DbModelBuilder modelBuilder)
{
throw new UnintentionalCodeFirstException();
}
public virtual DbSet<user> users { get; set; }
}
}
Notice that this a partial class. This means you can extend this class and add a new constructor.
Add the following class.
using System;
using System.Configuration ;
using System.Data.Entity ;
namespace what.ever.your.namespace.is
{
public partial class client_0Entities : DbContext
{
public client_0Entities(string dbName) : base(GetConnectionString(dbName))
{
}
public static string GetConnectionString(string dbName)
{
var connString = ConfigurationManager.ConnectionStrings["client_0Entities"].ConnectionString.ToString();
// obviously the next 2 lines could be done as one but creating and
// filling a string is better for debugging. You can see what happened
// by looking a conn
// return String.Format(connString, dbName);
string conn = String.Format(connString, dbName);
return conn ;
}
}
}
The class adds a new constructor which allows you to get the base connection string for the data entity model which from above looks like:
<add name="myclient_0Entities" connectionString="metadata=
res://*/Areas.Authorizations.Models.ClientDbUserUpdater.csdl|
res://*/Areas.Authorizations.Models.ClientDbUserUpdater.ssdl|
res://*/Areas.Authorizations.Models.ClientDbUserUpdater.msl;
provider=MySql.Data.MySqlClient;provider connection string="
server=localhost;user id=xxxx;password=yyyyy;
persistsecurityinfo=True;database={0}"" providerName="System.Data.EntityClient" />
and modfiy it at run time to change the schema.
The String.Format() call in the new partial class swaps out the database schema name in this connection string at run time.
At this point all configuration is done.
Step 5
Now you can make it go. For better understanding of this example it is nice to know what the model looks like for this entity. It is very simple because I was just testing and trying to make it go.
Drilling down through ClientDbUserUpdater.edmx and into into ClientDbUserUpdater.tt you will find your model in modelname.cs . My model is called "user" so my file name is called user.cs
namespace what.ever.your.namespace.is
{
using System;
using System.Collections.Generic;
public partial class user
{
public int UserId { get; set; }
public string Email { get; set; }
public string FirstName { get; set; }
public string LastName { get; set; }
public Nullable<bool> Active { get; set; }
}
}
Now you can generally access your model like this.
client_0Entities _client_0Entities = new client_0Entities("schemaName");
and this code can be anywhere in your solution that can see class client_0Entities.
which in practice is a line similar to any of the 3 below which are connection to databases client_19, client_47 and client_68 respectively.
client_0Entities _client_0Entities = new client_0Entities("client_19");
client_0Entities _client_0Entities = new client_0Entities("client_47");
client_0Entities _client_0Entities = new client_0Entities("client_68");
the following is an actual code example that works on my system. Obviously I am going to not hard code in "client_19" but its better for demo purposes.
here is actual code with real names that works and adds a new row to the user table on database client_19
string _newSchema = "client_19"
using(client_0Entities _client_0Entities = new client_0Entities(_newSchema))
{
user _user = new user();
_user.UserId = 201;
_user.Email = "someone#someplace.com"
_user.FirstName ' "Someone";
_user.LastName = "New";
_user.Active = true;
client_0Entities.users.Add ( _user ) ;
client_0Entities.SaveChangesAsync ( ) ;
}
Hopefully this helps some people. I spent about 20 hrs looking at different solutions which simply did not work or provide enough information to complete them. As I said, finding Franciso's example allowed me to get it working.
Regards,

Entity Framework Code First with non-user instance

these are my first steps with Entity Framework Code First workflow, so maybe I am missing something simple. I created my context class inheriting from DbContext
public class Context : DbContext
{
public DbSet<Book> Books { get; set; }
public DbSet<Author> Authors { get; set; }
}
And I executing it in a controller code:
var author = new Author()
{
FirstName = "Julia",
LastName = "Lerman"
};
using (var context = new Context())
{
context.Authors.Add(author);
context.SaveChanges();
}
I have no other code or configuration that i set for EF.
When I start the application and hit code in controller everything works fine, the only problem is that database is actually not created in SQL Server but it runs, somewhat, locally. That means I cannot really use SQL Server management studio to inspect the data when my application runs, which I want to do.
I figured out that in web.config automatically folowing key was added:
<add name="ApplicationServices" connectionString="data source=.\SQLEXPRESS;Integrated Security=SSPI;AttachDBFilename=|DataDirectory|aspnetdb.mdf;User Instance=false;" providerName="System.Data.SqlClient" />
How can I modify this config key (or is there anything else that I need to do) to make sure that this database will run on actual SQL Server express rather than in-process?
OK, I figured it out, thanks to "Programming Entity Framework: Code First" book, apparently it is the convention over configuration scenario, for the query string to be picked up the name of it needs to match the name of the Context class (the one you inherit from DbContext).
Therefore I need to add new connection string like below:
<add name="Context" connectionString="Data Source=.\SS2012EXPRESS;Initial Catalog=HomeLibrary;Integrated Security=True" providerName="System.Data.SqlClient" />
Original connection string can stay as it is, it is not used run-time. I am not entirely sure what is the purpose of it.

Categories