I have a ASP.NET Web application. The application connects three different databases. So I have defined three connection string in web.config with different database name and credentials.From the application code I am pointing to the relevant connection string and firing stored procedures. Sometimes the procedures are hitting the wrong database. My guess is that as .NET cache the web.config, somehow the framework is returning the wrong connection string from cache and the application hitting the wrong database. I have checked the application code and found it is pointing to the correct connection database in all cases. Is this happening due to web.config chancing? I cannot identify the root cause of the problem. Please help.
If you are using EF to connect to the database, you have to close the scope of the context and then initialize a new context with the required connection string and then use that context to execute the SP.
I don't think this is an issue with the caching !
If there is only one DAL which connecting to different databases then it is a high chance of application mistake somewhere.
Possible solution, as we have no idea of how is you data access code looks like, is to create 3 different DAL and in each of them realize logic to work only with specified connection string.
For example create 3 different classes inherited from DbContext with different connection strings in constructors.
Related
I have created an Amazon RDS DB instance. I can connect to it and perform operations with SSMS. I can also bind to it using Entity Framework DB-first and generate my model. However, when I run my app, using the same connection string that was generated in the data access project, I get a "the network path was not found" error while trying to establish a connection to the DB.
Let me be clear: the db exists, the right ports are open, and the connection string is correct. I am the only one connecting to the database and the status is "available".
So what's going on? Has anyone experienced something like this?
Let me also further mention that I have already checked the usual things like internet connectivity, firewall rules, state of the database, etc.
well it started working all of a sudden. So I guess this problem will only pop up again in production or something.
Hi I am working on ASP.NET MVC project. Currently I am using Model first approach, where i used to add database manually by using ADO.NET model. Currently I have 4 database and I have 4 connection strings in web.config file.
It was fine till now, since I was working on development environment. But now I need to move my code to live and problem is, in live we have like 40 to 50 databases.
So what I should do is, generate connection string dynamically when user wants to connect to particular database.
I have stored procedure for this which returns connection string and database name.
For example if I have 4 database name like db1, db2, db3 and db4, I need to compare this database name with my stored procedure results database name and if both are equal, then generate connection string equal to that database name.
And also I need to put this in session once i generate connection string, so no need to generate connection string again for particular session.
Can someone help me in this??
EF DbContext as a constructor parameter takes a name of a connection string or connection string itself. So there is no problem in generating any kind of connection string and supply this when creating DdContext.
In our application we have many tenants and have a database per tenant. For every request we lookup what tenant it is and from Settings DB provide a connection string to tenant's own database.
I've not worked with Ado.Net, but from what I see in Google, this is very similar (or based on) to Entity Framework. So down to your particular implementation, there must be a way to provide connection string to database context outwith web.config.
I have a WCF Data Services service that exposes a set of ICD codes. The primary key for the underlying table and the data set that WCF provides access to is a varchar or string in C#.
The service works properly if I have a query like this:
http://somehost/someService.svc/IcdSet('001')
If, however, the ICD code happens to have a . in the identifier as many do, the service fails. Here's an example of one that won't work (IIS gives a 404 - Not Found response):
http://somehost/someService.svc/IcdSet('001.1')
So the question is how can I escape the period or properly pass it to WCF Data Services? It must be interpreting it as a different type of filter condition.
Note: The code for the underlying class seems irrelevant to the question but I can provide it if needed.
Edit: My guess at the moment is that IIS is trying to find a file that ends with .1') which is then producing the 404 error. But how can I tell IIS that it shouldn't be looking for files as these are all data queries?
check this out http://blogs.msdn.com/b/peter_qian/archive/2010/05/25/using-wcf-data-service-with-restricted-characrters-as-keys.aspx
Also might be of interest if you're using .Net 3.5 http://www.microsoft.com/download/en/details.aspx?displaylang=en&id=5121
I have an asp.net mvc application, which uses the default user database. It all works pretty well, but I would like to create some tests for it. I Have a test project, I immediately stumble upon an exception
The specified named connection is either not found in the configuration, not intended to be used with the EntityClient provider, or not valid.
But the connection string is working perfectly (at least in the mvc project's web.config). The exception is thrown by my Entity DB access logic constructor
public ASPNETDBEntities() :
base("name=ASPNETDBEntities", "ASPNETDBEntities")
{
this.OnContextCreated();
}
Any ideas?
Thanks in advance.
The problem is that when running the unit tests, you need to have a connection string set in the App.config file of the test project in order for Entity Framework to find it.
However, if you're doing unit testing you most likely don't want to access the db at all, but rather mock up some dummy objects to test against. (If this is hard to do in your code as it is, you might need some refactoring of your code...)
A third possible scenario is that you're doing integration testing, and thus want to access a real db when testing - however, it doesn't have to be the real db. It can be any db with the same database schema. I'd recommend setting up a dummy db with some dummy records in it, which you can perform tests against (and which you put a connectionstring for in the App.config file in the test project) that will not grow and become slower when the "real" db does.
I'm using grails but I have lot's of stored procedures I'm trying to call from groovy.Sql.newInstance...
In all the examples I've found I never see anyone actually calling close on the connection. But when I tried running a bunch of methods within one response that each uses its own call to newInstance, then it got an error that there were too many connections. That leads me to believe that it isn't pooling the connections. That's a bummer. So do people create one connection and pass it around? Does grails and groovy close the connection at the end of the request?
I don't think that the connection is automatically closed after a request (which wouldn't be cool either), but you can manually close the connection used by the Sql instance:
Sql sql = Sql.newInstance("jdbc://...")
// some queries
sql.close()
See the JavaDoc.
If you would like to use pooled connections (which is surely a good idea), you should be able to create a pooled BasicDataSource (as Spring bean) and use the Sql(DataSource dataSource) constructor of Sql, rather than newInstance().