I have a c# windows form application that connects to databases dynamically where each user may connect to different databases.
The current implementation is as follows:
Connection Repository that contains a dynamically populated list of connections (per user).
When a user initiates a request that requires a database connection the respective connection is looked up from the connection repository ,opened , and then used in the user request .
Code Sample from the connection repository
public class RepoItem
{
public string databasename;
public SqlConnection sqlcnn;
}
public class ConnectionRepository
{
private List<RepoItem> connectionrepositroylist;
public SqlConnection getConnection(String dbname)
{
SqlConnection cnn = (from n in connectionrepositroylist
where n.databasename == dbname
select n.sqlcnn).Single;
cnn.Open();
return cnn;
}
}
sorry for any code errors i just improvised a small version of the implementation for demonstration purpose.
I'am not closing connections after a command execution because it may be used by another command simultaneously.
The questions are:
Should i be worried about closing the connections ?
Does connection close automatically if it is idle for a specific period ?
I have a method in mind to implement a timer in the created Connection Repository and check for idle connections through the Executing ConnectionState Enumeration and close them manually.
Any suggestions are welcome .
When i want a specific connection i call the getConnection function in the ConnectionRepository class and pass the database name as a parameter.
PS: I didn't post the complete implemented code because it is quite big and includes the preferences that affect the populating of the connection list.
I would suggest not to return the SQLConnection to the calling method at all.
Instead, create a method that will accept an Action<SqlConnection>, create the connection inside a using block, and execute the action inside that block
This way you know that the connection will always be correctly closed and disposed, while giving the using code the freedom to do whatever it needs:
public class RepoItem
{
public string databasename;
public SqlConnection sqlcnn;
}
public class DatabaseConnector
{
private List<RepoItem> connectionrepositroylist;
private SqlConnection GetConnection(String dbname)
{
return (from n in connectionrepositroylist
where n.databasename == dbname
select n.sqlcnn).SingleOrDefault();
}
public void Execute(String dbname, Action<SqlConnection> action)
{
using (var cnn = GetConnection(dbname))
{
if (cnn != null) // in case dbname is not in the list...
{
cnn.Open();
action(cnn);
}
}
}
}
Then, to execute an sql statement you can do something like this:
public void ExecuteReaderExample(string dbName, string sql)
{
Execute("dbName",
connection =>
{
using (var cmd = new SqlCommand(sql, connection))
{
using (var reader = cmd.ExecuteReader())
{
while (reader.Read())
{
// do stuff with data form the database
}
}
}
});
}
Of course, you can also wrap the SqlCommand in a method like this.
I've been working with this approach for quite some time now, and as far as I can tell it's working well. In fact, It's working so well I've published a project on git hub based on this approach.
It saves you a lot of the plumbing when dealing with ado.net, by wrapping the connection, command, reader and adapter much the same way.
Feel free to download it and adapt to your needs.
P.S.
To answer your questions directly:
Should i be worried about closing the connections ?
Yes, you should.
Does connection close automatically if it is idle for a specific period ?
No, it doesn't.
However, implementing a method like I suggested will handle closing and disposing the connection object for you, so you don't need to worry about it.
Update
As Yahfoufi wrote in his comment, this design has a flaw, since multiple commands are using the same instance of SqlConnection, you are risking closing the connection while other commands are running.
However, fixing this design flaw is very easy - instead of holding SqlConnection in RepoItem you can simply hold the connection string:
public class RepoItem
{
public string DatabaseName {get; set;}
public string ConnectionString {get; set;}
}
Then you change the GetConnection method like this:
private SqlConnection GetConnection(String dbname)
{
return new SqlConnection(from n in connectionrepositroylist
where n.databasename == dbname
select n.sqlcnn).SingleOrDefault());
}
Now each Execute method is working on it's own individual instance of SqlConnection so you don't need to worry about closing in the middle of some other command executing.
However, While we are on the subject of refactoring, I would suggest removing the RepoItem class all together and instead of using a List<RepoItem> to hold the connection strings simply use a Dictionary<string, string>, where the database name is the key and the connection string is the value. This way you can only have one connection string per database name, and your GetConnection method is simplified to this:
private Dictionary<string, string> connectionrepositroylist;
private string GetConnectionString(String dbname)
{
return connectionrepositroylist.ContainsKey(dbname) ? connectionrepositroylist[dbname] : "";
}
So, the complete DatabaseConnector class will look like this:
public class DatabaseConnector
{
private Dictionary<string, string> connectionrepositroylist;
private string GetConnectionString(String dbname)
{
return connectionrepositroylist.ContainsKey(dbname) ? connectionrepositroylist[dbname] : "";
}
public void Execute(String dbname, Action<SqlConnection> action)
{
var connectionString = GetConnectionString(dbname);
if(!string.IsNullOrEmpty(connectionString))
{
using (var cnn = new SqlConnection(connectionString))
{
cnn.Open();
action(cnn);
}
}
}
// Of course, You will need a way to populate your dictionary -
// I suggest having a couple of methods like this to add, update and remove items.
public bool AddOrUpdateDataBaseName(string dbname, string connectionString)
{
if(connectionrepositroylist.ContainsKey(dbname))
{
connectionrepositroylist[dbname] = connectionString;
}
else
{
connectionrepositroylist.Add(dbname, connectionString);
}
}
}
The good news is that ADO.Net manages your connection pools dynamically, so there's minimal overhead in you dynamically opening and closing connections in code. There's a good document here if you want to look through the detail.
To answer the specific questions you've raised:
Should i be worried about closing the connections ?
Yes, but not for the reasons you may think. Microsoft encourage you to close your connections, so as to return them to the pool for (re)use elsewhere in your code. Closing the connection doesn't actually close it - it merely returns the underlying connection to the pool. Failure to close your connections properly can lead to delays in them being returned to the pool, thus adversely affecting your applications performance as more connections need to be added to the pool to cope with demand.
Does connection close automatically if it is idle for a specific
period ?
A connection is only returned to the pool when it's Dispose or Finalise methods get called. If you create a connection and drop it into a static container then it will not be returned to the pool at all. As such, your ConnectionRepository may actually be harming performance.
I have a method in mind to implement a timer in the created Connection
Repository and check for idle connections
This is unnecessary - close your connections to allow them to return to the pool. This way they will be available for other threads to use
Personally, I'd suggest that you modify your RepoItem class to store connection strings, rather than connection objects, and let ADO.Net's pooling do all the heavy lifting.
public static class ConnectionRepository
{
private static readonly Dictionary<string, string> Connections = new Dictionary<string, string>(StringComparer.CurrentCultureIgnoreCase);
public static bool Contains(string key)
{
return Connections.ContainsKey(key);
}
public static void Add(string key, string connectionString)
{
Connections.Add(key, connectionString);
}
public static SqlConnection Get(string key)
{
var con = new SqlConnection(Connections[key]);
con.Open();
return con;
}
}
With this in place, you can query the database as follows:
public static void foo()
{
using (var con = ConnectionRepository.Get("MyConnection"))
using (var cmd = new SqlCommand("SELECT * FROM MyTable", con))
{
var dr = cmd.ExecuteReader();
//...
}
}
Once the query has executed and the connection is no longer required, the using() block calls its Dispose() method and releases the underlying connection back to the pool for re-use.
As #tinudu says, the SqlConnection class reuses existing connections automatically - you don't have to implement that yourself. See SQL Server Connection Pooling.
If you create the SqlConnection object in a using statement, C# will close the connection automatically as required.
Wrapping the whole thing (create connection, open, run query, close connection) in a function is the best idea. You can put the function in a repository base-class, so it is available to all your repositories.
You would need several functions for the different types of SQL query (select, update, stored proc) but you only need to write one of each - they will get reused.
if you worried about so many conditions let say parallel execution as
well so consider reserving connections for app and close all at once
when app is closing.
public class RepoItem
{
public string databasename;
public SqlConnection sqlcnn;
}
public class ConnectionRepository
{
private List<RepoItem> connectionrepositroylist;
public SqlConnection getConnection(String dbname)
{
SqlConnection cnn = (from n in connectionrepositroylist
where n.databasename == dbname
select n.sqlcnn).Single;
if (cnn!= null && cnn.State == cnn.Closed) // Impelement other checks as well
{
cnn.Open();
}
return cnn;
}
}
Implement CloseConnections and call while application closing i.e
Application.ApplicationExit event
public void CloseConnections()
{
foreach (var connection in connectionrepositroylist)
{
try
{
if (connection.State == System.Data.ConnectionState.Open) // check other conditions
{
connection.Close();
}
}
catch (Exception)
{
//logging or special handling
}
}
}
Points to be note
If some query is still executing and user tries to shut down or close the
app can consider following implementations
Wont allow the application shutdown . Callback delegate will help in this
case to ensure that query is returned.
Force stop and close the connection
It is a better Practice to close the sqlconnection manually, Since it can release the connection which can be used for other processes. Also note that you should open the connection as much as late you can and close it early as possible.
Related
Please give expert opinion, refer to below static sorted list based on key value pair.
Method1 for close connection uses approach of accessing sorted list using key.
Method2 for close connection uses lock statement on the Sorted List and access it by index.
Please guide which approach is better as thousands of users simultaneously creating thousands of connections on web application. Note, accessing by index without locking can raise Index out of bound exception.
internal class ConnA
{
static internal SortedList slCons = new SortedList();
internal static bool CreateCon(string ConnID)
{
string constring = "sqlconnectionstring_containing_DataSource_UserInfo_InitialCatalog";
SqlConnection objSqlCon = new SqlConnection(constring);
objSqlCon.Open();
bool connSuccess = (objSqlCon.State == ConnectionState.Open) ? true : false;
if (connSuccess && slCons.ContainsKey(ConnID) == false)
{
slCons.Add(ConnID, objSqlCon);
}
return connSuccess;
}
//Method1
internal static void CloseConnection(string ConnID)
{
if (slCons.ContainsKey(ConnID))
{
SqlConnection objSqlCon = slCons[ConnID] as SqlConnection;
objSqlCon.Close();
objSqlCon.Dispose();
objSqlCon.ResetStatistics();
slCons.Remove(ConnID);
}
}
//Method2
internal static void CloseConnection(string ConnID)
{
lock (slCons)
{
int nIndex = slCons.IndexOfKey(ConnID);
if (nIndex != -1)
{
SqlConnection objSqlCon = (SqlConnection)slCons.GetByIndex(nIndex);
objSqlCon.Close();
objSqlCon.Dispose();
objSqlCon.ResetStatistics();
slCons.RemoveAt(nIndex);
}
}
}
internal class UserA
{
public string ConnectionID { get { return HttpContext.Current.Session.SessionID; } }
private ConnA objConnA = new objConnA();
public void ConnectDB()
{
objConnA.CreateCon(ConnectionID));
}
public void DisConnectDB()
{
objConnA.CloseConnection(ConnectionID));
}
}
Access to the SortedList isn't thread safe.
In CreateCon, two threads could access this simultaneously:
if (connSuccess && slCons.ContainsKey(ConnID) == false)
Both threads could determine that the key isn't present, and then both threads try to add it, so that one of them fails.
In method 2:
When this is called - slCons.RemoveAt(nIndex); - the lock guarantees that another call to the same method won't remove another connection, which is good. But nothing guarantees that another thread won't call CreateCon and insert a new connection string, changing the indexes so that nIndex now refers to a different item in the collection. You would end up closing, disposing, and deleting the wrong connection string, likely one that another thread was still using.
It looks like you're attempting an orchestration which ensures that a single connection string will be used across multiple operations. But there's no need to introduce that complication. Whatever class or method needs a connection, there's no need for it to collaborate with this collection and these methods. You can just let each of them open a connection when it needs it and dispose the connection when it's done.
That's expensive, but that's why the framework implements connection pooling. From the perspective of your code connections are being created, opened, closed, and disposed.
But behind the scenes, the "closed" connection isn't really closed, at least not right away. It's actually kept open. If, in a brief period, you "open" another connection with the same connection string, you're actually getting the same connection again, which is still open. That's how the number of connections opened and closed is reduced without us having to manually manage it.
That in turn prevents us from having to do what it looks like you're doing. This might be different if we were opening a transaction on a connection, and then we had to ensure that multiple operations were performed on the same connection. But even then it would likely be clearer and simpler to pass around the connection, not an ID.
So the title is a bit vague, but the question is really this: in practice is it best to make a data access class, in this case access to a SQL Server.
Where all the static methods of the class need a connection string and a SQL statement?
Something along these lines:
public static void ExecuteSql(string connStr, string strSqlStatement)
{
SqlConnection conn = new SqlConnection(connStr);
conn.Open();
SqlCommand cmd = new SqlCommand(strSqlStatement, conn);
cmd.ExecuteNonQuery();
conn.Close();
conn.Dispose();
}
And then a method that does something similar except it returns data you queried for.
Or in practice, is it better to create an object for the specific application you are building, and code these items as the objects extensions. So the SQL to run or stored procedure to execute would be wrapped up in that class.
obj.GetSomethingViaQueryThatIsWrittenInTheClassLibrary()
So I guess the first one is more like a service library for accessing and writing data in a SQL Server database. What does one typically do in this situation?
Or could you even use the service library in tandem with the objects library?
There's a ton of, not conflicting, but different opinions on data access and I am really trying to see what is more common a practice.
I disagree with Terry, because:
The connections are pooled, so opening/closing connections is not a problem and you don't want to use a single connection to execute multiple concurrent queries. This is very usefull when having many concurrent threads (for example webservers) but this also applies on a normal application (which might uses Tasks to retrieve data on a separate thread to keep the UI responsive)
So I would create a ConnectionManager class that uses a connection string as constructor parameter, this way the connectionstring wouldn't 'travel' thru your program and is encapsulated in a 'manager' object.
This is a poor example, but I think it work just fine when using ADO.NET
public class ConnectionManager
{
private string _connectionString;
public ConnectionManager(string connectionString)
{
_connectionString = connectionString;
}
public SqlConnection GetConnection()
{
return new SqlConnection(_connectionString);
}
}
Then I would use it something like:
var connectionManager = new ConnectionManager(connectionString);
using(var con = connectionManager.GetConnection())
{
// not all operations require .Open()/.Close()
// multiple queries.
}
using(var con = connectionManager.GetConnection())
{
// not all operations require .Open()/.Close()
// multiple other queries.
}
By using using the connection will be disposed (put back in the pool)
You might even use something like:
public class ConnectionManager
{
private string _connectionString;
public ConnectionManager(string connectionString)
{
_connectionString = connectionString;
}
public void ExecuteNonQuery(string strSqlStatement)
{
using(var connection = new SqlConnection(_connectionString))
using(var command = new SqlCommand(strSqlStatement, connection))
{
connection.Open();
command.ExecuteNonQuery();
}
}
}
Which makes:
var connectionManager = new ConnectionManager(connectionString);
connectionManager.ExecuteNonQuery("SELECT * FROM Whatever");
Tip: You sure need to checkout the SqlParameter to prevent SQL Injection
I wouldn't create, open, and close the connection in an sql execute method like that. Instead, I recommend creating the connection and pass it in to the query method each time a query is needed and when all queries are finished, then close the connection.
I have developed an application which runs query against a DB located in another machine (let's call il SRV-SQL).
In certain cases the application can fail to communicate with the DB machine so my employer wants me to develop a new module which is capable of switching at run-time between the existent connection (to SRV-SQL) to a local DB instance (which is the very same copy of the one in the DB machine).
I have a couple of questions about that:
which is the best way to detect connection problems so that I can fire an "event" to switch from remote connection to the locale one? Would I need to trap SqlException or there is a better way?
Is there any way to switch between the 2 environments without breaking user's actions?
The second point is less trivial as it is acceptable to logout and ask the user to login again (even though i would try to avoid it).
Please let me know if you need more information... also, sorry but I cannot provide any code as this is a design question!
Thanks for taking the time to answer!
It would be good to abstract away the fact that there could possibly be multiple databases from the rest of your application.
public class WidgetRepository
{
private readonly string _primaryConnectionString;
private readonly string _secondaryConnectionString;
public WidgetRepository(string primaryConnectionString, string secondaryConnectionString)
{
_primaryConnectionString = primaryConnectionString;
_secondaryConnectionString = secondaryConnectionString;
}
public void AddWidget(Widget widget)
{
ExecuteAction(AddWidgetAction(widget));
}
public void UpdateWidget(Widget widget)
{
ExecuteAction(UpdateWidgetAction(widget));
}
private Action<string> AddWidgetAction(Widget widget)
{
return Action<string>(connectionString => {
using (var connection = new SqlConnection(connectionString))
{
connection.Open();
using (var command = connection.CreateCommand())
{
command.CommandText = "INSERT INTO Widgets(name, price) VALUES(#name, #price)";
command.Parameters.AddWithValue("#name", widget.Name);
command.Parameters.AddWithValue("#price", widget.Price);
command.ExecuteNonQuery();
}
}
});
}
private Action<string> UpdateWidgetAction(Widget widget)
{
// Logic here to update a widget
}
private void ExecuteAction(Action<string> action)
{
try
{
action(_primaryConnectionString);
}
catch (SqlException)
{
action(_secondaryConnectionString);
}
}
}
Usage:
var widgetRepository = new WidgetRepository("dbconn1", "dbconn2");
widgetRepository.AddWidget(new Widget("Cog", 15.99m));
I have a Linq-to-SQL DataClasses object that I utilize to make database calls. I've wrapped it like so:
public class DataWrapper {
private DataClassesDataContext _connection = null;
private static DataWrapper _instance = null;
private const string PROD_CONN_STR = "Data Source=proddb;Initial Catalog=AppName;User ID=sa;Password=pass; MultipleActiveResultSets=true;Asynchronous Processing=True";
public static DataClassesDataContext Connection {
get {
if (Instance._connection == null)
Instance._connection = new DataClassesDataContext(DEV_CONN_STR);
return Instance._connection;
}
}
private static DataWrapper Instance {
get {
if (_instance == null) {
_instance = new DataWrapper();
}
return _instance;
}
}
}
I have a couple of threads using this wrapper to make stored procedure calls, like this:
DataWrapper.Connection.Remove_Message(completeMessage.ID);
On very rare occasions, my DataClasses object will throw the exception:
ExecuteNonQuery requires an open and available Connection. The
connection's current state is closed.
I'm not managing the connection's state in any way -- I figured Linq-to-SQL should handle this. I could check the connection state of the Connection each time I make a call and open it if it has been closed but that seems like a hack.
I've tried putting MultipleActiveResultSets=true and Asynchronous Processing=True on the connection string to try to handle the possibility of SQL forcibly closing connections, but that hasn't seemed to help.
Any ideas?
You should not cache and re-use a DB connection object ... especially from multiple threads.
You should open a connection, execute your operation(s) and close your connection each time you need to access the DB.
The underlying database access infrastructure (ASP.NET/OLEDB) will manage connection pooling in such a way that reduces most re-connection costs to (effectively) zero.
I'm looking to add transactional support to my DB engine and providing to Abstract Transaction Handling down to passing in Guids with the DB Action Command. The DB engine would run similar to:
private static Database DB;
public static Dictionary<Guid,DBTransaction> Transactions = new ...()
public static void DoDBAction(string cmdstring,List<Parameter> parameters,Guid TransactionGuid)
{
DBCommand cmd = BuildCommand(cmdstring,parameters);
if(Transactions.ContainsKey(TransactionGuid))
cmd.Transaction = Transactions[TransactionGuid];
DB.ExecuteScalar(cmd);
}
public static BuildCommand(string cmd, List<Parameter> parameters)
{
// Create DB command from EntLib Database and assign parameters
}
public static Guid BeginTransaction()
{
// creates new Transaction adding it to "Transactions" and opens a new connection
}
public static Guid Commit(Guid g)
{
// Commits Transaction and removes it from "Transactions" and closes connection
}
public static Guid Rollback(Guid g)
{
// Rolls back Transaction and removes it from "Transactions" and closes connection
}
The Calling system would run similar to:
Guid g
try
{
g = DBEngine.BeginTransaction()
DBEngine.DoDBAction(cmdstring1, parameters,g)
// do some other stuff
DBEngine.DoDBAction(cmdstring2, parameters2,g)
// sit here and wait for a response from other item
DBEngine.DoDBAction(cmdstring3, parameters3,g)
DBEngine.Commit(g)
}
catch(Exception){ DBEngine.Rollback(g);}
Does this interfere with .NET connection pooling (other than a connection be accidently left open)?
Will EntLib keep the connection open until the commit or rollback?
The connection will be kept open until a commit or rollback. It is the transaction that is keeping the connection open.
It will not affect connection pooling, other than a connection held by a transaction will not be returned to the connection pool.
I would recommend that you look at the .net TransactionScope. This may be able to meet your needs, without you writing any of this custom code.