I'm using SQL Azure in a Windows Azure app running as a cloud service. Most of the time my database actions works completely fine (that is, after handling all sorts of timeouts and what not), however i'm running into a problem that seems
using (var connection = new SqlConnection(m_connectionString))
{
m_ConnectionRetryPolicy.ExecuteAction(() => connection.Open());
using (var command = connection.CreateCommand())
{
command.CommandText = "SELECT * FROM X WHERE Y = Z";
var reader = m_CommandRetryPolicy.ExecuteAction(() => command.ExecuteReader());
return LoadData(reader).FirstOrDefault();
}
}
The line that fails is the Command.ExecuteReader with an:
ExecuteReader requires an open and available Connection. The connection's current state is closed
Things that i have already considered
I'm not "reusing" an old connection or saving a connection is a member variable
There should be no concurrency issues - the repository class that these methods belong to is created each time it is needed
Have anyone else experienced this? I could of course just add this to the list of exception which would yield a retry, but I'm not very comfortable with that as
I had a bunch of these errors a few days ago (West Europe) on my production deployment, but they went away by themselves. At the same time I was seeing timeouts, throttling and other errors from SQL Azure. I assume that there was a temporary problem with the platform (or at least the server that I am running on).
You probably aren't doing anything wrong in your code, but are suffering from degraded performance on SQL Azure. Try and handle the errors, perform retries, exponential back-off, queues (to reduce concurrency), splitting your load across databases — that sort of thing.
write every thing within try and catch,finally block.
as follows:
try
{
con.open();
m_ConnectionRetryPolicy.ExecuteAction(() => connection.Open());
using (var command = connection.CreateCommand())
{
command.CommandText = "SELECT * FROM X WHERE Y = Z";
var reader = m_CommandRetryPolicy.ExecuteAction(() => command.ExecuteReader());
return LoadData(reader).FirstOrDefault();
}
con.close();
}
catch(exception ex)
{
}
finally
{
con.close();
}
Remember to close connection in finally block as well.
There is an Enterprise Library that MS has produced specifically for SQL Azure, here are some examples from their patterns and Practice.
It's similar to what you are doing, however it does more on the reliability (and these examples show how to get a reliable connection)
http://msdn.microsoft.com/en-us/library/hh680899(v=pandp.50).aspx
Are you sure it's the reader that's failing and not the opening of the connection? I'm encountering an exception when I wrap the connection.Open() in the m_ConnectionRetryPolicy.ExecuteAction().
However it works just fine for me if I skip the ExecuteAction wrapper and open the connection using connection.OpenWithRetry(m_ConnectionRetryPolicy).
And I'm also using command.ExecuteReaderWithRetry(m_ConnectionRetryPolicy) which is working for me.
I have no idea though why it's not working when wrapped in ExecuteAction though.
I believe this means that Azure has closed the connection behind the scenes, without telling the connection pooler. This is by design. So, the connection pooler gives you what it thinks is an available, open connection, but when you try to use it, it finds out it's not open after all.
This seems very clunky to me, but it's the way Azure is at the moment.
Related
I'm writing a small, simple web app in go using the gorm ORM.
Since the database can fail independently of the web application, I'd like to be able to identify errors that correspond to this case so that I can reconnect to my database without restarting the web application.
Motivating example:
Consider the following code:
var mrs MyRowStruct
db := myDB.Model(MyRowStruct{}).Where("column_name = ?", value).First(&mrs)
return &mrs, db.Error
In the event that db.Error != nil, how can I programmatically determine if the error stems from a database connection problem?
From my reading, I understand that gorm.DB does not represent a connection, so do I even have to worry about reconnecting or re-issuing a call to gorm.Open if a database connection fails?
Are there any common patterns for handling database failures in Go?
Gorm appears to swallow database driver errors and emit only it's own classification of error types (see gorm/errors.go). Connection errors do not currently appear to be reported.
Consider submitting an issue or pull request to expose the database driver error directly.
[Original]
Try inspecting the runtime type of db.Error per the advice in the gorm readme "Error Handling" section.
Assuming it's an error type returned by your database driver you can likely get a specific code that indicates connection errors. For example, if you're using PostgreSQL via the pq library then you might try something like this:
import "github.com/lib/pq"
// ...
if db.Error != nil {
pqerr, ok := err.(*pq.Error)
if ok && pqerr.Code[0:2] == "08" {
// PostgreSQL "Connection Exceptions" are class "08"
// http://www.postgresql.org/docs/9.4/static/errcodes-appendix.html#ERRCODES-TABLE
// Do something for connection errors...
} else {
// Do something else with non-pg error or non-connection error...
}
}
When trying to run the following in Redis using booksleeve.
using (var conn = new RedisConnection(server, port, -1, password))
{
var result = conn.Server.FlushDb(0);
result.Wait();
}
I get an error saying:
This command is not available unless the connection is created with
admin-commands enabled"
I am not sure how do i execute commands as admin? Do I need to create an a/c in db with admin access and login with that?
Updated answer for StackExchange.Redis:
var conn = ConnectionMultiplexer.Connect("localhost,allowAdmin=true");
Note also that the object created here should be created once per application and shared as a global singleton, per Marc:
Because the ConnectionMultiplexer does a lot, it is designed to be
shared and reused between callers. You should not create a
ConnectionMultiplexer per operation. It is fully thread-safe and ready
for this usage.
Basically, the dangerous commands that you don't need in routine operations, but which can cause lots of problems if used inappropriately (i.e. the equivalent of drop database in tsql, since your example is FlushDb) are protected by a "yes, I meant to do that..." flag:
using (var conn = new RedisConnection(server, port, -1, password,
allowAdmin: true)) <==== here
I will improve the error message to make this very clear and explicit.
You can also set this in C# when you're creating your multiplexer - set AllowAdmin = true
private ConnectionMultiplexer GetConnectionMultiplexer()
{
var options = ConfigurationOptions.Parse("localhost:6379");
options.ConnectRetry = 5;
options.AllowAdmin = true;
return ConnectionMultiplexer.Connect(options);
}
For those who like me faced the error:
StackExchange.Redis.RedisCommandException: This operation is not
available unless admin mode is enabled: ROLE
after upgrading StackExchange.Redis to version 2.2.4 with Sentinel connection: it's a known bug, the workaround was either to downgrade the client back or to add allowAdmin=true to the connection string and wait for the fix.
Starting from 2.2.50 public release the issue is fixed.
I have simple sql script:
Select * from student where score > 60
What i am trying to do is run this above script every 1 hour and getting notified on my computer in any way possibe that above condition was met. So basically i dont want to go in there and hit F5 every hour on the above statement and see if i get any result. I am hoping someone out here has something exactly for this, if you do please share the code.
You can use Sql Agent to create a job, Sql server 2008 also has mail functionality
Open SQL Management Studio and connect to your SQL Server
Expand the SQL Server Agent node (if you don't see it, use SQL configuration manager or check services and ensure that SQL Server Agent (SQLINSTANCENAME) is started)
Right click on Jobs and choose 'New Job'
You can run a SQL statement in a job. I'll let you figure out the rest of that part (it's pretty intuitive)
You may want to send your mail using xp_sendmail
Check out the SQL documentation for xp_sendmail
http://msdn.microsoft.com/en-us/library/ms189505(v=sql.105).aspx
You might need to turn the feature on (afaik it's off by default) and you need some server/machine to deliver the mail (so you might need IIS and SMTP installed if on a local machine)
Edit:
Assuming you can't access the server and want to do this on the client side, you can create a .NET framework app or windows service to do the work for you using a schedule or a timer approach:
Schedule approach:
Create a simple command line application which does the query and mails the results, and use the windows scheduler to invoke it every hour (or whatever your interval may be)
Timer approach:
Create a simple application or windows service that will run a timer thread which does the work every x number of minutes
I'd probably just go for the former. The code would be quite simple - new console app:
static void Main(string args[])
{
// No arguments needed so just do the work
using(SqlConnection conn = new SqlConnection("ConnectionString"))
{
using(SqlCommand cmd = new SqlCommand("sql query text", conn))
{
var dr = cmd.ExecuteReader();
List<myClass> results = new List<myClass>();
// Read the rows
while(dr.Read())
{
var someValue = dr.GetString(dr.GetOrdinal("ColumnName"));
// etc
// stuff these values into myClass and add to the list
results.Add(new myClass(someValue));
}
}
}
if(results.Count > 0) // Send mail
{
//Send the message.
SmtpClient client = new SmtpClient(server);
// Add credentials if the SMTP server requires them.
client.Credentials = CredentialCache.DefaultNetworkCredentials;
MailMessage message = new MailMessage(
"recipient#test.com",
"sender#test.com",
"Subject",
"Body");
// Obviously you'd have to read the rows from your list, maybe override ToString() on
// myClass and call that using a StringBuilder to build the email body and append the rows
// This may throw exceptions - maybe some error handling (in any of this code) is advisable
client.Send(message);
}
}
Disclaimer: probably none of this will compile :D
Edit 2: I'd go this way as it's much easier to debug than a windows service as you can just run it from the command line. You can also pass command line arguments so you don't need an application configuration file
I am working on some small project for the local firm and the following code runs fine on my machine, but it produces errors on their server. Currently I don't have access to that server, and this is not a field that I know a lot about, so I have to ask you guys.
The page is written in the classic ASP (javascript for scripting).
The logic goes like this:
conn.Open("myconnection");
bigQuery = "...";
rs = conn.execute(bigQuery);
while (!rs.eof) {
...
smallQuery = "..."
rssmall = conn.execute(smallQuery);
...
rssmall.close();
...
rs.movenext();
}
rs.close();
conn.close();
As I said this runs fine on my machine, but it returns some error (the worst thing is that I don't even know what error) on company's server if bigQuery returns more than ~20 rows. Is there something wrong with my code (this really isn't my field, but I guess it is ok to gather data in the loop like this?), or is there something wrong with their IIS server.
Thanks.
edit:
More info:
It 's Access database. Everything is pretty standard:
conn=Server.CreateObject("ADODB.Connection");
conn.Provider="Microsoft.Jet.OLEDB.4.0";
conn.Open("D:/db/testingDb.mdb");
Queries are bit long, so I wont post them. They are totally ordinary selects so they aren't the issue.
I had a legacy Classic ASP application which I inherited that was very similar (big queries with other queries running within the loop retrieving the first query's results) that ran fine until forty or more records were returned. The way I "solved" the problem was to instantiate another connection object to run the other query. So using your pseudo code, try --
conn.Open("myconnection");
conn2.Open("myconnection")
bigQuery = "...";
rs = conn.execute(bigQuery);
while (!rs.eof) {
...
smallQuery = "..."
rssmall = conn2.execute(smallQuery);
...
rssmall.close();
...
rs.movenext();
}
rs.close();
conn2.close();
conn.close();
What server are they actually running?
Most newer versions of Windows Server don't actually come with the Jet 4.0 driver for 64 bit at all so you can't use an access database with that driver if your app runs as a 64 bit app. You can try running as 32 bit which might solve the problem.
There is an alternative driver packaged as an office component which may be an option.
Try writing a simple test page that literally opens and closes the database connection like so:
<%
Dim conn
Set conn = Server.CreateObject("ADODB.Connection")
conn.Provider = "Microsoft.Jet.OLEDB.4.0"
conn.Open("D:/db/testingDb.mdb")
Response.Write("Database Opened OK")
conn.Close()
%>
Run this on the production server and if you see Database Opened OK then you'll know it's definitely the query rather than the database causing the issue.
If you get an error trying to open the database then you need to changed to using the newer driver or try the app in 32 bit mode
In the case that it is the query causing the issue then it may be that you'll need to use the various additional arguments to the Open() method to try using a different cursor (forward only if you only need to iterate over the results once) which will change how ADODB retrieves the data and hopefully mediate any performance bottleneck related to the query.
Edit
If you want to try debugging the code a bit more add the following at the start of the file. This causes the ASP script processor to continue even if it hits an error
On Error Resume Next
Then at intervals throughout the file where you expect an error might have happened do
If Err <> 0 Then
Response.Write(Err.Number & " - " & Err.Description)
End If
See this article from ASP 101 for the basics of error handling in ASP and VBScript
Background: I am using Nhibernate in an ASP.NET MVC application with an open-session-in-view pattern and I need to use raw ADO.NET to execute some performance-critical database operations.
I'm somewhat confused about how I should be getting my connection instance as I've seen two different methods in numerous blog posts.
Do I want to use:
var connection = Session.Connection;
Or:
var connection = ((ISessionFactoryImplementor)sessionFactory).ConnectionProvider.GetConnection();
I can't seem to find a conclusive answer anywhere and I'm hoping that someone with some extensive NHibernate experience can chime in here.
If you already have a session, use the connection from it.
That will also allow you to share the transaction (if one is open) by enlisting your commands on it.
i'm using something in the lines of (also uses the underlying already-open transaction)
SqlCommand command = new SqlCommand(updateString, (SqlConnection)NHibernateSession.Connection);
command.Parameters.AddRange(parameters.ToArray());
try
{
ITransaction tx = NHibernateSession.Transaction;
tx.Enlist(command);
command.ExecuteNonQuery();
}
catch (SqlException)
{
NHibernateSessionManager.Instance.RollbackTransaction();
throw;
}