I am attempting to learn Service Broker from Bob Beauchemin's book "A Developer's Guide to SQL Server" - Chapter 11. I'm finding it to be very good but I'm confused over the concept of closing a conversation. Could someone answer the following questions for me?
When a conversation is ended, can the conversation handle that was created when the conversation was created still be used? (I assume not)
Beauchemin says, on page 511, that when a conversation ends, "Any messages still in the queue from the other end of the conversation are deleted with no warning." Does this mean that if I send a message that expects a reply, but I end the conversation, the message is still sent, it is still received by the other endpoint, the other endpoint processes it, but I'll never receive the reply?
Beauchemin says that if no lifetime is specified, the conversation is active for the number of seconds which can be represented by the maximum size of an integer. Does this mean that if I don't specify a lifetime, a conversation is active for many, many years?
I've read Remus' article on Fire and Forget tactics when it comes to ending conversations, and I have to admit I am guilty of sending a message to a service and immediately ending the conversation. I've set up a stored procedure to end the conversation on the initiator queue, but I'm guessing it's never being fired because I am not ending the conversation in the target activated stored proc, which is a CLR stored proc.
Can you tell me how to end the conversation from a CLR stored proc?
Now that I think of it none of the code I use in my TSQL activated stored procs to handle different message types and error-checking is done in the CLR stored proc.
I'm using service broker and keep getting errors in the log even though everythig is working as expected
SQL Server 2005 Two databases Two end points - 1 in each database Two stored procedures: SP1 is activated when a message enters the sending queue. it insert a new row in a table SP2 is activated when a response is sent from the receiving queue. it cleans up the sending queue.
I have a table with an update trigger In that trigger, if the updted row meets a certain condition a dialogue is created and a message is sent to the sending queue. I know that SP1 and SP2 are behaving properly because i get the expected result. Sp1 is inserteding the expected data in the table SP2 is cleaning up the sending queue.
In the Sql Server log however i'm getting errors on both of the stored procs. error #1 The activated proc <SP 1 Name> running on queue Applications.dbo.ffreceiverQueue output the following: 'The conversation handle is missing. Specify a conversation handle.'
error #2 The activated proc <SP 2 Name> running on queue ADAPT_APP.dbo.ffsenderQueue output the following: 'The conversation handle is missing. Specify a conversation handle.'
I would appreceiate anybody's help into why i'm getting this. have i set up the stored procs in correctly?
i can provide code of the stored procs if that helps.
My service broker was working perfectly fine earlier. As I was testing...I recreated the whole service broker once again.
Now I am able to get the message at the server end from intiator. When trying to send message from my server to the intiator it gives this error in sql profiler.
broker:message undeliverable: This message could not be delivered because the Conversation ID cannot be associated with an active conversation. The message origin is: 'Transport'.
broker:message undeliverable This message could not be delivered because the 'receive sequenced message' action cannot be performed in the 'ERROR' state.
We have implemented our service broker architecture using conversation handle reuse per MS/Remus's recommendations. We have all of the sudden started receiving the conversation handle not found errors in the sql log every hour or so (which makes perfect sense considering the dialog timer is set for 1 hour). My question is...is this expected behavior when you have employed conversation recycling? Should you expect to see these messages pop up every hour, but the logic in the queuing proc says to retry after deleting from your conversation handle table so the messages is enqueued as expected?
Second question...i think i know why we were not receiving these errors before and wanted to confirm this theory as well. In the queuing proc I was not initializing the variable @Counter to 0 so when it came down to the retry logic it could not add 1 to null so was never entering that part of the code...I am guessing with this set up it would actually output the error to the application calling the queueing proc and NOT into the SQL error logs...is this a correct assumption?
I have attached an example of one of the queuing procs below:
Code Block DECLARE @conversationHandle UNIQUEIDENTIFIER, @err int, @counter int, @DialogTimeOut int, @Message nvarchar(max), @SendType int, @ConversationID uniqueidentifier select @Counter = 0 -- THIS PART VERY IMPORTANT LOL :) select @DialogTimeOut = Value from dbo.tConfiguration with (nolock) where keyvalue = 'ConversationEndpoints' and subvalue = 'DeleteAfterSec' WHILE (1=1) BEGIN -- Lookup the current SPIDs handle SELECT @conversationHandle = [handle] FROM tConversationSPID with (nolock) WHERE spid = @@SPID and messagetype = 'TestQueueMsg'; IF @conversationHandle IS NULL BEGIN BEGIN DIALOG CONVERSATION @conversationHandle FROM SERVICE [InitiatorQueue_SER] TO SERVICE 'ReceiveTestQueue_SER' ON CONTRACT [TestQueueMsg_CON] WITH ENCRYPTION = OFF; BEGIN CONVERSATION TIMER ( @conversationHandle ) TIMEOUT = @DialogTimeOut -- insert the conversation in the association table INSERT INTO tConversationSPID ([spid], MessageType,[handle]) VALUES (@@SPID, 'TestQueueMsg', @conversationHandle);
SEND ON CONVERSATION @conversationHandle MESSAGE TYPE [TestQueueMsg] (@Message)
END ELSE IF @conversationHandle IS NOT NULL BEGIN SEND ON CONVERSATION @conversationHandle MESSAGE TYPE [TestQueueMsg] (@Message) END SELECT @err = @@ERROR; -- if succeeded, exit the loop now IF (@err = 0) BREAK; SELECT @counter = @counter + 1; IF @counter > 10 BEGIN -- Refer to http://msdn2.microsoft.com/en-us/library/ms164086.aspx for severity levels EXEC spLogMessageQueue 20002, 8, 'Failed to SEND on a conversation for more than 10 times. Error %i.' BREAK; END -- We tried on the said conversation, but failed -- remove the record from the association table, then -- let the loop try again DELETE FROM tConversationSPID WHERE [spid] = @@SPID; SELECT @conversationHandle = NULL; END;
Hi Guys, Im having trouble stopping a transaction once the result on one or more of my sql statements rolls back the transaction. Take a look at one my first 'IF' statement (below). If the parameter '@Return' is 'False' i want the transaction rolleback and all sql execution to stop. But my problem is the rest of the sql gets executed and i end up getting rubbish data. Im still new to transactions so i think im doing something really wrong. Please take a look at my sql and help me out in what ever way you think is best. thanks in advance. MattBEGIN TRY SET TRANSACTION ISOLATION LEVEL SERIALIZABLE BEGIN TRANSACTION; BEGIN DECLARE @Return BIT; SELECT @Return = (SELECT IsReturnable FROM INVENTORY WHERE InventoryID = 6) IF (@Return = 'False') ROLLBACK TRANSACTION; DECLARE @Already_In INT; SELECT @Already_In = (SELECT ReturnedBy FROM InventoryInUse WHERE (CheckOutNumber = 375)) IF (@Already_In IS NOT NULL) ROLLBACK TRANSACTION; DECLARE @Outstanding As INT; SELECT @Outstanding = (SELECT (QtyTaken - 40) AS Outstanding FROM InventoryInUse WHERE (CheckOutNumber = 375)); DECLARE @OutstandingReason As VARCHAR(200); DECLARE @Get_Person_ID INT; Select @Get_Person_ID = (SELECT PersonID From [User] WHERE Username = 'matthewsk'); IF (@Outstanding < 0) ROLLBACK TRANSACTION; ELSE IF (@Outstanding > 0)BEGIN IF (@OutstandingReason = '') SELECT @RollBack = @RollBack + 1; UPDATE InventoryInUse SET ReturnedBy = @Get_Person_ID, ReturnDate = GETDATE(), QtyReturned = 40, QtyOutstanding = @Outstanding, WriteOffReason = 'Testing This' WHERE (CheckOutNumber = 375); UPDATE Inventory SET TotalQty = TotalQty - @Outstanding, QuantityInUse = QuantityInUse - (40 + @Outstanding) WHERE InventoryID = 6; END;ELSE IF (@Outstanding = 0) BEGIN UPDATE InventoryInUse SET ReturnedBy = @Get_Person_ID, ReturnDate = GETDATE(), QtyReturned = 50 WHERE CheckOutNumber = 375; UPDATE Inventory SET QuantityInUse = QuantityInUse - 50 WHERE InventoryID = 6; END; COMMIT TRANSACTION; END;END TRY BEGIN CATCH IF (XACT_STATE()) = -1 BEGIN ROLLBACK TRANSACTION;
Thanks for your help in my date problem. It worked perfectly. I was away and could not answer back promptly.
yesterday, i was inserting data from a data file into a SQL table. 1 record had a name field as varchar. But it accidently, ended with a &. For example, "New Construction &". when I am inserting, its taking all records except this record that ends with a &. Is there any way to include this one Or should I tell my client that this is an error. If its an error, what is the expalnation(Like string char does not end with a & and & is a ------). Thanks others in the forum.
I am sorry. "New Construction &". Its not a column name. Its just a record. The table has a column called Name and it accidently, got a record that has "New Construction &" for the Name column.
Its not inserting that record with "New Construction &" in the Name field. Other records are fine.
Alright here's my situation. I have two servers both running Eval copies of SQL Server 2005 Enterprise. This is only because I had the promise that I would be getting a shiney new server with a fully licensed copy of the server soon so I wasn't worried when I setup these two programs to use these. Now they are dragging their feet to approve the server and software so I'm scared. First question - What will happen when the eval period runs out? Second question - Can I tell what the date was when I installed the server?
These both will help me push management to approve atleast the purchase of an SQL license if I can't get the server.
I'm using SQL SERVER 2000 and in the process of establishing recoveryguide.Tried to rebuild master using rebuildm.exe but the "Configuring Server"never ending. Then I tried to copy original databases from SQLServerInstallation CD to hard disk as the "Browse for folder", but gave thesame problem. Need advice what need to be done.
I have a process that seems to leave many orphaned sessions over a period of time. The software is by a 3rd party vendor and they can't seem to fix it. It is safe to end the orphaned sessions and I do that on a regular basis. My questions is: What is the best way to do this via T-SQL?
I can select the orphaned sessions by using a simple query on the sys.dm_exec_sessions table. How do I then run a kill command for each session_id from that query?
What i have tried has'nt even been close to a solution. UPDATE dbo.t_Work_Hours SET WeekEnding = DATETIME ( ??? , StartTime) . I have no idea where to go!
I have a job in SQL Server agent with 199 SSIS steps and growing. In the job log only 99 steps are shown, which is a problem if a step higher than 99 fails to run. I need the steps to run in sequence and preferable would like to split the job into several jobs each containing no more than 99 steps. The only thing I don't know is if it is possible to have the next job in sequence start only when the previous job has finished.
Does anyone know if this is possible, and if so how it is done?
Hi, I have a simple web application which calls a stored procedure. The stored procedure operates as a transaction and runs for several minutes. I've created a partial class to set the SQLcommand timeout property to avoid any timeouts, which works fine. Unfortunately though, when the application is run in the production environment, it ends in an error after a certain amount of time (maybe a couple of minutes - not exactly sure), which seems to be the same each run. It doesn't appear to end the stored procedure though, which results in locking the tables. It runs fine in the development environment, and it doesn't appear as though any error information is provided when the application crashes. I'm assuming that the ASP.NET application is timing out for some reason, but the stored procedure itself is fine. I can run it directly from SQL server without any dramas. In the Virtual Directory configuration within IIS, I have the script timeout period set to 1200 seconds. The Default Web-Site timeout property is set to 120 seconds, but I'm assuming that this is only for internet connection timeout, not database transaction timeouts. Any information as to what may be causing this is appreciated. Thanks
Morning Guys, I have a database where I need to kill a remote connection. If I was manually logged into the box i would you a netstat command to stop the service. I would like to end a process from another machine before starting a restore process
I need to write VS2005 C# code using SQL OLE DB to access SQL Server 2005. I have no choice in that matter. I can create a database user with a password like COMPANY", which is a string of uppercase characters ending in an unmatched double-quote mark.
Using Microsoft SQL Server Server Management Studio I login using Windows Authentication, create an account with the password, COMPANY", check the "Enforce password policy", click "OK", and then exit.
Setup
I launch Microsoft SQL Server Server Management Studio again, select "SQL Server Authentication", type in the account name and the COMPANY" password, click the Connect button, and I'm in.
Problem
Now, I need to connect programmatically and run a stored procedure. The password is stored in clear text in hte Registry€”not my choice, it's a legacy application, and changing that is not an option open to me. (We have probably all seen company safes where the combination is scribbled on the wall in case you forget it!)
If I set the connection string in the registry to COMPANY", I get an error like this:
Server Error in '/' Application.
Response is not available in this context. Description: An unhandled exception occurred during the execution of the current web request. Please review the stack trace for more information about the error and where it originated in the code.
Exception Details: System.Web.HttpException: Response is not available in this context.
Source Error:
Line 60: catch (System.Exception e) Line 61: { Line 62: Response.Write(e.Message.ToString()); Line 63: } Line 64: Source File: C:sourceProductNameDataCenterAdm&ReportsWebPagesProductNameWebCommonCodeCommon.cs Line: 62
Stack Trace:
[HttpException (0x80004005): Response is not available in this context.] System.Web.UI.Page.get_Response() +2077605 SiteIQWeb.CommonCode.Common..ctor() in C:ProductNameDataCenterAdm&ReportsWebPagesProductNameWebCommonCodeCommon.cs:62 SiteIQWeb.MasterPage..ctor() in C:ProductNameDataCenterAdm&ReportsWebPagesProductNameWebMasterPage.master.cs:21 ASP.masterpage_master..ctor() in c:WINDOWSMicrosoft.NETFrameworkv2.0.50727Temporary ASP.NET Files oot 255b545c8a400c7App_Web_hfj8popy.0.cs:0 __ASP.FastObjectFactory_app_web_hfj8popy.Create_ASP_masterpage_master() in c:WINDOWSMicrosoft.NETFrameworkv2.0.50727Temporary ASP.NET Files oot 255b545c8a400c7App_Web_hfj8popy.3.cs:0 System.Web.Compilation.BuildResultCompiledType.CreateInstance() +49 System.Web.UI.MasterPage.CreateMaster(TemplateControl owner, HttpContext context, VirtualPath masterPageFile, IDictionary contentTemplateCollection) +250 System.Web.UI.Page.get_Master() +48 System.Web.UI.Page.ApplyMasterPage() +18 System.Web.UI.Page.ProcessRequestMain(Boolean includeStagesBeforeAsyncPoint, Boolean includeStagesAfterAsyncPoint) +685
Version Information: Microsoft .NET Framework Version:2.0.50727.1433; ASP.NET Version:2.0.50727.1433
The closes to success I have been able to come is to set the password to "COMPANYNAME""" or 'COMPANYNAME"', which results in a System.Data.OleDb.OleDbConection Exception with a HResult of 0x80040e4d, and the message Login failed for user 'username'.
Questions
Can the password be formatted differently in the registry, or somehow processed after retrieving it, so that SQL Server 2005 will accept it?
Is this a bug in .NET Framework or SQL OleDb?
Is this simply a case of "Is it hurts, don't do it"?
Thanks in advance. As with all postings, my job, promotion, product success, company future, or some combination thereof is on the line.
I have an very long ntext field, made up of many sentences that I append a full stop to every one, I also strip out any line breaks within the text. However I get this error, when I look it up it comes up with "Failed to locate the ending boundary of a sentence."
I have a system that will post a message to a queue, but does not need to wait for a response - just needs to make sure the message arrived properly in the queue, not that is was processed at the receiving end. A second service will poll the queue to retrieve outstanding messages and will then move the message to an outside system. The movement of the message to the outside system will be wrapped in a transaction and if the process is successful, then the transaction will be commited otherwise it will be rolled back.
1) is it appropriate for the service that posts the message to send an END CONVERSATION ? This way the sending service will not be waiting for a response.
2) in the data movement phase, is it appropriate to issue and END CONVERSATION when commiting and not issue when ROLLBACK occurs. Or should ROLLBACK occur with a following END CONVERSATION with error message?
I want to reuse conversations to minimize overhead during bursts of activity. Remus' article on reusing conversations (http://blogs.msdn.com/remusrusanu/archive/2007/05/02/recycling-conversations.aspx) is great. (I know you are reading this Remus, thanks.)
I was wondering if there is a simpler way of ending a cached conversation - Quiesce the conversation (Stop using it), then after some period of time, end it.
I create a conversation, cache it in RLY_Conversations, and use it for 50 seconds. After 1 minute, the dialog timer servicing proc ends the conversation. There will be no messages sent around the time the End Conversation takes place, thus no race conditions.
Do you see any problems with this method?
Select @DialogHandle = [conversation_handle] From RLY_Conversations Where TableName = @TableName and IsActive = 1 And
CreatedTmstp > dateadd(ss, -50, getdate())
if @DialogHandle is null Begin -- initialize a conversation and record it in our reuse table BEGIN DIALOG CONVERSATION @DialogHandle FROM SERVICE FirstHostRelayService TO SERVICE 'SecondHostRelayService' ON CONTRACT RelayContractSentByAny WITH ENCRYPTION=OFF ;
-- cache the dialog handle to minimize dialog creation overhead. Insert into RLY_Conversations ( TableName, conversation_handle, conversation_id, is_initiator, service_contract_id, conversation_group_id, service_id, lifetime, state, state_desc, IsActive, CreatedBy, CreatedTmstp ) Select @TableName, conversation_handle, conversation_id, is_initiator, service_contract_id, conversation_group_id, service_id, lifetime, state, state_desc, 1, 'Setup', getdate() From sys.conversation_endpoints Where conversation_handle = @DialogHandle;
-- initiate housekeeping process BEGIN CONVERSATION TIMER ( @DialogHandle ) TIMEOUT = 60; End
when you move a conversation to a conversation group, that conversation_group has to have been created previously, ie, you cant specify a non-existing conversation_group, right?
I ask because I am trying to develop an application where I use optimally one conversation related to many given conversation_groups, so that when I receive, I lock only a small determined subset of messages. What I could have used was a way to send messages on a conversation, specifying a conversation_group_id.
Message ordering is of utmost importance in our application.
As i found in testing the only way to ensure message ordering is if they are in the same conversation.If you send multiple messages in different conversations there is no garantee which will be processed first.
Therefore i will be creating conversations that last "forever", that is using a single conversation.
I plan on doing a BEGIN DIALOG CONVERSATION when an inititator site is setup and writing the conversation handle guid to a table.
I will them simply SEND ON SONVERSATION using the guid, i will never issue a end conversation from target or initiator.
Is this theory solid, ie: is there a better way or best practice to do this?
I know that conversatons persist with sql server restarts, however what happens if an initiator site db is restored ?
I was thinking of adding logic to first check if a conversation endpoint exists with the specified guid if not , then start another conversation. But is this the best way?
I am thinking of updating my SQL monitoring application to use Service Broker.
Right now I loop through my list of servers performing various checks on each server. Things like 'check last database backup', 'check for new databases', 'check for server restart'. I loop through, one server at a time, doing one check at a time. The more servers I have the longer it is taking.
So, I want to multi-thread the servers, but single-thread the checks on each individual server. This way I can check say, 5 servers at a time, but on each server I will only do one check at a time. This way I won't flood an individual server with multiple checks.
Is this possible? It looks like Conversation groups might be the way to go but I'm not sure.
I'm having some troubles with conversation groups. I need to send two messages on the same conversation group so I have the following in my SP....
BEGIN DIALOG CONVERSATION @providerConversationHandle FROM SERVICE [ProviderDataService] TO SERVICE 'CalculatedDataService' ON CONTRACT [ProviderDataContract] WITH ENCRYPTION = OFF , LIFETIME = 600;
BEGIN DIALOG CONVERSATION @curveConversationHandle FROM SERVICE [ProviderDataService] TO SERVICE 'CalculatedDataService' ON CONTRACT [ProviderDataContract] WITH RELATED_CONVERSATION = @providerConversationHandle , ENCRYPTION = OFF , LIFETIME = 600;
SEND ON CONVERSATION @providerConversationHandle MESSAGE TYPE [ProviderDataMessage] ( @providerMessage );
SEND ON CONVERSATION @curveConversationHandle MESSAGE TYPE [ProviderCurveMessage] ( @curveMessage );
When I query the queue I see two messages, but they don't have the same conversation_group_id.
I have not been successfull in getting conversation group to work. My understanding is that I can specify a 'guid' for a conversation group id in the create dialog and when I send a message on this conversation it will have that specific guid for its conversation group. When I do this it does not appear this way in the "target" queue.
I am looking for an example to help me understand how to use a conversation group. The MSDN has not really provided that run able example that I can run and verify and tweak.
The idea that I would like to try is that the initiator must send 5 different XML messages to a target. These 5 messages are all related and must exist together. What I assume is that if I want the target to get all 5 messages together out of the queue all messages must be sent in their own conversation but all linked with the same conversation group Id. I have not been able to get this to work.
The communication is really a one way where the initiator sends the data to the target and does not process or need a message sent back from the target.
I have a conversation that I want to know has not ended so I am using LIFETIME.
When the conversation times out I then have three records in the queue.
1. The original conversation record that has not been received.
2. Error message to the initiator.
3. Error message to the target.
Both message bodies on the error records say that it was a lifetime error.
If I end the conversation on the initiator side after it is sent, I still get the target error record but the message_body field is null.
So say I don't end the conversation on the initator side. My next receive on the target side will pull the original record. Then it will pull the initator record and then it will pull the target record. Nothing on that record says that it had timed out.
What is the best practice for handling lifetime errors?
END CONVERSATION DIALOG CONVERT(uniqueidentifier, '58C1A7AA-C0D7-DB11-B4C6-005056C00008')
using ICommand.Execute method, I got an error:
Incorrect syntax near the keyword 'END'.
I use SQL Native Client as OLE DB Provider.
In profiler I can see that my statement becomes like this:
exec END CONVERSATION CONVERT(uniqueidentifier, '903586ED-C0D7-DB11-B4C6-005056C00008')
So provider add 'exec' before my statement, maybe because NCLI don't know such statetment and thinks that it is a stored procedure call with the name 'END'?
Is there is a way to avoid it except of using following construction:
I am new to service broker and would like a little help please. I have a SP which gathers information from a collection of tables. Depending on the data gathered it may or may not begin a dialog conversation with a service broker queue. What i'm needing to know is should at the end of the SP once the required message has been sent should i end the conversation or not?
