Are you the publisher? Claim or contact us about this channel

Embed this content in your HTML


Report adult content:

click to rate:

Account: (login)

More Channels


Channel Catalog

Channel Description:

Discussions on SQL Server Replication

older | 1 | .... | 74 | 75 | (Page 76) | 77 | 78 | .... | 181 | newer

    0 0
  • 11/10/08--01:04: data validation alerts
  • Hello guys,
    I have a problem with triggering a "Replication: Subscriber has passed data validation" alert. The replication configuration is as follows:

    Bidirectional Transactional Replication
    SQL Server Mail is well configured and an Operator is present and is confirmed to send emails (this operator is used in the Response tab of the alert)

    The error number of
    alert's properties is 20575 (Alerts will be raised on: error number 20575). After I mark a publication to validate its subscription I see in the Distribution agent details that the subscription passes validation and the table "sysreplicationalerts" contains the rows (for each published table) with alert_error_code=20575. But I don't see any corresponding event in the event viewer -> application. I think this is the reason of why the alert is not triggered. As far as I understood, for SQL Agent to be able to trigger an allert there should be an event inscription with the corresponding event id.

    Could anybody give me a tip why there are no event insciptions and in which situations are sql server events generated? Or may be there is another problem?

    Best regards

    0 0

    I have been replicating happily between 2008 R2 SQL Servers for some time using Transactional replication.

    Replication stopped yesterday with the following messages on the source server:

    MSSQL_ENG10054Communication link failure
    MSSQL_ENG10054TCP Provider: An existing connection was forcibly closed by the remote host.
    MSSQL_ENGFailed to send batch after max errors
    MSSQL_ENGUnspecified error
    MSSQL_ENGAn existing connection was forcibly closed by the remote host.
    MSSQL_REPL0 The process could not bulk copy into table '"pro"."ZZM012X"'.

    In the Event Viewer on the target server this message appears:

    "A fatal error occurred while reading the input stream from the network. The session will be terminated (input error: 10054, output error: 0)."

    So this happened after it had been replicating for some time.  I get the same error now when I try to restart it or if I rebuild the publication/subscription completely.

    There are 7 tables in the publication - 6 of them bulk copy and then replicate without issue.  The error is specific to one table.  The table has a column in that is varchar(132) and it contains recipe instructions to our manufacturing plant.  If I replicate the table and exclude this field, it doesn't fail.  I initially thought it might be spurious characters in the text column but I have removed these and still get the same error.  I have set up a test version of the table and publication/subscription to enable me to test outside of the live environment.  Same errors occur.

    Does anyone have any clues or pointers to what might be causing it?

    Many thanks


    0 0
  • 01/05/14--22:41: Replication Error Message

  • k

    0 0
  • 12/19/13--07:51: Merge Replication -Analysis
  • Hi

         We thought of moving our project to use replication.

    What are the steps needs to be addressed before implementing merge replication.

    Please help.

                        1.Two web servers

                        2.Two Application database servers  (replication will be implement in this server)

                        3.Four Content Read only servers.

               The connection will be start from content to application server.


    Muthukumar S


    0 0

    I am attempting to set up a very simple snapshot replication and I am encountering this error in the SQL logs: Login failed for user 'Domain\ReplicationUser'. Reason: Attempting to use an NT account name with SQL Server Authentication. [CLIENT: <local machine>].

    The interesting thing is that this is a mixed security box running SQL 2008 and I set up the exact same scenario on a SQL 2012 instance on a test server using the same domain account and it worked flawlessly.

    The account is valid (and unlocked).  If I enter the wrong passwords into the replication property dialogs then the processes abort immediately.  With the correct password, they run and put the below information into Job History.   Internet searches indicated this could be related to SPN not being registered but my SQL log shows it registered correctly.

    Examples here have been edited to remove actual user and server names.  Any help is appreciated.

    Date  1/6/2014 1:56:03 PM
    Log  Job History (XXXX03-blahblahblah)

    Step ID  2
    Server  XXX03
    Job Name  XXX03-blahblahblah
    Step Name  Run agent.
    Duration  00:00:07
    Sql Severity 0
    Sql Message ID 0
    Operator Emailed 
    Operator Net sent 
    Operator Paged 
    Retries Attempted 0

    2014-01-06 19:56:08.48 Microsoft (R) SQL Server Snapshot Agent
    2014-01-06 19:56:08.48 [Assembly Version =, File Version = 10.0.1600.22 ((SQL_PreRelease).080709-1414 )]
    2014-01-06 19:56:08.48 Copyright (c) 2008 Microsoft Corporation.
    2014-01-06 19:56:08.48 The timestamps prepended to the output lines are expressed in terms of UTC time.
    2014-01-06 19:56:08.48 User-specified agent parameter values:
    2014-01-06 19:56:08.48 --------------------------------------
    2014-01-06 19:56:08.48 -Publisher XXX03
    2014-01-06 19:56:08.48 -PublisherDB A
    2014-01-06 19:56:08.48 -Publication A
    2014-01-06 19:56:08.48 -Distributor XXX03
    2014-01-06 19:56:08.48 -DistributorSecurityMode 1
    2014-01-06 19:56:08.48 -XJOBID 0x0E5F2FE6FFBBC0478EC2D3BEFDB20B04
    2014-01-06 19:56:08.48 --------------------------------------
    2014-01-06 19:56:08.48 Connecting to Distributor 'XXX03'
    2014-01-06 19:56:08.72 Parameter values obtained from agent profile:
    2014-01-06 19:56:08.72 ---------------------------------------------
    2014-01-06 19:56:08.72 -BcpBatchSize 100000
    2014-01-06 19:56:08.72 -HistoryVerboseLevel 2
    2014-01-06 19:56:08.72 -LoginTimeout 15
    2014-01-06 19:56:08.72 -QueryTimeout 1800
    2014-01-06 19:56:08.72 ---------------------------------------------
    2014-01-06 19:56:08.73 Connecting to Publisher 'XXX03'
    2014-01-06 19:56:08.76 The replication agent had encountered an exception.
    2014-01-06 19:56:08.76 Source: Replication
    2014-01-06 19:56:08.76 Exception Type: Microsoft.SqlServer.Replication.ConnectionFailureException
    2014-01-06 19:56:08.76 Exception Message: Login failed for user 'Domain\ReplicationUser'.
    2014-01-06 19:56:08.76 Message Code: 18456
    2014-01-06 19:56:08.76

    0 0

    One SQL server is publisher in site A.

    the other sql server is subscriber in site B.

    Site A connect to Site B through network.

    Replication failed sometime. Any idea why it fails from the following log? Thanks

    I find following log from verbose log file:

    2014-01-06 06:35:03.20 [0%] The replication agent had encountered an exception.

    2014-01-06 06:35:03.20 Source: Replication

    2014-01-06 06:35:03.20 Exception Type: Microsoft.SqlServer.Replication.SqlCommandTimeoutException

    2014-01-06 06:35:03.20 Exception Message: Timeout expired.  The timeout period elapsed prior to completion of the operation or the server is not responding.

    2014-01-06 06:35:03.20 Message Code: -2

    2014-01-06 06:35:03.20 

    2014-01-06 06:35:03.20 Call Stack:

    2014-01-06 06:35:03.20 Microsoft.SqlServer.Replication.SqlCommandTimeoutException: Timeout expired. The timeout period elapsed prior to completion of the operation or the server is not responding. ---> System.Data.SqlClient.SqlException: Timeout expired.  The timeout period elapsed prior to completion of the operation or the server is not responding.

    2014-01-06 06:35:03.20    at System.Data.SqlClient.SqlConnection.OnError(SqlException exception, Boolean breakConnection)

    0 0

    Hi , everyone I have odd issue with the SQL replication wizard when creating subscribers. When I first try to create an subscription I get the message below.

    Any Idea what it means and how to cure it ?


    Then when I go into the wizard when creating a subscription I can see my subscriber server but I can not see any of the databases on it , as per image


    Any advice gratefully recieved as I am struggling to understand and resolve this issue.

    Many thanks

    0 0

    I have added a new subscriber to a transactional replication publication. The distribution agent applied the snapshot (showed 'delivered snapshot' in status) but then failed with an access error (The distribution agent failed to create temporarily files....System returned errorcode 5). I added the agent account to the com directory referenced in the error and reran the distribution job. This started the snapshot over from the beginning (this was a 9 hour process to apply the snapshot). I received the same failure message and am researching permissions. 

    Once I fix the permissions, is there any way to restart the agent without having to reapply the entire snapshot? If not, is there any way to test the step that was failing without waiting for the 9 hour process to complete again?



    0 0

    Hi all

    I am attempting to replicate a database based on 2008 R2 to an SQL 2012 server.  I have successfully made a transactional distribution on the source server with the snapshot created.  The issue is when I create a subscription in the destination server, it seems to create ok but upon monitoring the subscription, an error is displayed with the message 'Assembly 'smdiagnostics' could not be installed because existing policy would keep it from being used.'  Any idea on what this means and how I can overcome this?

    Thanks in advance.

    0 0

    Hi All,

    We have to build transnational replication with updatable  subscribers between two different domains .

    Is it possible ,if it's possible please share the any document or process .....

    Thanks in advance. 


    0 0


    I want to copy / replicate some data from one database to other database.

    My main server should have data of all branches .

    My each branch has SQL database with schema same as main server database.

    1. I want to copy some master data from main server to branch server again and again after some interval . ( I can give conditions for selection of rows )

    2. After above updation process will check from addition of transaction data in branch & update the same  in main server.

    please note : My branches & Main server is connected to each other

    I am using sql server 2008 on my main server & branches

    How I can do this.

    Any suggestions related to this working are welcome.

    ANAND ERP (INDIA) PVT.LTD. email : Site :

    0 0
  • 01/07/14--20:00: Replication error
  • Replication was setup a while back and was working fine. Replication was setup from ServerA(Publisher) to ServerC(Subscriber\distributor). There is a mirror setup between ServerA(Principal) and ServerB(Mirror), we tested the mirror few days back and we failed over from ServerA to ServerB, and then from ServerB back to ServerA. In that process replication failed.
    Now when I am trying to reinitialize my subscription, the snapshot agent job is failing with the below error:
    Exception Message: Named Pipes Provider: Could not open a connection to SQL Server [5]

    I checked the publisher for the TCP\IP which was already enabled, I can ping ServerA(publisher) from ServerC(Subscriber\distributor) etc. Actually replication was working fine, before the mirror failover and just by testing the failover I am getting the above error. Any ideas? Thanks.


    0 0

    Attempting to start a replication relation, however,
    when initializing, it gives me the following error:

    09:23:02.225 The schema script 'FillMercaTempTab_56.sch' could not be propagated
    to the subscriber.
    2014-01-08 09:23:02.234 Category:NULL
    Source: Merge
    Replication Provider
    Number: -2147201001
    Message: The schema script
    'FillMercaTempTab_56.sch' could not be propagated to the
    2014-01-08 09:23:02.236 Category:NULL
    Source: Microsoft SQL
    Server Native Client 10.0
    Number: 213
    Message: Column name or number of
    supplied values does not match table definition.

    How do I figure out
    which statement is the culprit?

    Please visit

    0 0


    We had many conflicts and I did multiple article changes. I wanted to reinitialize all the subscriberswithout uploading data from subscribers first. I created a new snapshot and every time I reinitialize a subscriber I leave the "upload changes" textbox unchecked. But as soon as the process start I see subscribers uploading data on the server (there are a few bidirectional articles in the database but most are download only).

    Anyone has a clue on how this can happen? Unless I missed something I couldn't find anything in the BOL and when I try to search here and on the web I get result from people having the opposite effects (uploads doesn't work).



    0 0

    Looks like my replication set up is all messed up. I like to cleanup all the existing setup and startover. How do I do that?

    First I am not able to delete the subscriptions,
    Getting the error "Invalid Object Name 'dbo.MSmerge_agents'

    0 0

    I am replicating a database across domains (call them A and B) through a VPN.  The Publisher and Distributor are 2012 SP1 in domain A and the subscriber in domain B is 2005 SP3.  Since the VPN connection is extremely slow I am using the FTP option for the publication.  I now want to use that same publication to replicate to a another subscriber within domain A.  What I want to know is if I change the location of the snapshot folder and don't use the FTP option, because this will be faster, will it reinitialize the subscription for the subscriber in domain B?

    0 0


    I've just started learning SQL Server database synchronization and my goal is to make an application that syncs its database to cloud (Azure). This is probably the most basic question but I can't seem to find any simple answers while googling. Is there a way to synchronize new columns that are added to the client (my app) database so that the change is also made to the "cloud" database? And naturally vice versa.

    I have learned synchronization using this article:

    I've also read similar articles from the web. Haven't found any simple articles that would answer my question, though... :)

    I have created two local databases and I have managed to sync new/deleted rows and column value changes but is there a way to sync "structural changes" like new or deleted columns? This is probably something like it would cause data loss etc if clients could alter the database any way the want but, hey, I'm new with this stuff so bear with me. :)

    And a follow-up question: ... and what then if it's not possible? How to add new columns to the "master" database if you need new information stored?

    Any ideas and tips are very welcome and especially links to tutorials etc.

    Best regards, 


    0 0

    We are planning to upgrade(not in place, moving to a new machine) the publisher(2005 EE) to 2008R2 EE but the subscriber(2005 EE) will be the same.

    Replication Type: Transactional replication
    Number of replicated databases: 1
    SQL Server 2005 SP4 (Publisher & Distributer) with 8 publications
    SQL Server 2005 SP4 (Subscriber)
    SQL Server 2005 SP4 (Publisher & Distributer) this will be upgrading to SQL Server 2008R2 SP2
    SQL Server 2005 SP4 (Subscriber) this will be same.

    After moving the database to new SQL Server 2008R2, how to re-configure the replication with the existing 2005 subscriber?

    There are hundreds of articles (objects) involving in the replication so it is not easy to reconfigure in GUI mode. So looking for plan to script the existing replication and plan to run those scripts on new server to re-configure is this possible?

    0 0

    Hi All,

    I am facing the different issue in my environment .

    we have grant the read and write access to one group on database, after that we have added the one user to that group.

    now user was unable to connect to database and his get below error....

    Msg 916, Level 14, State 1, Line 1
    The server principal "xxxx\xxxx" is not able to access the database "testDB1" under the current security context.

    Please help me on this issue asap.....

    Thanks in advance .....


    0 0


    I have browsed through a few threads about this error and the suggested answer(s) do help addressing the issue.  However, I like someone to help me understand why this issue arised in our situation.  Let me layout the situation out.

    We are using sql 2012 enterprise edition and moved our databases from one datacenter to a different one six months ago using backup and restore method.  The dbs are involving in SQL 2012 Always-On  A push transaction replication was set up after the move and replication has been working ever since.  The logreader security is using SQL job agent service account to run on and for the connection to publisher we use 'By impersonating the process account'.   

    Out of the blue a couple days ago, we received replication error messages as follows:

    "'The process could not execute 'sp_replcmds' on server xxx"

    "Cannot execute as the database principal because the principal "dbo" does not exist, this type of principal cannot be impersonated, or you do not have permission. (Source: MSSQLServer, Error number: 15517)".

    For the above error, the 'exec sp_changedbowner'  command on the db addresses the issue.   However, the puzzling thing is that if the db owner is the issue, it has been there since we moved six months ago, why this happened now?   The log reader is using SQL agent service account to run the process.   This account is a domain account which has 'sa' permission.   When I saw the error "'The process could not execute 'sp_replcmds' on server xxx", my iniital thought was that for some reason Logreader was having difficulty using sQL agent service account to authenticate or having permission issue.     This prompts me to the following questions:

    1.  If indeed SQL agent service account is having persmission issue, what would cause it since it has 'sa' permission already?  Behind the scene, would it need to authenticate with the Active Directory before it proceeds with its process? If so, when would this happen?  If there is some networking or domain authentication issue, would it cause this 'sp_replcmds' error?

    2. Say logreader is having permission issue to execute 'sp_replcmds', would it try other method(s) such as impersonating in other context( sicj as other domain login) to try to execute the command?  I am trying to understand what route(or process) logreader is trying bheind the scene to cause this error 'the principal "dbo" does not exist' to show up.  Again, this so-call db owner issue has existed for six months.  If it is the issue, why doesn't it happen before?

    It seems when things are working, logreader is working as normal but if something interrupted logreader process, logreader found some hidden problem from its behind the scene process.  If someone can shred some light on this, it would be very much appreciated.


    Ocean Deep

older | 1 | .... | 74 | 75 | (Page 76) | 77 | 78 | .... | 181 | newer