Quantcast
Channel: SQL Server Replication forum
Viewing all 4054 articles
Browse latest View live

Merge Replication -> SQL CE - Subscriber coordination and limit?

$
0
0

Hi

I have a client / server architecture. Write / Read ratio about 20/80

Client has to do lots of analytic stuff - so i want to provide a read cache. My question

1.) Can a SQL Server instance handle 200 subscribers? (10 tables, every table about 20 writes / sec, data size: normal)
2.) If i limit the subscriber count in the SQL Agent - do the clients coordinate themselves who is next?
3.) Is it possible to merge from multiple SQL Server (different disjunct data) into one single SQL Server compact table?

Thanks a lot
Michael


Could not delete publication?

$
0
0

I got the following error message when deleting a publication. The user 'Domian\deletedUser' is deleted from AD by IT. Although the deleted user originally created the database DB1. The publication is created by me and the SQL Agent is running using another active domain user.

Is there any way to delete the publication?

Could not obtain information about Windows NT group/user 'Domian\deletedUser', error code 0x534.
Changed database context to 'DB1'. (Microsoft SQL Server, Error: 15404)

Error message:

Msg 15404, Level 16, State 11, Proceduresp_replcmds, Line 1
Could not obtain information about Windows NT group/user 'Domain\deletedUser', error code 0x534.

Different server db name on the same Database

$
0
0

Hello i  have problem with configure distirbution (2008 R2)

SQL Server replication requires the actual server name to make a connection to the server

Am i missing something ?

i did two thing

1. sp_helpserver got the answer ip-0AC46E75

2. select @@SERVERNAME got the answer IP-0A4CDA2B

Two different thing 

I tried lot of thinkg

like
sp_dropserver  'IP-0A4CDA2A'
go
sp_addserver 'ip-0AC46E72', LOCAL
go

Merge Replication 3 Instances

$
0
0
Is it possible to setup Merge replication between 3 instances?

John M. Couch

RMO and Parallel

$
0
0

Hi,

I'm trying to synchronize 2 publication in parallel using 2 background workers but has soon as one reaches the mergeSubscription.SynchronizationAgent.Synchronize() line, i stop receiving feedback from the other until the first finishes.

Only if i start 2 instance of the application i'm able to synchronize concurrently.

Help anyone ?

SQL server replication error Cannot find the dbo or user defined function........

$
0
0

Hi there,

I ran into a problem while deploying sql 2008 transactional replication between two servers lets assume server A and B.

Actually I created  both distributor and publisher on server A and after that i created a subscriber which is pointed to server B, then i launch the replication monitor and found that their is a big red cross. When i click on the tab "distributor to subscriber history" their is an error says "cannot find either column "dbo" or the user defined-function or aggregate "dbo.ufnLeadingZeros", or the name is ambiguous". Can some body help regarding this issue.


Regards, Ishuv

Oracle to sql replication -log reader agent no.of commands in batch process keeps on increasing for hours

$
0
0
Oracle to sql server replication set up sql 2012. suddenly it started lagging.log reader agent said. batch process in progress.xx no.of commands..0 trans.xxx goes on increasing for hours ..but no updates yet.

Merge Replication Stored Procedure Custom Resolver

$
0
0
Hi All,

Running SQL Server 2008 R2 in a merge replication topology.  Publisher and distributor are on the same machine.  All subscribers running SQL Server 2008 R2 Express.

I have a requirement to implement a stored procedure custom resolver, however searching BOL and Online has not lead me to any examples of how one would look.  DDL and Sample Data are below.

CREATE TABLE [dbo].[Records](
    [RecordID] [varchar](50) NOT NULL,
    [ValueA] [varchar](50) NULL,
    [ValueB] [varchar](50) NULL,
    [StatusCode] [varchar](50) NULL)
GO

-- Publisher Records Table Data
INSERT INTO [dbo].[Records]
        ([RecordID], [ValueA], [ValueB], [StatusCode])
VALUES
    ('1', 'Tested', 'Fixed', 'S05')

-- Subscriber Records Table Data
INSERT INTO [dbo].[Records]
    ([RecordID], [ValueA], [ValueB], [StatusCode])
VALUES
    ('1', 'Tested', 'Not Good', 'S01')

What I'm looking for is a stored procedure that would resolve the conflict based on the StatusCode.  As an example, the Publisher would win the row conflict because it has a StatusCode of 'S05' or greater, so ValueA would remain as 'Tested' and ValueB would remain as 'Fixed'.  If this is not the case the conflict should resolve itself according to the default resolver.  In addition, if this resolver is placed on an article which has join filters attached to it, will the conflict resolution filter down to these sub-articles?

Any ideas/examples would be greatly appreciated!

Thanks.

All subscribers getting renitialized when I generate a new snapshot and try to reinitialize only 1 subscriber

$
0
0

If I want to only reinitialize ONE subscriber db at a time....either due to errors or a publication article change,  etc. )and I belive it also happens when I add a subscriber) ALL the subscribers are getting reinitialzied.  I only want a single subscriber to get reinitialized.  Why is it reinitializing all subscribers.

Here is what I do:

Open replication monitor

choose publisher and then publication

right click on the subscriber I want to reinitialize and choose reinitialize subscriber

then I choose Use a a new snapshot and check Generate a new snapshot.

Choose mark for reinitialization

Here is my configuration

SQL server 2008 r2 transactional replication push to subscribers on sql server 2008 r2. The distributor runs on the publisher. 

replication problem from sql2000 to sql2008

$
0
0
Hi,

We are in planning to migrate from 2000 standard to 2008 standard.
As part of migration, am working on replication but i believe that sql 2000 replication scripts are not working in sql 2008.

So, what i thought of is to setup the replication in 2008 using GUI rather using sql 2000 scripts.

During this process, i have question on Delivery format of the sql statements
In 2000, i can find options line

- Replace INSERT commands with the stored procedure call
- Replace UPDATE commands with the stored procedure call
- Replace DELETE commands with the stored procedure call
- Send parameters in binary format
- USe column names in commands that are not replaced by the stored procedure.


In 2008,
Below are the list of properties i can see it for INSERT,UPDATE & DELETE operations
Properties for the Articles

Statement delivery
INSERT delivery format | Call <stored procedure>
Do not replicate INSERT Statement
INSERT Statement
INSERT statement without column list


UPDATE delivery format | Call <stored procedure>
Do not replicate UPDATE statements
UPDATE Statement
<MCALL storeprocedure>
XCALL stored procedure
SCALL stored procedure

DELETE delivery format | Call <stored procedure>
Do not replicate the DELETE statements
DELETE Statement
XCALL <stored procedure>


But could not find any option to send the parameters in BINARY format where can i find it is my question ??????????????

Thanks,

Transactional replication slow (distributor to subscriber)

$
0
0

Hi,

We have a increase in load on our production database that is also a publisher in transactional replication (distribution is also on same machine), currently we see >3 days latency from distributor to subscriber (log reader looks fine and this is push replication), what can we check or modify to improve the performance without moving distubitor to different machine(this will take some time for us).

Thanks

--Harvinder

MSmerge_sessions table

$
0
0

I need some clarification on what the values in this table mean, based on this article: http://msdn.microsoft.com/en-us/library/ms173753.aspx

The "durstion" is the doifference between "start_time" and "end_time".  I have calcualted that.  However, when you add in the "delivery_time", "upload_time", and "download_time", we get a larger number. 

What is an accurate representation of how long it took in total for a synchronization to occur between the publisher and the subscriber - from the time I hit the sync button, to the time all the data is synchronized and accessible on the subscriber.

Thanks!

 

Clint

Big Red X on Replication Monitor (sql server 2000) after dropping replication on a DB. How to remove this Red X? Why is it there?

$
0
0

For years we had a DB under replication (sql server 2000).  We don't need to replicate anymore, so I dropped replication with

sp_removedbreplication 'myDB'

I had to detach the DB to get rid of the corresponding log file that had grown to nearly 30 gigs.  I re-attached the DB with a new ldf  (500 megs), and now there is a big Red X on the Replication monitor.  Is this significant?  How do I get rid of this Big Red X?

Thanks


Rich P

Replication fails with weird error

$
0
0

Hi All,

I have peer to peer replication configured between 2 SQL server 2008 R2 enterprise servers. While replication is running fine from Node A to Node B, it fails from Node B to Node A with error below. I have verified and found that all configurations are fine. can you let me know if you have faced the same error?

Database 'dbo' does not exist. Make sure that the name is entered correctly.


SugeshKumar

Subscription are still showing in replication monitor after being dropped

$
0
0

Hello,

We have an app that uses sql server merge replication. I dropped a subscription at both publisher side and subscriber side. But when I look at replication monitor after successfully dropping, the subscription is still showing in subscription list. My question is the subscription list in replication monitor will keep dropped subscription until it expires (default is 14 days)? if so, how can I know which subscriptions are dropped manually before it expires?


help with setting up PROD to TEST data refresh with existing transactional replication

$
0
0

Hello, this is my first post to the forums.  I am hoping to get some guidance about setting up a weekly refresh of data from production to test.  I want to use SSIS but I am open to whatever method makes sense with the following scenario: 

prod server one (database named A) is the publisher and distributor to a database on prod server 2 (A_REPL). 

prod server 2 (database B) is the publisher and distributor to another database on prod 1  (B_REPL) (so 4 databases so far on two servers).

the test server has two databases that replicate to two other databases all on this test server (4 databases and 1 server) (named A and A_REPL and B and B_REPL)

Now I want to refresh the data from prod 1 one of the databases (database A)) to the test server (one of the publisher databases (database named A on this test server)). 

How can I refresh the data without messing up replication? Is SSIS the way to go?  I would do this after hours.

Thanks in anticipation of any advice anyone may have. General suggestions or specific details appreciated.  Whatever you have time for :-)

CLR Function Replication breaks in sql 2008 r2

$
0
0

The CRL function replication break seems to be due to the following...

A CLR function containing default parameter value in qoutes was added.  Now, when I try to generate a create function statement, the qoutes are stripped from the default parameter value.

For instance

I run this statement to add the function .... CREATE FUNCTION [dbo].[func](@list [nvarchar](max), @delim [nchar](1) = ',')

when generating create function sql, I get CREATE FUNCTION [dbo].[func](@list [nvarchar](max), @delim [nchar](1) = ,)

Any help on the matter is appreciated

USE (?what) database for sp_droppullsubscription -- what is the proper usage for sp_droppullsubscription?

$
0
0

Last week I ran   sp_removedbreplication 'myDB'  so that I could dettach it to get rid of a bloated ldf file (we're not replicating myDB anymore).  I ran sp_removedbreplication without running

sp_droppullsubscription

sp_dropmergepullsubscription


beforehand  (I ran checkDB on myDB -- everything is fine).  Now I have a big red x on the replication monitor.  I (sort of) understand that I need to deal with the distribution database when removing replication on a replicated DB.  From my readings, it says to use the Subscriber database to exec sp_droppullsubscription.

My question is this:  what is the subscriber database?  what do I USE?   Do I just go to master?  Can someone give an example of the usage of sp_droppullsubscription -- including USE ...   like

USE ....

sp_droppullsubscription  arg1, arg2, arg...


--and while I'm at it -- what is the correct usage of  sp_dropmergepullsubscription  ?


Thanks


Rich P

Error: Replication agent has not logged

$
0
0

Hi All,

I deleted and recreated publisher and subscriber but still i am getting below error. can anyone tell me what is the cause.

Error messages:
The replication agent has not logged a progress message in 10 minutes. This might indicate an unresponsive agent or high system activity. Verify that records are being replicated to the destination and that connections to the Subscriber, Publisher, and Distributor are still active.

Thanks

Agha

Merge Replication Resolving Conflicts for SQL Server 2008

$
0
0

Hi,

Can anyone help me with my issue...

I have one (merge) publication (data are managed from, in my case, web application) with several subscribers (data are managed by PDA devices).
There are around 35 tables that need to be synchronized.
Because records in each of those tables can be managed in both publication and subscribers, for eachsp_addmergearticle, I've set "@article_resolver = N'Microsoft SQL Server Subscriber Always Wins Conflict Resolver'" because data collected from PDA are more relevant that ones managed from web application, thus Subscribers always must win.

BUT when I sync. there is still conflicts that requires manually resolving (through "Microsoft Replication conflicts Viewer"). Normally, all I need to do is to "Submit Winner". All I need is that "submitting winner" to be automatically because these kind of "situation" is normal in my application.

To be more specific, I'll put one scenario for one table (with values from one set of parameters):

Table settings in DB and in script from publication:

Table1 (NPID, ParamID, ParamValue)

exec sp_addmergearticle @publication = N'MyDatabase_Rep', @article = N'Table1', @source_owner = N'dbo', @source_object = N'Table1', @type = N'table', @description = N'', @creation_script = N'', @pre_creation_cmd = N'drop', @schema_option = 0x000000B208034FF1, @identityrangemanagementoption = N'auto', @pub_identity_range = 1000000, @identity_range = 1000, @threshold = 80, @destination_owner = N'dbo', @force_reinit_subscription = 1, @column_tracking = N'false', @article_resolver = N'Microsoft SQL Server Subscriber Always Wins Conflict Resolver', @subset_filterclause = N'', @vertical_partition = N'false', @verify_resolver_signature = 0, @allow_interactive_resolver = N'false', @fast_multicol_updateproc = N'true', @check_permissions = 0, @subscriber_upload_options = 0, @delete_tracking = N'true', @compensate_for_errors = N'false', @stream_blob_columns = N'false', @partition_options = 0

through web application I've put following values:

NPID = 98
ParamID = 299
ParamValue = 5.00000000

through one of the PDAs for the same parameter I've put following values:

NPID = 98
ParamID = 299
ParamValue = 35.00000000

According to the table setting in the replication that subscriber always wins I thought that ParamValue = 35.0000000 will overwrite ParamValue = 5.0000000 BUT values in Table1 was same and when I check "View conflicts" I've notice that there were conflict :"The same row was updated at both 'WM_EU - 7c67c309222d' and 'SERVER.MyDataBase'. The resolver chose the update from 'WM_EU - 7c67c309222d' as the winner." and I need to confirm "Submit Winner"...

All I need to do is to, somehow, set the replication to handle this conflicts automatically.

Thanks,

Stefan

Viewing all 4054 articles
Browse latest View live


<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>