We have a server with one default collation, but a DB that needs to be published which has another collation.
When I configure Distribution, the db Distribution gets the system default collation. When I then try to make a publisher for this DB that runs another collation, the system whines about different collations. How can I publish a DB which has a different collation than the system and thus the distribution db?
We have Transactional Replication running from SQL Server 2005 to 2012 (both standard editions). The distributor sits with the subscriber on the SQL2012 machine. Since we fired it up we have been gettingsporadic failures of data import SPs. By sporadic I mean cca. 1 failure in 20-30 executions. The error message is:
Schema change failed on object '[dbo].[TableA]'. Possibly due to active snapshot or other schema change activity.
But...
Due to active snapshot? - We don't use the Snapshot Agent; it is disabled.
Other schema change activity? - We are getting these failures since we started using replication and we are doing just as much schema changes at the moment as we had been doing before replication was started.
The schema changes the SPs do are along these lines:ALTER TABLE dbo.TableA NOCHECK CONSTRAINT Constr1 --insert some stuff etc. ALTER TABLE dbo.TableA CHECK CONSTRAINT Constr1
or
ALTER TABLE dbo.TableA DISABLE TRIGGER ALL
Triggers, check constraints are NOT replicatied.
The question is then: Why are we getting this?
The parameters of the publication for your reference:
pubid 1
restricted 0
status 1
task 1
replication frequency 0
synchronization method 3
immediate_sync 0
enabled_for_internet 0
allow_push 1
allow_pull 0
allow_anonymous 0
independent_agent 1
immediate_sync_ready 0
allow_sync_tran 0
autogen_sync_procs 0
snapshot_jobid 0xFFFF
retention 0
has 1
subscription
allow_queued_tran 0
snapshot_in_defaultfolder 1
alt_snapshot_folder NULL
pre_snapshot_script NULL
post_snapshot_script NULL
compress_snapshot 0
ftp_address NULL
ftp_port 21
ftp_subdirectory NULL
ftp_login anonymous
allow_dts 0
allow_subscription_copy 0
centralized_conflicts NULL
conflict_retention 14
conflict_policy NULL
queue_type NULL
backward_comp_level 40
publish_to_AD 0
allow_initialize_from_backup 0
replicate_ddl 1
enabled_for_p2p 0
publish_local_changes_only 0
enabled_for_het_sub 0
Hi,
I tried on building transactional replication connecting the systems within same domain
while creating the publisher and subscriber given the domain credentials as in the windows account.
Snapshot Agent and Log Reader Agent are not throwing any error. But when checking on Job history the following error is thrown
[SQLSTATE 22001] (Error 8152).
Detect nonlogged agent shutdown.,,Executed as user: NT SERVICE\SQLSERVERAGENT. String or binary data would be truncated. [SQLSTATE 22001] (Error 8152). The step failed.,00:00:00,16,8152,,,,0
The exception type is
Exception Type: Microsoft.SqlServer.Replication.ConnectionFailureException
Someone please help on this....
Thanks,
Anitha.B
| 575706 |
Hello helpful community, :)
I have a question about replication / logshipping.
Our situation: We have old, big databases on an current production servers (~900GB). (We need to keep many years of data because of FDA) We build a complete new system with new databases, new applications etc. The new databases will go live end of the year
and we plan to run both systems parallel for some time. The issue I'm facing are the reports. We need to pull data from both (old and new systems) during this transition period before we can archive the old system.
My Question: We have a new and fast report server. I would like to replicate the current data to the new report server to be able to pull data. Is it possible to setup a replication or logshipping for a certain date range? I thought about having two years worth of data in the subscription / secondary database. 2 years is the range we need for our internal reports.
Additional information: I would prefer to use a replication or logshipping instead of linked servers. The new SQL servers are faster than the old once.
Old systems: SQL 2005 SP4
New systems: SQL 2012 SP1
Thanks in advance
I have a sql server failover cluster with 2 nodes (named DBS). The database runs normally. The OS is Windows Server 2012 Standard Edition. The version of DBS is sql server 2012 enterprise.
And I have 4 stand-alone databases (DB1, DB2, DB3, Dist). The Dist is set up to a distribution database. DBS is a publisher. DB1, DB2, DB3 are subscribers.
But the DBS ran very slowly. Actually, it very hard to use Management Studio. I usually have got error message: Time out.
Therefore, I have to install another stand-alone database, and make it to a publisher. And the replication is fine.
So what 's happened with my sql cluster ???
I can not find any errors or warnings in Event viewer and Sql server logs. Anybody has the same problem ?
We've merge replication configured in our environment(1 Publisher, 3 Subscribers) and We know by default merge replication is bi-directional. Its working fine in case of inserts i.e, if a row is inserted at publisher then it's getting updated at subscriber and vice versa. The real issue is, if row gets deleted at publisher then its getting deleted at subscriber but, if a row gets deleted in any of 3 subscribers then its not getting deleted at publisher.
When a row is deleted at publisher then it's getting updated in MSmerge_tombstone(Publisher). But, in case of subscriber, the corresponding row(rowguid) is not getting updated in MSmerge_tombstone(Subscriber). I think that might be the issue, even merge delete trigger exist on these tables(Subscriber). Please help me out, as it is a production issue.
Note: delete_tracking is enabled(set to '1')
Publisher(2008R2 sp1-Enterprise), Subscribers(2008R2 sp1-Standard)
HI Team,
I do have 23 publishers and some of the publishers have 1 to 15 subscribers.
All are transnational( one way)/push subscriptions .
when I browse the below error table in distribution database i do get error codes. But is there any way to find the publisher/subscriber which is causing those errors?
I was using the below command to get the error details.
SELECTtop 100 * FROM distribution.dbo.MSrepl_errorsORDERBY TIME
Please help me ...
Hi,
For example, if my replication having some 100 Tables. Out of 100 if i want to replication only single table what to do? because if i reinitialize i all 100 tables will going to reload.
Hi,
I have a database with merge replication. there are four tables with bulk insert option. because of this those tables are not in complete sync. so every time I am doing dummy updates by finding through tablediff cmd, before i done everytime like that. now we are planning to fix the issue by putting fire triggers to solve this. Now i planned to do manual sync for the last time, three tables i found and done the dummy update but one table i am not able to find through tablediff, it is showing like both the table are identical. but the row count it is showing like more than one lack records difference. So, I m not able to solve the issue...
Can anybody please help me that how to figure out those records.
Thank you,
Aparna.Ch
We have SQL Server 2008 R2 running on our Windows 2008 R2 server that is the Publisher and the Distributor. We have SQL Server 2008 R2 express (Windows 7) running on our desktops/laptops with Pull Subscriptions running. We have 6 subscriptions configured/running and one of those is getting the error below. We are using the default distribution manager profile and it is not designating OLEDB streaming. Windows Sync Center on the laptop/desktop is being used to Sync the subscriptions. Subscriptions can be reinitialized successfully but the pull of updates results in the error.
Here is the TSQL used to build the subscription:
use [DMLR]
GO
exec sp_addpullsubscription @publisher = N'WSQ00796', @publisher_db = N'DMLR', @publication = N'DmlrEnf', @independent_agent = N'true', @subscription_type = N'anonymous', @description = N'Transactional publication of DMLR database from Publisher WSQ00796.',
@update_mode = N'read only', @immediate_sync = 1
exec sp_addpullsubscription_agent @publisher = N'WSQ00796', @publisher_db = N'DMLR', @publication = N'DmlrEnf', @distributor = N'WSQ00796', @subscriber_security_mode = 1, @distributor_security_mode = 1, @frequency_type = 64, @frequency_interval = 1, @frequency_relative_interval
= 1, @frequency_recurrence_factor = 0, @frequency_subday = 4, @frequency_subday_interval = 5, @active_start_date = 0, @active_end_date = 0, @active_start_time_of_day = 0, @active_end_time_of_day = 235959, @enabled_for_syncmgr = N'true', @use_ftp = N'false',
@publication_type = 0, @dts_package_location = N'subscriber', @offloadagent = N'false'
GO
In the past, the subscription has been an MSDE subscription interfacing w/SQL 2000.
The distribution agent failed to create temporary files in 'C:\windows\system32' directory. System returned errorcode 5. (Source: MSSQL_REPL, Error number: MSSQL_REPL21100)
Error messages:
I inherited a server that has pieces of replication remaining. I cannot remove this.
I have tried to disable through the GUI but I get errors for missing tables (syssubscriptions, then syspublications, then a missing column error).
I have tried to run sp_dropdistribution @no_checks=1,@ignore_distributor=1. This results in error for missing syssubscriptions table. Create that table manually and get error on invalid column srvname. This errors but also leaves the transaction open.
I have tried to run sp_replicationdboption @dbname='xxdbname',@optname='publish',@value='false'. This gives an error trying to connect to the old distributor server.
I have tried to run sp_removedbreplication 'dbname', @type='tran' but this causes blocking on my server and hasn't been able to run.
Is there any way to manually clean this up? Since the pieces are there, my log file is growing. dbcc opentran() shows an oldest non-distributed LSN and log_reuse_wait_desc = 'REPLICATION'.
Thanks for any help or suggestions!
HI all,
When I checked from "help subsciption stored procedure" I could not find any entry on initialization/mark for reinitialzation .
Is there any way to track the reinitialization history( even for article/entire subscription)
regards,
I'm trying to get merge replication working with a small test database with just one table.
- I've setup the distributor, publisher(local distributor), created merge publications and publication snapshots. When I run the snapshot agent, I can see that the snapshots are getting created in the share.
- I've setup IIS based on the instructions for IIS7 and the Web URL is working fine.
- Created a merge pull subscription and agent at the subscriber.
But when the merge agent run I get following error consistently,
The table schema seems to copied over but no data comes thru. I've run the merge agent with verbose output level. Can anyone please see what I'm doing wrong. The following in the snippet from the merge agent output file, I have the full output file also, if someone needs it.
...
...
2013-09-23 21:17:57.317 Bulk copying data into table 'demo'My organization has 3 branches, a Main branch and 2 sub branches, in 3 different locations. All branches have similar platforms of Windows server 2008 R2 and SQL Server 2008 R2.
All branches maintain its area information (eg. customer info for that particular area).
Now what i want is to let the Main branch allow access to the other 2 sub branches data and i want the sub branches to only maintain their area information.
Please let me know if this is possible via replication. And if not what other options are there?
What i do not want is to overload the sub branches with all branches data as these sub branches are very small scale and located at very remote areas where high bandwidth connectivity is costly. I've attached a diagram with this post... please refer it for a more details.
Thanx all in advance
Hi!
If I use transactional replication retention period as "never expires", how it will effect on the dick space occupied by "repldata" folder. Will SQL server store transaction files for replication for ever in this case?
Thank you!
Hello,
I have replicated database R1 to database R2.As per the concept any changes made in R1 should reflect in R2 automatically. Replication process is working fine.But the problem i encounterd was, when i created a new table in R1 database,That table did not got replicated to R2 database . Please give any solution to my problem . What setting i need to do when i keep for replication.
Regards
Manoj
We recently retired a replication publication, deleting each subscription (push subscriptions) and then deleting the publication, all from within SSMS. However, we continue to get the following message (once every minute) in the four days since deleting this publication.
Replication-Replication Distribution Subsystem: agent (null) failed. The publication 'Replicate_ASG_Reference_Data' does not exist.
Why is it still trying to run a deleted publication, and how do I stop it?
I have added new tables and stored procedures in to the database which is already replicated now i need to run the sync to replicate changes,
please advise the procedure.
Regards
k
