Ask Ray Miao Question For Transaction Log Replication
Sep 13, 1999
Thanks. I worry about my setup information may be have some issue.I can't express it in word.so please tell me you e-mail address,i will send a document to you.
Some information may be useful,i had did same thing with another database,it's normal in 1 day.
I have not aliased pager as dbo. Are you suggesting to try this? I want the replicated tables to show a db owner of 'pager'. Here is another description of my situation (I think the first one may have been confusing):
I have a Database named Page with 'sa' as the db owner. The objects (tables and sps) of this database have the owner 'pager'. This is the way the database is defined on the Publication and Subscription Servers. But, when the replication process is completed, the replicated tables on the Subscription Server have the owner 'dbo'. I want my replicated tables on my Subscription Server to have the owner 'pager'. So, that the 4 replicated tables will have the same db owner as the existing tables on the Page Database.
Is this happening because I have the NT trusted connection which points to 'sa' on the SQL Server? If so, can I change the trusted connection to point to 'pager'? I am just guessing at this point. I will continue to research this issue. Someone has suggested that I go into EM, Manage Logins, select 'pager', and put 'dbo' in the Alias column for the replicated database (Page Db) row. I will try this.
Thanks in advance, Kevin
-----------
If you alias pager as dbo, the object created by pager will show owner of dbo. This is the way sql6.5 works, but sql7 will show pager as owner.
------------ Kevin G. at 11/22/99 5:14:22 PM
I am in the process of replicating a table within a workgroup setup (not a Domain).
I notice that when the table is created on the Subscription Server the replication process is creating the tables with a db owner of (dbo). I want the tables to be created with 'pager' as the owner. 'pager' is already defined as a db owner in the source and destination database. In the Publication Server Source Database, 'pager' is the owner of the tables selected for replication.
How can I accomplish this? Currently, I am doing the following on both the Publication and Subscriber Servers: logging on NT as 'Administrator' and starting both services (MSSQLServer and SQLExecutive) with the 'sqlnt2' NT Login which is a member of the NT Administrators Group. The NT Administrators Group has been trusted to SQL Server on both machines.
I am using the NT4.0, SP3 and SQL Server 6.5 with SP5a.
I will try the mixed security tomorrow. I have in the mean time discovered that I am not able to use the odbcping.exe utility successfully. It returns the same 08001 error message (Specified SQL Server not found). Does this means that the problem is within ODBC? What are somethings that I can try?
Thanks, Kevin
-----------
You need mixed security, also check password of the sql service account on nt.
------------ Kevin G. at 3/8/00 3:21:47 PM
I have set up replication on two SQL Servers (6.5), SP5a, on NT 4.0 (SP3). The Distribution Task on the Publisher is failing with the following error:
08001[Microsoft][ODBC SQL Server Driver][dbnmpntw]Specified SQL Server not found.
I am using standard security in a workgroup environment. I have my trusted connection setup and I am using named pipes. I had this process working on our test servers but when I tried to implement it into production I received the above message. Please give me some ideas or things to try. What source can I use to look up the 08001 error?
And thanks to Ray for another frustrating response. Gentle suggestion, if you are going to hang out here you may find people appreciate a tad more detail than 'script it'. I consistently get a chuckle out of your responses.
Ray Miao is at it again ...( refer thread on " SQL 7.0 TEMPDB " for laughs )
John says his tempdb is growing, he cannot find a way of shrinking it other than restarting SQL Server. Ray Miao's $1.7 million answer ? Restart SQL Service !!
A simple theory should apply to these kind of forums. Be CONSTRUCTIVE !!? If you do not have a reasonably good answer for a problem, keep out !!!
I see you answering questions out here all the time like I do, and had one question for you.
Are you really moving all of your user data from the primary (or default) filegroup into another named group like is discussed in the swynk column for up-to-the-minute recovery?
I am not clear on the benefit of doing this. It appears to me that the inability to restore to current would only happen if the primary group was corrupt, and I don't read that this happens very often.
Ray, I'm sending you this message because you have helped me (and many others) in the swynk forum, and I appreciate that. I just joined this new forum, and can't do a search using "SQL" because it doesn't have at least 4 letters. Is it possible that we have a SQL forum where SQL is not allowed in searches?? Are they kidding? Respectfully, Robert Boyd
Can anyone help me with handy scripts/stored procedures to capture blocking info on the server? We have SQL 7.0 w/sp3. Ray replied earlier saying that I can use so_who2!!, I think everyone knows that,please reply with some valuable info!!Please.. Thanks. Sonali.
Dear all, I set up a transaction replication (from server A to server B) for tables and stored procedures. When I update the data at server A, the data at server B will be updated by replication. However, the stored procedure can't update at server B no matter what I changed at server A's stored procedure. How can I update the stored procedure using transaction replication? The other question is the table schema. As I know, the transaction replication will be created a snapshot to project all needed tables and stored procedures at the first time. If I created a new table, stored procedure or changed the table schema (e.g. add/delete a new column), the transaction replication didn't update the new objects to the server B. What can I do to update the transaction replication except creating a new subscription?
Good Morning to all, I am new to SQL Server.Here is my question. I have set up transactional replication with immediately updating subscriber between to servers A & B. A is publisher/Distributor and B is subscriber. In the event A fails the applications will be directed towards B. My problem is suppose A has faliled and my data is now written to B. How can i syncronise data in A and b when A is again online and what should i do ?
Hi all, I am using SQL 2k on two seperate servers on win 2k advance servers without any service packs. My problem is i am using transactional replication between two servers (SML1 and SML2), SML1 is publisher/Distributor and SML2 is subscriber. My db name on both the server is same. when i am inserting data in db on SML1 it replicates immediately on SML2 but vice versa in not happening. Any help in setting up two way replication is highly appreciated. (There is no timestamp column on table which is used as article and there is PK).
My Transaction Replication fail with this error message pls help.
This is the error message The process could not create file 'SQLMS1d$MSSQL7ReplDatauncSQLMS1_AFT_National_EMN _National - Transaction - Changes at Both Sides - EMN to Local Only20000607090006Benefit.sch'.
I it possible to setup transaction replication between two sql servers if both Sqlerveragents runs under different accounts and the servers are in different domains
I am attempting to set up transactional replication between a production server (SQL1) and a “reports” server (SQL2). The production server is the publisher, the “reports” server is the distributor and the first subscriber, other subscribers will be added later. There are multiple databases on the production server; selected tables from each will be replicated to corresponding databases on the other servers. Both are SQL7, no SP (yet).
I have set up the replication on two databases so far. The initial snapshot replication runs fine; the tables and data are duplicated on the reports server.
For one DB the transaction replication fail the other still reports “No replicated transactions are available”, which is probably true.
The failing log reader agent history reports: “The process could not execute ‘sp_replcmds’ on SQL1“ (five time, 10 minutes apart). The session details shows 2 items, “Initializing” and the same error message. The error details show it as an ODBC error timeout.
After the fifth time the log agent history reports “The process could not execute ‘sp_Msadd_repl_commands27hp’ on SQL2 ” The session details show 3 items , “Initializing”, “No replicated transactions are available” and the same error message. The error details show the following (Category, Source, Number, Message): Agent, blank, 2000, “IDistPut Interface has been shut down.”; Agent, SQL1-JCL1, 20007,” No replicated transactions are available.” ( 26 times); Data Source, SQL2, 1007 “Cannot insert duplicate key row in object 'MSrepl_transactions' with unique index 'ucMSrepl_transactions'.Cannot insert duplicate key row in object 'MSrepl_transactions' with unique index 'ucMSrepl_transactions'.”
I can restart the agent, but after two hours the same thing appears. These were set up with EM using the wizzards
I had established a transaction log replication ,it's not immediate-updated. on next 2 days,it's normal.but now,it's not available ,and haven't any warning,error,informatin. I have very trouble about the 'log reader' is running but i change the data in articles it's no effect. please tell how i do?
Hi, Every one Can I add coulmn in my existing table which i am using for replication. This article is i am using for transaction replication. I added column to both articles , publisher & subscriber, But now it is giving me error that, Insert error. I checked in Create & Manage publications property on E.M , that cloumn is not checked & it is not allowing me to do from E.M , Does any one know which store proc I require to execute from QA.
I am trying to setup real time transaction replication between two databases on two different sql server 7.0.The ideas is to replicate all the objects from the first database to the second database including table, stored procedures and views. Both SQL servers’ runs under different service NT account. The sql server that I want to replicate runs an application that create additional tables when the application is running.
How do I schedule a transaction replication so that all the objects plus those additional tables and stored procedures that the application creates at run time are also going to get replicated in real time?
I have been asked to take over the maintenance on dB servers that I have not originally 'set up'. The request came as a result of the database log files growing at such a rapid pace its taking up hard drive space. I ran sp_dboption and noticed that the log files are NOT set up to truncate on checkpoint. As in SQL 2000 I wanted to determine if the dB was set up in 'simple' or 'full' recovery mode. It appears they are set up in 'full' recovery mode (which we can modify, since this is not a transactional dB and we rarely if ever have to restore from the transaction log). Hence my question,
Is there anything I need to worry about (regarding replication)if I change the dboption to truncate on checkpoint and set dB in simple recovery mode?
And after I backup the transaction log and try to shrink it will that have any adverse affects? I feel like this is all I need to do but replication is new ground for me, any information regarding it would be greatly appreciated. Also, any advice on how to maintain the transaction logs would be helpful.
Do you know the function of system store procedures for replication in publish database and distrubtion database? please tell me. I'd established 'transaction log' replication, it's like something is full or can't active,i don't know what process is unvalid. How i do!
Hi, Every one I have added new column in my Publication - Article Table , Same column I have added in my Subscriber Table , I am doing Transaction replication , but now its giving me error Insert Error - Column Name or Supplied Value doesn't Match, If any one has any idea please suggest me Thanks Nirmal
Hi all, I am using SQL 2k on two seperate servers on win 2k advance servers without any service packs. My problem is i am using transactional replication between two servers (SML1 and SML2), SML1 is publisher/Distributor and SML2 is subscriber. My db name on both the server is same. when i am inserting data in db on SML1 it replicates immediately on SML2 but vice versa in not happening. Any help in setting up two way replication is highly appreciated. (There is no timestamp column on table which is used as article and there is PK).
Besides loopback detection, is there anything else that is needed to prevent the replication command from sending back to the source. Cos i am having problem with the subscriber B sending back the replication to A in the form of B as the publisher and A as the subscriber.
Two Servers,1st 2nddatabase a -------> after replication a'b' <------- after replication bon 1st server a is replicated to 2ndon 2nd server b is replicated to 1stI tried with merge replication it is working fine.But if i tried to usetransactional replication it give some error like "accessviolation".Using tansactional replication is it possible.plz help me toslove this problem.Regards,Senthil prabu R
Hi all, using transactional replication in SQL 05, I have a schema that is different at the subscriber and wanted to know the best approach for transforming the published data. For example, the subscriber is more normalized than the data being published causing 1 published record broken out into to many tables at the subscriber. Also need to applying scalar funtions to a few published columns before committing at subscriber.
I've read DTS, although backwards compatible, is being deprecated in a future release. I suppose I can use either SSIS or custom procedures. I didn't see any tie-in to SSIS in the system procedures to setup replication (only DTS). Any advice on which would work in this type of scenario?
:mad: I have one Publisher! (SQL 2005) Two Subscribers. (SQL 2005) Each subscriber requires different row filter from the published tables. So, i need to create Two publications for each subscriber for the same db. When i setup first publication, and apply filters to published tables. All goes fine. As soon i create an other publication with same published tables but DIFFERENT row filter; the log reader agent fails!
The process could not execute 'sp_replcmds' on 'DB1'. (Source: MSSQL_REPL, Error number: MSSQL_REPL20011) Get help: http://help/MSSQL_REPL20011 Execution of filter stored procedure 2076702271 failed. See the SQL Server errorlog for more information. (Source: MSSQLServer, Error number: 18764) Get help: http://help/18764 The Log Reader Agent failed to construct a replicated command from log sequence number (LSN) {0008443a:000003b5:003e}. Back up the publication database and contact Customer Support Services. (Source: MSSQLServer, Error number: 18805) :confused:
I have tried to re-initialze the subscriptions and re-run snapshots but no luck. As i remove the row filter from the second publication; log reader starts working fine.
Previously i was using Merge replication and all was going BUT, on very large DML on publisher, we use to receive deadlock errors on application. I decided to use Tran repl instead because log reader reads the changes from tran-log instead of locking the published tables; NOW I AM STUCK!
I wanted to schedule the transaction replication. How do I do it? Currently I have set up a transaction replication which runs continuously and synchronizes the changes with immediate effect.
I need to configure a replication which will gather logs from the publication once in a day.