Showing posts with label distributer. Show all posts
Showing posts with label distributer. Show all posts

Wednesday, March 7, 2012

Replication not updating the subscriber

I am having some trouble getting merge replication working. I have 1 server
acting as a publisher and distributer and another server acting as my
subscriber. Both are running SQL 2000 SP4.
I need both the publisher and subscriber to be able to update the database.
One database is for internal company apps and the other is for our web
server both of which need access to the same database.
The steps are as follows.
1. Create a new publication using the "Create Publication Wizare" and using
all the defaults. When I come to the "Specify Articles" dialog I add my
table articles using the defaults and a few minor changes as follows:
- Keep the existing table unchanged.
- Automatically assign and maintain a unique range for each
subscription.
- Range size at pub = 10000
- Range size at sub = 10000
- Threshold = 90
2. I run the snapshot agent
3. I PUSH a new subscription using the "Push Subscription Wizard", using all
the defaults.
So far I have not gotten this to work at all. I get one of two major errors.
1. Violation of Primary Key constraint 'tablename'. Cannot insert duplicate
key in object 'tablename'
2. (If I select "Drop the existing table and re-create it" when adding
articles) I get an error complaining that the table cannot be dropped
because of a foreign key relationship (which it does have).
3. If I try and add any data at the subscriber I get an error telling me I
have to run Sp_adjustpublisheridentityrange.
What is the correct process here.
1) can we get more info on the primary key violation. Where does this show
up? Do you see it in the conflict viewer?
2) Ideally your subscriber would be an empty database when applying the
subscriber. Can you drop the problem table on the subscriber before
distributing the snapshot.
3) Check the table on the subscriber and drop the replication check
constraint on the identity column.
"Robert Sheppard" <sheprts@.cox.net> wrote in message
news:%23xX47jx5FHA.884@.TK2MSFTNGP14.phx.gbl...
>I am having some trouble getting merge replication working. I have 1 server
> acting as a publisher and distributer and another server acting as my
> subscriber. Both are running SQL 2000 SP4.
> I need both the publisher and subscriber to be able to update the
> database.
> One database is for internal company apps and the other is for our web
> server both of which need access to the same database.
> The steps are as follows.
> 1. Create a new publication using the "Create Publication Wizare" and
> using
> all the defaults. When I come to the "Specify Articles" dialog I add my
> table articles using the defaults and a few minor changes as follows:
> - Keep the existing table unchanged.
> - Automatically assign and maintain a unique range for each
> subscription.
> - Range size at pub = 10000
> - Range size at sub = 10000
> - Threshold = 90
> 2. I run the snapshot agent
> 3. I PUSH a new subscription using the "Push Subscription Wizard", using
> all
> the defaults.
> So far I have not gotten this to work at all. I get one of two major
> errors.
> 1. Violation of Primary Key constraint 'tablename'. Cannot insert
> duplicate
> key in object 'tablename'
> 2. (If I select "Drop the existing table and re-create it" when adding
> articles) I get an error complaining that the table cannot be dropped
> because of a foreign key relationship (which it does have).
> 3. If I try and add any data at the subscriber I get an error telling me I
> have to run Sp_adjustpublisheridentityrange.
> What is the correct process here.
>
>

Saturday, February 25, 2012

Replication Issues

Hello,
I am replicating databases from a sql2000 server with a
seperate distributer running sql2000 server and a MSDE subscriber. I
am using merge replication. The subscribers pull a subscription.
We want to have this setup across the internet for multiple
subscribers
each of whom might be updating as frequently as 1 hour. There are a
couple of issues I would like to find out about.
1.Security: I understood that the initial snapshots can be
published via ftp ( or any other means ofcourse). Can the merge
process be accomplished via ftp too? i.e The merge agent running on
the subscriber, writes to the ftp space reserved for it and the
publisher uses these files to bring the publisher up to speed and put
the updates back to all the subscribers. (sounds very slow). What are
our options as far as channel security if we are not using ftp ( and
is it possible in the first place across the internet).If we use a
secure ftp site, can the agents manage that?
2.Load: Our publisher has about 20 databases right now, with a
total of 1Gig file space occupied by mdf and log files. One database
is being replicated. Its consuming 1Gig of RAM Memory. Is that normal?
Is it because of the replication. How can I find out? Can anybody with
experience in this area, tell how much more load we can expect to
see? How much processing is needed for 40 users each updating about
500KB of data every one hour?
Thanks a lot,
Hananiel
1) I'm confused by what you are referring to by ftp. Are you talking about
the ftp port (21) or the ftp service. IE can the merge agent download the
commands using FTP? The answer is no, only the snapshot. If you are talking
about using port 21 the answer is yes. The merge agent will connect to the
publisher/distributor/subscriber using port 1433 or whatever port your SQL
Server is configured to run on, which could be port 21.
2) SQL Server will consume as much memory as is accessible on the machine or
whatever you configure SQL Server to consume using sp_configure 'max server
memory (MB)'
Hilary Cotter
Looking for a SQL Server replication book?
http://www.nwsu.com/0974973602.html
"Hananiel" <hananiel@.yahoo.com> wrote in message
news:e338e321.0409170710.68893ce9@.posting.google.c om...
> Hello,
> I am replicating databases from a sql2000 server with a
> seperate distributer running sql2000 server and a MSDE subscriber. I
> am using merge replication. The subscribers pull a subscription.
> We want to have this setup across the internet for multiple
> subscribers
> each of whom might be updating as frequently as 1 hour. There are a
> couple of issues I would like to find out about.
> 1.Security: I understood that the initial snapshots can be
> published via ftp ( or any other means ofcourse). Can the merge
> process be accomplished via ftp too? i.e The merge agent running on
> the subscriber, writes to the ftp space reserved for it and the
> publisher uses these files to bring the publisher up to speed and put
> the updates back to all the subscribers. (sounds very slow). What are
> our options as far as channel security if we are not using ftp ( and
> is it possible in the first place across the internet).If we use a
> secure ftp site, can the agents manage that?
> 2.Load: Our publisher has about 20 databases right now, with a
> total of 1Gig file space occupied by mdf and log files. One database
> is being replicated. Its consuming 1Gig of RAM Memory. Is that normal?
> Is it because of the replication. How can I find out? Can anybody with
> experience in this area, tell how much more load we can expect to
> see? How much processing is needed for 40 users each updating about
> 500KB of data every one hour?
> Thanks a lot,
> Hananiel
|||Hilary Thanks!,
Does SqlServer natively support encryption of
communication between its remote merge agents and itself? Or should we use a
tool like secure shell to make an encrypted connection?
Thanks,
Hananiel
"Hilary Cotter" <hilary.cotter@.gmail.com> wrote in message
news:eXjhIAOnEHA.644@.tk2msftngp13.phx.gbl...
> 1) I'm confused by what you are referring to by ftp. Are you talking about
> the ftp port (21) or the ftp service. IE can the merge agent download the
> commands using FTP? The answer is no, only the snapshot. If you are
talking
> about using port 21 the answer is yes. The merge agent will connect to the
> publisher/distributor/subscriber using port 1433 or whatever port your SQL
> Server is configured to run on, which could be port 21.
> 2) SQL Server will consume as much memory as is accessible on the machine
or
> whatever you configure SQL Server to consume using sp_configure 'max
server
> memory (MB)'
> --
> Hilary Cotter
> Looking for a SQL Server replication book?
> http://www.nwsu.com/0974973602.html
>
> "Hananiel" <hananiel@.yahoo.com> wrote in message
> news:e338e321.0409170710.68893ce9@.posting.google.c om...
>

Tuesday, February 21, 2012

Replication in LAN

Can Replication works in share network, I mean if i make a client as
publisher and server as distributer and subscriber willbe on remote location,
then will it work? if yes then how will i define the subscriber?
Manish,
if the networks are trusted then it is very much as per usual. if not, then
this article should help:
http://www.replicationanswers.com/InternetArticle.asp
Cheers,
Paul Ibison SQL Server MVP, www.replicationanswers.com
(recommended sql server 2000 replication book:
http://www.nwsu.com/0974973602p.html)