| Author |
Replicate between to MSSQL 2000 Servers
|
|
|
| Hello
I have 2 locations, that each has an MSSQL 2000 server.
Since the locations are to far from each other and the connection is
very slow.
My thought was to have a database on both servers that should replicate
“on the fly” or as asynchronous sync.
The problem is that both locations work in the database frequently. So
it’s very important that there are no duplicate IDs.
Is that possibly and how?
~Peter
| |
| lightning-dave 2006-03-05, 8:33 pm |
| Peter,
What datatype are you using for your IDs and how are you generating
your IDs?
Dave
| |
|
| Peter wrote:
> Hello
>
> I have 2 locations, that each has an MSSQL 2000 server.
> Since the locations are to far from each other and the connection is
> very slow.
>
> My thought was to have a database on both servers that should replicate
> “on the fly” or as asynchronous sync.
>
> The problem is that both locations work in the database frequently. So
> it’s very important that there are no duplicate IDs.
>
> Is that possibly and how?
>
>
> ~Peter
the database drives an HelpDesk/ServiceDesk system.
since it is to slow to use on the same database from the other location
, the plan was to make a dtabase on both locations that replicates.
How it generates the id and where/wich type. I do not no.
There must be over 300 tables on that database.
~Peter
| |
| lightning-dave 2006-03-05, 8:33 pm |
| The problem is that there is no way (until you work out how keys are
generated) of ensuring that duplicate keys are not created. If they're
created as a uniqueidentifier you could potentially use merge
replication but you need to understand the schema of your database to
do this and work out which tables you need to replicate.
I think the key to solving this problem is working out how your keys
are generated.
| |
| Greg D. Moore \(Strider\) 2006-03-05, 8:33 pm |
|
"Peter" <SPAM@Changer.dk> wrote in message
news:4406c2f8$0$7828
5$157c6196@dreader1.cybercity.dk...
> Hello
>
Look at Merge Replication and at a mechanism that ensures your keys are
unique.
If it's only two locations somtimes it's easiest to have location one start
at 1 and count up and location two start at -1 and count down.
> I have 2 locations, that each has an MSSQL 2000 server.
> Since the locations are to far from each other and the connection is
> very slow.
>
> My thought was to have a database on both servers that should replicate
> “on the fly” or as asynchronous sync.
>
> The problem is that both locations work in the database frequently. So
> it’s very important that there are no duplicate IDs.
>
> Is that possibly and how?
>
>
> ~Peter
|
|
|
|