On SQL 2000 all inserts after exceeding the allotted range would be rolled
back. In SQL 2005 there is a 100% buffer before the inserts get rolled back.
range should be incremented.
"Polaris431" <polaris431@gmail.com> wrote in message
news:1172416552.106940.19180@h3g2000cwc.googlegroups.com...
> >From what I read on "automatic identy range management", you define a
> range of IDs for each subscriber. Of course, what happens once your
> run out of IDs on the client side?
>
>
> On Feb 25, 2:59 pm, "Hilary Cotter" <hilary.cot...@gmail.com> wrote:
>> No, unless you use automatic identity range management or some form of
>> partition all subscribers will enter the same value and you will have
>> primary key collisions which will result in conflicts.
>>
>> --
>> Hilary Cotter
>>
>> Looking for a SQL Server replication
>> book?
http://www.nwsu.com/0974973602.html >>
>> Looking for a FAQ on Indexing Services/SQL
>> FTS
http://www.indexserverfaq.com >>
>> "Polaris431" <polaris...@gmail.com> wrote in message
>>
>> news:1172397392.406716.20470@h3g2000cwc.googlegroups.com...
>>
>>
>>
>> > When using merge replication and all subscribers use the same
>> > subscription and when two or more subscribers insert new records, does
>> > replication on the server automatically modify the unique ID field for
>> > that record. In other words, if mobile user A inserts a record and the
>> > ID field automatically gets set to 5 and then mobile user B also
>> > inserts a record into the same table and the ID also gets set to 5,
>> > will replication at the server automatically reassign the ID so that
>> > no conflicts take place. After replication is completed, the record
>> > that gets sent to mobile user A might be 10 and the one sent to mobile
>> > user B might be 11. Is this how replication works when inserting data
>> > using merge replication?
>>
>> > Thanks for you help
>> > Polaris- Hide quoted text -
>>
>> - Show quoted text -
>
>
