When you say "edit the replication stored procedures" are you talking about
the sp_MSins_TableName, sp_MSupd_TableName, etc, procs? If such is the case,
the section to comment out is after the 'else' statement?
If I go with Queued Updating, can I still keep the identity attribute
property of the column set to Yes on the subscriber?
Roger.
[quoted text, click to view] "Paul Ibison" wrote:
> Roger,
> on the subscriber there shouldn't be the identity attribute at all. You
> could remove this attribute (identity - No), or edit the replication stored
> procedures and comment out the second section. If you want it all to work on
> failover you could use Queued Updating Subscribers, in which case the
> replication stored procedures are coded differently and you won't have this
> issue. Also, you'll have to consider the identity range, and the queued
> updating option will do this for you if you select automatic range
> management.
> HTH,
> Paul Ibison SQL Server MVP,
www.replicationanswers.com > (recommended sql server 2000 replication book:
>
http://www.nwsu.com/0974973602p.html)
>
>
Roger,
yes this is the section I was referring to, but I would definitely go with
queued updating subscribers instead. The Identity property should be Yes,
Not For Replication on the publisher and it'll be transferred in this way to
the subscriber. Also don't forget to enable automatic range management to
make your life easier, otherwise you'll have to reseed each identity column
after failover.
HTH,
Paul Ibison SQL Server MVP,
www.replicationanswers.com (recommended sql server 2000 replication book:
http://www.nwsu.com/0974973602p.html)