missing records on subscribing database (transactional replication)

Hi all,
I have a publishing database and a subscriber database on two
different machines. Transactional replication is configured between
them. Everything is straigforward at the start, then after a few days
I got reports that the same tables on both databases are not in synch,
and some tables in the subscriber database has missing rows.
Has anyone experienced this? I realize that it is possible to delete
rows on the subscriber database, but other than that, is it possible
for the transactional replication to actually fail to copy some rows
from the publishing database?
There are no filters set, btw.
Thanks!
Aramid
If you don't have any errors in the replication engine, then, no it will not
fail to send changes. You would have to interfere with it in some way such
as within the distribution database. I would suggest doing some auditing at
your subscriber and see if there is a process which is making changes there.
Mike
http://www.solidqualitylearning.com
Disclaimer: This communication is an original work and represents my sole
views on the subject. It does not represent the views of any other person
or entity either by inference or direct reference.
"Aramid" <aramid@.hotmail.com> wrote in message
news:8fueu1to6vadlh5dd4q1be9509cteqnbdk@.4ax.com...
> Hi all,
> I have a publishing database and a subscriber database on two
> different machines. Transactional replication is configured between
> them. Everything is straigforward at the start, then after a few days
> I got reports that the same tables on both databases are not in synch,
> and some tables in the subscriber database has missing rows.
> Has anyone experienced this? I realize that it is possible to delete
> rows on the subscriber database, but other than that, is it possible
> for the transactional replication to actually fail to copy some rows
> from the publishing database?
> There are no filters set, btw.
> Thanks!
> Aramid
|||Aramid,
these are 2 separate possibilities that you're describing. In one, the rows
are deleted from the subscriber, while in the other, some inserts are not
sent from the publisher to the subscriber. For the first, this is unlikely
to occur through the replication engine, so I'd go along with Mike's
suggestion and use Lumigent's LogExplorer to investigate. This is also
possible to occur naturally through the replication, if you are replicating
stored procedure execution. In the other case, I'd use sp_browsereplcmds to
see if the commands are still waiting to be sent down.
Cheers,
Paul Ibison SQL Server MVP, www.replicationanswers.com
(recommended sql server 2000 replication book:
http://www.nwsu.com/0974973602p.html)