We are doing a proof of concept of heterogeneous transactional replication
from Sql 2005 to Oracle. I have some questions about the snapshot.
Q1) If the schema at the subscriber has been intitialized why do you need
to have the automatically created daily job that runs the snapshot?
I am concerned what happens if our DB grows to be very large, and sometime
after replication has already been initialized, gets out of sync.
Q2) After the subscriber has been initialized do any further snapshots
transfer over all the data or just what has changed since the previous
snapshot?
Thank you for any replies.
1) The daily snapshot is only run if the subscriber needs it. So the job
will wake up, figure out if any subscribers need the snapshot and go to
sleep if none of them do. If one of them does, it will generate the
snapshot. I am not sure why it is set up to run daily though as opposed to
once a week or something like that.
2) When the snapshot starts a water mark is placed in the distribution
database. All transactions occurring after this watermark are replicated to
the subscriber as transactions. This way after the snapshot is applied and
the distribution agent is run both databases will be identical or
consistent.
Hilary Cotter
Director of Text Mining and Database Strategy
RelevantNOISE.Com - Dedicated to mining blogs for business intelligence.
This posting is my own and doesn't necessarily represent RelevantNoise's
positions, strategies or opinions.
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
"DBAdan" <DBAdan@.discussions.microsoft.com> wrote in message
news:4B776D9B-259D-42FE-853A-FCD51156B353@.microsoft.com...
> We are doing a proof of concept of heterogeneous transactional replication
> from Sql 2005 to Oracle. I have some questions about the snapshot.
> Q1) If the schema at the subscriber has been intitialized why do you need
> to have the automatically created daily job that runs the snapshot?
> I am concerned what happens if our DB grows to be very large, and
> sometime
> after replication has already been initialized, gets out of sync.
> Q2) After the subscriber has been initialized do any further snapshots
> transfer over all the data or just what has changed since the previous
> snapshot?
> Thank you for any replies.
|||I always disable the snapshot job after initialization and run it manually
if needed. It's useful for heterogenous subscribers but generally I have
names ones so this is not relevant, and in your case it seems you have only
1 subscriber so the same applies.
Once the snapshot has applied, only inserts/updates/deletes will replicate,
rather than BCP flat-file data.
Cheers,
Paul Ibison SQL Server MVP, www.replicationanswers.com .
Monday, March 12, 2012
Explain daily snapshot in transactional replication
Labels:
concept,
database,
explain,
heterogeneous,
microsoft,
mysql,
oracle,
proof,
replication,
replicationfrom,
server,
snapshot,
sql,
transactional
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment