reinitialization of subscriptions

If I have a subscription to a publication (transactional)
and the Distribution Agent tells me I need to
reinitialize.
What are the steps to do this?
Do I need to delete all the data on the subscriber? If
so, is there a way around this? If not, what is the best
practice of accomplishing this?
What if there are changes made at the subscriber do I
loose them if the subscription is reinitializatized?
Thanks...
Jimmy,
reinitialize means apply a new snapshot, so, you'll need to run the snapshot
agent first then run the distribution agent. When you say will it delete all
the data on the subscriber, I'm not sure why this is a concern. For
transactional replication, the subscriber data is treated as RO data. If
however someone has edited the data, then you could use binary checksums to
see what has changed and then do the same changes on the publisher. If the
subscriber is using updatable transactional replication then the changes are
already on the publisher. If you're using a queue, then be sure to run the
queue reader to send up the changes.
HTH,
Paul Ibison
|||expand your publishing database, right click on your publication, select
properties, click on your subscriber tab, highlight your subscriber, and
click the reinitialize button.
It will delete all the data of the replication objects and their schemas and
then replace them.
"Jimmy" <anonymous@.discussions.microsoft.com> wrote in message
news:9dc801c43447$43f3a5a0$a601280a@.phx.gbl...
> If I have a subscription to a publication (transactional)
> and the Distribution Agent tells me I need to
> reinitialize.
> What are the steps to do this?
> Do I need to delete all the data on the subscriber? If
> so, is there a way around this? If not, what is the best
> practice of accomplishing this?
> What if there are changes made at the subscriber do I
> loose them if the subscription is reinitializatized?
> Thanks...
|||Paul,
You said...
then you could use binary checksums to see what has
changed and then do the same changes on the publisher
How?
Larry...

>--Original Message--
>Jimmy,
>reinitialize means apply a new snapshot, so, you'll need
to run the snapshot
>agent first then run the distribution agent. When you
say will it delete all
>the data on the subscriber, I'm not sure why this is a
concern. For
>transactional replication, the subscriber data is
treated as RO data. If
>however someone has edited the data, then you could use
binary checksums to
>see what has changed and then do the same changes on the
publisher. If the
>subscriber is using updatable transactional replication
then the changes are
>already on the publisher. If you're using a queue, then
be sure to run the
>queue reader to send up the changes.
>HTH,
>Paul Ibison
>
>.
>
|||Larry,
there's a good explanation of BINARY_CHECKSUM in books online (BOL). If you
have any problems with the explanation, please post back.
Regards,
Paul Ibison

REINDEX in a replicated environment

Hello,
I have an environment where I have a DB "source" being published from server
"A" to a DB "destination" on server "B". The "distribution" database in on
server "A". The replication agents are running on server "A".
I have a downtime where no users or jobs will be running on the "source" DB.
My queries are:
1) Will my replication be impacted if I run a DBREINDEX command on the
indexes in the "source" DB.
2) Will my replication be impacted if I run a DBREINDEX command on the
indexes in the "destination" DB.
3) Do I need to disable replication to be able to run the DBREINDEX command
on the subscriber and publisher end?
4) What would be the ideal steps to do a DBREINDEX on both the publisher and
subscriber end.
Thank you for your responses.
Regards,
Salil.
The answer to all of these is no, however DBReindex is logged and your log
reader agent won't work as efficiently as it has more log to chew through.
Use indexdefrag instead as it is an online operation.
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
"Salil" <Salil@.discussions.microsoft.com> wrote in message
news:041E047F-D8C9-4BDF-A7BC-764365EDA489@.microsoft.com...
> Hello,
> I have an environment where I have a DB "source" being published from
server
> "A" to a DB "destination" on server "B". The "distribution" database in on
> server "A". The replication agents are running on server "A".
> I have a downtime where no users or jobs will be running on the "source"
DB.
> My queries are:
> 1) Will my replication be impacted if I run a DBREINDEX command on the
> indexes in the "source" DB.
> 2) Will my replication be impacted if I run a DBREINDEX command on the
> indexes in the "destination" DB.
> 3) Do I need to disable replication to be able to run the DBREINDEX
command
> on the subscriber and publisher end?
> 4) What would be the ideal steps to do a DBREINDEX on both the publisher
and
> subscriber end.
> Thank you for your responses.
> Regards,
> Salil.
|||Thanks very much for your response...
I am doing DBREINDEX because INDEXDEFRAG will not affect Extent
Fragmentation much and my indexes are badly in need of a rebuild.
"Hilary Cotter" wrote:

> The answer to all of these is no, however DBReindex is logged and your log
> reader agent won't work as efficiently as it has more log to chew through.
> Use indexdefrag instead as it is an online operation.
> --
> 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
> "Salil" <Salil@.discussions.microsoft.com> wrote in message
> news:041E047F-D8C9-4BDF-A7BC-764365EDA489@.microsoft.com...
> server
> DB.
> command
> and
>
>