Process for Dropping & Readding articles in Trasnsreplication

  • Hi,

    I have a Publisher (Pub_A) with two subscribers (Sub_A/Sub_B), both are transactional replication with pull subscriptions.

    The publication has ten articles. I have a requirement to drop two articles, add a column to each, then add them back in to the publication. Just looking for a sanity check for the correct process.

    1. Drop subscription SubA & SubB

    2. Drop articles at PubA

    3. Make changes

    4. Add articles to PubA

    5. Add pull subscriptions on SubA/SubB

    6. Generate snapshot for new articles only.

    Thanks

    qh

    [font="Tahoma"]Who looks outside, dreams; who looks inside, awakes. – Carl Jung.[/font]
  • Seems about right but would strongly encourage you to verify that the anonymous/immediate_sync options have been set properly, if they are not set to 0, when you trigger the snapshot...it will create snapshots for all articles (not just the two you want).

    You can check this by running /* On the distributor run to verify the publishers - make CERTAIN allow anonymous/immediate_sync are all ZEROS */

    SELECT publisher_id, publication_id,

    publisher_db ,

    publication , allow_anonymous, immediate_sync

    FROM distribution. dbo.MSpublications WHERE publication_type = 0 --> 0 for transactional, 1 for snapshot

    /* If IMMEDIATE_SYNC is not zero, run the following on the ditributor */

    EXEC sp_changepublication @publication = 'YourPub', @property = 'allow_anonymous', @value = 'false'

    GO

    EXEC sp_changepublication @publication = 'YourPub', @property = 'immediate_sync', @value = 'false'

    GO

    ______________________________________________________________________________Never argue with an idiot; Theyll drag you down to their level and beat you with experience

  • Cheers MDJ.

    qh

    [font="Tahoma"]Who looks outside, dreams; who looks inside, awakes. – Carl Jung.[/font]
  • Will data be modified during the process and are you just adding a column?

    If the data is static and will remain static then you can skip the snapshot generation step depending on the size of the table.

  • MysteryJimbo (7/31/2014)


    Will data be modified during the process and are you just adding a column?

    If the data is static and will remain static then you can skip the snapshot generation step depending on the size of the table.

    No, data will remain static and snaphots are not used here.

    All sorted now, thanks.

    qh

    [font="Tahoma"]Who looks outside, dreams; who looks inside, awakes. – Carl Jung.[/font]

Viewing 5 posts - 1 through 4 (of 4 total)

You must be logged in to reply to this topic. Login to reply