April 19, 2005 at 9:28 am
Has anyone come across a problem where they have seen MSrepl_commands table growing very rapidly. Our distribution cleanup job runs every 10 mins but somehow this table doesnt seem to be getting cleaned up. I narrowed it down to replication procedure sp_MSdelete_publisherdb_trans which cleans up MSrepl_commands table . Anyone has any solution to this? Any help will be greatly appreciated.
TIA
April 19, 2005 at 2:39 pm
are you sure the transactions are being delivered to all subscribers?
* Noel
April 20, 2005 at 6:56 am
Yes i am, as i checked the count(*) of rows for few tables which existed on both source and destination and they looked more or less the same. ANy other thoughts why this is happening.
May 2, 2007 at 6:12 am
Did you find a solution? I have the same problem.
May 3, 2007 at 6:11 am
Under your miscellaneous agents in the replication monitor, there is distributor clean up agent. This agent runs sp_MSdistribution_cleanup which runs sp_MSdistribution_delete which in turn runs sp_MSdelete_publisherdb_trans.
Make sure the agent has been running without error. Also make sure it is getting the correct dbid when sp_MSdistribution_delete loops through it's cursor.
May 20, 2007 at 2:25 pm
I have similar problem; the scheduled job is taking 45 min to complete though it scheduled to run for every 10 minutes. I haven't find any solution so far but now I scheduled this job to run continuously.
Bhushan
July 18, 2011 at 12:27 pm
I know this thread is super old, but in the interest of future individuals Googling and finding this, the below link may be the solution for you....
http://www.replicationanswers.com/TransactionalOptimisation.asp
Good luck...
-Patrick
Viewing 7 posts - 1 through 6 (of 6 total)
You must be logged in to reply to this topic. Login to reply