The replication agent has not logged a progress message in 10 minutes
-
Tuesday, December 04, 2012 10:02 PMI previously had a replication issue where the number of rows in msmerge_genhistory (3 million+) was causing initialization from a snapshot to fail. In order to fix this I had to increase my querytimeout from the default of 5 minutes to 30 minutes. Now a client is trying to synchronize and initializing from the snapshot. The querytimeout doesn't throw the error now but I still get an error that the agent has not logged a progress message in 10 minutes. I read somewhere that changing the heartbeat_interval on the distributor to a higher value will increase the time before an error is thrown. Will this have any adverse affects on others? There are many other databases on our server using replication that use this distributor. Is there some way to check the current hearbeat_interval? The error message is indicating 10 minutes so if I could confirm the current heartbeat_interval is set to 10 that would be nice.
All Replies
-
Tuesday, December 04, 2012 10:16 PMModerator
sp_helpdistributor_properties will return the current heartbeat_interval.- Proposed As Answer by Brandon WilliamsMicrosoft Community Contributor, Moderator Monday, December 17, 2012 12:46 AM
- Marked As Answer by Maggie LuoMicrosoft Contingent Staff, Moderator Wednesday, December 19, 2012 9:53 AM
-
Tuesday, December 04, 2012 10:35 PMModerator
The heartbeat_interval can be changed using sp_changedistributor_property.
The heartbeat_interval is the max number of minutes the agent can go without logging a progress message. Errors will still be reported regardless of the heartbeat_interval.- Proposed As Answer by Brandon WilliamsMicrosoft Community Contributor, Moderator Monday, December 17, 2012 12:47 AM
- Marked As Answer by Maggie LuoMicrosoft Contingent Staff, Moderator Wednesday, December 19, 2012 9:53 AM

