Taiyeb Zakir Microsoft SQL Server Escalation Support Services When replicating identity columns https://docs.microsoft.com/en-us/sql/relational-databases/replication/publish/replicate-identity-columns and using Auto Identity management, you need to make sure ranges are defined properly. The default “ranges” for automatic identity range management, while good 20 years ago, may need to be tuned for today’s higher workloads.
The Distribution database is a cache of undistributed commands that the Distribution SQL Agent jobs are responsible to deliver to their subscribers. This cache is populated by the Publisher LogReader Agent job as new data changes are detected. The primary tables holding the cache of undistributed data is MSrepl_transactions and MSrepl_commands.
Problem of the day: Transactional Replication was configured on a database in AlwaysOn Availability Group. When the group was failed over to synchronized secondary node Transactional Log Reader stopped working.
Customer asked: “We need to reload a table weekly\nightly, batch job takes few minutes, but Replication takes long time to move the batch changes to Subscriber, how can we speed this up”? If Subscriber is mirror of Publisher, perhaps Replication is the wrong technology! A better solution may be to configure the database for AlwaysOn
Read More »
Every write an update statement touching +50 million commands in a single transaction? Ever on a table being Published in Transaction Replication? How about to a subscriber half way around the world or on a satellite connected network? While nothing in Replication design prevents, there may be challenges that should be considered when making large
Read More »