In my company, we have a central database and 4 remote databases. the
central database replicate same database into 4 remote database in parallel.
the replication is transaction replication and the data is input in the
central database and replicated into remote databases. there are some
triggers in the remote database that retrieve data that was replicated from
central database into other table. the design is like this. it is wierd that
one of remote database is very slow and replication locking the table and
client application is blocked out. the client application is hang. the
replication is very slow, 4 rows/minute. other 3 remote db are ok. I don't
know why only one remote db is ofen happened such thing. The other remote
databases were no problem, and the replication doesn't lock the table. I know
the triggers are bad, especially work with replication. But why other remote
db were ok. I don't know if there are some other problem. can dbcc dbreindex
be help? Thanks.
I would make sure indexes on the keys exist. Replication is not magic
either, you can monitor it just like normal database activity. I would
suggest running the profiler. Use the duration template, but filter
for say >2000ms.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment