I have a problem with replication between 3 servers. I made something like this: server A replicate (transactional replication) to server B (to 'mydb' database), then server B replicate 'mydb' (using transactional replication) to server C. On the beginning it looks and works fine, but something wrong is going on (about 2-3 month later) and replication break up. SQL say that hi can replicate db because db is allready use to replicate. Any one had similar broblem? Mayby someone knows hot can I make alternative configuration to have similar funcionality?


I would recommend making both Subscribers (B and C) replicate directly from the original Publisher (A). This is a common scenario and should work correctly.

They only way this wouldn't work is if the replicated data is being modified on server B before replicating to C.

Edit: Not sure if replication is best solution for disaster recovery scenarios like you describe. I would consider using log shipping instead (transaction log backups taken on server A throughout the day are restored to servers B and C). I'd have to look up the details, but I would think that if you failed over to B, you could continue to do log shipping from B to C in case of a second failure.

For even higher-availibility, you would probably have to consider clustering.

  • OK. I understand. unfortunately my scenario is not a easy to replace. On server A I have system with db that should be usable anytime on server B (its my stand-by system). server C is cold reserve against B server failure. Mayby U can propose some alternative solution? – hubertus Mar 17 '10 at 22:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.