Hi All,
I am in the procoess of setting up merge replicaiton on sql 2008 that invloves 3 servers. Publisher (P), Subscriber (S) and distributor (D). I have successfully created all tyhe steps witha a domain user with SQL sysadmin access on all three servers so no permission issue. When I run the snapshot I can the folder ReplData\unc\dataDB_MERGE\20091007105412 created but empty although I can see the activity in Repl Monitor but it seems to stop at one table which is kind of big about 20 million records and failed with transaction log full on publisher. I am wondering why transaction log is getting used to this extent that it is getting full. I do have limited space on these test vm servers but its well enough for databases invloved. the database size is 17GB and 5GB log and after snapshot failed, i can see it tried to take all space on the drive and grew the log to 20GB and failign with space error.
Both subscirber and publisher are set for full recovery model for merge replication. Is there any setting or configuration I am missing, In sql 2000 and 2005 you could see the snapshot being created by seeing the unc folder with table and schema scripts but here it remains empty although repl monitor show sactivitiy for abotu 20 tables for 10 minutes than waiting on one table ( the big one) and failed in the end.
Any help is greatly appreciated.
Thanks!
dba60
I am in the procoess of setting up merge replicaiton on sql 2008 that invloves 3 servers. Publisher (P), Subscriber (S) and distributor (D). I have successfully created all tyhe steps witha a domain user with SQL sysadmin access on all three servers so no permission issue. When I run the snapshot I can the folder ReplData\unc\dataDB_MERGE\20091007105412 created but empty although I can see the activity in Repl Monitor but it seems to stop at one table which is kind of big about 20 million records and failed with transaction log full on publisher. I am wondering why transaction log is getting used to this extent that it is getting full. I do have limited space on these test vm servers but its well enough for databases invloved. the database size is 17GB and 5GB log and after snapshot failed, i can see it tried to take all space on the drive and grew the log to 20GB and failign with space error.
Both subscirber and publisher are set for full recovery model for merge replication. Is there any setting or configuration I am missing, In sql 2000 and 2005 you could see the snapshot being created by seeing the unc folder with table and schema scripts but here it remains empty although repl monitor show sactivitiy for abotu 20 tables for 10 minutes than waiting on one table ( the big one) and failed in the end.
Any help is greatly appreciated.
Thanks!
dba60