AnsweredAssumed Answered

SLOW remote end of Cluster: cause is remote DB

Question asked by unknown-user on May 11, 2010
Latest reply on Jul 7, 2010 by unknown-user
Hi all,

I have 2 servers in a cluster running Alf 3.2 Community with 11,000 pieces of content on Fedora 12 x32 with Sun Java 1.6u16 & a mySQL DB.

Users on the server that the database is installed on have no issues, Alf runs fine and their content is synced to the remote Alf installation via replicating-content, multi-casting & clustering.

The problem is: The remote location that points to the the Database host for its Database, is so slow it has become unusable. It takes 25-40 seconds just to open a space with 40 pieces of content in it. When a user needs to quickly view a file to get a name or number, opening a piece of content is painfully slow.

I then took a copy of the host DB and used it locally on the remote server, and the speed was as it should be, almost instantaneous, but of course, no replicating can be accomplished this way.

Does anyone have any suggestions to help speed up the read of the remote server to the host DB? Is there a way to have replication work with 2 local databases?

Thanks guys.

Outcomes