HumanScaleDev
asked on
SMB transfers causing issues on Server 2012/SQL 2012 cluster
We recently upgraded our web infrastructure to Server 2012, SQL 2012 and Cold Fusion 10. Our old set-up was Server 2008, SQL 2008 and Cold Fusion 10. We have FusionReactor configured so we can actively monitor web requests and SNMP monitoring on everything involved. In our new set-up we introduced a DR environment and set-up log-shipping in SQL which is where we started to see our issue. Log-shipping and simple drag and drop SMB file transfers cause web requests to hang. The users will get page timeouts and we can see queue of web requests in FusionReactor build up. We don't see any resource exhaustion in the SNMP monitoring. If we use FTP rather than SMB the issue doesn't happen. Also, this issue seems to be related to load in some way. During slow times we can complete an SMB transfer with no issue. Any help in troubleshooting would be greatly appreciated.
Are you running SQL and ColdFusion on the same server? SQL by default will try to use 100% of the CPU and RAM unless configured to do otherwise. Set some system resource alerts or monitor the RAM, CPU, and NETWORK limits while loading testing.
ASKER
SQL and ColdFusion are on separate boxes. We have alerts monitoring resources on both of them and don't see any issues with available resources. We tested dropping the RAM allocated to SQL but the issue still happened.
Are you log-shipping to a remote server? How far is the server? You may need a dedicated network only for that.
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.