July 14, 2014 at 7:01 am
I'm pretty weak on networking concepts. I'll say that up front. Hopefully I can ask my question intelligently enough for someone to understand and know the answer.
We have a series of SQL Server boxes. Each server has 2 dedicated boxes:
SEP01 runs on SEP01A and SEP01B
HR01 runs on HR01A and HR01B
There are a bunch more but I'm sure you get the point.
My question involves the drive and directory structures on the nodes. SEP01 has a database named CM and it's mdf/ldf files are defined as being in F:\DATA and F:\LOGS. If I remote to 1 node, the F drive is exactly as I expect to see. If I remote to the other node, an F drive doesn't even exist. Can someone give me the 5 cent explanation of why this works or if it's bad and I should fix it? Should I make the nodes match each other?
July 14, 2014 at 7:14 am
robin.pryor (7/14/2014)
I'm pretty weak on networking concepts. I'll say that up front. Hopefully I can ask my question intelligently enough for someone to understand and know the answer.We have a series of SQL Server boxes. Each server has 2 dedicated boxes:
SEP01 runs on SEP01A and SEP01B
HR01 runs on HR01A and HR01B
There are a bunch more but I'm sure you get the point.
My question involves the drive and directory structures on the nodes. SEP01 has a database named CM and it's mdf/ldf files are defined as being in F:\DATA and F:\LOGS. If I remote to 1 node, the F drive is exactly as I expect to see. If I remote to the other node, an F drive doesn't even exist. Can someone give me the 5 cent explanation of why this works or if it's bad and I should fix it? Should I make the nodes match each other?
What you seeing is entirely correct. What you have is your SQL server instances hosted on a 2 node failover cluster. At any given time only 1 node of your cluster is running the active SQL services. When the node is active the shared storage drives (F:\ in your case) is logically attached to the active node hence why you can see the drive in windows explorer on the active node but not on the passive node.
In the event of a failover the drives and SQL services will shutdown \ detach from the active node and attach \ start up on what was the passive node.
All storage in a failover cluster are shared storage and generally will be SAN storage rather than physical drives within the host server.
MCITP SQL 2005, MCSA SQL 2012
July 14, 2014 at 7:32 am
Excellent explanation. Thank you so much!
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply