Re: Using sql server failover clustering

You're limited by the OS when it comes to memory:

http://msdn2.microsoft.com/en-us/library/ms143685.aspx

Re where the DB is, consider the following:

<local disk>-<NODE1>-<shared disk>-<NODE2>-<local disk>

This says that NODE1 has a local disk and NODE2 has a local disk. Those
disks are independent of each other. IOW, NODE1 can write to its local disk
and NODE2 does not see it, and vice versa . Now, in the middle, you have a
shared disk. For a clustered instance of SQL Server, this is the only place
where a DB can be stored. It is accessible by both nodes, *but not at the
same time*. In the above scenario, the local disks would be, say, C:. The
binaries - i.e. SQL Server executables - would go on the C: drive of each of
the two nodes. The data would be on the *shared* drive; let's call it X:.
Whether you call it X: as viewed by NODE1 or X: as viewed by NODE2, it is
the *same* X: drive, unlike the C: drives.

In the above, each node can access its own C: drive all the time. However,
only the node that currently owns the X: drive can access that drive.
During a failover, ownership changes to the other node.

Check out www.clusterhelp.com. I took the course and am very grateful that
I did.

HTH


--
Tom

----------------------------------------------------
Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
SQL Server MVP
Toronto, ON Canada
https://mvp.support.microsoft.com/profile/Tom.Moreau


"Roy Goldhammer" <roy@xxxxxxxxxxx> wrote in message
news:uikfhKf0HHA.1164@xxxxxxxxxxxxxxxxxxxxxxx
Questions for you tom about it below:
"Tom Moreau" <tom@xxxxxxxxxxxxxxxxxxxx> wrote in message
news:%23vktche0HHA.4476@xxxxxxxxxxxxxxxxxxxxxxx
Answers below:

--
Tom

----------------------------------------------------
Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
SQL Server MVP
Toronto, ON Canada
https://mvp.support.microsoft.com/profile/Tom.Moreau


"Roy Goldhammer" <roy@xxxxxxxxxxx> wrote in message
news:OkyPzXe0HHA.464@xxxxxxxxxxxxxxxxxxxxxxx
Whell Hilary:

I convince my boss to work on sql server failover clustering for better
high
avilability and performance.

For what i've seen is: and here are my questions
1. I need to create windows clustering between the nodes as windows
failover
clustering and not NLB clustering. Is it true?

Correct.

2. I need sql server 2005 enterprise edition on both servers?

You install SQL Server only once. The binaries are replicated to all
other
nodes. You need EE if you have > 2 nodes in the cluster. You can use Std
Ed if you have only 1 or 2 nodes.
My database size is more then 20Gb and in some cases more then 100Gb. and i
need more then 2Gb ram for normal actions. Do i still can use Standart
edition?

3. Does the data which entered to the database on main node is being
transfered by MS-DTC to passive nodes?

The data is not stored in the nodes. It's stored in the DB and there is
only one copy of it.
On which server the database is on in case of failover clustering? i didn't
understand and it is very important because i need backup of the database as
well

DTC is not required unless you are doing distributed
transactions across 2 or ore instances.

4. After the main server has revived, how does it return to be primary?

An administrator manually fails back to the primary. It should not matter
anyway, since both nodes are identical.


"Hilary Cotter" <hilary.cotter@xxxxxxxxx> wrote in message
news:OASsnUd0HHA.1184@xxxxxxxxxxxxxxxxxxxxxxx
I am not sure what a shered database is. Perhaps you mean a shared
database. The databases and other resources are shared between nodes in a
cluster. At one time the primary node will "own" all the clustered
resources, and the secondary node (the inactive node) will own none. When
you failover the roles will be reversed and the former secondary/inactive
node will become the primary active node and now own all the resources.



--
Looking for a SQL Server replication book?
http://www.nwsu.com/0974973602.html

Looking for a FAQ on Indexing Services/SQL FTS
http://www.indexserverfaq.com
"Roy Goldhammer" <roy@xxxxxxxxxxx> wrote in message
news:%23lMTC8a0HHA.4652@xxxxxxxxxxxxxxxxxxxxxxx
Hello there

For my knowlage to implement sql server failover clustering i need at
least 2 servers: one as primary node. and second as secondery node

But in all diagrams i've seen Shered database. What is it? another
server?







.

Relevant Pages

  • Re: Active/Active configuration
    ... Other than a read-only shared database model for reporting, there is no provision in SQL Server for multiple SQL instances runing on different nodes to access the same data store. ... It is a holdover from SQL 7.0 clustering and does not accurately reflect the capabilities of the current product. ... >>> networks ...
    (microsoft.public.sqlserver.clustering)
  • Re: Using sql server failover clustering
    ... I convince my boss to work on sql server failover clustering for better ... Does the data which entered to the database on main node is being ...
    (microsoft.public.sqlserver.clustering)
  • Re: Clustering in Active/Active
    ... SQL Server clustering is a failover technology, ... If you want more database processing power, ...
    (microsoft.public.sqlserver.clustering)
  • Re: Question about setup
    ... Senior Database Administrator ... Microsoft SQL Server MVP ... MVP - Windows Server - Clustering ...
    (microsoft.public.sqlserver.clustering)
  • Re: Question about setup
    ... Senior Database Administrator ... Microsoft SQL Server MVP ... MVP - Windows Server - Clustering ...
    (microsoft.public.sqlserver.clustering)