Viewing 15 posts - 7,816 through 7,830 (of 9,244 total)
in my opinion and experience, management who are lacking the key understandings are the real issue. They dont necessarily need to be a DBA themselves but they should have a...
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
August 6, 2010 at 6:42 am
wow this is really going off topic now, the OP only asked about cluster resources failing over from node to node 😀
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
August 6, 2010 at 6:33 am
Mike Hinds (8/6/2010)
Perry Whittle (8/6/2010)
Mike Hinds (8/5/2010)
but manually bringing either online does nothing for the other.then you do not have your dependencies set correctly!
see my attached screenshot for details
Thanks for...
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
August 6, 2010 at 5:47 am
Mike Hinds (8/5/2010)
but manually bringing either online does nothing for the other.
then you do not have your dependencies set correctly!
see my attached screenshot for details
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
August 6, 2010 at 2:19 am
Even with locked pages the OS can still ask ever so nicely for SQL to free some memory it just can't force it to as the buffer pool pages are...
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
August 5, 2010 at 1:35 pm
With locked pages in memory sql server will never release below the min memory setting. It is important to set both the min and max memory when using locked pages.
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
August 5, 2010 at 1:11 pm
Bringing the SQL Server agent online first will automatically bring SQL Server up due to the dependencies between these resources 😉
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
August 5, 2010 at 7:53 am
That's an excellent link. IMHO anybody who hooks a SQL Server system up to a storage system without completing IO tests needs their brain looked at 😀
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
August 4, 2010 at 3:09 pm
Thank you, I could possibly add the witness setup to the VMWare based doc but I don't have or use Hyper-V.
Its very easy, just configure the cluster as usual using...
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
August 4, 2010 at 2:29 pm
neither, in my opinion RAID group 2 is the weak point. I would opt for multiple smaller arrays!
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
August 4, 2010 at 2:21 pm
Robert
remember as you are implementing a cluster you will inherently have more than 2 machines on the network (a cluster is dependant upon DNS and AD).
You cluster the 2...
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
August 4, 2010 at 1:49 pm
OK, taking RAID group2 from your diagram link above you have 1 RAID 10 group carved up to service
TEMPDB data
user DB data
Backup data
Sharing storage amongst 3 LUNs will likely not...
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
August 4, 2010 at 1:43 pm
sivaprasad (7/28/2010)
Single Node cluster
SQL Server 2005 with default instance
O.S: Windows 2003 with KB937444, Microsoft .NET Framework 3.5 Service Pack 1 & Windows Installer 4.5
The upgrade of SQL 2005 cluster...
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
August 4, 2010 at 12:04 pm
Jamie Wormald (8/3/2010)
The way I understand it only one server if live at any one time and this server is the one with access to the quorum drive.
is this...
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
August 4, 2010 at 12:02 pm
for an OLTP system shared storage is a bad idea!
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
August 4, 2010 at 12:00 pm
Viewing 15 posts - 7,816 through 7,830 (of 9,244 total)