September 11, 2015 at 12:55 pm
Can someone help me out? Before we break our new servers, I want to check something.
We have 2 windows clusters ABCD-Reg and ABCD-DR with 4 servers each. They are clustered with the names ABCD-Reg-V and ABCD-DR-V, set up with DTC, the quorum drive and all appropriate disks. When we go to do the very first SQL install we get all the way to the Instance Configuration page where it asks for the SQL Server Network Name and there we get our error:
"The SQL Server failover cluster instance name 'ABCD-Reg-V' already exists as a clustered resource. Specify a different failover cluster instance name."
Google points us to a bunch of articles that say to delete the clustered resource. But the clustered resource is the windows cluster. No other resource in Failover Cluster Management has that exact name. So I'm worried if we delete it, we'll break the windows cluster. My coworker is worried that if we do something like name the SQL cluster ABCDRegV that it'll add it as a sub-item under the first node and break the virtual naming we have set up.
Can someone please advise us on what is going on with this and why SQL is doing this with 2012 when it didn't (we think) do it with 2008 or 2005?
September 11, 2015 at 1:07 pm
I did find this link that says it deletes an orphaned user (resource), but my coworker is worried that entering the below code will trash the cluster.
Cluster RES "" /Delete
Every other bit of code we're finding says to put the instance name between those double-quotes. Not this one. And I'm wondering if it'll clean up a ghost record without trashing our cluster.
The problem is, we don't have a sandbox on which to try this. So I'm really hoping someone can assist us with this.
September 11, 2015 at 2:46 pm
You need to give it a new name or destroy the cluster and recreate it with a new name. The cluster name is registered as a virtual computer name, similarly is the instance name so they cannot be the same.
You may have luck renaming the cluster, you would have to double check that DNS and AD match as well. I have never tried that.
September 13, 2015 at 12:51 pm
Brandie Tarvin (9/11/2015)
Can someone help me out? Before we break our new servers, I want to check something.We have 2 windows clusters ABCD-Reg and ABCD-DR with 4 servers each. They are clustered with the names ABCD-Reg-V and ABCD-DR-V, set up with DTC, the quorum drive and all appropriate disks. When we go to do the very first SQL install we get all the way to the Instance Configuration page where it asks for the SQL Server Network Name and there we get our error:
"The SQL Server failover cluster instance name 'ABCD-Reg-V' already exists as a clustered resource. Specify a different failover cluster instance name."
Google points us to a bunch of articles that say to delete the clustered resource. But the clustered resource is the windows cluster. No other resource in Failover Cluster Management has that exact name. So I'm worried if we delete it, we'll break the windows cluster. My coworker is worried that if we do something like name the SQL cluster ABCDRegV that it'll add it as a sub-item under the first node and break the virtual naming we have set up.
Can someone please advise us on what is going on with this and why SQL is doing this with 2012 when it didn't (we think) do it with 2008 or 2005?
The names ABCD-Reg-V and ABCD-DR-V have been assigned to the Windows server failover cluster, when creating new Failover cluster instances of sql server you will need to supply new\unique network names. So for clustered instance 1 use
ABCD-Reg-V1
ABCD-DR-V1
Read through my stairway to HA[/url] if you have any other uncertainties
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
September 14, 2015 at 4:32 am
Perry Whittle (9/13/2015)
Brandie Tarvin (9/11/2015)
Can someone help me out? Before we break our new servers, I want to check something.We have 2 windows clusters ABCD-Reg and ABCD-DR with 4 servers each. They are clustered with the names ABCD-Reg-V and ABCD-DR-V, set up with DTC, the quorum drive and all appropriate disks. When we go to do the very first SQL install we get all the way to the Instance Configuration page where it asks for the SQL Server Network Name and there we get our error:
"The SQL Server failover cluster instance name 'ABCD-Reg-V' already exists as a clustered resource. Specify a different failover cluster instance name."
Google points us to a bunch of articles that say to delete the clustered resource. But the clustered resource is the windows cluster. No other resource in Failover Cluster Management has that exact name. So I'm worried if we delete it, we'll break the windows cluster. My coworker is worried that if we do something like name the SQL cluster ABCDRegV that it'll add it as a sub-item under the first node and break the virtual naming we have set up.
Can someone please advise us on what is going on with this and why SQL is doing this with 2012 when it didn't (we think) do it with 2008 or 2005?
The names ABCD-Reg-V and ABCD-DR-V have been assigned to the Windows server failover cluster, when creating new Failover cluster instances of sql server you will need to supply new\unique network names. So for clustered instance 1 use
ABCD-Reg-V1
ABCD-DR-V1
Read through my stairway to HA[/url] if you have any other uncertainties
I'm looking at the HA article now, but to clarify... We were able to use the same names in 2008 for both the windows & sql cluster. Is this change because of HA?
September 14, 2015 at 1:39 pm
Brandie Tarvin (9/14/2015)
Perry Whittle (9/13/2015)
Brandie Tarvin (9/11/2015)
Can someone help me out? Before we break our new servers, I want to check something.We have 2 windows clusters ABCD-Reg and ABCD-DR with 4 servers each. They are clustered with the names ABCD-Reg-V and ABCD-DR-V, set up with DTC, the quorum drive and all appropriate disks. When we go to do the very first SQL install we get all the way to the Instance Configuration page where it asks for the SQL Server Network Name and there we get our error:
"The SQL Server failover cluster instance name 'ABCD-Reg-V' already exists as a clustered resource. Specify a different failover cluster instance name."
Google points us to a bunch of articles that say to delete the clustered resource. But the clustered resource is the windows cluster. No other resource in Failover Cluster Management has that exact name. So I'm worried if we delete it, we'll break the windows cluster. My coworker is worried that if we do something like name the SQL cluster ABCDRegV that it'll add it as a sub-item under the first node and break the virtual naming we have set up.
Can someone please advise us on what is going on with this and why SQL is doing this with 2012 when it didn't (we think) do it with 2008 or 2005?
The names ABCD-Reg-V and ABCD-DR-V have been assigned to the Windows server failover cluster, when creating new Failover cluster instances of sql server you will need to supply new\unique network names. So for clustered instance 1 use
ABCD-Reg-V1
ABCD-DR-V1
Read through my stairway to HA[/url] if you have any other uncertainties
I'm looking at the HA article now, but to clarify... We were able to use the same names in 2008 for both the windows & sql cluster. Is this change because of HA?
No, nothing new. You must use a separate network name and IP address for the cluster Client Access Point and each subsequent clustered instance Client Access Point.
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
September 15, 2015 at 6:25 am
Thanks, Perry. I appreciate the assistance.
September 15, 2015 at 11:13 am
So apparently I misunderstood my problem. It wasn't that I needed to use the windows cluster name for the network name, I wanted to use the instance name for the SQL Server Network Name.
And the reason it was failing? Someone in corporate created Cluster Roles using our instance names and assigned the cluster IPs to each of those roles. That made installation hang and fail with all sorts of weird messages. Took me several days before I figured out that I needed to remove the storage from those roles, delete the roles, uninstall the partial installs of SQL, reboot, then start from scratch (SQL-wise, not OS-wise).
#headdesk
But at least I'm past the first hurdle now. The first node is officially installed.
January 18, 2017 at 11:18 am
you need to run this line
Remove-ClusterResource "ResourceNameGoesHere"
Viewing 9 posts - 1 through 8 (of 8 total)
You must be logged in to reply to this topic. Login to reply