August 24, 2017 at 7:12 am
Does SQLServer 2012 SE support clustering between an A-P cluster (SE) and another A-P cluster(SE). I know that SE support only A-P setup when it comes to clustering but with disk being a single point of failure I was just wondering if it would allow us to create a cluster between 2 clusters and automatically failover to another cluster when a disk goes offline in one cluster.
Thanks in Advance.
“If your actions inspire others to dream more, learn more, do more and become more, you are a leader.” -- John Quincy Adams
August 24, 2017 at 8:04 am
sasken - Thursday, August 24, 2017 7:12 AMDoes SQLServer 2012 SE support clustering between an A-P cluster (SE) and another A-P cluster(SE). I know that SE support only A-P setup when it comes to clustering but with disk being a single point of failure I was just wondering if it would allow us to create a cluster between 2 clusters and automatically failover to another cluster when a disk goes offline in one cluster.Thanks in Advance.
with std edition, each instance may have only 2 failover partners, so you can install and run 2 separate instances
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
August 24, 2017 at 8:41 am
OK Thanks. So a disk failure in this case should perform an automatic failover as the instances are different and so are the SANs. correct?
“If your actions inspire others to dream more, learn more, do more and become more, you are a leader.” -- John Quincy Adams
August 24, 2017 at 9:04 am
sasken - Thursday, August 24, 2017 8:41 AMOK Thanks. So a disk failure in this case should perform an automatic failover as the instances are different and so are the SANs. correct?
if you have muliptle LUNs from different lovcations per instance then effectively yes, but i have no idea on the configuration of your system.
I'm relying on what you are telling me
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
August 25, 2017 at 5:30 am
May be reading this the wrong way but hopefully I got this right.
Sounds to me like you have two clusters already (say Clust1 with Srv1 and Srv2, Clust2 with Srv3 and Srv4) you have SQL installed on both clusters so say Clust1\SQL and Clust2\SQL. You want to cluster the two existing clusters to prevent a disk failure at the cluster level.
From that you can't cluster an existing cluster and create a giant cluster (well not to my knowledge anyway).
You might be best splitting the cluster disks out from shared cluster disks to stand alone disk and going down the Always On route or something like Log Shipping/Replication, or even mirroring (appreciate is deprecated) but will give you the automatic failover you need between 2 separate windows clusters.
August 25, 2017 at 6:53 am
anthony.green - Friday, August 25, 2017 5:30 AMMay be reading this the wrong way but hopefully I got this right.Sounds to me like you have two clusters already (say Clust1 with Srv1 and Srv2, Clust2 with Srv3 and Srv4) you have SQL installed on both clusters so say Clust1\SQL and Clust2\SQL. You want to cluster the two existing clusters to prevent a disk failure at the cluster level.
From that you can't cluster an existing cluster and create a giant cluster (well not to my knowledge anyway).
You might be best splitting the cluster disks out from shared cluster disks to stand alone disk and going down the Always On route or something like Log Shipping/Replication, or even mirroring (appreciate is deprecated) but will give you the automatic failover you need between 2 separate windows clusters.
Thanks for the reply. Dont have any clusters ready yet. But I was thinking of the giant cluster way. So is there a way to prevent disk as a single point of failure and rely on automatic failover in Standard Edition of SQLServer 2012 (no manual intervention)?
“If your actions inspire others to dream more, learn more, do more and become more, you are a leader.” -- John Quincy Adams
August 25, 2017 at 6:54 am
Perry Whittle - Thursday, August 24, 2017 9:04 AMsasken - Thursday, August 24, 2017 8:41 AMOK Thanks. So a disk failure in this case should perform an automatic failover as the instances are different and so are the SANs. correct?if you have muliptle LUNs from different lovcations per instance then effectively yes, but i have no idea on the configuration of your system.
I'm relying on what you are telling me
so how does data get replicated from SAN1 to SAN2?
“If your actions inspire others to dream more, learn more, do more and become more, you are a leader.” -- John Quincy Adams
Viewing 7 posts - 1 through 6 (of 6 total)
You must be logged in to reply to this topic. Login to reply