May 17, 2016 at 11:46 am
I have configured HA always On on WSFC-1 with two nodes SQLNODE-1 and SQLNIDE-2 on sql server 2012 R2 and tested SQL failover successfully. I want to add one more node as DR site on another WSFC-2 which is on another site. When i am trying to add SQLNODE-3 which is on WSFC-2, from the primary HA SQLNODE-1, it does not find the SQLNODE-3 and says this node is not RUNNING under WSFC. Can anyone help me sharing some document with which i can follow and set this up for my DR, FAILOVER setup?
SQLNODE-1 is setup read/write, SQLNODE-2 is setup for read only. I want SQLNODE-3 as Read only but it will be on a separate cluster such that incase of disaster on primary cluster we can failover on DR and avail the feature. Please share your thoughts and any document if availabe.
May 17, 2016 at 4:03 pm
With AlwaysOn (FCIs and AGs) all nodes must be a part of the same WSFC. You can have an AG with multiple FCIs in different locations, but they still have to be a part of the same Windows cluster.
An alternative would be to have WSFCs in each location and do something like log shipping or mirroring between the clusters. But there are more moving parts there and the clusters will not know about each other - they will treat them the same as two stand-alone instances.
Joie Andrew
"Since 1982"
May 18, 2016 at 12:01 am
Joie Andrew (5/17/2016)
With AlwaysOn (FCIs and AGs) all nodes must be a part of the same WSFC. You can have an AG with multiple FCIs in different locations, but they still have to be a part of the same Windows cluster.
We have VM environment under SAN. In such case the windows cluster (WSFC) may fail only because of hardware failure on primary data center. The purpose of DR is to keep safe our database 24X7 available on secondary data center using Always On feature. In your comments, it seems we have to rely only on one windows cluster which does not seems to commit Always On.
An alternative would be to have WSFCs in each location and do something like log shipping or mirroring between the clusters. But there are more moving parts there and the clusters will not know about each other - they will treat them the same as two stand-alone instances.
I tried to setup mirroring on a database that is already part of AlwaysOn HA (Replica) and synchronizing with another node under HA. It does not work and says that the database is already enabled for database mirroring.
The database "TestAlwaysOn" is already enabled for database mirroring. (Microsoft SQL Server, Error: 1405)
I need to setup Always On on another windows cluster nodes, How it can be done. Please share your solution?
May 18, 2016 at 1:36 am
When you say "AlwaysOn HA" I am assuming you are meaning an availability group, correct? Both FCIs and AGs can provide HA. If you are using an AG, you can still have both nodes in different data centers. However, like I stated before all nodes in an AlwaysOn configuration (both FCI and AG) need to be a part of the same WSFC. This is non-negotiable; it is a requirement of AlwaysOn.
Joie Andrew
"Since 1982"
May 18, 2016 at 3:22 am
Yes, I got it. Always On has a limitation of all nodes need to be under one WSFC. Can you please share me a document which states steps how to implement a DR with AlwaysOn with some diagram for better understanding and presenation?
May 18, 2016 at 4:02 pm
Viewing 6 posts - 1 through 5 (of 5 total)
You must be logged in to reply to this topic. Login to reply