Always On - SQL Server 2012 Enterprise edition - Does Requires to pay for an extra license?

Question

Always On - SQL Server 2012 Enterprise edition - Does Requires to pay for an extra license?

APA0876

Hall of Fame

Points: 3853
More actions
February 10, 2015 at 9:14 am

#315088

Hi,
I have three questions:
Always On - SQL Server 2012 Enterprise edition -
1- Does Requires to pay for an extra license of SQL Server Enterprise Edition?
I have a SQL Server Failover Cluster with two nodes and we want to implement Always ON
2- Will this AO failover be transparent for the application?
3-I have read that when Always On is implemented over a Windows Failover Cluster the Always On Failover can not be automatic. Is this a manual process on this scenario
Thanks for your help

Viewing 12 posts - 1 through 11 (of 11 total)

You must be logged in to reply to this topic. Login to reply

Perry Whittle SSC Guru Points: 234013 More actions · Answer 1

APA0876 (2/10/2015)
Always On - SQL Server 2012 Enterprise edition -
1- Does Requires to pay for an extra license of SQL Server Enterprise Edition?

Licensing is a grey area, download the latest guide from the Microsoft site for further info. It does actually state that replicas that are not serving readable\accessible copies of the database are subject to "relaxed licensing".

APA0876 (2/10/2015)
I have a SQL Server Failover Cluster with two nodes and we want to implement Always ON
2- Will this AO failover be transparent for the application?

AlwaysOn groups or AlwaysOn failover cluster instances?

The terminology has changed and can get confusing, what do you intend to deploy. As you mention Enterprise edition required I'm assuming AlwaysOn groups.

APA0876 (2/10/2015)
3-I have read that when Always On is implemented over a Windows Failover Cluster the Always On Failover can not be automatic. Is this a manual process on this scenario
Thanks for your help

That's not correct, When an AlwaysOn group configuration uses a SQL Server Failover cluster instance as a replica, automatic failover is not available 😉

-----------------------------------------------------------------------------------------------------------

"Ya can't make an omelette without breaking just a few eggs" 😉

TheSQLGuru SSC Guru Points: 134017 More actions · Answer 2

WHY do you want to implement Always On?? There are SOOO many ways you can mess that up, and lots of caveats, provisos, limitations, gotchas, etc. I have seen WAY more people/clients screw things up than get them 100% correct.

Best,
Kevin G. Boles
SQL Server Consultant
SQL MVP 2007-2012
TheSQLGuru on googles mail service

APA0876 Hall of Fame Points: 3853 More actions · Answer 3

HI I'm new on this, but the business wants to implement it as a DR alternative:

Please can you guide me? or provide me a reference where I can find steps to configure this.

Scenario- we want to implement this for Disaster Recovery (In case of a SAN failure)

Microsoft SQL Server 2012 - 11.0.5058.0 (X64) Enterprise Edition (64-bit) on Windows NT 6.1 <X64> (Build 7601: Service Pack 1) (Hypervisor)

SQL Server Failover Cluster (2 nodes) Active – Passive (Hosting the Analytical Database) – hosted over SAN Storage

Microsoft SQL Server 2012 - 11.0.5058.0 (X64) Standard Edition (64-bit) on Windows NT 6.1 <X64> (Build 7601: Service Pack 1) (Hypervisor)

SQL Server Failover Cluster (2 nodes) Active-Passive (Hosting the CMS database) a dependency of the reporting solution hosted over SAN Storage

I know that there are two technologies:

?Instance-level High Availability with AlwaysOn failover cluster instances

?Database-level High Availability with AlwaysOn Availability groups

Questions

1-Based on the purpose DR what will be SQL Server License associated cost (regular cost or special cost)

2-In our case what will be the best option or which one better applies? Instance level High Availability or Database Level High Availability?

3-Do I have to re-point all my applications to the new Availability Group Listener instead of my current Failover SQL Server Instance

What will be an overview of the steps for this kind of setup:

1-Request the Extra Storage capacity on a Different SAN to allocate the Drives for the Databases

2-Request a Windows Server to install the Standalone SQL Server that will be hosting the Secondary Replica (SQL SERVER Enterprise Extra license ???)

4-Request an IP address for creating the availability group listener (network name)

5-Enable the AlwaysOn Feature on the two Nodes of the Failover Cluster Nodes and on the Standalone Instance

6-Configure the AlwaysOn (steps pending – defining the Availability Groups ??? etc) how will be this on a Cluster environment.

7-Decide the Availability Modes (Asynchronous preferably)

8-Decide the Type of Failover (Manual Failover on our scenario as the SQL Server Failover Cluster does not support automatic Failover)

9-Redirect Applications to the new network name of the Availability Group Listener

Thanks

TheSQLGuru SSC Guru Points: 134017 More actions · Answer 4

Sorry, I cannot guide you on this. It is WAY beyond a forum post, and in fact would be tens of hours of consulting at a minimum. Honestly this stuff has been screwed up by SOOO many people and clients, and there are SOOO many ways you can go wrong or suboptimally and things you just gotta do right to have success that I can only VERY STRONGLY advise you to get professional help in this endeavor.

Remember, this is DR so if it isn't done correctly you could quite literally cause the company to go out of business!!

Best,
Kevin G. Boles
SQL Server Consultant
SQL MVP 2007-2012
TheSQLGuru on googles mail service

JeepHound SSCrazy Points: 2963 More actions · Answer 5

First thing to understand is that "Always On" is a marketing term. Based on what you stated in the original post and what you described as "Instance-level High Availability with AlwaysOn failover cluster instances" This is what you already have implemented (SQL Server Failover Clustering Instance). In SQL2012 they changed the terminology to include the "Always On" branding.

At the same time they released "Always On" Availability Groups or what you describe as: "Instance level High Availability with AlwaysOn Availability groups" and this provides something different.

Can you give us more background on your company and the reasons you might have for going to Availability groups? We know too little about the situation to make even a guess as to what might be the best for you.

APA0876 Hall of Fame Points: 3853 More actions · Answer 6

Hi,

The purpose of implementing AlwaysOn is to keep the services available in case of a SAN failure.

Even the SAN is redundant, they want to be prepared for an unforeseen situation with the SAN.

Even we are reading a lot of documentation we would like to validate the overall steps or at a reference to follow with some one that have configured this:

Thanks (We appreciate your help), that's the purpose of the Forum!

Perry Whittle SSC Guru Points: 234013 More actions · Answer 7

APA0876 (2/11/2015)
HI I'm new on this, but the business wants to implement it as a DR alternative:
Please can you guide me? or provide me a reference where I can find steps to configure this.
Scenario- we want to implement this for Disaster Recovery (In case of a SAN failure)
Microsoft SQL Server 2012 - 11.0.5058.0 (X64) Enterprise Edition (64-bit) on Windows NT 6.1 <X64> (Build 7601: Service Pack 1) (Hypervisor)
SQL Server Failover Cluster (2 nodes) Active – Passive (Hosting the Analytical Database) – hosted over SAN Storage
Microsoft SQL Server 2012 - 11.0.5058.0 (X64) Standard Edition (64-bit) on Windows NT 6.1 <X64> (Build 7601: Service Pack 1) (Hypervisor)
SQL Server Failover Cluster (2 nodes) Active-Passive (Hosting the CMS database) a dependency of the reporting solution hosted over SAN Storage
I know that there are two technologies:
?Instance-level High Availability with AlwaysOn failover cluster instances
?Database-level High Availability with AlwaysOn Availability groups
Questions
1-Based on the purpose DR what will be SQL Server License associated cost (regular cost or special cost)
2-In our case what will be the best option or which one better applies? Instance level High Availability or Database Level High Availability?
3-Do I have to re-point all my applications to the new Availability Group Listener instead of my current Failover SQL Server Instance
What will be an overview of the steps for this kind of setup:
1-Request the Extra Storage capacity on a Different SAN to allocate the Drives for the Databases
2-Request a Windows Server to install the Standalone SQL Server that will be hosting the Secondary Replica (SQL SERVER Enterprise Extra license ???)
4-Request an IP address for creating the availability group listener (network name)
5-Enable the AlwaysOn Feature on the two Nodes of the Failover Cluster Nodes and on the Standalone Instance
6-Configure the AlwaysOn (steps pending – defining the Availability Groups ??? etc) how will be this on a Cluster environment.
7-Decide the Availability Modes (Asynchronous preferably)
8-Decide the Type of Failover (Manual Failover on our scenario as the SQL Server Failover Cluster does not support automatic Failover)
9-Redirect Applications to the new network name of the Availability Group Listener
Thanks

As Kevin has said this really goes beyond a forum post. However, I do have a stairway series to AlwaysOn available on this site starting at this link[/url]

JeepHound (2/11/2015)
At the same time they released "Always On" Availability Groups or what you describe as: "Instance level High Availability with AlwaysOn Availability groups" and this provides something different.

AlwaysOn Availability groups provide database level redundancy not instance level

-----------------------------------------------------------------------------------------------------------

"Ya can't make an omelette without breaking just a few eggs" 😉

EdVassie SSC Guru Points: 60476 More actions · Answer 8

Going back to the OP, this is what I understand about SQL licencing after many hours of negotiation with resellers and Microsoft. However it is only a guide, you absolutely need to speak to your licence reseller, tell them exactly what you want to do, do not assume they will get the right answer for anything you miss out, and you will probably get a variety of ways and prices for how the licencing can be done.

* All of the capabilities of Always On are included in the Enterprise Edition licence

* If you are using legacy SQL Server clustering, the passive node in a two-node cluster does not need an additional SQL licence. If you have three or more nodes in the SQL Cluster then each additional node above two needs its own licencing

* If you use availability groups, each node in the availability group needs its own SQL licencing

I think it is best to look at the business problem and work out what are the best technical solutions, and only after this look at license implications. If you have a solution that has a good business justification then the money will be found for the licencing. If you aim only for a solution with what you think are low licence costs then you may not meet the needs of the business.

Original author: https://github.com/SQL-FineBuild/Common/wiki/ 1-click install and best practice configuration of SQL Server 2019, 2017 2016, 2014, 2012, 2008 R2, 2008 and 2005.

When I give food to the poor they call me a saint. When I ask why they are poor they call me a communist - Archbishop Hélder Câmara

APA0876 Hall of Fame Points: 3853 More actions · Answer 9

Perry Whittle (2/12/2015)
APA0876 (2/11/2015)
HI I'm new on this, but the business wants to implement it as a DR alternative:
Please can you guide me? or provide me a reference where I can find steps to configure this.
Scenario- we want to implement this for Disaster Recovery (In case of a SAN failure)
Microsoft SQL Server 2012 - 11.0.5058.0 (X64) Enterprise Edition (64-bit) on Windows NT 6.1 <X64> (Build 7601: Service Pack 1) (Hypervisor)
SQL Server Failover Cluster (2 nodes) Active – Passive (Hosting the Analytical Database) – hosted over SAN Storage
Microsoft SQL Server 2012 - 11.0.5058.0 (X64) Standard Edition (64-bit) on Windows NT 6.1 <X64> (Build 7601: Service Pack 1) (Hypervisor)
SQL Server Failover Cluster (2 nodes) Active-Passive (Hosting the CMS database) a dependency of the reporting solution hosted over SAN Storage
I know that there are two technologies:
?Instance-level High Availability with AlwaysOn failover cluster instances
?Database-level High Availability with AlwaysOn Availability groups
Questions
1-Based on the purpose DR what will be SQL Server License associated cost (regular cost or special cost)
2-In our case what will be the best option or which one better applies? Instance level High Availability or Database Level High Availability?
3-Do I have to re-point all my applications to the new Availability Group Listener instead of my current Failover SQL Server Instance
What will be an overview of the steps for this kind of setup:
1-Request the Extra Storage capacity on a Different SAN to allocate the Drives for the Databases
2-Request a Windows Server to install the Standalone SQL Server that will be hosting the Secondary Replica (SQL SERVER Enterprise Extra license ???)
4-Request an IP address for creating the availability group listener (network name)
5-Enable the AlwaysOn Feature on the two Nodes of the Failover Cluster Nodes and on the Standalone Instance
6-Configure the AlwaysOn (steps pending – defining the Availability Groups ??? etc) how will be this on a Cluster environment.
7-Decide the Availability Modes (Asynchronous preferably)
8-Decide the Type of Failover (Manual Failover on our scenario as the SQL Server Failover Cluster does not support automatic Failover)
9-Redirect Applications to the new network name of the Availability Group Listener
Thanks
As Kevin has said this really goes beyond a forum post. However, I do have a stairway series to AlwaysOn available on this site starting at this link[/url]
JeepHound (2/11/2015)
At the same time they released "Always On" Availability Groups or what you describe as: "Instance level High Availability with AlwaysOn Availability groups" and this provides something different.
AlwaysOn Availability groups provide database level redundancy not instance level

Hi I don't think this goes beyond a forum Post: the scenario is well defined -

A Failover Cluster Instance (SQL Server 2012) two nodes Active/Passive - on this scenario the only weak link is the storage; the storage subsystem (SAN) becomes the Single Point of Failure and the business decided to implement AO to make the more resilient High Availability platform. (based on your link).

The initial question having that scenario: Can you just review the Overall Steps on how to implement or incorporate the Always On Availability Groups -

Steps I have gathered based on the documentation read:

- Request IP for the Group Listener

- Request Storage on a Different SAN

- Request a server to install a Stand Alone SQL to host the secondary replica (Databases) and install SQL Server Enterprise Edition (as on the nodes of the FCI)

- After deciding the Availability Modes (Synchronous or Asynchronous) -Configure the availability groups on the FCI

I'm not mentioning the details - I just want to have the overall or foundation steps clear - and I'm pretty sure that there are many members on this forum that has implemented AO and can review the Steps and provide feedback in case I'm missing something critical.

Thanks Again!

Ford Fairlane SSCertifiable Points: 7664 More actions · Answer 10

We have implemented Always on, so my answers are based on my experience.

Always On - SQL Server 2012 Enterprise edition -

1- Does Requires to pay for an extra license of SQL Server Enterprise Edition?

Enterprise is more expensive than STD

I have a SQL Server Failover Cluster with two nodes and we want to implement Always ON

2- Will this AO failover be transparent for the application?

Use listeners and yes it is pretty much transparent to the application

3-I have read that when Always On is implemented over a Windows Failover Cluster the Always On Failover can not be automatic. Is this a manual process on this scenario

You can only implement Alwayson with in a Windows Failover Cluster and it can be automatic. WHere it cant be automatic is if you have a SQL cluster built with in the Windows cluster and you try and fail over to the SQL Cluster, this needs to be a manual process.

Hope this helps...

Ford Fairlane
Rock and Roll Detective

Perry Whittle SSC Guru Points: 234013 More actions · Answer 11

APA0876 (2/16/2015)
Hi I don't think this goes beyond a forum Post: the scenario is well defined -

Despite what you may believe, this is well beyond a forum thread. Have you read my stairway series?

AlwaysOn and what it encompasses is huge, there are so many caveats and pre reqs to take into account.

APA0876 (2/16/2015)
A Failover Cluster Instance (SQL Server 2012) two nodes Active/Passive - on this scenario the only weak link is the storage;

the other weak link here is if you plan to put an Availability group on the this FCI and include other replicas, you'll need more cluster nodes.

APA0876 (2/16/2015)
Steps I have gathered based on the documentation read:
- Request IP for the Group Listener
- Request Storage on a Different SAN
- Request a server to install a Stand Alone SQL to host the secondary replica (Databases) and install SQL Server Enterprise Edition (as on the nodes of the FCI)
- After deciding the Availability Modes (Synchronous or Asynchronous) -Configure the availability groups on the FCI

Why storage on a different SAN, you can use storage area networks but alternatively for replicas that are not FCIs use fast local storage.

You'll also need to configure your readonly routing scripts and reconfigure your cluster witness to account for the new cluster node. You'll also need to carefully size your tempdb especially if using readable secondarys

-----------------------------------------------------------------------------------------------------------

"Ya can't make an omelette without breaking just a few eggs" 😉