1

Topic: AlwaysOn on Windows Server Failover Cluster and it listener: understanding of the concept

Hello,
One of my teachers told to me that it is possible to state about understanding of the concept, only if can explain it to everybody, and he understands.
And so, I in understanding of technology AlwaysOn have one essential gap. Listener
In the beginning all like is clear:
1. For an equipment failure at physical level answers Windows Cluster
2. For a failure of service SQL AlwaysOn
3. All copies Sql Server for usage AlwaysOn should be installed on nodes of one cluster (not  the domain)
4. Group AlwaysOn creation
4. Further: creation Listner.
Here at me a gap. What is in effect listener? Under the documentation it "network connectivity endpoint". On determination endpoint this device. This physical device? Or record in ActiveDirectory forms automatically at creation listener and IP (statico) does not correspond to any real "iron" (that is any free address is entered)? That is, while server ActiveDirectory is live is live and listener?
Thus, whether will correctly tell, what at an iron failure there will be an automatic switching on the following server and AlwaysOn in case of automatic switching makes a secondary remark of active?
And if IP listener corresponds or is  primary server SQL on which iron refused? What happens? Windows produces switching on the following node, and AlwaysOn? Whether it will be accessible? Whether possibility will activate manually a secondary remark?
I will be grateful for answers and corrections

2

Re: AlwaysOn on Windows Server Failover Cluster and it listener: understanding of the concept

Eugenia;
2. For a failure of service SQL AlwaysOn - is not present, for a failure answers cluster services RHS, which  periodically service SQL
What is in effect listener? Is an object  services AlwaysOn to which parameters, such as objects in DNS, IP are anchored

wrote:

Thus, whether will correctly tell, what at an iron failure there will be an automatic switching on the following server and AlwaysOn in case of automatic switching makes a secondary remark of active?

- not always if at AlwaysOn it is necessary to check Healt databases then yes passes Failover, from iron  Failover if iron is indirectly connected to cluster service.
If you had disk or a grid which do not influence in any way cluster services Failover will not be.