MSSQL Interview Question Clustering- Part3

Q6 - Why is this needed to manage the max memory on failover cluster?
Ans: So if you have a two-node cluster, each currently configured with 8 GB of memory as well as two instances of SQL Server (one with 5 GB of memory and the other with 7 GB), 7 + 5 does not equal 8; it equals 12. If they both happen to run on the same node, one will probably be able to get the memory it needs and the other will not. You need to adjust the amount of memory your instances are using so that, in a failover scenario, you do not starve one instance or possibly have it not start up after failover

Q7-Why antivirus should not be run on?
Ans: If there is no reason to put antivirus software on the cluster nodes dedicated to SQL Server, very secure, and without file shares on it, do not install the software. If your corporate security policy dictates that antivirus software must be configured on all servers, you must set the filtering of the antivirus program to exclude the scanning of all SQL Server drives that contain data and log files. You do not want the scanner on another node in the event of a failover to detect the drive and prevent SQL Server from starting because the virus scanner is now scanning your database and log files. Also remember, as noted in Chapter 5, to exclude the \MSCS directory on the quorum if you are using a disk-based quorum

Q8 - How to see the drives?
Ans: Execute the following Transact-SQL query:
select * from ::fn_servershareddrives()

Q9- Fail Resources on a group level?
Ans: To prevent an automatic failback, select Prevent Failback. To allow automatic failback, select Allow Failback, and then one of the following options:

•Immediately This means that the moment Windows Clustering detects that the preferred cluster node is online, it fails back any resources. This is not advisable because it could disrupt clients and applications, especially at peak times in the business day.

•Failback Between N And N1 Hours This option allows a controlled failback to a preferred node (if it is online) during a certain period. The hours are set using numbers from 0 through 23.

Q10- If you dont want the group to fail in case of any resource failure what you shoud do?

Ans: Clear "Affect the group" For that resource.

Q11- what are the parameters for "Affact The Group"?

Ans: Threshold- where you can define the retries of failover-- No of time cluster service will restart the resources
     Period - The amount of time between the retries.
Reader of This blog also showed their interest in following blogs:
  1. SQL Server DBA “Interview Questions And Answers”
  2. SQL Azure Interview Questions and Answers Part – 1
  3. SQL Azure Interview Questions and Answers Part – 2
  4. Powershell Interview Questions and Answers
Along with technical learning I would like to share some great articles for anyone interested in the betterment of his/her family life



Popular posts from this blog

Understand Query Execution Plan- Index Scan, Index Seek And Key Lookup

MSSQL Interview Question Clustering- Part1

DBCC CheckDB The operating system returned error 21(The device is not ready.)