SQL Server supports Windows Authentication and mixed-mode. Mixed-mode allows you to use both Windows Authentication and SQL Server Authentication to log into your SQL Server. It’s important to note that if you use Windows Authentication, you will not be able to log in as sa.
Yes it is possible to have various configurations in a Replication environment.
No, Transparent Data Encryption (TDE) does not encrypt the data during transfer over a communication channel.
Failover Clustering, Database Mirroring, Log Shipping and Replication are the High-Availability features available in SQL Server. I would recommend reading this blog of mine which explains the differences between these 4 features.
In a clustered index, the leaf level pages are the actual data pages of the table. When a clustered index is created on a table, the data pages are arranged accordingly based on the clustered index key. There can only be one Clustered index on a table.
In a Non-Clustered index, the leaf level pages does not contain data pages instead it contains pointers to the data pages. There can multiple non-clustered indexes on a single table.
Checkpoints, whether scheduled or manually executed, cause the transaction log to be truncated up to the beginning of the oldest open transaction (the active portion of the log). That is, the dirty pages from the buffer cache are written to disk. Storing committed transactions in the cache provides a performance gain for SQL Server. However, you do not want the transaction log to get too big because it might consume too many resources and, should your database fail, take too long to process to recover the database.
One important thing to note here is that SQL Server can only truncate up to the oldest open transaction. Therefore, if you are not seeing the expected relief from a checkpoint, it could very well be that someone forgot to commit or rollback their transaction. It is very important to finalize all transactions as soon as possible.
In-place upgrade and Side-by-Side Upgrade.
Fill Factor is a setting that is applicable to Indexes in SQL Server. The fill factor value determines how much data is written to an index page when it is created / rebuilt.
Introduced in SQL Server 2008 Transparent Data Encryption (TDE) is a mechanism through which you can protect the SQL Server Database files from unauthorized access through encryption. Also, TDE can protect the database backups of the instance on which TDE was setup.
There are quite a lot of changes and enhancements in SQL Server 200@Few of them are listed here :
Support for Analysis Services on a a Failover Cluster.
The easiest way to find and change the default fill factor value is from Management Studio, right-click the SQL Server and choose properties. In the Server Properties, choose Database Settings, you should see the default fill factor value in the top section. You can change to a desired value there and click OK to save the changes. The other option of viewing and changing this value is using
GRANT, DENY and REVOKE.
Virtual network name for the SQL Server, Virtual IP address for SQL Server, IP addresses for the Public Network and Private Network(also referred as Hearbeat) for each node in the failover cluster, shared drives for SQL Server Data and Log files, Quorum Disk and MSDTC Disk.
SQL Server has 2 Authentication modes: Windows Authentication and SQL Server and Windows Authentication mode also referred as Mixed Mode.
In the Cluster Administrator, rick click on the SQL Server Group and from the popup menu item choose Take Offline.
From Start -> Run and type CluAdmin (case insensitive) and the Cluster Administrator console is displayed OR you can also go to Start -> All programs -> Administrative Tools -> Cluster Administrator.
There are 3 recovery models available for a database. Full, Bulk-Logged and Simple are the three recovery models available.
The 3 main components in Replication are Publisher, Distributor and Subscriber. Publisher is the data source of a publication. Distributor is responsible for distributing the database objects to one or more destinations. Subscriber is the destination where the publishers data is copied / replicated.
You can set the fill factor on your indexes. This tells SQL Server how much free space to leave in the index pages when re-indexing. The performance benefit here is fewer page splits (where SQL Server has to copy rows from one index page to another to make room for an inserted row) because there is room for growth built in to the index.
System databases are the default databases that are installed when the SQL Server is installed. Basically there are 4 system databases: Master, MSDB, TempDB and Model. It is highly recommended that these databases are not modified or altered for smooth functioning of the SQL System.
A user database is a database that we create to store data and start working with the data.
I would go to the SQL Server Configuration Manager.In the left pane of the tool, I would select SQL Server Services, the right side pane displays all of the SQL Server Services / components that are installed on that machine. If the Service is displayed as (MSSQLSERVER), then it indicates it is a default instance, else there will be the Instance name displayed.
Row compression and Page compression.
The simplest answer to this is “Clustered and Non-Clustered Indexes”. There are other types of Indexes what can be mentioned such as Unique, XML, Spatial and Filtered Indexes. More on these Indexes later.
Primarily, recovery model is chosen keeping in view the amount of data loss one can afford to. If one expects to have minimal or no data loss, choosing the Full recovery model is a good choice. Depending on the recovery model of a database, the behavior of database log file changes. I would recommend you read more material on log backups and log file behavior and so on to understand in depth.
DCL stands for Data Control Language.
By default the fill factor value is set to 0.
Update Statistics is used to force a recalculation of query optimization statistics for a table or indexed view. Query optimization statistics are automatically recomputed, but in some cases, a query may benefit from updating those statistics more frequently. Beware though that re-computing the query statistics causes queries to be recompiled. This may or may not negate all performance gains you might have achieved by calling update statistics. In fact, it could have a negative impact on performance depending on the characteristics of the system.
Inside SSMS, in Object explorer under SQL Server Agent look for Job Activity Monitor. The job activity monitor displays the current status of all the jobs on the instance. Choose the particular job which failed, right click and choose view history from the drop down menu. The execution history of the job is displayed and you may choose the execution time (if the job failed multiple times during the same day). There would information such as the time it took to execute that Job and details about the error occurred.
This is a generic question often asked by many interviewers. Explain what are the different SQL Server Versions you have worked on, what kind of administration of those instances has been done by you. Your role and responsibilities carried out in your earlier projects that would be of significance to the potential employer. This is the answer that lets the interviewer know how suitable are you for the position to which you are being interviewed.
Setting an autogrowth in multiples of MB is a better option than setting autogrowth in percentage (%).
The answer would be depending on the versions you have worked on, I would say I have experience working in SQL Server 7, SQL Server 2000, 2005 and 200@If you have worked only the some version be honest in saying that, remember, no one would be working on all versions, it varies from individual to individual.
A correlated sub-query is a nested query that is linked to the outer query. For instance, say I wanted to find all the employees who have not entered their time for the week. I could query the Employee table to get their first and last name, but I need to look at the TimeEntry table to see if they’ve entered their time or not. I can’t do a straight join here because I’m looking for the absence of time data, so I’ll do a correlated sub-query similar to this:
SELECT FirstName, LastName
FROM EMPLOYEE e
WHERE NOT EXISTS (SELECT 1 FROM TimeEntry te
WHERE te.EmpID = e.EmpID
AND te.WeekID = 35)
Notice that the inner query relates to the outer query on the employee ID, thus making it a correlated sub-query. The inner query will be evaluated once per outer query row.
There are basically 3 types of replication: Snapshot, Transactional and Merge Replication. The type of Replication you choose, depends on the requirements and/or the goals one is trying to achieve.For example Snapshot Replication is useful only when the data inside the tables does not change frequently and the amount of data is not too large, such as a monthly summary table or a product list table etc. Transactional Replication would useful when maintaining a copy of a transactional table such as sales order tables etc. Merge Replication is more useful in case of remote / distributed systems where the data flow can be from multiple sites, for example sales done at a promotional events which might not be connected to the central servers always.
Database Mirroring runs in 2 operating modes High-Safety Mode and High-Performance Mode.
DBCC statements are Database Console Commands and come in four flavors:
Maintenance, Informational, Validation, and Miscellaneous. Maintenance commands are those commands that allow the DBA to perform maintenance activities on the database such as shrinking a file. Informational commands provide feedback regarding the database such as providing information about the procedure cache. Validation commands include commands that validate the database such as the ever-popular CHECKDB. Finally, miscellaneous commands are those that obviously don’t fit in the other three categories. This includes statements like DBCC HELP, which provides the syntax for a given DBCC command.
Drop database deletes the database along with the physical files, it is not possible to bring back the database unless you have a backup of the database. When you take a database offline, you the database is not available for users, it is not deleted physically, it can be brought back online.
Replication can be configured in any topology depending keeping in view of the complexity and the workload of the entire Replication. It can be any of the following:
SQL profiler is the SQL Server utility you can use to trace the traffic on the SQL Server instance. Traces can be filtered to narrow down the transactions that are captured and reducing the overhead incurred for the trace. The trace files can be searched, saved off, and even replayed to facilitate troubleshooting.
Replication is a feature in SQL Server that helps us publish database objects and data and copy (replicate) it to one or more destinations. It is often considered as one of the High-Availability options. One of the advantages with Replication is that it can be configured on databases which are in simple recovery model.
SQL Agent is the job scheduling mechanism in SQL Server. Jobs can be scheduled to run at a set time or when a specific event occurs. Jobs can also be executed on demand. SQL Agent is most often used to schedule administrative jobs such as backups.
No you cannot have a Distributor on a previous version than the Publisher.
On SQL Server 2005, installing SQL Server failover cluster is a single step process whereas on SQL Server 2008 or above it is a multi-step process. That is, in SQL Server 2005, the Installation process itself installs on all of the nodes (be it 2 nodes or 3 nodes). In 2008 or above this has changed, we would need to install separately on all the nodes. 2 times if it is a 2 node cluster or 3 times in a 3 node cluster and so on.
The model database, as its name implies, serves as the model (or template) for all databases created on the same instance. If the model database is modified, all subsequent databases created on that instance will pick up those changes, but earlier created databases will not. Note that TEMPDB is also created from model every time SQL Server starts up.