Azure Data Studio Connections and SSMS Registered Servers

Azure Data Studio and SQL Server Management Studio share a great deal of functionality, especially if you add the functionality of the ADS extensions.  While they both are great tools, very little information is passed from one tool to the other. For example, when you make a connection to a SQL Server in either tool, it will query the sysdatabases table to get a list of databases.   You will see the same list of databases in both tools.  This is something that looks like the two tools share, but in reality do not.

Now, let’s take a look at something that does get passed from Azure Data Studio to SSMS. To see this you will first need to open the Registered Servers window.  This can be found under the View menu item. It is here that you will find something that is passed from Azure Data Studio to SQL Server Management Studio.

 

Registered servers in SQL Server Management Studio can be very useful, especially if you need to run the same block of code on multiple servers.  This window does not exist in Azure Data Studio.  When viewing the Registered Servers window in SQL Server Management Studio, you will see at least two groups, Local Server Groups and Azure Data Studio.

Notice in the above image, you see several folders under both Local Server Groups and Azure Data Studio.  The items under Local Server Groups were created using SSMS.  If you right clock on Local Server Groups, you will see a context menu pop up and you can then add another group or register a server.

However, if you right click on Azure Data Studio a context menu will NOT appear.  If you want to add a group here, you will need to do so using Azure Data Studio instead.

To do this in Azure Data Studio, you will need to create a new Server Group.  When in the Connections panel, use the button highlighted below to create a new Server Group.

You do not need to restart Azure Data Studio to see the new group in the Connections pane.  However, you will need to restart SQL Server Management Studio to see the new group in the Registered Servers window.

Thanks for visiting my blog!!!

PoorSQL Formatter in SSMS

Formatting T-SQL code is very important.  There are many tools that we can use to help us.  One of my favorites is the PoorSQL Formatter.  There is a website that you can use, www.poorsql.com.  I also did a blog post a while ago on the PoorSQL extension for Azure Data Studio, you can find it here.  Now I have found out that there is also a plug in for SQL Server Management Studio.  This apparently has been around for a while. but I just became aware of it recently. The version of PoorSQL Formatter we are going to talk about in this post will work with SSMS 2014 or newer.

So what is PoorSQL Formatter?  Well, as you probably know, formatting is very important when writing T-SQL code.  While we try to format our code as we write it, sometimes the code we need to review was written by someone else and maybe doesn’t follow formatting standards. This is where PoorSQL Formatter will come in handy.  It will help us take the poorly written code and format it in a manner that is more in line with T-SQL formatting best practices.  The best part, it will do the formatting in seconds.

In order to install this extension, we will need to download it first.  To do so, go to PoorSQL,com.  Once you are there, find the link that says “SSMS/Visual Studio plugin, this will be toward the top of the page.  As in the image below.

Once you download the MSI file, double click it and install it.  Then open SSMS.  When you do, under the Tools menu, you will find two new items.  “Format T-SQL Code” and “T-SQL Formatting Options…”.  Of course the first one is what you will use to format your code.  While the second item is where the rules that the tool will follow when formatting code.

Let’s take a look at the options first. When you click the menu item, a window similar to the image below will appear.

As you can see there are a number of options.  I usually keep the defaults, which is what you see above.  I think they are pretty good default settings.

Once you have opened your poorly formatted code in SSMS, you can now use PoorSQL Formatter to fix some of the issues.  To do this, now click on the first menu item, “Format T-SQL Code”.  Once you click it, it will now format your code based on the preferences you have define.  Again, I think the defaults are pretty good.  Well, they work for me.

The more I use this tool, the more I like it.  I even have choices in what rules for formatted and what development tool I can use.  I can use it in both Azure Data Studio and SQL Server Management Studio.

Thanks for visiting my blog!!

TSQL Formatting – My Thoughts

During my SQL Server classes I am asked many times about how to format T-SQL.  If you do a Google search on How to Format T-SQL, you will get a ton of results.  There are many great posts on this topic. This post will identify some industry standards for formatting and my personal thoughts on formatting.

As you may know, formatting is very important.  If we have a production outage that is caused by a stored procedure, as the DBA you might be asked to review the code in the stored proc.  This is expected, however what might not be expected is poorly formatted code.  This poorly formatted code could lead to a longer outage because now you will need to take more time to read the code to figure out what it is doing.  This is true especially if you did not write the code or you did write the code but don’t remember it because it was a long time ago.

So here are my thoughts……

Capitalize all Keywords

While this in no way is a requirement, I believe that by doing so the code is cleaner and easier to read.  I also think the keywords jump out more when they are in caps.

select *
from production.Product
where ProductID > 350

With keywords in Caps:

SELECT *
FROM production.Product
WHERE ProductID > 350

Alias All Tables

I like to alias all tables, even if the query only uses only one table. The reason for using an alias even with one table is that if that query evolves into a query with more than one table, that initial table already has an alias and is set up for the addition of more tables.

When I use a table alias, I have two simple rules I follow.

          • It needs to be somewhat descriptive – The reason for this is straight forward.  I feel it makes it easier to determine which table all the columns are coming from.
          • All aliases should have the same number of characters – When I write T_SQL code, I find it easier to read if the dots between the alias  and the column name.  If the alias is the same length this is easier to do.

The code below has three table aliases, all different length.  To me it just seems busier and more difficult to read.

Alias All Columns

When there is a table alias, we should be using it for all columns.  When a column exists in both tables in the join, you will get error like below.

Msg 209, Level 16, State 1, Line 2
Ambiguous column name ‘ProductID’.

Using an alias on all columns, you can prevent this error.  You can also make is easier to figure out what table each column comes from.  Using the example below, it can be a bit challenging to figure out which table the ListPrice column comes from.  When looking at the tables involved in the query, it could logically come from two tables, Production.Product and Sales.SalesOrderDetail.  Because of the lack of an alias, this task becomes more difficult.

Have Vertical Lines

I like to vertically line up elements of the statements.  If you look below you will see an example of what my code will typically look like.

By doing this, for me it is easier to identify which SELECT, FROM and WHERE are part of the same statement.  As well as which ON goes with which JOIN.  I also feel that by doing this, the code is cleaner and easier to read.

Place Column Names Vertically

Over years I have had to review code that was written by others, and honestly sometimes myself, that placed the columns horizontally.  Similar to the first image below.  What I have found is that by placing the columns in this manner, it becomes more difficult to review the code.  Especially if there is a function or a CASE statement.

By doing this, it will be easier to add or remove columns from the result set.

Place the Comma before the Column Name

If you can place the comma before, of course you can also place it after the column name.  While some believe placing it after is the way to go, I have found that placing it before the column name works better for me.  While you may develop your own preference, I think the most important thing here is that you have a standard and be consistent following it.

Looking at the example below, you can see that by having the commas at the front it is a bit easier to comment out a column.  The only except would be the first column.  If you comment out the first column, there is still the comma at the front of the second line that will cause an error.

Steps in the FROM and WHERE

When working in the FROM or WHERE clauses I like to format the code in a way that resembles steps.  I like to line up the INNER or OUTER key words in the join, but on the ON keyword.

By doing this, I have found it easier to pair up the ON and the JOINs.

I also like to do something similar in the WHERE clause.  By placing the AND keywords on different lines and off setting them, again similar to stairs, I think it is easier to read what the criteria for the query is.

Derived Tables

Sub-queries are not usually at the top of my list of potential solutions, however derived tables I look at a bit differently.  Because of this, I have used them from time to time in production code.  Just like another code, they should be formatted in a way that allows it to be easily read.

Kathi Kellenberger defines a derived table in her post at Apress.com in this way:

“Derived tables are subqueries that are used in the FROM clause instead of named tables”

When writing the code for a derived table, I still try to follow all the same rules, especially since it is still a SELECT statement.

CASE Statement

When using a CASE statement, I like to line up the WHEN keywords as well as the CASE and END.  As seen below.  I feel this just makes it easier to see your options in the CASE.

Indent

I think indenting is very important for any well formatted T-SQL code.  This can easily be done by using the TAB key.  There is a setting in both SSMS and Azure Data Studio that will define the number of spaces the cursor will move each time you press TAB.  The default is 4.  I find this to be the ideal number.  I have seen code that has used 2 and for me, 4 makes it easier to see the indent.  Therefore, in my opinion the code is easier to read.

In SQL Server Management Studio, you can set the number of spaces in the Options, which can be found under that Tools menu.

While in Azure Data Studio, this same setting can be found in the Preferences, which can be found under the File menu.

UNION\UNION ALL

With these two statements, I would usually follow all the formatting guidelines for a SELECT statement.  I do like to place a line before and after the UNION or UNION ALL.  I feel this makes it easier to identify the two statements.

SELECT Name
             , City
             , State
             , ‘Customer’ AS ‘Source’
FROM Customers

UNION

SELECT Name
             , City
             , State
             , ‘Employees’
FROM Employees

Comments

Comments are critical to any well written code.  T-SQL must be self-documented. Comments is how this can be accomplished.

There are two methods you can use to comment your code.  The first is by using two dashes.  This will comment out any code that is to the right of the two dashes.  In the image below, there are two examples of comments using the dashes.

The second method is to use a set of characters, /* and */.  Any text between these will be commented out.  As you can see below.

When I really want a comment to pop out, I like to use the * to define a start and end of the comment.  I like to do this more so when the code or procedure is very long.  I think this is a great way to break a larger block of code into more readable sections.

/*****************************************************

Insert your comment code here.

*****************************************************/

If you are creating a stored procedure, it should include a “flower box”.  This is a part of the code that provides critical information about the stored procedure.

I like to flower box to include the following information

          • Procedure Name
          • Date created
          • Who created it
          • Purpose
          • List of Parameters
          • Sample code
          • Historical modifications

Below is an example:

Calling a Stored Procedure

Calling a stored procedure is usually a relatively simple piece of code to write.  Especially if there are not any parameters involved.

If no parameters are involved, this code will just be a single line.

EXEC StoredProcName

However, if there are parameters involved, you have a few options to consider.  Parameters can be called either by name or by position.  My preference here is to call by name.

When calling by position, this is how the code would look.  By looking at it, you can probably assume that the 34 is the customerID and the dates are the start and end dates for the range. I have found that assuming something gets me “unexpected results” sometimes, so I don’t like to assume.

EXEC GetSalesByCustAndDate 34, ’02/01/2020′, ’02/29/2020′

I find that calling the parameters by name works better for me.  I also like to place each parameter on a separate line with the @ lined up.  While this in no way a requirement, it just works for me.

EXEC GetSalesByCustAndDate
                 @CustID = 34,
                 @StartDate = ’02/01/2020′
                 @EndDate =  ’02/29/2020′

Tools to help format

PoorSQL Formatter – This is an extension for Azure Data Studio.  I really like this extension and it is very easy to use.  In order to utilize this, you will need Azure Data Studio and the extension is self.  When you use this extension, there are a number of settings in Azure Data Studio that you can define the rules for PoorSQL Formatter to follow. Here is a link to my blog post in the topic.

In addition to being an extension for ADS, there is also a website that you can use.  Like the extension, you can define some of the rules it will follow when formatting code.  Here is the link.

I think it is important to mentions, both the extension and the web site have a great price…..they are FREE!!!

Redgate SQL Prompt – This tool is very nice for formatting.  It allows the developer to utilize pre-defined styles or you can create your own.  SQL Prompt will be installed right into SQL Server Management Studio.  After the installation is complete, you will see a new menu item, SQL Prompt. When open, you will see a number of menu items that allow you to access the functionality.

This is a great tool!!!  While there is a cost with this tool, it is definitely worth it.

SQL Prompt can be downloaded here, https://www.red-gate.com.

Code Beautifier – This is a nice online tool. Here is the link to this tool, https://codebeautify.org/sqlformatter.  This tool is also available as an extension in Visual Studio Code.

As with many of the online formatters, there are options.  In this case, you can Beautify your code as in the image below.

Or you can “minify” your code as in the image below.

Additional Websites for formatting – these are just a few sites I found on line.  Please review them carefully and use at your own risk.

http://www.dpriver.com/pp/sqlformat.htm

https://sqlformat.org/

While following industry standards is important, it is also important just to have a standard any standard.  Hopefully, the standard you follow will have it’s roots in what considered best practice..

These are in no way an all-inclusive list.  There are other guidelines as well.  These are just the basic ones I have followed for a number of years. I in no way comes up with these, these are the guidelines that I have learned over the years from many different sources. Again, these are just my thoughts and have worked for me for quite a few years.

Thanks for visiting my blog!!!

Backup History and Information

How long a database takes to backup is something that over the years I have been asked to get.  These requests come for different reasons, sometimes it could be to find out how much it has increased over time, sometimes it could be to see if the backup job is interfering with other jobs and sometime it isn’t about duration at all, it is more about showing the backups were completed.  Over the years I have had several auditors ask for backup history.

In order to get this information, we need to pull data from two tables in the MSDB database, backupset and backupmediafamily.

Backupset

Below is how Microsoft describes that Backupset table.  This table is in the MSDB database.

“Contains a row for each backup set. A backup set contains the backup from a single, successful backup operation.”

Since most audits usually are for a date range, the start date of the backup is how we will limit the rows returned to only the ones we need.

This table also has a column named Type.  This is where you can identify the type of backup.  Usually audits, at least the ones I have been involved in, usually only care about Full and Differential backups.  Because of this we will use this column to limit the rows returned to just that two back up types.  You can adjust as needed for your environment.

According to Microsoft’s documentation, these are the backup types you might find in this table.  This can be found here.

 

Backupmediafamily

This table has many columns, however for our purposes we will only use one, physical_device_name.  This column will store the destination of the backup.

I like to include this column because if we need to do a restore, this will help us find the proper back up.  In addition, it could also help us determine if a third part tool is being used to back up the database.

In the physical_device_name column you will usually see something like this:

C:\Program Files\Microsoft SQL Server\MSSQL14.MSSQLSERVER\MSSQL\Backup\100MBDB_backup_2020_02_01_061030_4137875.trn

As you can see the above path not only identifies the destination folder, it also has the file name.

Sometimes, you will see something like below:

{FCE8C7A1-C810-4F29-A612-2D39308039E3}4

Most of the time, although not all, this is because the database if being backed up by a third-party tool such as COMMVAULT.

Here is the code that pulls the data from the two tables mentioned above. If you look at the join, you will see that the two tables are joined on the media_set_id column in both tables.

SELECT s.database_name
, m.physical_device_name ‘DestinationLocation’
, CAST(DATEDIFF(SECOND, s.backup_start_date, s.backup_finish_date) AS VARCHAR(4))
+ ‘ ‘ + ‘Seconds’ AS ‘Duration’
, s.backup_start_date
, CASE WHEN ROW_NUMBER() OVER(PARTITION BY s.database_name ORDER BY s.backup_start_date ASC) = 1
THEN 0
ELSE
DATEDIFF(DAY,LAG(s.backup_start_date, 1,0)
OVER(ORDER BY DAY(s.backup_start_date)), s.backup_start_date)
END ‘DaysSinceLastBackup’
, CASE s.[type]
WHEN ‘D’ THEN ‘Full’
WHEN ‘I’ THEN ‘Differential’
WHEN ‘L’ THEN ‘Transaction Log’
END ‘BackupType’
FROM msdb.dbo.backupset s
INNER JOIN msdb.dbo.backupmediafamily m
ON s.media_set_id = m.media_set_id
WHERE s.backup_start_date >= ’04/01/2020′
AND s.backup_start_date <= ’06/30/2020′
AND s.type IN (‘D’, ‘I’)
ORDER BY s.database_name, s.backup_start_date

The above code will return a result set similar to below.  Notice the results include, Database Name, Destination, duration, start time, Days since last backup and Backup Type.

This is much of the information that I have had to gather for various reasons over the years.

Notice the column DaysSinceLastBackup.  This uses the LAG function to calculate the number of days between the current row back up start time and the previous row start time. This can be used to help identify if there are any missed backups.  We have used this as the foundation of an alert that will email us if a backup is missing.

Over the years I have found this code to be very useful.  While this focuses mostly on full and differential backups, it can easily be modified to return other backups times as well.  However, keep in mind the impact that might have on the DaysSinceLastBackup.  If you also return transaction log backups, this number might not be as useful as you might like it to be.

Unlike a DMV, the data in these tables will persist after a reboot.

Thanks for visiting my blog!!!