Tag Archives: find

How to assess T-SQL code quickly

I’m sure you’re an excellent SQL coder writing beautiful efficient queries, but your predecessor . . . well they might have just been lucky to have a job.

Going through someone else’s bad code is usually tiresome, tedious and often very confusing.

I’ve created the T-SQL Assessor excel file to help in this task.

DOWNLOAD (Dropbox link)

The assessor will colour code the sql to highlight the lines of importance. With the Key Word column you can then simply filter to words like INSERT, UPDATE, MERGE and EXEC to see where the data is going or filter the column by the word FROM to see where the data has come from.

To use the T-SQL Assessor file you will first have to format your code using Poor Man’s T-SQL Formatter. This excellent tool can be installed in Visual Studio, SQL Server Management Studio or Notepad++.

http://architectshack.com/PoorMansTSqlFormatter.ashx

You can also use the online option:

http://poorsql.com/

Poor Man’s T-SQL Formatter makes text that contains a SQL command a new line, so you can’t have INSERT and FROM on one line. This is what allows Excel formula’s to highlight the lines with key words as each line can only contain one key word, excluding comments.

Once the code is formatted simply paste it into the first sheet of the file, “SQL”.

That’s it, all the work is then done for you on the second sheet of the file, “SQL Assessed”

T-SQL Assessor is also great at preparing a report from a schema compare script created by Visual Studio. It’s very annoying Microsoft didn’t provide a way of exporting the comparison directly into excel the way Redgate did but this will help. Simply filter the file to only include the keywords.

DOWNLOAD (Dropbox link)

Advertisements

How to identify and remove unused tables in SQL Server with T-Sql

In a perfect world an organisation should never get itself into the situation where tables need to be identified as functionally obsolete and removed in bulk. Schema changes should be stepped through the cycle of development, test, staging and live with developers cleaning as they go, but we don’t live in a perfect world as you may have noticed.

What do I mean by functionally obsolete? This means the tables are no longer being interacted with by user generated objects like stored procedures, functions or views. There may also be tables that could be classified as business redundant. That is they are being referenced by user generated objects frequently but they no longer have a use to the business, i.e. one job may have been replaced by another without the former being disabled or dropped. Business redundant objects are more difficult to determine and finding them may require input from multiple stakeholders.

A good approach for removing objects is to rename the objects first. This makes it easier to put the environment back the way it was if there are any problems encountered. After a set period of time if there is no impact on the overall environment script out the object then drop it. (Obviously do this in a test environment first if possible)

The script below is mostly a light weight SELECT statement that can be run on any environment. It does not execute any of the code it generates. It uses the sys.dm_db_index_usage_stats dynamic management view to determine when the tables were last interacted with. Interaction being defined as the following actions being applied against the object, update, seek, scan, lookup.

Caveat: Entries in this view reset to NULL after a Server reboot. Also the DMV has been known to be a bit unreliable with earlier versions of SQL Server with cases of the view being reset when a full index rebuild is carried out against a table. So don’t execute any code without first reviewing it.

Tables with NULL values for the fields below should be tables that have not been referenced at all or at least since the last time the server was rebooted.

LastUserUpdate
LastUserSeek
LastUserScan
LastUserLookup

Removing these objects is the low hanging fruit of cleaning up an environment. The script also provides stats on how many times these interactions happened and the size of the object. All these stats together should help you determine if an object is functionally redundant or business redundant and can be removed.

To aid further in the cleanup the script also creates the fields Action, Comments, Renamed, RenamedDate, RenameForDeletion, RestoreOriginalName, DropTable, DroppedDate.

The script output can then be copied and pasted into an Excel spread sheet and used to coordinate and track the cleanup progress.

SET NOCOUNT ON

DECLARE @Database TABLE ([DbName] [sysname])
DECLARE @DbName AS [sysname]
DECLARE @Sql AS [varchar] (max)

IF OBJECT_ID('tempdb..#TableStats', 'U') IS NOT NULL
	DROP TABLE #TableStats

IF OBJECT_ID('tempdb..#IndexStats', 'U') IS NOT NULL
	DROP TABLE #IndexStats

IF OBJECT_ID('tempdb..#TableUsageStats', 'U') IS NOT NULL
	DROP TABLE #TableUsageStats

IF OBJECT_ID('tempdb..#TableSizeStats', 'U') IS NOT NULL
	DROP TABLE #TableSizeStats

CREATE TABLE #TableStats (
	[DbName] [sysname]
	,[SchemaName] [sysname]
	,[ObjectId] [bigint]
	,[TableName] [sysname]
	,[ModifiedDate] [datetime]
	);

CREATE TABLE #IndexStats (
	[DbName] [sysname]
	,[ObjectId] [bigint]
	,[HasIndex] [bit]
	);

CREATE TABLE #TableSizeStats (
	[DbName] [varchar](255) NULL
	,[SchemaName] [varchar](255) NULL
	,[ObjectId] [bigint]
	,[TableName] [varchar](255) NULL
	,[RowCount] [bigint] NULL
	,[AvailableSpacePercentage] [numeric](6, 2) NULL
	,[UnusedSpaceGb] [numeric](10, 3) NULL
	,[UsedSpaceGb] [numeric](10, 3) NULL
	,[TotalSpaceGb] [numeric](10, 3) NULL
	,[UnusedSpaceMb] [numeric](13, 3) NULL
	,[UsedSpaceMb] [numeric](13, 3) NULL
	,[TotalSpaceMb] [numeric](13, 3) NULL
	,[UnusedSpaceKb] [bigint] NULL
	,[UsedSpaceKb] [bigint] NULL
	,[TotalSpaceKb] [bigint] NULL
	)

CREATE TABLE #TableUsageStats (
	[DbName] [sysname]
	,[ObjectId] [bigint]
	,[TableName] [nvarchar](128) NULL
	,[LastUserUpdate] [datetime] NULL
	,[LastUserSeek] [datetime] NULL
	,[LastUserScan] [datetime] NULL
	,[LastUserLookup] [datetime] NULL
	,[UserUpdateCount] [bigint] NOT NULL
	,[UserSeekCount] [bigint] NOT NULL
	,[UserScanCount] [bigint] NOT NULL
	,[UserLookupCount] [bigint] NOT NULL
	);

SET @DbName = '';

INSERT INTO @Database (DbName)
SELECT NAME
FROM master.dbo.sysdatabases
WHERE NAME NOT IN (
		'tempdb'
		,'master'
		,'mode'
		,'model'
		)
ORDER BY NAME ASC;

WHILE @DbName IS NOT NULL
BEGIN
	SET @DbName = (
			SELECT MIN(DbName)
			FROM @Database
			WHERE DbName > @DbName
			);
	SET @Sql = '
INSERT INTO #TableStats (
	DbName
	,schemaName
	,ObjectId
	,TableName
	,ModifiedDate
	)
SELECT DbName
	,SchemaName
	,ObjectId
	,TableName
	,ModifiedDate
FROM (
SELECT DISTINCT ' + '''' + @DbName + '''' + ' AS DbName
	,s.NAME AS SchemaName
	,t.object_id AS ObjectId
	,t.NAME AS TableName
	,t.modify_date AS ModifiedDate
FROM ' + QUOTENAME(@DbName) + '.sys.tables AS t
INNER JOIN ' + QUOTENAME(@DbName) + '.sys.schemas AS s ON t.schema_id = s.schema_id
LEFT JOIN ' + QUOTENAME(@DbName) + '.sys.extended_properties AS ep ON ep.major_id = t.[object_id] /*Exclude System Tables*/
WHERE t.NAME IS NOT NULL
	AND s.NAME IS NOT NULL
	AND (ep.[name] IS NULL OR ep.[name] <> ''microsoft_database_tools_support'')
	) AS rd
WHERE rd.SchemaName IS NOT NULL
ORDER BY DbName ASC
	,TableName ASC;
'

	EXEC (@Sql)

	SELECT @Sql = '
INSERT INTO #IndexStats (
	DbName
	,ObjectId
	,HasIndex
	)
SELECT ' + '''' + @DbName + '''' + ' AS DbName
	,OBJECT_ID AS ObjectId
	,IndexCheck AS HasIndex
FROM (
	SELECT DISTINCT OBJECT_ID
		,CASE 
			WHEN (
					[TYPE] > 0
					AND is_disabled = 0
					)
				THEN 1
			ELSE 0
			END AS IndexCheck
	FROM ' + QUOTENAME(@DbName) + '.sys.indexes
	) AS rd
WHERE rd.IndexCheck = 1
'

	EXEC (@Sql)

	SET @Sql = 
		'
INSERT INTO #TableSizeStats (
	[DbName]
	,[SchemaName]
	,[ObjectId]
	,[TableName]
	,[RowCount]
	,[AvailableSpacePercentage]
	,[UnusedSpaceGb]
	,[UsedSpaceGb]
	,[TotalSpaceGb]
	,[UnusedSpaceMb]
	,[UsedSpaceMb]
	,[TotalSpaceMb]
	,[UnusedSpaceKb]
	,[UsedSpaceKb]
	,[TotalSpaceKb]
	)
SELECT DISTINCT rd.[DbName]
	,rd.[SchemaName]
	,rd.[ObjectId]
	,rd.[TableName]
	,rd.[RowCount]
	,CASE 
		WHEN TotalSpaceKb > 0
			THEN ((UnusedSpaceKb / TotalSpaceKb) * 100)
		ELSE 0
		END AS AvailableSpacePercentage
	,CONVERT(NUMERIC(10, 3), (rd.[UnusedSpaceKb] / 1024.) / 1024.) AS UnusedSpaceGb
	,CONVERT(NUMERIC(10, 3), (rd.[UsedSpaceKb] / 1024.) / 1024.) AS UsedSpaceGb
	,CONVERT(NUMERIC(10, 3), (rd.[TotalSpaceKb] / 1024.) / 1024.) AS TotalSpaceGb
	,CONVERT(NUMERIC(13, 3), (rd.[UnusedSpaceKb] / 1024.)) AS UnusedSpaceMb
	,CONVERT(NUMERIC(13, 3), (rd.[UsedSpaceKb] / 1024.)) AS UsedSpaceMb
	,CONVERT(NUMERIC(13, 3), (rd.[TotalSpaceKb] / 1024.)) AS TotalSpaceMb
	,rd.[UnusedSpaceKb]
	,rd.[UsedSpaceKb]
	,rd.[TotalSpaceKb]
FROM (
	SELECT ' 
		+ '''' + @DbName + '''' + ' AS DbName
		,t.Object_id AS ObjectId
		,s.[name] AS [SchemaName]
		,t.[name] AS [TableName]
		,p.[rows] AS [RowCount]
		,SUM(a.[used_pages]) * 8 AS [UsedSpaceKb]
		,(SUM(a.[total_pages]) - SUM(a.[used_pages])) * 8 AS [UnusedSpaceKb]
		,SUM(a.[total_pages]) * 8 AS [TotalSpaceKb]
	FROM ' + QUOTENAME(@DbName) + '.sys.tables AS t
	INNER JOIN ' + QUOTENAME(@DbName) + '.sys.schemas AS s ON t.schema_id = s.schema_id
	INNER JOIN ' + QUOTENAME(@DbName) + '.sys.indexes AS i ON t.OBJECT_ID = i.object_id
	INNER JOIN ' + QUOTENAME(@DbName) + '.sys.partitions AS p ON i.object_id = p.OBJECT_ID
		AND i.[index_id] = p.[index_id]
	INNER JOIN ' + QUOTENAME(@DbName) + '.sys.allocation_units a ON p.[partition_id] = a.[container_id]
	WHERE t.[is_ms_shipped] = 0
		AND i.OBJECT_ID > 255
	GROUP BY s.[name]
		,t.[name]
		,t.[object_id]
		,p.[rows]
	) AS rd
ORDER BY DbName ASC
	,SchemaName ASC
	,TableName ASC;
'

	EXEC (@Sql)

	SET @Sql = '
INSERT INTO #TableUsageStats (
	[DbName]
	,[ObjectId]
	,[TableName]
	,[LastUserUpdate]
	,[LastUserSeek]
	,[LastUserScan]
	,[LastUserLookup]
	,[UserUpdateCount]
	,[UserSeekCount]
	,[UserScanCount]
	,[UserLookupCount]
	)
SELECT DbName
	,ObjectId
	,TableName
	,LastUserUpdate
	,LastUserSeek
	,LastUserScan
	,LastUserLookup
	,UserUpdateCount
	,UserSeekCount
	,UserScanCount
	,UserLookupCount
FROM (
	SELECT DISTINCT ROW_NUMBER() OVER (
			PARTITION BY ius.Object_Id ORDER BY last_user_update DESC
			) AS RN
		,' + '''' + @DbName + '''' + ' AS DbName
		,ius.OBJECT_ID AS ObjectId
		,o.NAME AS TableName
		,ius.last_user_update AS LastUserUpdate
		,ius.last_user_seek AS LastUserSeek
		,ius.last_user_scan AS LastUserScan
		,ius.last_user_lookup AS LastUserLookup
		,ius.user_updates AS UserUpdateCount
		,ius.user_seeks AS UserSeekCount
		,ius.user_scans AS UserScanCount
		,ius.user_lookups AS UserLookupCount
	FROM ' + QUOTENAME(@DbName) + 
		'.sys.dm_db_index_usage_stats AS ius
	INNER JOIN ' + QUOTENAME(@DbName) + '.sys.objects AS o ON ius.OBJECT_ID = o.OBJECT_ID
		AND o.NAME IS NOT NULL
	) AS rd
WHERE rd.RN = 1
ORDER BY rd.DbName ASC
	,rd.TableName ASC
	,rd.LastUserUpdate DESC
	,rd.LastUserSeek DESC
	,rd.LastUserScan DESC
	,rd.LastUserLookup DESC;
'

	EXEC (@Sql)
END;
GO

SELECT DISTINCT ROW_NUMBER() OVER (
		ORDER BY ts.[DbName] ASC
			,ts.[SchemaName] ASC
			,ts.[TableName] ASC
		) AS Row
	,ts.[DbName]
	,ts.[SchemaName]
	,ts.[TableName]
	,'' AS Action --Rename, keep etc.
	,'' AS Comments
	,'' AS Renamed --boolean flag
	,'' AS RenamedDate 
	,i.[HasIndex] --Tables without an index are heaps
	,tss.[TotalSpaceMb] AS TableSizeInMb
	,ts.[ModifiedDate]
	,tus.[LastUserUpdate]
	,tus.[LastUserSeek]
	,tus.[LastUserScan]
	,tus.[LastUserLookup]
	,tus.[UserUpdateCount]
	,tus.[UserSeekCount]
	,tus.[UserScanCount]
	,tus.[UserLookupCount]
	,tss.[AvailableSpacePercentage]
	,tss.[UnusedSpaceGb]
	,tss.[UsedSpaceGb]
	,tss.[TotalSpaceGb]
	,tss.[UnusedSpaceMb]
	,tss.[UsedSpaceMb]
	,tss.[TotalSpaceMb]
	,tss.[UnusedSpaceKb]
	,tss.[UsedSpaceKb]
	,tss.[TotalSpaceKb]
	,'USE ' + QUOTENAME(ts.[DbName]) + '; EXEC sp_rename ' + '''' + ts.[SchemaName] + '.' + ts.[TableName] + '''' + ', ' + '''' + '_DELETE_' + ts.[TableName] + '''' + ';' AS RenameForDeletion
	,'USE ' + QUOTENAME(ts.[DbName]) + '; EXEC sp_rename ' + '''' + ts.[SchemaName] + '.' + '_DELETE_' + ts.[TableName] + '''' + ', ' + '''' + ts.[TableName] + '''' + ';' AS RestoreOriginalName
	,'USE ' + QUOTENAME(ts.[DbName]) + '; DROP TABLE ' + QUOTENAME(ts.[SchemaName]) + '.' + '[' + '_DELETE_' + ts.[TableName] + ']' + ';' AS 'DropTable'
	,'' AS DroppedDate
FROM #TableStats AS ts
LEFT JOIN #TableSizeStats AS tss ON ts.ObjectId = tss.ObjectId
	AND tss.DbName = ts.DbName
LEFT JOIN #IndexStats AS i ON ts.ObjectId = i.ObjectId
	AND i.DbName = ts.DbName
LEFT JOIN #TableUsageStats AS tus ON ts.ObjectId = tus.ObjectId
	AND tus.DbName = ts.DbName
ORDER BY ts.[DbName] ASC
	,ts.[SchemaName] ASC
	,ts.[TableName] ASC

	-- REF: https://docs.microsoft.com/en-us/sql/relational-databases/system-dynamic-management-views/sys-dm-db-index-usage-stats-transact-sql
Picture showing SQL Server Agent Properties Window

How to setup how much Agent Job History is retained

I was asked to do a historic analysis of SQL Server jobs to determine their statistics, such as the number of runs, average run duration, successes/failures etc. As part of the daily checks I do on a server I’d only check the job runs for the past 24 hours. So I was surprised when I started the analysis and saw that averages for the same job across servers varied significantly.

When I looked at the msdb.dbo.sysjobhistory table of the server with the lowest averages I noted that there were only 999 rows and the history only went back a day or two.

So I had effectively no history on that server. When I looked at another the sysjobhistory table had almost 10 million entries, roughly 4 Gb of data! (The joys of taking over someone else’s environment)

This got me looking at the history settings for the SQL Agent. To open the SQL Server Agent Properties while in SQL Server Object Explorer right click on SQL Server Agent and then click on properties.

Picture showing SQL Server Agent Properties Window

For the first server the history settings were set to the default of 1000 lines for the log as a whole and 100 lines for the individual jobs. If you have a job that runs 4 steps that will use 5 lines, one line for the job and one line for each of the steps that are run. When I looked at the history settings on the second, server as you would expect, no settings had been applied.

Some sensible thresholds needed to be applied across the environment. I’m not going to dictate to you what your retention policy should be regarding job run history but I will say it can be as much as business decision as a server maintenance one so ask around if people need a record that something has run.

You may be in the position that you can be specific enough to enter row values but most organisations will have jobs that run weekly or monthly. In that case you may be better off using the “Older than” option and setting a retention period rather than using rows.

Always with logging, if you have the space, you’re better off having too much than having too little and missing data you need.  The size of each row in the sysjobhistory table is at most 4.5KB so keeping 10k rows is only around 45 Mb. If you don’t have 45 Mb to spare on your server then job history retention is the least of your worries. The worst thing that is likely to happen in storing more rows than you need is that it takes a bit longer to bring up the job history viewer. But here’s a link to some queries to help with that.

The moral of the story is just because it’s in the MSDB doesn’t mean Microsoft are taking care of it. Check your settings and make sure you’re retaining an appropriate amount of history now.

How to identify all user objects in the SQL Server Master database

What is the SQL Server Master Database used for?

The master database is used by SQL Server to contain all of the system level information:

  • Logins
  • Linked servers
  • Endpoints
  • Configuration settings
  • Information about the other databases on this instance and the location of their files

If the master database is not present, SQL Server cannot start.

This is way you should always take regular backups of the master database.
as if SQL Server suffers a failure, those changes will be lost and you’ll be in a lot of trouble.

So the master database is needed for SQL Server to work. Logic and objects needed to enable your application should not be tangled up in there so you should always avoid creating objects in the master database. Use a user database instead.

In saying that everyone has made the mistake of not specifying the database they want to USE in a script and by default objects get written to the Master database. Or you might have inherited the responsibility for an application which did not conform to best practices and it uses objects created in Master.

To gain visibility of these objects just run the script below.

SELECT o.object_id AS ObjectId
	,o.NAME AS ObjectName
	,o.type_desc AS ObjectType
	,o.create_date AS CreateDate
	,o.modify_date AS ModifyDate
	,SUM(st.row_count) AS RowCnt
	,CAST(SUM(st.used_page_count) / 128.0 AS DECIMAL(36, 1)) AS DataSize_MB
FROM master.sys.objects o
LEFT JOIN master.sys.dm_db_partition_stats st ON st.object_id = o.object_id
	AND st.index_id < 2
GROUP BY o.object_id
	,o.NAME
	,o.type_desc
	,o.create_date
	,o.modify_date
	,o.is_ms_shipped
HAVING o.is_ms_shipped = 0
	AND o.NAME <> 'sp_ssis_startup'
	AND o.type_desc NOT LIKE '%CONSTRAINT%'
ORDER BY CAST(SUM(st.used_page_count) / 128.0 AS DECIMAL(36, 1)) DESC
	,RowCnt DESC

Query To Find Columns From All Tables Of Database

If your job is to create reports using SQL chances are you have or will encounter this situation:

You’ve been asked to prepare a report, but the person who has asked for the report simply has a list of fields they want and they have no idea where those fields come from. They may have received previous reports in the past, so they know the fields exist, but cannot provide any of the SQL queries used to create these reports as an example.

You, the developer, may not be familiar with that particular area of the business or associated data sources. Possibly because you typically prepared financial reports and this request has come from the operations or marketing departments.

So the first step is to locate these columns within the database.

The following query will return the Table Name, Schema Name and Column Name from the database.

In the example below all instances where the column name equals CustomerID, OrderID, OrderDate will be returned. Also Column names that contain the word Status or Promotion will also be return. Simply change or add additional columns names as needed.

That should help get you started in preparing the report.