Monitoring | SQL Freelancer Blog

Collect SQL Server Performance Counters and Build Reports with SSRS

By bradyupton in Business Intelligence, Monitoring, Reporting Services (SSRS) February 26, 2014 1 Comment

Third party tools are awesome for capturing performance metrics, but some small shops (and even large shops) don’t budget for this type of software leaving it up to the DBA to create their own monitoring solution.

There are a few different ways to capture certain performance metrics, but in this post I’ll focus on using the sys.dm_os_performance_counters DMV and how to view this data in a more readable form using SQL Server Reporting Services graphs. The DMV doesn’t include all the counters as Performance Monitor, but it does show the SQL Server related counters. (Note that some of the counters in this DMV are of cumulative values since the last reboot.)

You can query this DMV using the following query:

SELECT * FROM sys.dm_os_performance_counters

SQL Freelancer SQL Server Performance Counters SSRS

As you can see from the screenshot above, this can be hard to read. Also, it only shows the current values at the time the query is executed so you don’t know anything about the past and it makes it hard to see how the data fluctuates during the business day.

Collecting the SQL Server Monitoring Report Data

For the purpose of simplicity, this tip will focus on one counter, Page Life Expectancy (PLE). I’ll show you how to capture data and create a graph for analysis.

First, we’ll need to create a table that will store our metrics:

CREATE TABLE [dbo].[CounterCollections](
[ID] [int] IDENTITY(1,1) NOT NULL,
[object_name] [varchar](128) NOT NULL,
[counter_name] [varchar](128) NOT NULL,
[cntr_value] [bigint] NOT NULL,
[collection_datetime] [datetime] NOT NULL )

Next, we’ll need to create a script that will insert our data into the table we created above:

INSERT INTO CounterCollections
SELECT object_name, counter_name, cntr_value, GETDATE() collection_datetime
FROM sys.dm_os_performance_counters
WHERE object_name = ‘SQLServer:Buffer Manager’
AND counter_name = ‘Page life expectancy’

Finally, we’ll need to create a SQL Server Agent Job that will run the script above on a specified schedule:

I’ll run this job every 5 minutes:

Creating the SQL Server Monitoring Report

While the table gathers data we can switch over to Business Intelligence Development Studio (BIDS) or the SQL Server Data Tools (SSDT) and create a SSRS Report that will display our data.

In this example, I’m using SSDT. Choose File, New Project. On the New Project dialog choose Report Server Project and name the Project:

Click here to view the rest of this post.

Fixing SQL Server error in-row data RSVD page count is incorrect

By bradyupton in Monitoring September 3, 2013 0 Comment

I ran DBCC CHECKDB (Integrity Checks) in my maintenance plan and I also ran it manually, but I keep getting the error “The In-row data RSVD page count for object “table_name” , index ID 0, partition ID 58037252456448, alloc unit ID 58037252456448 (type In-row data) is incorrect.” How can I fix this error?

Solution

Before we discuss the simple steps to fixing this error, let’s discuss what might have happened in this situation.

Anytime you see an integrity check error it might make you a little queasy. No need to worry on this error. If you have recently upgraded from SQL Server 2000 then this error is somewhat common. I know what you’re saying, “Why is there a tip regarding SQL Server 2000?” Trust me, a lot of people are still running SQL Server 2000 which is now out of compliance with Microsoft, so everyone is in a rush to upgrade. Anyway, SQL Server 2000 used to update the page space used metadata. Once SQL Server 2005 came along, this was no longer the case so if you didn’t run DBCC UPDATEUSAGE after the upgrade/migration this error is likely to appear when you run DBCC CHECKDB.

Here is what the error message looks like:

Msg 2508, Level 16, State 3, Line 1

The In-row data RSVD page count for object "table_name", index ID 0, partition ID 58037252456448,
alloc unit ID 58037252456448 (type In-row data) is incorrect. Run DBCC UPDATEUSAGE.

Running DBCC UPDATEUSAGE

If you notice in the maintenance plan or the DBCC CHECKDB results above, it displays “Run DBCC UPDATEUSAGE” after the error message. Ta da! After backing up your databases and establishing a maintenance window run DBCC UPDATEUSAGE (databasename). In a perfect world, SQL would notice this error and run DBCC UPDATEUSAGE on the appropriate table. This is not a perfect world, so we have to run it manually.

DBCC UPDATEUSAGE reports and corrects pages and row count inaccuracies in the catalog views. These inaccuracies may cause incorrect space usage reports returned by the sp_spaceused system stored procedure – MSDN.

 DBCC UPDATEUSAGE (Apollo32_Fixed)

After running DBCC UPDATEUSAGE (databasename) on my questionable database you can see that all the page counts have been updated:


DBCC UPDATEUSAGE: Usage counts updated for table 'UserList' (index 'UserList', partition 1):
        USED pages (In-row Data): changed from (5) to (4) pages.
DBCC UPDATEUSAGE: Usage counts updated for table 'UserRights' (index 'UserRights', partition 1):
        USED pages (In-row Data): changed from (72) to (24) pages.
        RSVD pages (In-row Data): changed from (80) to (33) pages.
DBCC UPDATEUSAGE: Usage counts updated for table 'UserRights' (index 'UserRights', partition 1):
        USED pages (In-row Data): changed from (5) to (4) pages.
DBCC UPDATEUSAGE: Usage counts updated for table 'UserGroups' (index 'UserGroups', partition 1):
        USED pages (In-row Data): changed from (4) to (3) pages.
DBCC UPDATEUSAGE: Usage counts updated for table 'Application' (index 'Application', partition 1):
        USED pages (In-row Data): changed from (3) to (2) pages.

You only need to run DBCC UPDATEUSAGE once. After the page counts have been corrected your SQL Server will be back to normal. If you do not fix this issue right away, page counts will continue to get worse when data is added and will result in abnormalities in space usage reports.

After the update, running DBCC CHECKDB gives me a clean database again:

CHECKDB found 0 allocation errors and 0 consistency errors in database ”database_name’. DBCC execution completed. If DBCC printed error messages, contact your system administrator.

In essence, DBCC UPDATEUSAGE updates the used, reserved, and rows columns of the sysindexes (later named sys.indexes in SQL Server 2005) table.

In this example I have two databases named Apollo32 and Apollo32_Fixed. When running an integrity check against Apollo32 I get the error described above. Apollo32_Fixed is a copy of Apollo32 after running DBCC UPDATEUSAGE.

Before and After Comparison

If I compare Apollo32.dbo.sysindexes to Apollo32_Fixed.dbo.sysindexes I can see I have 534 differences.

 (SELECT id, rowcnt, used, reserved, rows FROM Apollo32.dbo.sysindexes
EXCEPT
SELECT id, rowcnt, used, reserved, rows FROM Apollo32_Fixed.dbo.sysindexes)
UNION ALL
(SELECT id, rowcnt, used, reserved, rows FROM Apollo32_Fixed.dbo.sysindexes
EXCEPT
SELECT id, rowcnt, used, reserved, rows FROM Apollo32.dbo.sysindexes)

Click here to view the rest of this post.

Using the SSIS WMI Task to Gather System Information

By bradyupton in Integration Services (SSIS), Monitoring April 22, 2013 0 Comment

Finding server information is vital when monitoring servers and is usually pretty easy to find on SQL Servers, however when monitoring web or app servers this may be a bit of a challenge when SQL querying is not at your disposal. This post will talk about the SSIS task “WMI Data Reader” and how it can help you monitor your non-SQL Servers.

First, a short introduction to WMI….

WMI (Windows Management Instrumentation) is used for accessing management information in an enterprise environment. The ability to obtain management data from remote computers is what makes WMI useful. It provides a consistent approach to carrying out day-to-day management tasks with programming or scripting languages – MSDN

Some things you can do with WMI may include:

Rebooting a computer remotely
Get a list of applications installed on a local or remote computer
Find OS info such as the version of Windows and service pack levels
Query the event logs on a local or remote computer
Find system information such as Manufacturer and Model of a computer

As you can see, WMI is a very powerful tool and can be used locally or remotely.

There are tons of different metrics you can capture using WMI scripts, but for this example I’ll capture disk space information using the Win32_LogicalDisk class and insert it into a database table using a SSIS package.

First, let’s create our database table that will store our disk information:

CREATE TABLE DiskInfo
(
Drive varchar(4),
DriveSize numeric(18,2),
FreeSpace numeric(18,2),
[Date] Datetime Not NUll Default GetDate()
)

Once our table is created, we can create the SSIS package. Open BIDS or SSDT and create a new Integration Services project.

From the SSIS toolbox, drag the WMI Data Reader Task into the Design view.

Right click in the Connection Manager window and select New Connection and select WMI. Click Add:

This will open the WMI Connection Manager Editor. Give the connection manager a name and description. If you are creating the package for a local computer then use \\localhost to connect. If you are creating the package for a remote server then use \\RemoteServer to connect. For this example, I’m connecting locally using Windows Authentication:

Next we’ll need to right click on the WMI Data Reader Task and choose edit. Click the WMI Options tab. This tab is where we will need to input information such as the WMI connection string and the WMI query. For this example, I have used the following:

WMI connection – This is the connection we created in the previous step.
WqlQuerySourceType – Direct input
WqlQuerySource – Select Name, Size, FreeSpace From Win32_LogicalDisk Where DriveType = 3
OutputType – Data table
OverwriteDestination – Overwrite destination
DestinationType – File connection
Destination – New destination (see below)

Click here to view the rest of this post.

Detecting CPU Pressure

By bradyupton in Dynamic Management Views, Monitoring, Performance Tuning February 28, 2013 0 Comment

There are a few different ways to detect CPU pressure, for example, Performance Monitor or SQL Profiler, but in this tip we’ll discuss the wait type SOS_SCHEDULER_YIELD and the DMV sys.dm_os_schedulers.

Some of you might have seen the wait type SOS_SCHEDULER_YIELD and wondered what in the world is this? Books Online (BOL) describes this wait type as:

Occurs when a task voluntarily yields the scheduler for other tasks to execute. During this wait the task is waiting for its quantum to be renewed.

SOS_SCHEDULER_YIELD is a fairly common wait type and occurs when there is CPU pressure. SQL Server runs multiple threads and tries to allow all threads to run without problems. However, if all the threads are busy on each scheduler and can’t let other threads run, it will yield itself for another thread, which in turn creates the SOS_SCHEDULER_YIELD wait type.

You can see the top wait stats and the percentage of waits at the current point in time by running the following query by Tim Ford and Glenn Berry.

WITH Waits AS 
( 
SELECT 
wait_type, 
wait_time_ms / 1000. AS wait_time_s, 
100. * wait_time_ms / SUM(wait_time_ms) OVER() AS pct, 
ROW_NUMBER() OVER(ORDER BY wait_time_ms DESC) AS rn 
FROM sys.dm_os_wait_stats 
WHERE wait_type 
NOT IN 
('CLR_SEMAPHORE', 'LAZYWRITER_SLEEP', 'RESOURCE_QUEUE', 
'SLEEP_TASK', 'SLEEP_SYSTEMTASK', 'SQLTRACE_BUFFER_FLUSH', 'WAITFOR', 
'CLR_AUTO_EVENT', 'CLR_MANUAL_EVENT') 
) -- filter out additional irrelevant waits 
SELECT W1.wait_type, 
CAST(W1.wait_time_s AS DECIMAL(12, 2)) AS wait_time_s, 
CAST(W1.pct AS DECIMAL(12, 2)) AS pct, 
CAST(SUM(W2.pct) AS DECIMAL(12, 2)) AS running_pct 
FROM Waits AS W1 
INNER JOIN Waits AS W2 ON W2.rn <= W1.rn 
GROUP BY W1.rn, 
W1.wait_type, 
W1.wait_time_s, 
W1.pct 
HAVING SUM(W2.pct) - W1.pct < 95; -- percentage threshold;

After running this query on one of my servers I can see that only 2.22% of my waits are due to CPU pressure so I can rule that out as a cause for poor performance. (This is only current since the last SQL Service restart):

If you see a high percentage of this wait type then one way of troubleshooting this is to use the DMV sys.dm_os_schedulers. BOL describes sys.dm_os_schedulers as:

Returns one row per scheduler in SQL Server where each scheduler is mapped to an individual processor. Use this view to monitor the condition of a scheduler or to identify runaway tasks.

This DMV has several columns including number or workers, active tasks, and the status of each scheduler, etc., which can help troubleshoot certain problems, but the most important column is the one used for measuring queued tasks, the runnable_tasks_count column. This column displays the count of tasks yielding to other tasks resulting in a SOS_SCHEDULER_YIELD wait type. If this column is frequently greater than 0 then CPU pressure may be present and blocking may occur.

To see an average of current tasks and current waiting tasks you can use the following query:

SELECT AVG(current_tasks_count) AS [Avg Current Task], 
AVG(runnable_tasks_count) AS [Avg Wait Task]
FROM sys.dm_os_schedulers
WHERE scheduler_id < 255
AND status = 'VISIBLE ONLINE'

Click here to view the rest of this post.

Report to Capture Table Growth Statistics for SQL Server

By bradyupton in Monitoring, Reporting Services (SSRS) December 3, 2012 0 Comment

There are a few tools that monitor table size growth, but with the use of sp_spaceused and SQL Server Reporting Services (SSRS) you can make custom reports.

Using sp_spaceused

The system stored procedure sp_spaceused displays the number of rows, disk space reserved, and disk space used by a table, indexed view, or Service Broker queue in the current database in an object is provided or it displays the disk space reserved and used by the whole database.

This procedure, sp_spaceused can be used at the database or table level. For example, sp_spaceused at the database level returns the following:

And sp_spaceused at the table level displays:

Creating the Report

To create our custom report, first we will need to create a table that will store the data from sp_spaceused. For this example, I’ll create a table named TableSizeGrowth.

CREATE TABLE [dbo].[TableSizeGrowth](
[id] [int] IDENTITY(1,1) NOT NULL,
[table_schema] [nvarchar](256) NULL,
[table_name] [nvarchar](256) NULL,
[table_rows] [int] NULL,
[reserved_space] [int] NULL,
[data_space] [int] NULL,
[index_space] [int] NULL,
[unused_space] [int] NULL,
[date] [datetime] NULL
) ON [PRIMARY]

I’m also going to add a default constraint that will automatically update the date column to the current date for each new row.

ALTER TABLE [dbo].[TableSizeGrowth] ADD CONSTRAINT [DF_TableSizeGrowth_date]  
DEFAULT (dateadd(day,(0),datediff(day,(0),getdate()))) FOR [date]
GO

Next, we need to create a stored procedure that moves the data from sp_spaceused to the table we created above. I have put comments throughout the stored procedure to help you figure out what is going on. Also make sure you change the table names to match your tables. For this example, I want to monitor HumanResources.Employee, Production.Product, and Purchasing.Vendor tables.

SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
CREATE PROCEDURE [dbo].[sp_TableSizeGrowth] 
AS
BEGIN
 SET NOCOUNT ON

 --DECLARE VARIABLES
 DECLARE
 @max INT,
 @min INT,
 @table_name NVARCHAR(256),
 @table_schema NVARCHAR(256),
 @sql NVARCHAR(4000)

 --DECLARE TABLE VARIABLE
 DECLARE @table TABLE(
 id INT IDENTITY(1,1) PRIMARY KEY,
 table_name NVARCHAR(256),
 table_schema NVARCHAR(256))

 --CREATE TEMP TABLE THAT STORES INFORMATION FROM SP_SPACEUSED
 IF (SELECT OBJECT_ID('tempdb..#results')) IS NOT NULL
 BEGIN
  DROP TABLE #results
 END

 CREATE TABLE #results
 (
  [table_schema] [nvarchar](256) NULL,
  [table_name] [nvarchar](256) NULL,
  [table_rows] [int] NULL,
  [reserved_space] [nvarchar](55) NULL,
  [data_space] [nvarchar](55) NULL,
  [index_space] [nvarchar](55) NULL,
  [unused_space] [nvarchar](55) NULL
 )

 --LOOP THROUGH STATISTICS FOR EACH TABLE
 INSERT @table(table_schema, table_name)
 SELECT  
  table_schema, table_name
 FROM
  information_schema.tables 
 WHERE table_schema + '.' + table_name IN ('HumanResources.Employee','Production.Product', 'Purchasing.Vendor') --INSERT TABLE NAMES TO MONITOR

 SELECT
  @min = 1,
  @max = (SELECT MAX(id) FROM @table)

 WHILE @min = @max
 BEGIN
  SELECT 
   @table_name = table_name,
   @table_schema = table_schema
  FROM
   @table
  WHERE
   id = @min

  --DYNAMIC SQL
  SELECT @sql = 'EXEC sp_spaceused ''[' + @table_schema + '].[' + @table_name + ']'''

  --INSERT RESULTS FROM SP_SPACEUSED TO TEMP TABLE
  INSERT #results(table_name, table_rows, reserved_space, data_space, index_space, unused_space)
  EXEC (@sql)

  --UPDATE SCHEMA NAME
  UPDATE #results
  SET table_schema = @table_schema
  WHERE table_name = @table_name
  SELECT @min = @min + 1
 END

 --REMOVE "KB" FROM RESULTS FOR REPORTING (GRAPH) PURPOSES
 UPDATE #results SET data_space = SUBSTRING(data_space, 1, (LEN(data_space)-3))
 UPDATE #results SET reserved_space = SUBSTRING(reserved_space, 1, (LEN(reserved_space)-3))
 UPDATE #results SET index_space = SUBSTRING(index_space, 1, (LEN(index_space)-3))
 UPDATE #results SET unused_space = SUBSTRING(unused_space, 1, (LEN(unused_space)-3))

 --INSERT RESULTS INTO TABLESIZEGROWTH
 INSERT INTO TableSizeGrowth (table_schema, table_name, table_rows, reserved_space, data_space, index_space, unused_space)
 SELECT * FROM #results

 DROP TABLE #results
END

Now that the stored procedure has been created, let’s go ahead and execute it.

Once the stored procedure finishes we can view the table:

Click here to view the rest of this post.

Finding what port SQL Server is listening on

By bradyupton in Monitoring August 15, 2012 0 Comment

SQL Server communicates over TCP/IP and listens for incoming connections over what is called a port. The default TCP/IP port for SQL Server is 1433. The port doesn’t have to be 1433, but 1433 is the official Internet Assigned Number Authority (IANA) socket number for SQL Server. (http://support.microsoft.com/kb/287932)

Named instances are configured for dynamic ports. This means when the SQL Server service is started it selects a random port. Best practice is to configure named instances to listen on a specific port.

You need to know what port your SQL Server is using so you can allow the ports through the firewall otherwise you can run into connection problems between the client and the server.

There are a few different ways to find the port number that SQL Server is listening on. Below you will find T-SQL statements that will allow you to find the port number for default and named instances.


--SQL Server 2000 Default Instance 
DECLARE @tcp_port nvarchar(5)

EXEC xp_regread
 @rootkey    =    'HKEY_LOCAL_MACHINE',
 @key        =    'SOFTWARE\Microsoft\MSSQLServer\MSSQLServer\SuperSocketNetLib\TCP',
 @value_name    =    'TcpPort',
 @value        =    @tcp_port OUTPUT

select @tcp_port

--SQL Server 2000 Named Instance
DECLARE @tcp_port nvarchar(5)

EXEC xp_regread
 @rootkey    =    'HKEY_LOCAL_MACHINE',
 @key        =     'SOFTWARE\Microsoft\Microsoft SQL Server(InstanceName)\MSSQLServer\SuperSocketNetLib\TCP',
 @value_name    =    'TcpPort',
 @value        =    @tcp_port OUTPUT

select @tcp_port

--SQL Server 2005 Default Instance
DECLARE @tcp_port nvarchar(5)

EXEC xp_regread
 @rootkey    =    'HKEY_LOCAL_MACHINE',
 @key        =    'SOFTWARE\Microsoft\Microsoft SQL Server\MSSQL.1\MSSQLServer\SuperSocketNetLib\TCP\IPAll',
 @value_name    =    'TcpPort',
 @value        =    @tcp_port OUTPUT

select @tcp_port

--SQL Server 2005 Named Instance
DECLARE @tcp_port nvarchar(5)

EXEC xp_regread
 @rootkey    =    'HKEY_LOCAL_MACHINE',
 @key        =    'SOFTWARE\Microsoft\Microsoft SQL Server\MSSQL.(#)\MSSQLServer\SuperSocketNetLib\TCP\IPAll',
 @value_name    =    'TcpPort',
 @value        =    @tcp_port OUTPUT

select @tcp_port

--SQL Server 2008 R2 Default Instance
DECLARE @tcp_port nvarchar(5)

EXEC xp_regread
 @rootkey    =    'HKEY_LOCAL_MACHINE',
 @key        =    'SOFTWARE\Microsoft\Microsoft SQL Server\MSSQL10_50.MSSQLSERVER\MSSQLServer\SuperSocketNetLib\TCP\IPAll',
 @value_name    =    'TcpPort',
 @value        =    @tcp_port OUTPUT

select @tcp_port

--SQL Server 2008 R2 Named Instance
DECLARE @tcp_port nvarchar(5)

EXEC xp_regread
 @rootkey    =    'HKEY_LOCAL_MACHINE',
 @key        =    'SOFTWARE\Microsoft\Microsoft SQL Server\MSSQL10_50.(InstanceName)\MSSQLServer\SuperSocketNetLib\TCP\IPAll',
 @value_name    =    'TcpPort',
 @value        =    @tcp_port OUTPUT

select @tcp_port

 --SQL Server 2012 Default Instance
DECLARE @tcp_port nvarchar(5)

EXEC xp_regread
 @rootkey    =    'HKEY_LOCAL_MACHINE',
 @key        =    'SOFTWARE\Microsoft\Microsoft SQL Server\MSSQL11.MSSQLSERVER\MSSQLServer\SuperSocketNetLib\TCP\IPAll',
 @value_name    =    'TcpPort',
 @value        =    @tcp_port OUTPUT

select @tcp_port

--SQL Server 2012 Named Instance
DECLARE @tcp_port nvarchar(5)

EXEC xp_regread
 @rootkey    =    'HKEY_LOCAL_MACHINE',
 @key        =    'SOFTWARE\Microsoft\Microsoft SQL Server\MSSQL11.(InstanceName)\MSSQLServer\SuperSocketNetLib\TCP\IPAll',
 @value_name    =    'TcpPort',
 @value        =    @tcp_port OUTPUT

select @tcp_port

Managing SQL Server Extended Events in Management Studio

By bradyupton in Extended Events, Monitoring July 19, 2012 0 Comment

SQL Server 2012 introduces a GUI in SQL Server Management Studio to create and manage extended events. Prior to the SQL Server 2012 integration, Extended Events could only be created using T-SQL. In this tip, I’ll show you step by step process on how to create a simple Extended Event in SQL Server 2012 using the new GUI in SQL Server Management Studio.

Creating an Extended Event has never been easier with SQL Server 2012. Open SSMS, and drilldown to Management, Extended Events, Sessions as shown in the image below. By default, you should see an AlwaysOn_health and a system_health session already created. You will notice the AlwaysOn_health session is disabled and the system_health session is running. The system_health session collects system data that you can use to help troubleshoot performance issues. For the most part, SQL Server Extended Events use very little resources.

There are two ways to create a session. Right click on the Sessions folder and you can choose New Session or New Session Wizard. In this tip, we’ll step through using the wizard.

After clicking New Session Wizard, an Introduction window will appear that will give you a brief introduction. Click the “Next” button to continue.

The next window, Set Session Properties, is where you can specify the session name and whether or not you want the session to start on server start-up. In this tip, I’ll name the session DB Monitor and choose to start the event session at server start-up. Click the “Next” button to continue.

The next screen will allow us to choose a preconfigured template or create our own. If you’ve ever used SQL Server Profiler’s built in template, these function the same way. In this tip, we’ll create our own. Choose the “Do not use a template” option and click the “Next” button to continue.

The “Select Events To Capture” window is an important one. This is where we select the events we want to capture. For this example, I want to monitor when my DB goes offline and when it becomes available, so I’ll choose the events that relate to this: database_attached, database_created, database_detached, database_started, and database_stopped. Once you select the events from the “Event library” (on the left), click the right arrow to move them to the “Selected events” (on the right). Click the “Next” button to continue.

Click here to view the rest of this post.

sp_who

By bradyupton in Monitoring November 8, 2011 0 Comment

There are numerous ways you can view blocking information but I’ll share the most popular in sp_who and sp_who2.

Like their name implies sp_who and sp_who2 give information about “who” is in your SQL Server as well as other useful information including processes and blocking info.

Below is an example of sp_who running against a SQL Server:

As you can see, sp_who will return SPID, status, loginname, blk, dbname, etc.

spid – The system process ID.
ecid – The execution context ID of a given thread associated with a specific SPID.
status – The process status.
loginame – The login name associated with the particular process.
hostname – The host or computer name for each process.
blk – The system process ID for the blocking process, if one exists. Otherwise, this column is zero.

When a transaction associated with a given SPID is blocked by an orphan distributed transaction, this column will return a ‘-2′ for the blocking orphan transaction.
dbname – The database used by the process.
cmd – The SQL Server command (Transact-SQL statement, SQL Server internal engine process, and so on) executing for the process.