Introduction
This is the second article of SQL Monitor, the first one is here: sqlmon.aspx
For background infomation, please refer to the above mentioned article.
SQL Monitor can monitor sql server processes and jobs, analyse performance, object version control, view executing sql query, kill process / job, object explorer, database shrink/log truncate/backup/detach/attach etc.
Ok, for a better understanding, first let's look at the following picture:
Well, we can see it shows the real time IO/CPU/Network info of the SQL Server. In fact, it can do a lot more. For infomation about object explorer, activities and version control, please the above mentioned first article.
Background
In the previous article, we discussed about how I implemented the object explorer, activities and version control, in this article, I will talk about the newly introduced performance graph and analysis features.
We have been wondering like "What is going on with my SQL Server?", "How well is it running?", "Where is the bottleneck", "Can I make it faster?" etc. Well, it is time to get the answer, ok, to some extent, somehow, so far ;)
Performance Graph
Ok, maybe this is not the most exciting functionality you wish for the coming Christmas :-D.
With performance graph, you can keep tracking what is going on with your SQL Server, it shows the following real time infomation:
- Total IO busy time since last server start
- Total CPU busy time since last server start
- Total IO Reads (in byte) since last server start
- Recent n seconds IO Reads (in byte)
- Total IO Writes (in byte) since last server start
- Recent n seconds IO Writes (in byte)
- Total Packet Reads (in byte) since last server start
- Recent n seconds Packet Reads (in byte)
- Total Packet Writes (in byte) since last server start
- Recent n seconds Packet Writes (in byte)
- Total connections since last server start
- Recent n seconds connections
n second is the monitor refresh interval, which you can set in the options.
Where does the infomation come from?
Well, because I am a very boring guy, so I really want to do something to kill time, from time to time, I will dig through different kinds of resources, like the whole SQL Server databases, including all functions, stored procedures, views, and one day I found a system stored procedure called sp_monitor. mmmmmm...... That means everything, right? Thank you very much, SQL Server team ;). When you dig deeper, you can find out that it uses the following system variables:
@@cpu_busy
@@io_busy
@@idle
@@pack_received
@@pack_sent
@@connections
@@packet_errors
@@total_read
@@total_write
@@total_errors
I modified the stored procedure to accomplish my need:
declare @now datetime
declare @cpu_busy int
declare @io_busy int
declare @idle int
declare @pack_received int
declare @pack_sent int
declare @pack_errors int
declare @connections int
declare @total_read int
declare @total_write int
declare @total_errors int
declare @oldcpu_busy int
declare @interval int
declare @mspertick int
select @mspertick = convert(int, @@timeticks / 1000.0)
select
@now = getdate(),
@cpu_busy = @@cpu_busy,
@io_busy = @@io_busy,
@idle = @@idle,
@pack_received = @@pack_received,
@pack_sent = @@pack_sent,
@connections = @@connections,
@pack_errors = @@packet_errors,
@total_read = @@total_read,
@total_write = @@total_write,
@total_errors = @@total_errors
select @oldcpu_busy = cpu_busy
from master.dbo.spt_monitor
if @oldcpu_busy > @cpu_busy
begin
update master.dbo.spt_monitor
set
lastrun = @now,
cpu_busy = @cpu_busy,
io_busy = @io_busy,
idle = @idle,
pack_received = @pack_received,
pack_sent = @pack_sent,
connections = @connections,
pack_errors = @pack_errors,
total_read = @total_read,
total_write = @total_write,
total_errors = @total_errors
end
set nocount on
select @interval = datediff(ss, lastrun, @now)
from master.dbo.spt_monitor
if @interval = 0
select @interval = 1
select last_run = lastrun, current_run = @now, seconds = @interval,
cpu_busy_total = convert(int, ((@cpu_busy * @mspertick) / 1000)),
cpu_busy_current = convert(int, (((@cpu_busy - cpu_busy)
* @mspertick) / 1000)),
cpu_busy_percentage = convert(int, ((((@cpu_busy - cpu_busy)
* @mspertick) / 1000) * 100) / @interval),
io_busy_total = convert(int, ((@io_busy * @mspertick) / 1000)),
io_busy_current = convert(int, (((@io_busy - io_busy)
* @mspertick) / 1000)),
io_busy_percentage = convert(int, ((((@io_busy - io_busy)
* @mspertick) / 1000) * 100) / @interval),
idle_total = convert(int, ((convert(bigint,@idle) * @mspertick) / 1000)),
idle_current = convert(int, (((@idle - idle)
* @mspertick) / 1000)),
idle_percentage = convert(int, ((((@idle - idle)
* @mspertick) / 1000) * 100) / @interval),
packets_received_total = @pack_received,
packets_received_current = @pack_received - pack_received,
packets_sent_total = @pack_sent,
packets_sent_current = @pack_sent - pack_sent,
packet_errors_total = @pack_errors,
packet_errors_current = @pack_errors - pack_errors,
total_read = @total_read,
current_read = @total_read - total_read,
total_write = @total_write,
current_write = @total_write - total_write,
total_errors = @total_errors,
current_errors = @total_errors - total_errors,
connections_total = @connections,
connections_current = @connections - connections
from master.dbo.spt_monitor
update master.dbo.spt_monitor
set
lastrun = @now,
cpu_busy = @cpu_busy,
io_busy = @io_busy,
idle = @idle,
pack_received = @pack_received,
pack_sent = @pack_sent,
connections = @connections,
pack_errors = @pack_errors,
total_read = @total_read,
total_write = @total_write,
total_errors = @total_errors
You can see that it uses a table named spt_monitor to hold last snapshot of the variables, when next time comes, use current variables to subtract the previous ones.
You can find the detail of the stored procedure here: http://technet.microsoft.com/en-gb/library/ms188912.aspx
How to show the information?
I uses the .NET 4.0 in-built Chart under System.Windows.Forms.DataVisualization. You can find samples here: http://archive.msdn.microsoft.com/mschart/Release/ProjectReleases.aspx?ReleaseId=4418
Analysis
Currently SQL Monitor can analyse disk/database/table/index space relationship and give possible suggestions.
How you did it?
First it finds out all databases, for each database, get size of all files (data, log), then use master.sys.xp_fixeddrives to find out free space for each disk. The it will get the total size of all database files and group them by disk that the file resides. Based on the ratio use defined in the options, SQL Monitor will decide whether a certain database need to shrink or truncate.
var databases = GetDatabasesInfo();
var files = new List<tuple<bool, />>();
databases.AsEnumerable().ForEach(d =>
{
var database = GetDatabaseInfo(d["name"].ToString());
database.AsEnumerable().ForEach(f =>
{
files.Add(new Tuple<bool, />(Convert.ToInt32(f["type"]) == 1, f["physical_name"].ToString(), Convert.ToInt64(Convert.ToDecimal(f["Size"]) / Size1K)));
}
);
});
var spaces = new Dictionary<string, />>();
var driveSpaces = Query("EXEC master.sys.xp_fixeddrives");
driveSpaces.AsEnumerable().ForEach(s =>
{
spaces.Add(s[0].ToString(), new KeyValue<long, />(Convert.ToInt64(s[1]), 0));
});
files.ForEach(f =>
{
try
{
var drive = f.Item2.Substring(0, 1);
if (spaces.ContainsKey(drive))
{
spaces[drive].Value += f.Item3;
}
}
catch (Exception)
{
}
});
spaces.ForEach(s =>
{
if (s.Value.Key < s.Value.Value / 100 * Settings.Instance.DatabaseDiskFreeSpaceRatio)
{
analysisResult.Add(new AnalysisResult { ResultType = AnalysisResultTypes.DiskFreeSpace, ObjectName = s.Key, ReferenceValue = s.Value.Key, CurrentValue = s.Value.Value, Factor = Settings.Instance.DatabaseDiskFreeSpaceRatio + SizePercentage });
}
});
databases.AsEnumerable().ForEach(d =>
{
var name = d["name"].ToString();
if (!systemDatabases.Contains(name))
{
var database = GetDatabaseInfo(name);
var databaseSpace = new Dictionary<databasefiletypes, /> { { DatabaseFileTypes.Data, 0 }, { DatabaseFileTypes.Log, 0 } };
database.AsEnumerable().ForEach(f =>
{
var key = (DatabaseFileTypes)Convert.ToInt32(f["type"]);
databaseSpace[key] += Convert.ToInt64(Convert.ToDecimal(f["Size"]) / Size1K);
}
);
bool? shrink = null;
if (databaseSpace[DatabaseFileTypes.Log] > databaseSpace[DatabaseFileTypes.Data] / 100 * Settings.Instance.DatabaseDataLogSpaceRatio)
shrink = false;
else
{
var logSpaces = SQLHelper.Query("DBCC SQLPERF(LOGSPACE)", GetServerInfo(name));
var logSpace = logSpaces.Select(string.Format("[Database Name] = '{0}'", name));
if (logSpace.Length > 0)
{
var logSpacedUsed = Convert.ToDouble(logSpace[0]["Log Space Used (%)"]);
if (logSpacedUsed < Settings.Instance.DatabaseDataLogSpaceRatio)
shrink = true;
}
}
if (shrink != null)
analysisResult.Add(new AnalysisResult { ResultType = AnalysisResultTypes.DatabaseLogSpace, ObjectName = name, ReferenceValue = databaseSpace[DatabaseFileTypes.Log], CurrentValue = databaseSpace[DatabaseFileTypes.Data], Factor = Settings.Instance.DatabaseDataLogSpaceRatio + SizePercentage, Key = (bool)shrink ? 1 : 0 });
}
});
For table space, system stored procedure called sp_spaceused to get the space usage of a table:
var tables = GetObjects(KeyTables);
tables.AsEnumerable().ForEach(t =>
{
var name = t[KeyName].ToString();
var space = Query(string.Format("EXEC sp_spaceused '{0}'", name), CurrentServerInfo);
if (space.Rows.Count > 0)
{
var row = space.Rows[0];
var dataSize = ToKB(row["data"]) / Size1K;
var indexSize = ToKB(row["index_size"]) / Size1K;
if (indexSize > dataSize / 100 * Settings.Instance.TableDataIndexSpaceRatio)
analysisResult.Add(new AnalysisResult { ResultType = AnalysisResultTypes.TableIndexSpace, ObjectName = name, ReferenceValue = dataSize, CurrentValue = indexSize, Factor = Settings.Instance.DatabaseDataLogSpaceRatio + SizePercentage, Key = (int)TableIndexSpaceRules.DataIndexSpaceRatio });
}
});
Points of Interest
Well, I really learned a lot during the digging (sounds like a construnction worker:D) into SQL Server, the deeper I dig, the more I realize SQL Server is very complicated, and powerful: it can easily handle over 10 billion records in one table. However, in my daily database development, I found quite a few problems with SQL Server (even in SQL Server 2008 R2), especially when it try to yield resources to other worker, it just hangs there not doing anything. Thus whhy I develop SQL Monitor, trying to figure out what actually going on inside SQL Server.
Please, stay tuned, in the next version, will be more exciting, ambitous, hardcore and comprehensive features coming.
Finally, Happy New Year :)
History
