通过DMVS采集并存储SQL Server性能计数器数据
2024-04-07 21:31:23
通过DMVS采集并存储SQL Server性能计数器数据
问题描述
我想定期去采集并存储一些SQL Server性能计数器数据，因为服务器太多，又不想用Windows系统自带的性能监视器在每一台服务器上去部署。
解决方案
通过动态性能视图sys.dm_os_performance_counters，编写采集SQL代码，通过SQL Server Agent作业定期采集性能数据。
第一步：
创建注册服务器组，将要采集的服务器分组。可供后面批量部署脚本。
先通过下面的语句熟悉性能计数器的存储内容：
SELECT [object_name], [counter_name], [instance_name], [cntr_value]
FROM sys.dm_os_performance_counters
WHERE counter_name = 'Buffer cache hit ratio';第二步：
我利用一个专门的数据库在所有的被监控数据库服务器上来放置采集逻辑，包括表、视图、存储过程和函数，然后在每个数据库实例上用统一的规则运行。
这里，我创建iDBA数据库，以及一些处理sys.dm_os_performance_counters数据采集的对象：
· MetaBOT schema
· MetaBOT.dm_os_performance_counters表 -- 存储从sys.dm_os_performance_counters采集的结果
· MetaBOT.watched_counters表 – 存储打算采集的计数器
USE [iDBA];
GO
CREATE SCHEMA MetaBOT AUTHORIZATION dbo;
GO
IF NOT EXISTS (SELECT name FROM iDBA.sys.[tables] T WHERE name = 'dm_os_performance_counters')
CREATE TABLE [MetaBOT].[dm_os_performance_counters]
(
[object_name] NVARCHAR(128),
[counter_name] NVARCHAR(128),
[instance_name] NVARCHAR(128),
[cntr_value] bigint,
[date_stamp] DATETIME
);
IF NOT EXISTS (SELECT name FROM iDBA.sys.[tables] T WHERE name = 'watched_counters')
CREATE TABLE [iDBA].[MetaBOT].[watched_counters]
(
[object_name] NVARCHAR(128),
[counter_name] NVARCHAR(128),
[active] bit
);
GO然后初始化MetaBOT.watched_counters表，最后创建存储过程，通过SQL Server Agent作业去存储到MetaBOT.dm_os_performance_counters。关于命名实例在object_name列的影响，动态地通过基于对象名和实例名构建，通过数据库系统变量@@SERVICENAME：
--+-- Now populate the watched counters table based upon instance properties
DECLARE @NamedInstance bit
DECLARE @ObjectNamePrefix VARCHAR(50)
SELECT @NamedInstance = 1
IF @@SERVICENAME = 'MSSQLSERVER'
BEGIN
SELECT @NamedInstance = 0 --This is the default instance
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Buffer Manager', 'Buffer cache hit ratio', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Buffer Manager', 'Buffer cache hit ratio base', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Buffer Manager', 'Database pages', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Buffer Manager', 'Free pages', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Buffer Manager', 'Page life expectancy', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Buffer Manager', 'Page lookups/sec', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Buffer Manager', 'Page reads/sec', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Buffer Manager', 'Page writes/sec', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Buffer Manager', 'Reserved pages', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Buffer Manager', 'Stolen pages', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Buffer Manager', 'Target pages', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Buffer Manager', 'Total pages', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Cursor Manager by Type', 'Active cursors', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Databases', 'Active Transactions', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Databases', 'Data File(s) Size (KB)', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Databases', 'Log File(s) Size (KB)', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Databases', 'Log File(s) Used Size (KB)', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Databases', 'Percent Log Used', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Databases', 'Transactions/sec', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Exec Statistics', 'DTC calls', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Exec Statistics', 'OLEDB calls', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_namme], [counter_name], [active])
VALUES ('SQLServer:General Statistics', 'Transactions', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:General Statistics', 'User Connections', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Latches', 'Latch Waits/sec', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Latches', 'Total Latch Wait Time (ms)', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Locks', 'Average Wait Time (ms)', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Locks', 'Lock Wait Time (ms)', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Memory Manager', 'Target Server Memory (KB)', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Memory Manager', 'Total Server Memory (KB)', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Plan Cache', 'Cache Hit Ratio', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Plan Cache', 'Cache Hit Ratio Base', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Plan Cache', 'Cache Pages', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:SQL Statistics', 'Batch Requests/sec', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Transactions', 'Free Space in tempdb (KB)', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Transactions', 'Longest Transaction Running Time', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Transactions', 'NonSnapshot Version Transactions', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Transactions', 'Snapshot Transactions', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Transactions', 'Transactions', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Transactions', 'Update Snapshot Transactions', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Transactions', 'Version Store Size (KB)', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Wait Statistics', 'Lock waits', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Wait Statistics', 'Log buffer waits', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Wait Statistics', 'Log write waits', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Wait Statistics', 'Memory grant queue waits', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Wait Statistics', 'Network IO waits', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Wait Statistics', 'Non-Page latch waits', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Wait Statistics', 'Page IO latch waits', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Wait Statistics', 'Page latch waits', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Wait Statistics', 'Thread-safe memory objects waits', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Wait Statistics', 'Transaction ownership waits', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Wait Statistics', 'Wait for the worker', 1);
INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])
VALUES ('SQLServer:Wait Statistics', 'Workspace synchronization waits', 1);
END
ELSE
BEGIN--Account for named instance when adding object names to watchSELECT @ObjectNamePrefix = 'MSSQL$' + @@SERVICENAMEINSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Buffer Manager', 'Buffer cache hit ratio', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Buffer Manager', 'Buffer cache hit ratio base', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Buffer Manager', 'Database pages', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Buffer Manager', 'Free pages', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Buffer Manager', 'Page life expectancy', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Buffer Manager', 'Page lookups/sec', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Buffer Manager', 'Page reads/sec', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Buffer Manager', 'Page writes/sec', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Buffer Manager', 'Reserved pages', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Buffer Manager', 'Stolen pages', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Buffer Manager', 'Target pages', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Buffer Manager', 'Total pages', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Cursor Manager by Type', 'Active cursors', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Databases', 'Active Transactions', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Databases', 'Data File(s) Size (KB)', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Databases', 'Log File(s) Size (KB)', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Databases', 'Log File(s) Used Size (KB)', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Databases', 'Percent Log Used', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Databases', 'Transactions/sec', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Exec Statistics', 'DTC calls', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Exec Statistics', 'OLEDB calls', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':General Statistics', 'Transactions', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':General Statistics', 'User Connections', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Latches', 'Latch Waits/sec', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Latches', 'Total Latch Wait Time (ms)', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Locks', 'Average Wait Time (ms)', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Locks', 'Lock Wait Time (ms)', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Memory Manager', 'Target Server Memory (KB)', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Memory Manager', 'Total Server Memory (KB)', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Plan Cache', 'Cache Hit Ratio', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Plan Cache', 'Cache Hit Ratio Base', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Plan Cache', 'Cache Pages', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':SQL Statistics', 'Batch Requests/sec', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Transactions', 'Free Space in tempdb (KB)', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Transactions', 'Longest Transaction Running Time', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Transactions', 'NonSnapshot Version Transactions', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Transactions', 'Snapshot Transactions', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Transactions', 'Transactions', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Transactions', 'Update Snapshot Transactions', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Transactions', 'Version Store Size (KB)', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Wait Statistics', 'Lock waits', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Wait Statistics', 'Log buffer waits', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Wait Statistics', 'Log write waits', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Wait Statistics', 'Memory grant queue waits', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Wait Statistics', 'Network IO waits', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Wait Statistics', 'Non-Page latch waits', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Wait Statistics', 'Page IO latch waits', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Wait Statistics', 'Page latch waits', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Wait Statistics', 'Thread-safe memory objects waits', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Wait Statistics', 'Transaction ownership waits', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Wait Statistics', 'Wait for the worker', 1);INSERT INTO [MetaBOT].[watched_counters] ([object_name], [counter_name], [active])VALUES (@ObjectNamePrefix + ':Wait Statistics', 'Workspace synchronization waits', 1);
END你可能注意到我跟踪了大量的性能计数器，我期望部署的计数器能覆盖到问题所以我宁愿多采集而不会不够。根据你的环境和数据库的数量，你可能希望采集更少的计数器信息。一些计数器在你的数据库计数器中是1：1的关系。我基于我的实例去做计算，决定每行数据存储平均232字节；基于此，我然后评估多久运行一次我的数据采集作业。运行采集进程每10分钟，估计每月存储需要1.5GB。当每两分钟运行一次，每月存储增加到8GB，并且这只是单实例的情况。
现在创建要在SQL Server Agent作业里调用的存储过程。它是一个简单的INSERT通过sys.dm_os_performance_counters过滤行INNER JOIN关联MetaBOT.watched_counters表。我没有创建任何索引在这个表，因为表数据量太小。遵循规则一条没有被使用的索引是不需要的。
CREATE PROCEDURE MetaBOT.usp_collect_perfmon_counters AS
DECLARE @datestamp DATETIME
SELECT @datestamp = GETDATE()
INSERT INTO MetaBOT.[dm_os_performance_counters]
(
[object_name],
[counter_name],
[instance_name],
[cntr_value],
[date_stamp]
)
SELECT
DOPC.[object_name],
DOPC.[counter_name],
DOPC.[instance_name],
DOPC.[cntr_value],
@datestamp
FROM sys.[dm_os_performance_counters] DOPC
INNER JOIN iDBA.[MetaBOT].[watched_counters] WC
ON [DOPC].[object_name] = [WC].[object_name]
AND [DOPC].[counter_name] = [WC].[counter_name]
ORDER BY [object_name], [counter_name];第三步：
在每一个实例上创建用于采集性能计数器值的作业。
最简单的方式，是通过SQL Server Management Studio图形管理界面创建作业，然后生成脚本。我就是这样做的。然后，有一项需要去修改。你必需要么移除或注释掉job_id GUID的输出代码行。这是不需要的，会导致脚本失败。有一个变量@RunEveryXMinutes在代码中允许你设置作业的运行频率；设置这个值为作业运行的间隔分钟。
USE [msdb]
GO
/****** Object: Job [Metadata_Collect_SQL_Perfmon_Counters] Script Date: 12/09/2010 12:14:38 ******/
BEGIN TRANSACTION DECLARE @ReturnCode INT DECLARE @RunEveryXMinutes TINYINT
SELECT @ReturnCode = 0
SELECT @RunEveryXMinutes = 10
/****** Object: JobCategory [Tuning and Optimization] Script Date: 12/09/2010 12:14:39 ******/
IF NOT EXISTS (
SELECT name
FROM msdb.dbo.syscategories
WHERE name=N'Tuning and Optimization'
AND category_class=1)
BEGIN EXEC @ReturnCode = msdb.dbo.sp_add_category @class=N'JOB', @type=N'LOCAL', @name=N'Tuning and Optimization'
IF (@@ERROR <> 0
OR @ReturnCode <> 0) GOTO QuitWithRollback
END DECLARE @jobId BINARY(16)
EXEC @ReturnCode = msdb.dbo.sp_add_job @job_name=N'Metadata_Collect_SQL_Perfmon_Counters',
@enabled=1,
@notify_level_eventlog=0,
@notify_level_email=0,
@notify_level_netsend=0,
@notify_level_page=0,
@delete_level=0,
@description=N'Poll and store output from sys.dm_os_performance_counters',
@category_name=N'Tuning and Optimization',
@owner_login_name=N'SPECTRUM-HEALTH\svcSQLNotify', @job_id = @jobId OUTPUT
IF (@@ERROR <> 0
OR @ReturnCode <> 0) GOTO QuitWithRollback
/****** Object: Step [Collect Metadata] Script Date: 12/09/2010 12:14:39 ******/
EXEC @ReturnCode = msdb.dbo.sp_add_jobstep @job_id=@jobId, @step_name=N'Collect Metadata',
@step_id=1,
@cmdexec_success_code=0,
@on_success_action=1,
@on_success_step_id=0,
@on_fail_action=2,
@on_fail_step_id=0,
@retry_attempts=0,
@retry_interval=0,
@os_run_priority=0, @subsystem=N'TSQL',
@command=N'EXEC MetaBOT.usp_collect_perfmon_counters;',
@database_name=N'iDBA',
@flags=0
IF (@@ERROR <> 0
OR @ReturnCode <> 0) GOTO QuitWithRollback
EXEC @ReturnCode = msdb.dbo.sp_update_job @job_id = @jobId, @start_step_id = 1
IF (@@ERROR <> 0
OR @ReturnCode <> 0) GOTO QuitWithRollback
EXEC @ReturnCode = msdb.dbo.sp_add_jobschedule @job_id=@jobId, @name=N'dm_os_perfmon_counters Job',
@enabled=1,
@freq_type=4,
@freq_interval=1,
@freq_subday_type=4,
@freq_subday_interval=@RunEveryXMinutes,
@freq_relative_interval=0,
@freq_recurrence_factor=0,
@active_start_date=20101209,
@active_end_date=99991231,
@active_start_time=30,
@active_end_time=235959
-- ,@schedule_uid=N'519325e6-3114-453c-bfeb-6597300a45d2' <---Comment this line out of auto-generated script
IF (@@ERROR <> 0
OR @ReturnCode <> 0) GOTO QuitWithRollback
EXEC @ReturnCode = msdb.dbo.sp_add_jobserver @job_id = @jobId, @server_name = N'(local)'
IF (@@ERROR <> 0
OR @ReturnCode <> 0) GOTO QuitWithRollback
COMMIT TRANSACTION GOTO EndSave QuitWithRollback:
IF (@@TRANCOUNT > 0)
ROLLBACK TRANSACTION
EndSave:
GO一旦启动作业，你就可以在MetaBOT.dm_os_performance_counters表查询相关计数。我推荐你设置清理命令，这已经被作为MetaBOT.usp_collect_perfmon_counters存储过程的一部分，去从MetaBOT.dm_os_performance_counters表删除date_stamp列小于你设置的阈值的的记录。根据你的性能需要，我也推荐在该列创建索引。
转载于:https://blog.51cto.com/ultrasql/1580103
通过DMVS采集并存储SQL Server性能计数器数据相关推荐
- [20141121]无法通过powershell读取sql server性能计数器问题
[20141121]无法通过powershell读取sql server性能计数器问题 背景: 全新服务器,需要增加性能监控,发现无法通过powershell读取性能指标 解决方法: Open the ...
- asp毕业设计——基于vb+VB.NET+SQL Server的数据存储与交换系统设计与实现（毕业论文+程序源码）——数据存储与交换系统
基于vb+VB.NET+SQL Server的数据存储与交换系统设计与实现(毕业论文+程序源码) 大家好,今天给大家介绍基于vb+VB.NET+SQL Server的数据存储与交换系统设计与实现,文章 ...
- SQL Server 迁移数据到MySQL
一.背景 由于项目开始时候使用的数据库是SQL Server,后来把存储的数据库调整为MySQL,所以需要把SQL Server的数据转移到MySQL:由于涉及的表比较多,所以想在MySQL中生成对应 ...
- 不同版本的SQL Server之间数据导出导入的方法及性能比较
原文:不同版本的SQL Server之间数据导出导入的方法及性能比较 工作中有段时间常常涉及到不同版本的数据库间导出导入数据的问题,索性整理一下,并简单比较下性能,有所遗漏的方法也欢迎讨论.补充. 0 ...
- 05.SQL Server大数据群集小试牛刀--HDFS查询
05.SQL Server大数据群集小试牛刀--HDFS查询 SQL Server大数据群集查询HDFS ,利用之前创建好的大数据群集,使用官方提供的测试数据进行测试.脚本是官方的脚本,要知道干了什么 ...
- 03.三月风雨愁断肠，学习大数据特别忧伤——谈谈AKS创建SQL Server大数据群集创建愁断肠
03.三月风雨愁断肠,学习大数据特别忧伤--谈谈AKS创建SQL Server大数据群集创建愁断肠 [TOC] 前言 使用AKS创建SQL Server大数据群集有几种方法,由于产品不断的更新,SQL ...
- SQL Server 2019 新特性之 SQL Server大数据群集 介绍（一）
SQL Server 2019 新特性之 SQL Server大数据群集 介绍(一) 从开始SQL Server 2019 预览,SQL Server 大数据群集允许你部署的 Kubernetes 上 ...
- 使用SSIS包导入SQL Server FILESTREAM数据
初始配置 (Initial configuration) We have been exploring the SQL Server FILESTREAM feature in this ongoin ...
- RDA实现SQL CE与SQL Server间数据存取
本文介绍如何利用 RDA(RemoteDataAccess远程数据存取)实现掌上电脑Pocket PC(简称PPC)上的SQL Server CE 2.0与桌面SQL Server 2000数据库间的 ...
最新文章
- mysql not in优化_MySQL性能优化 — 实践篇2
- java aio聊天_JAVA aio简单使用
- oracle 时间按季度,Oracle按日期季度查询
- cesiumjs开发实践(七) 3D模型
- 不实例化图片，获取图片宽高的方法（vb.net）
- 黑客利用智能灯泡窃取用户数据！
- Turn to Stone
- 如何在回调中访问正确的“ this”？
- 《统计学习方法》——朴素贝叶斯法
- 小白数据分析师的快速上手指南
- CodeForces - 729D
- kettle 使用教程（入门）
- 对象不支持“attachEvent”属性或方法的解决办法
- 电力系统微型计算机继电保护试题及答案,全国7月高中等教育自学考试电力系统微型计算机继电保护试题及答案...
- 电力系统决策支持系统
- 2019年计算机一级考试pdf,2019年计算机一级考试试题及答案.pdf
- HDU 3966 Aragorns Story
- 如何做一个基于JAVA医院预约挂号系统毕业设计毕设作品(springboot框架)
- Java图形界面文字乱码
- 网页设计语言html做思维导图,HTML思维导图