未知性能問題

Question

在我們的sql server box（X64機器用8核16G Ram）時，我們發現性能真的很差，在我們產生了一堆數據之後，有時候我們甚至無法將RDP發送到這個盒子裏，在SQL錯誤日誌的幾個錯誤消息如下：

*2009-06-26 12:11:09.92 spid63  Error: 14151, Severity: 18, State: 1. 
2009-06-26 12:11:09.92 spid63  Replication-Replication Distribution Subsystem: agent ##job## failed. The process could not access database 'db_STR' on server 'Test01'. 
2009-06-26 12:11:30.66 spid57  Replication-Replication Distribution Subsystem: agent ##job## scheduled for retry. The process could not connect to Subscriber 'Test01'. 
2009-06-26 17:42:01.48 spid54  Replication-Replication Distribution Subsystem: agent ##Job## scheduled for retry. TCP Provider: An operation on a socket could not be performed because the system lacked sufficient buffer space or because a queue was full. 
2009-06-26 17:50:05.96 spid69  Replication-Replication Distribution Subsystem: agent ##Job## scheduled for retry. TCP Provider: An operation on a socket could not be performed because the system lacked sufficient buffer space or because a queue was full. 
2009-06-26 17:59:04.64 spid84  Replication-Replication Distribution Subsystem: agent ##Job## scheduled for retry. TCP Provider: An operation on a socket could not be performed because the system lacked sufficient buffer space or because a queue was full. 
2009-06-26 18:00:02.62 spid83  Error: 14151, Severity: 18, State: 1. 
2009-06-26 18:00:02.62 spid83  Replication-Replication Distribution Subsystem: agent ##Job## failed. Not enough storage is available to process this command. The step failed. 
2009-06-26 18:00:02.76 Server  Error: 26040, Severity: 17, State: 1. 
2009-06-26 18:00:02.76 Server  Server TCP provider has stopped listening on port [ 1433 ] due to a failure. Error: 0x2747, state: 2. The server will automatically attempt to reestablish listening. 
2009-06-26 18:00:03.17 spid94  Error: 14151, Severity: 18, State: 1.* 

我檢查所有磁碟機和所有有充足的自由空間，內存和CPU使用率都看起來不錯，內存上升到15G和CPU使用率是正常的少40％。我懷疑是頁面/非頁面池泄漏，但是對於x64機器，非頁面池的限制是> 6G，它只用了大約100M，有人知道系統有什麼問題嗎？

謝謝

Answer 1

看起來您正在生成太多的數據庫處理請求。

Answer 2

這也可能是IO滯後問題。

檢查以下頁面：

• Latches - Obtaining Information
• This might help too

Answer 3

你已經配置了RAM限制/即最大內存設置SQL服務器？如果不是，它可能會從操作系統中獲取更多的RAM，而不是有用的。我會嘗試將最大內存設置得足夠低，以便爲操作系統和其他進程提供幾GB的RAM。此外，你是企業與內存鎖定頁面或標準版？如果是標準的，並且SQL Server已經佔用了比它應該多的內存，它可能是分頁，這是一個全能殺手。

如果是這樣的問題，這是一個偉大的文章：

http://blogs.msdn.com/slavao/archive/2006/11/13/q-a-does-sql-server-always-respond-to-memory-pressure.aspx

另外這裏：

http://www.johnsansom.com/index.php/2009/03/sql-server-memory-configuration-determining-memtoleave-settings/

Answer 4

這可能是一個I/O損壞問題。 請看this MSDN forum post。