转自:http://djb1008.itpub.net/post/42280/518420
一. 问题描述
在AIX 6.1+ORACLE 10.2.0.5 的多个数据库的udump,bdump 目录产生了多个trace文件,不断报警告:
WARNING:Could not lower the asynch I/O limit to 6144 for SQL direct I/O. It is set to -1
二. 问题分析
在metalink上搜索'WARNING:Could not lower the asynch I/O limit to',找到了一篇文章
"WARNING:Could not lower the asynch I/O limit to 224 for SQL direct I/O. It is set to -1" After Upgrading To 10.2.0.5 [ID 1155445.1]",其内容如下:
This document is being delivered to you via Oracle Support's Rapid Visibility (RaV) process and therefore has not been subject to an independent technical review.
Applies to:
Oracle Server - Enterprise Edition - Version: 10.2.0.5 and later [Release: 10.2 and later ]
This problem can occur on any platform.
Symptoms
Warning below appears after upgrading to 10.2.0.5
*** SESSION ID:(538.48) 2010-07-15 08:41:30.649
WARNING:Could not lower the asynch I/O limit to 224 for SQL direct I/O. It is set to -1
WARNING:Could not lower the asynch I/O limit to 192 for SQL direct I/O. It is set to -1
WARNING:Could not lower the asynch I/O limit to 224 for SQL direct I/O. It is set to -1
WARNING:Could not lower the asynch I/O limit to 192 for SQL direct I/O. It is set to -1
WARNING:Could not lower the asynch I/O limit to 160 for SQL direct I/O. It is set to -1
Changes
Recently upgraded to 10.2.0.5.
Cause
This seems due to bug 9772888 10.2.0.5 RDBMS 10.2.0.5 BUFFER CACHE PRODID-5 PORTID-23 Abstract: WARNING:COULD NOT LOWER THE ASYNCH I/O LIMIT TO 160 FOR SQL DIRECT I/O. IT IS SE
Same behavior. appears and on the same platform. 10.2.0.5
Solution
1. Upgrade to Oracle 11.2 and install on top patchset 11.2.0.2.0 (or higher version patchset) where issue is fixed
OR
2. Install one-off patch on top of 10.2.0.5.0 if available to your platform.
References
BUG:9772888 - WARNING:COULD NOT LOWER THE ASYNCH I/O LIMIT TO 160 FOR SQL DIRECT I/O. IT IS SE
三.问题解决
当前数据库因为应用系统的问题,不方便升级到11.2.0.2,这里采用打9772888 补丁的方法来解决.
问题解决操作步骤
1.在metalink上下载补丁p9772888_10205_AIX64-5L.zip
2.停止数据库实例,使用opatch升级补丁
3.启动数据库实例,检查报警信息不再出现,问题得到解决
另外:
如果是rac环境,需要执行下面的步骤:
1. Stop instances running on node 1.
2. Run 'opatch apply /9772888' on node 1.
3. When the apply finishes, opatch asks for confirmation
before going on to apply the patch to node 2.
4. Start instances on node 1.
5. Stop instances running on node 2.
6. Ask opatch to continue to applying the patch to node 2.
7. Start instances on node 2.
来自 “ ITPUB博客 ” ,链接:http://blog.itpub.net/12932950/viewspace-759697/,如需转载,请注明出处,否则将追究法律责任。
转载于:http://blog.itpub.net/12932950/viewspace-759697/