在Linux 中为oracle配置hugepage的方法

在Linux 中为oracle配置hugepage的方法  

在Linux中配置hugepage可以提高oracle的性能，减少oracle sga的页交换，类似于aix中的lagepage。

下面演示配置hugepage的方法。

系统环境：

OS：CentOS5 x86

Oracle：10.2.0.1

配置过程  18010133657 

检查hugepagesize

ora_test@root[/root]> grep Hugepagesize /proc/meminfo

Hugepagesize: 4096 kB

配置分配hugepage的数量

nr_hugepages的计算公式:nr_hugepages>=sga(mb)/Hugepagesize(mb)

ora_test@root[/root]> echo 50 > /proc/sys/vm/nr_hugepages

ora_test@root[/root]> sysctl -w vm.nr_hugepages=50

vm.nr_hugepages = 50

ora_test@root[/root]> echo "vm.nr_hugepages=50" >> /etc/sysctl.conf

检查hugepage的分配情况

ora_test@root[/root]> grep HugePages_Total /proc/meminfo

HugePages_Total: 6

这里显示的值为6，小于我配置的数量50，所以需要重启

ora_test@root[/root]> reboot

Broadcast message from root (pts/1) (Mon Jan 25 09:38:04 2010):

The system is going down for reboot NOW!

重启之后，检查hugepage的分配情况

ora_test@root[/root]> grep HugePages_Total /proc/meminfo

HugePages_Total: 50

现在确认hugepage已经分配了50个单位（50×4096kb=200mb）
http://cnhtm.itpub.net/post/39970/496273
配置oracle用户的memlock

需要修改/etc/security/limits.conf文件，加入标记为红色的两行

计算公式为：>=HugePages_Total×1024，我这里设置了2倍的值，即：2×50×1024=102400

ora_test@root[/root]> cat /etc/security/limits.conf|grep lock

# - memlock - max locked-in-memory address space (KB)

# - locks - max number of file locks the user can hold

oracle soft memlock 102400

oracle hard memlock 102400

检查memlock情况

切换到oracle用户身份，使用ulimit -l命令检查memlock情况

ora_test@root[/root]> su - oracle

ora_test@oracle[/home/oracle]> ulimit -l

102400

启动数据库

ora_test@oracle[/home/oracle]> sqlplus / as sysdba

SQL*Plus: Release 10.2.0.1.0 - Production on Mon Jan 25 09:50:33 2010

Copyright (c) 1982, 2005, Oracle. All rights reserved.

Connected to an idle instance.

idle> startup

ORACLE instance started.

Total System Global Area 167772160 bytes

Fixed Size 1218292 bytes

Variable Size 67111180 bytes

Database Buffers 92274688 bytes

Redo Buffers 7168000 bytes

Database mounted.

Database opened.

idle> exit

Disconnected from Oracle Database 10g Enterprise Edition Release 10.2.0.1.0 - Production

With the Partitioning, OLAP and Data Mining options

检查是否已经使用了hugepage

ora_test@oracle[/home/oracle]> grep HugePages_Free /proc/meminfo

HugePages_Free: 30

这里显示剩余的hugepage为30，小于HugePages_Total（50），证明oracle已经使用了hugepage

--end--

为何oracle一起来，系统就会有swap？

db是11g，total memory 是16g ，swap 是8g，分给sga_target=8g,pga_aggregate_target=3g.按理说，这个内存分配也不会导致oracle吞噬过多内存导 致系统发生swap，可现在只要db起来，用vmsta 命令观察就会出现swap， 类似如下。

procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu------

r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa st

0  0 168524  87336    340  26356  128 3014   178  3054 1305  417  0  1 94  5  0

0  3 168524  81336    480  29052 1798    0  3166     8 1243  687  0  0 94  6  0

0  0 171024  85292    496  28984  880 1500  1176  1500 1225  612  0  0 89 11  0

0  0 171024  85292    496  28960   24    0    24     0 1055  348  0  0 99  0  0

0  3 172284  81960    512  34860  372 1158   430  1230 1307 1496  0  2 94  4  0

0  1 167072  84424    320  38828 1564  210  1716   230 1545 2796  0  3 85 12  0

0  1 167072  84008    340  39152  286    0   362    18 1115  504  0  0 96  4  0

0  0 167072  83764    352  39392   34    0    94     6 1069  344  0  0 99  0  0

0  0 167072  83764    352  39580   32    0    32     0 1072  347  0  0 100  0  0

0  0 167072  82896    352  39568  410    0   410     0 1075  367  0  0 99  1  0

我看了alert log,没有其他报错。现在系统非常慢，大伙帮我分析分析。

QUOTE:

原帖由 oracle_kai 于 2010-12-7 18:28 发表 

系统使用的就是hugepage，但问题也就出现在hugepage 参数的设置上面。

原先vm.nr_hugepages = 41984  

cat /proc/memory|grep HugePages_Total

HugePages_Total=7925

7925*2m 几乎就是整个os内存了。

后来修改该参数，参考sga=8g，所以新设vm.nr_hugepages =4096

重启机器，oracle起来后memory swap 消失，系统反应也正常。

#!/bin/bash

#

# hugepages_settings.sh

#

# Linux bash script to compute values for the

# recommended HugePages/HugeTLB configuration

#

# Note: This script does calculation for all shared memory

# segments available when the script is run, no matter it

# is an Oracle RDBMS shared memory segment or not.

#

# This script is provided by Doc ID 401749.1 from My Oracle Support

# http://support.oracle.com
# Welcome text

echo "

This script is provided by Doc ID 401749.1 from My Oracle Support

(http://support.oracle.com) where it is intended to compute values for

the recommended HugePages/HugeTLB configuration for the current shared

memory segments. Before proceeding with the execution please make sure

that:

* Oracle Database instance(s) are up and running

* Oracle Database 11g Automatic Memory Management (AMM) is not setup

   (See Doc ID 749851.1)

* The shared memory segments can be listed by command:

     # ipcs -m

Press Enter to proceed..."

read

# Check for the kernel version

KERN=`uname -r | awk -F. '{ printf("%d.%d\n",$1,$2); }'`

# Find out the HugePage size

HPG_SZ=`grep Hugepagesize /proc/meminfo | awk '{print $2}'`

# Initialize the counter

NUM_PG=0

# Cumulative number of pages required to handle the running shared memory segments

for SEG_BYTES in `ipcs -m | awk '{print $5}' | grep "[0-9][0-9]*"`

do

   MIN_PG=`echo "$SEG_BYTES/($HPG_SZ*1024)" | bc -q`

   if [ $MIN_PG -gt 0 ]; then

      NUM_PG=`echo "$NUM_PG+$MIN_PG+1" | bc -q`

   fi

done

RES_BYTES=`echo "$NUM_PG * $HPG_SZ * 1024" | bc -q`

# An SGA less than 100MB does not make sense

# Bail out if that is the case

if [ $RES_BYTES -lt 100000000 ]; then

   echo "***********"

   echo "** ERROR **"

   echo "***********"

   echo "Sorry! There are not enough total of shared memory segments allocated for

HugePages configuration. HugePages can only be used for shared memory segments

that you can list by command:

   # ipcs -m

of a size that can match an Oracle Database SGA. Please make sure that:

* Oracle Database instance is up and running

* Oracle Database 11g Automatic Memory Management (AMM) is not configured"

   exit 1

fi

# Finish with results

case $KERN in

   '2.4') HUGETLB_POOL=`echo "$NUM_PG*$HPG_SZ/1024" | bc -q`;

          echo "Recommended setting: vm.hugetlb_pool = $HUGETLB_POOL" ;;

   '2.6') echo "Recommended setting: vm.nr_hugepages = $NUM_PG" ;;

    *) echo "Unrecognized kernel version $KERN. Exiting." ;;

esac

# End