Verify that the iSCSI software packages are installed.

# pkginfo SUNWisciu SUNWiscsir
system SUNWiscsiu Sun iSCSI Device Driver (root)
system SUNWiscsir Sun iSCSI Management Utilities (usr)

Verify that you are running a Solaris release that supports the iSCSI protocol.

Solaris 10 7/05 release or late

% cat /etc/release
Solaris 10 7/05 X86
Copyright 2005 Sun Microsystems, Inc. All Rights Reserved.
Use is subject to license terms.
Assembled 03 March 2005

Solaris 10 release with the iSCSI patch

On a SPARC system:

# showrev -p | grep 119090

On an x86 system:

# showrev -p | grep 119091

Configure the SendTargets device discovery method or the static discovery method:

# iscsiadm add discovery-address 10.0.0.1:326

Enable the iSCSI target discovery method using one of the following:

# iscsiadm modify discovery --sendtargets enable

Create the iSCSI device links for the local system.

# devfsadm -i iscsi

Note: How to Remove Discovered iSCSI Targets

Disable an iSCSI target discovery method using one of the following:

iscsiadm modify discovery --sendtargets disable

Remove an iSCSI device discovery entry:

 iscsiadm remove discovery-address 10.0.0.1:3260

Accessing iSCSI Disks

After the devices have been discovered by the Solaris iSCSI initiator, the login negotiation occurs automatically. The Solaris iSCSI driver determines that the number of LUNs available and creates the device nodes. Then, the iSCSI devices can be treated as any other SCSI device.

You can view the iSCSI disks on the local system with the format utility.

AVAILABLE DISK SELECTIONS:
       0. c0t0d0 <SUN146G cyl 14087 alt 2 hd 24 sec 848>
          /pci@780/pci@0/pci@9/scsi@0/sd@0,0
       1. c0t1d0 <SUN146G cyl 14087 alt 2 hd 24 sec 848>
          /pci@780/pci@0/pci@9/scsi@0/sd@1,0
       2. c0t2d0 <SUN146G cyl 14087 alt 2 hd 24 sec 848>
          /pci@780/pci@0/pci@9/scsi@0/sd@2,0
       3. c0t3d0 <SUN146G cyl 14087 alt 2 hd 24 sec 848>
          /pci@780/pci@0/pci@9/scsi@0/sd@3,0
       4. c2t6090A078407041F52DE03441A471CD96d0 <EQLOGIC-100E-00-5.0 cyl 65533 alt 2 hd 16 sec 200>
          /scsi_vhci/ssd@g6090a078407041f52de03441a471cd96

Make UFS filesystem and mount it.

newfs /dev/dsk/c2t6090A078407041F52DE03441A471CD96d0/s6
mount c2t6090A078407041F52DE03441A471CD96d0/s6 /mnt/iscsi

User Creation and Environment Settings

1)Create groups for Oracle account

#groupadd oinstall
#groupadd dba
#groupadd oper

2)Create Oracle Default Home directory

# mkdir /export/home
# mkdir /export/home/oracle

3)Create Oracle user

# useradd -g oinstall -G dba -d /export/home/oracle -s /usr/bin/bash oracle
# chown oracle:oinstall /export/home/oracle

4)Create Project for Oracle for setting the kernel parameters

In case of Solaris 10, you can use projects to configure the kernel parameters instead of /etc/system file. This can be done as following

# projadd -U oracle -K "project.max-shm-memory=(priv,4g,deny)" oracle
# projmod -sK "project.max-sem-nsems=(priv,256,deny)" oracle
# projmod -sK "project.max-sem-ids=(priv,100,deny)" oracle
# projmod -sK "project.max-shm-ids=(priv,100,deny)" oracle

There are many more ways of creating project entries such as group.group-name or user.user-name. For more details refer to Solaris Administration documents.

Last three settings made by projmod command are not required as these values are lower than the default. This was pointed by Mike Madland and he also gave a Sun documentation link

http://docs.sun.com/app/docs/doc/819-2724/6n50b0795?l=en&a=view#chapter1-33

You can check the values for max-sem-ids and max-shm-ids with this command:

prctl -n project.max-sem-ids -i task `ps -o taskid= -p $$`

5)Create .bash_profile for Oracle user

#Oracle Environment Settings
TMP=/tmp; export TMP
TMPDIR=$TMP; export TMPDIR
ORACLE_BASE=/opt/oracle; export ORACLE_BASE
ORACLE_HOME=$ORACLE_BASE/product/11g; export ORACLE_HOME
ORACLE_SID=TESTDB11G; export ORACLE_SID
PATH=$PATH:/usr/local/bin:/usr/ccs/bin:/usr/sfw/bin:$ORACLE_HOME/bin

Now Set the Display to a X-windowing enabled system.

$ export DISPLAY=192.168.4.47:0.0

Also allow the host to accept the connection by

$xhost +

Oracle Software Installation

Go to the Oracle dump location and run runInstaller as Oracle user

$./runInstaller

Oracle 11g R2 Installation Error with Patch 124861-15

The Installation of 11gR2 is not proceeding due to missing OS patch 124861-15 on Solaris 10.

When we are trying to Install this patch it gives error:

bash-3.00# patchadd 124861-15
Validating patches...
Loading patches installed on the system...
Done!
Loading patches requested to install.
Version of package SPROlang from directory SPROlang in patch 124861-15 differs from the package installed on the system.
Version of package SPROlangx from directory SPROlangx in patch 124861-15 differs from the package installed on the system.
Version of package SPROsunms from directory SPROsunms in patch 124861-15 differs from the package installed on the system.
Version of package SPROmrcom from directory SPROmrcom in patch 124861-15 differs from the package installed on the system.
Done!
The following requested patches do not update any packages installed on the system
No Packages from patch 124861-15 are installed on the system.
No patches to dependency check.

The packages included in this patch are already Installed and with higher version. These packages are bundled with Sun Studio 12.1 and 12.2 , if you Installed Sun Studio 12.0 then it will allow you to Install the required patch 124861-15. This patch is for C compiler which is already installed with Sun Studio 12.x. So here Installing this patch doesn't make any sense as we already have these packages with higher version.

You can verify the packages bundled with this patch installed on your system or not:

bash-3.00# pkginfo -i SPROlang SPROlangx SPROsunms SPROmrcom
application SPROlang  Sun Studio 12 update 1 Compilers Common Components
application SPROlangx Sun Studio 12 update 1 Common 64-bit Components
application SPROmrcom Sun Studio 12 update 1 Common Compiler Man Pages/Online Info
application SPROsunms Sun Studio 12 update 1 Unbundled Shared libsunmath

Is these packages are Installed on your system, then you can avoid installing 124861-15 patch.

Here how you will avoid Installing this patch ?????

Just you need to comment pre-requiste file from 11g stage area.

11gR2stage/stage/cvu/cvu_prereq.xml

Find entry for patch 124861-15 :

original entry
!-- C Compiler Patches --
OSPATCH VALUE="124861-15" SEVERITY="IGNORABLE"     

Update Entry
!-- C Compiler Patches --
!-- OSPATCH VALUE="124861-15" SEVERITY="IGNORABLE" --

save this file and re-run the Installer it will not check this patch as a pre-requisite.

如果是通过Console口访问ALOM（Advanced Lights Out Manager），在启动系统时，在ALOM的命令行提示符下，输入break命令，就可以将系统重新引导到OK提示符状态。

• #. -- 退出Console
• sc> break -- 重新引导到OK提示符状态
• sc> console -- 进入Console

如果已经安装了操作系统，那么init 0或者shutdown命令会重启系统进入OK提示符状态。

如果接了键盘，那么L1-A (Stop-A) 键或Break键也可以进入OK提示符状态。

加注：OK提示符状态其实是进入OpenBoot PROM。

设置root密码：

set /SP/users/root password=

访问串口：

start /SP/console

注：Sun X4240 控制台是图形方式的，所以串口访问的话非常不清楚，通常我们通过web方式通过Applet访问。

设置IP地址以后，可以通过WEB界面来管理或者远程控制服务器。

----------------------------
第一章 AIX开机启动过程介绍
----------------------------
不从最开始说起了，只说从AIX kernel被装载到内存开始。
  1、AIX系统存储在BLV的kernel在内存中创建/(root),/usr/和/var文件系统。
       要注意，这里所说的文件系统，包括kernel在启动过程中都是存储在RAM（内存）中的。
       他们没有权限访问BLV之外的任何内容。
  2、文件系统被装载到RAM之后，kernel开始执行init进程，这个时候init进程开始接管系统启动进程。
  3、AIX kernel把init进程的PID设为1，这个进程是作为parent,root进程存在的，是AIX系统中运行的其他所有进程的父进程.当init进程被装载并在启动过程运行,init调用rc.boot.
  4、rc.boot文件在AIX系统启动过程中有三个重要的作用。       
      （1）、第一部分是rc.boot初始化了系统的硬件，使其为系统的启动做好准备。通过系统设置命令cfgmgr，一小部分系统启动需要的驱动被配置。       
      （2）、第二部分是文件系统/,/usr，/var和页面空间被装载。在这些文件系统被装载以后，init被硬盘上的PID为1的init进程所替代，同时RAM被清空。      

      （3）、第三部分和最后部分，rc.boot,实际的init进程是从硬盘上运行的。

              当init被执行后，/etc/inittab文件被堵去，同时里面的每一条都被执行。

              在这个过程中，/tmp文件系统正被装载到硬盘上。

              现在，系统正在启动的最后过程中，cfgmgr命令再一次运行把剩下的在第一部分没有被rc.boot运行的驱动运行起来。    
    5、init进程被执行以后，下一步就是init会打开/etc/inittab，并且执行里面的内容。

SQL Server

安装SQL Server 2008 Enterprise版本，将数据验证方式设置为SQL Server and Windows Authentication Mode，因为vCenter和SQL Server并不在一台服务器上，所以需要通过SQL Server账户验证。

为了避免Transaction Log占用过多的磁盘空间，建议将数据库的Recovery Mode改为Simple。参考： SQL Server Recovery Model Affects Transaction Log Disk Space Requirements

32bit ODBC DSN

To set up a 32-bit DSN launch the 32-bit version of the Data Source Administrator. It is located at:

systemdrive\Windows\SysWoW64\Odbcad32.exe

注意：在64位操作系统中，默认的ODBC DSN是64位的，而VMware Update Manager只支持32位的ODBC，在开始->运行中输入以上命令，开启32位ODBC配置界面。

error: database unavailable or has network problems

VMware Update Manager是以Plug-in的方式，整合在vSphere Client中，打开菜单Plug-ins -> Manage Plug-ins，选中VMware vCenter Update Manager Extension，然后Enable。

如果Enable Plug-in成功，会多出来一个Update Manager的页面，可以对Host或者Cluster安装补丁或者升级。

在搜寻流量控制软件时，发现了Panabit这款相当不错的软件。Panabit软件运行在FreeBSD操作系统之上，支持Live CD和一键安装到本地硬盘的方式。对于一个基本不懂Unix的用户，安装起来也不会又太大的困难。

下载安装ISO文件，刻成光盘，使用光盘启动，启动之后输入root用户名，口令root，即FreeBSD提示界面，输入：
Panabit8#./setup回车 (大约1分钟左右，时间视硬盘大小格式化时间，安装过程首先是自动查找盘，显示查到的磁盘设备名，本机使用建议选择安装在第一个盘，盘可以是硬盘、CF卡、U盘等(盘容量要大于256M，建议512M以上。其中主要过程是运行自动分区、格式化文件系统、安装精简FreeBSD 8.0、安装Panabit。)

安装成功后，重启机器，即可https登录管理界面https://X.X.X.X，Web管理用户名admin，口令panabit。

进入Web管理界面，第一件事选择"网络配置"-->"数据接口"配置网桥，如选择网桥1，将一个网卡定义为接内网，将另一个网卡定义为接外网，分别点击提交即可，配置网桥2、网桥3、网桥4重复同样步骤，如下图：

Gentoo系统安装OpenVPN前，建议先将系统做个Update，然后安装OpenVPN，安装时注意选上examples。

emerge -uDN -av world
emerge -av openvpn

证书

OpenVPN自带的ESAY RSA工具，可以方便的创建根证书与服务器和客户端的密钥，工具安装在/usr/share/openvpn/easy-rsa目录。先编辑vars文件，设置变量的默认值。

读取环境变量，并清除已有的证书和密钥文件。

. vars
./clean-all

创建根证书

linode easy-rsa # ./build-ca
Generating a 1024 bit RSA private key
............++++++
......++++++
writing new private key to 'ca.key'

Country Name (2 letter code) [CN]:
State or Province Name (full name) [SH]:
Locality Name (eg, city) [Shanghai]:
Organization Name (eg, company) [LiaoJL]:
Organizational Unit Name (eg, section) []:
Common Name (eg, your name or your server's hostname) [LiaoJL CA]:
Name []:
Email Address [liaojl@liaojl.com]:

创建服务器证书

linode easy-rsa # ./build-key-server server
Generating a 1024 bit RSA private key
..........................++++++
..............................................++++++
writing new private key to 'server.key'

Country Name (2 letter code) [CN]:
State or Province Name (full name) [SH]:
Locality Name (eg, city) [Shanghai]:
Organization Name (eg, company) [LiaoJL]:
Organizational Unit Name (eg, section) []:
Common Name (eg, your name or your server's hostname) [server]:
Name []:
Email Address [liaojl@liaojl.com]:

Please enter the following 'extra' attributes
to be sent with your certificate request
A challenge password []:
An optional company name []:
Using configuration from /usr/share/openvpn/easy-rsa/openssl.cnf
Check that the request matches the signature
Signature ok
The Subject's Distinguished Name is as follows
countryName           :PRINTABLE:'CN'
stateOrProvinceName   :PRINTABLE:'SH'
localityName          :PRINTABLE:'Shanghai'
organizationName      :PRINTABLE:'LiaoJL'
commonName            :PRINTABLE:'server'
emailAddress          :IA5STRING:'liaojl@liaojl.com'
Certificate is to be certified until Apr  8 08:54:47 2021 GMT (3650 days)
Sign the certificate? [y/n]:y


1 out of 1 certificate requests certified, commit? [y/n]y
Write out database with 1 new entries
Data Base Updated

创建客户端证书

linode easy-rsa # ./build-key i9000
Generating a 1024 bit RSA private key
.........................++++++
..............................++++++
writing new private key to 'i9000.key'

Country Name (2 letter code) [CN]:
State or Province Name (full name) [SH]:
Locality Name (eg, city) [Shanghai]:
Organization Name (eg, company) [LiaoJL]:
Organizational Unit Name (eg, section) []:
Common Name (eg, your name or your server's hostname) [i9000]:
Name []:
Email Address [liaojl@liaojl.com]:

Please enter the following 'extra' attributes
to be sent with your certificate request
A challenge password []:
An optional company name []:
Using configuration from /usr/share/openvpn/easy-rsa/openssl.cnf
Check that the request matches the signature
Signature ok
The Subject's Distinguished Name is as follows
countryName           :PRINTABLE:'CN'
stateOrProvinceName   :PRINTABLE:'SH'
localityName          :PRINTABLE:'Shanghai'
organizationName      :PRINTABLE:'LiaoJL'
commonName            :PRINTABLE:'i9000'
emailAddress          :IA5STRING:'liaojl@liaojl.com'
Certificate is to be certified until Apr  8 08:55:44 2021 GMT (3650 days)
Sign the certificate? [y/n]:y


1 out of 1 certificate requests certified, commit? [y/n]y
Write out database with 1 new entries
Data Base Updated

将证书与密钥文件复制到/etc/openvpn目录。

cp -R keys /etc/openvpn

OpenVPN配置文件

将Keys目录中的四个证书密钥文件ca.crt, server.crt, server.key, dh1024.pem复制到/etc/openvpn目录。然后从/usr/share/doc/openvpn-2.1.4/examples/sample-config-files目录复制配置文件server.conf.bz2，解压缩并重命名为openvpn.conf，配置如下：

port 11994
proto udp
dev tun

# SSL/TLS
ca ca.crt
cert server.crt
key server.key

# Diffie hellman parameters
dh dh1024.pem

# VPN subnet
server 10.8.0.0 255.255.255.0
ifconfig-pool-persist ipp.txt
push "redirect-gateway def1 bypass-dhcp"

将默认端口改为11994，高端端口比较不容易被防火墙阻挡。

防火墙的配置

客户端拨入系统后，只建立了客户端到Linode主机之间的网络连接，要想实现翻墙或者隐藏IP的功能，还需要在Linode主机上通过iptables启动NAT功能。

iptables -A FORWARD -m state --state RELATED,ESTABLISHED -j ACCEPT
iptables -A FORWARD -s 10.8.0.0/24 -j ACCEPT
iptables -A FORWARD -j REJECT
iptables -t nat -A POSTROUTING -s 10.8.0.0/24 -o eth0 -j MASQUERADE

DNS

如果你希望将客户端的DNS设置为Linode主机（10.8.0.1），然后安装dnsmasq包 ，并且在OpenVPN配置文件中，PUSH DNS记录。

push "dhcp-option DNS 10.8.0.1"

在Linode主机上开启DNSMASQ服务：

/etc/init.d/dnsmasq start
rc-update add dnsmasq default

或者，也可以直接用Google的DNS服务器，推荐使用外部DNS而不是DNSMASQ方式：

push "dhcp-option DNS 8.8.8.8"

客户端配置

下载 OpenVPN Community Project ，然后将ca.crt, i9000.crt, i9000.key文件复制到OpenVPN安装目录的config子文件夹。编辑client.ovpn配置文件：

client
dev tun
proto udp
remote linode.liaojl.com 11994

ca ca.crt
cert i9000.crt
key i9000.key

comp-lzo
verb 3

双击client.ovpn文件，或者通过OpenVPN GUI连接到VPN服务器。

1. 将两个控制器中的其中一块拔出，避免重置密码时，两边控制器信息不同步导致冲突。
2. 通过串口线连接到控制器，重启控制器，按Ctrl+P进入安装模式。
3. 设置RESETPASSWORD变量，然后reload控制器。

CFE> setenv RESETPASSWORD 1
CFE> reload

1. 默认密码是grpadmin，登录系统后重置密码。

> account select grpadmin passwd
Enter New Password: xxxxxxx
Retype password: xxxxxxx

确保串行连接具备以下特性：
• 9600 波特
• 1 个停止位
• 无奇偶校验
• 8 个数据位
• 无硬件流控制

下面附上Dell的原文档，写的更加详细：

If you are unable to log in to a PeerStorage group and perform group modifications because you do not know the password for an account that has read-write permission, you can temporarily reset the grpadmin account password to the default factory-set password, which is grpadmin. This gives you the opportunity to log in to the group with read-write permission. You can then change the grpadmin account password to a known value.

Be sure that no other group administrator is changing the grpadmin account password when you are following the password recovery procedure, because this will cause the password reset to fail.

Note: Because you must power-cycle one group member as part of the password reset procedure, volumes with data stored on that member will be unavailable and active iSCSI connections to those volumes will be lost until the member is restarted. You may want to warn users of any impending offline volumes and iSCSI disconnections before resetting the password.

To temporarily reset the grpadmin account password to the default factory-set password, follow these steps:

1. On one group member, connect a Micro-D adapter cable and null modem cable to serial port 0 on the active control module and to a console. The active control module is indicated by the green control module status LED labeled ACT, which is located on the left side of the control module, to the left of the Ethernet port 0 connector.
2. Turn off power to the member (if you have dual power supplies, turn off both power supplies). Volumes with data located on the member will be offline and iSCSI connections to those volumes will be lost until the member is restarted.
3. If the member has two control modules, after it is shut down, locate the lock on the control module that is not connected to the Micro-D adapter. This is the control module that was the secondary control module, not the active control module. The lock is on the far left side of the control module, to the left of the text for the three status LEDs. Loosen the Phillips screw that holds the lock in place, and then slide the lock to the right so that it partially covers the LED text. This will temporarily disable the control module and prevent it from interfering with the password reset procedure.
4. Turn on power to the member by turning on all the power supplies.
5. While the member is restarting, press Ctrl/p when the following message appears on the console:

Press Ctrl/p to enter setup mode.

This will halt the boot process and allow you to enter commands to the boot monitor.
6.At the CFE> prompt, enter the following commands, which are case-sensitive and must be typed exactly as shown:

CFE> setenv RESETPASSWORD 1
CFE> reload

When the member restarts, the PeerStorage login prompt appears. After a short pause, the following message should appear:

WARNING:Password recovery mode... Temporarily resetting grpadmin password.

If you get the previous message, log in the the group using the grpadmin account and its factory-set password, grpadmin:

Login: grpadmin
Password: grpadmin

At this point, you are logged in to the group with read-write permission and can perform any group administration task. At this point, you should set the grpadmin account password to a known value. Use the procedure described in the Modifying Accounts section in the Group Administration manual or use the following command:

> account select grpadmin passwd
Enter New Password: xxxxxxx
Retype password: xxxxxxx

Note: Unless you set the grpadmin account password within five minutes after the password recovery mode message appears, the grpadmin password will revert back to the value it had prior to the password reset procedure (that is, you will no longer be able to log in with the factory-set password). However, as long as you remain logged in to the grpadmin account, you can set the password to a known value.

If the password recovery mode message does not appear, the password reset procedure did not succeed (for example, because you did not enter the CFE commands exactly as shown). In this case, allow the member to completely start up, and then retry the password reset procedure, shutting down the member and following the steps above.

If the member has two control modules, after logging in to the group and setting the grpadmin password, slide the lock on the disabled control module to the left. Tighten the Phillips screw gently to secure the lock. This restarts the secondary control module and allows it to be used.

Within one minute, you should see a console message, indicating that the secondary control module is operational.

To install VMWare ESX in BL460c G7, you need download network adapter driver as below:

http://www.bitcon.be/?p=796

The HP ESX and ESXi 4.0 U2 and 4.1 images that support the ProLiant BL465c G7, BL685c G7 and BL460c G7 are posted at the following link:

https://h20392.www2.hp.com/portal/swdepot/displayProductsList.do?category=SVIRTUAL

As a reminder, the Emulex (ServerEngines) driver that enables the BE2 & BE3 devices on the BL465c G7, BL685cG7 and BL460c G7 has been posted to the VMware website. To get this driver:

1. Go to: http://downloads.vmware.com/d/info/datacenter_downloads/vmware_vsphere_4/4#drivers_tools
2. Select "Driver CDs"
3. Select VMware ESX/ESXi 4.x Driver CD for ServerEngines BladeEngine 10Gb Ethernet Controller version 2.102.440.0

The HP Offline Bundle for VMware ESXi and the HP Management SNMP Agents for VMware ESX 4.0 U2 and 4.1 are posted at the following links:

Blade Servers links:
http://h20180.www2.hp.com/apps/Nav?h_pagetype=s-001&h_lang=en&h_cc=us&h_product=241641&h_client=S-A-R163-1&h_page=hpcom&lang=en&cc=us

ML/DL links:
http://h20180.www2.hp.com/apps/Nav?h_pagetype=s-001&h_lang=en&h_cc=us&h_product=241435&h_client=S-A-R163-1&h_page=hpcom&lang=en&cc=us

Also, HP has posted a VMware driver for our PCIe IO Accelerator:
http://h20000.www2.hp.com/bizsupport/TechSupport/SoftwareDescription.jsp?lang=en&cc=us&prodTypeId=3709945&prodSeriesId=4151345&prodNameId=4157323&swEnvOID=4040&swLang=13&mode=2&taskId=135&swItem=MTX-5263b0cf1350472a8a1930dda4

先是怀疑是不是存储空间满了，不过DataStore剩余空间还很多，应该是VM文件本身超出了大小限制。在创建DataStore的时候，选的Block Size是1MB，因此单个文件的大小限制是256GB。

VMFS Size Limitations

还好公司新装的ESX Server都选择了4MB的Block Size，支持的最大文件大小达到1TB。经过与VMware TS确认，将文件传到支持超过256GB文件的ESX Server上，VM就应该能正常启动。

通过SCP传文件时，发现从一台ESX Server不能访问另外一条ESX Server的SSH端口。查下来是防火墙阻止了SSH OUT访问，最简单的解决方法是关闭防火墙服务。

/etc/init.d/iptables stop

将文件复制过来以后，执行下面的指令，注册VM：

vmware-cmd -s register name.vmx

最后，记得将防火墙重新打开。

系统默认已经安装了gmake和gcc，不过并没有将其加入$PATH环境变量中。

# pkginfo  | grep gmake
system      SUNWgmake gmake - GNU make

# pkginfo | grep gcc
system      SUNWgcc                          gcc - The GNU C compiler
system      SUNWgccruntime                   GCC Runtime libraries

将GNU相关包（gmake/gcc等）加入$PATH环境变量中：

PATH=/usr/local/bin:$PATH:/usr/ccs/bin:/usr/sfw/bin

系统默认安装的Perl 5.8.4版本，即使安装了gmake和cc还是无法正常安装模块，如果下载源代码直接手工编译安装倒是没有问题，最后只能到SunFreeware网站下载Perl 5.8.8版本才能正常工作。

pkgadd -d perl-5.8.8-sol10-sparc-local

并将Perl的路径添加到$PATH环境变量中，注意要将Perl的安装路径/usr/local/bin加在最前面，编辑/etc/profile文件：

PATH=/usr/local/bin:$PATH

至此，系统环境已经配置好，打开CPAN的Shell命令，先更新bundle::cpan模块。

# perl -MCPAN -e shell
cpan> install bundle::cpan
cpan> reload cpan

然后安装DBD::mysql等模块：

install DBI
DBD::mysql