直到VMware ESX(i) 4.0 U2，官方的VMware-tools都未提供对FreeBSD 8.0 amd64的支持，因此改为安装Open Virtual Machine Tools。
[root@yaoge123 ~/vmware-tools-distrib]# uname -a
FreeBSD yaoge123 8.0-RELEASE-p3 FreeBSD 8.0-RELEASE-p3 #0: Tue May 25 20:54:11 UTC 2010 root@amd64-builder.daemonology.net:/usr/obj/usr/src/sys/GENERIC amd64
[root@www ~/vmware-tools-distrib]# ./vmware-install.pl
Creating a new VMware Tools installer database using the tar4 format.
Installing VMware Tools.
……
Before running VMware Tools for the first time, you need to configure it by
invoking the following command: “/usr/local/bin/vmware-config-tools.pl”. Do you
want this program to invoke the command for you now? [yes]
Initializing…
Stopping VMware Tools services in the virtual machine:
Guest operating system daemon: done
Unable to copy the source file
/usr/local/lib/vmware-tools/modules/binary/FreeBSD8.0-amd64/vmxnet.ko to the
destination file /boot/modules/vmxnet.ko.
Execution aborted.
[root@www ~/vmware-tools-distrib]# cd /usr/local/lib/vmware-tools/modules/binary/
[root@www /usr/local/lib/vmware-tools/modules/binary]# ls
FreeBSD6.0-amd64 FreeBSD6.3-amd64 FreeBSD7.0-amd64 FreeBSD7.3-amd64
FreeBSD6.0-i386 FreeBSD6.3-i386 FreeBSD7.0-i386 FreeBSD7.3-i386
首先确认是否安装了内核源代码，如果您的系统中没有 /usr/src/sys 这样一个目录， 则说明没有安装内核源代码。安装它最简单的方法是通过以 root 身份运行 sysinstall， 选择 Configure – Distributions – src，选中其中的 base 、sbin和sys。如果提示Warning: Can’t find the `8.0-RELEASE-p3′ distribution on this……，在Configure – Distributions – Options中修改Release Name为8.0-RELEASE，安装完成后用freebsd-update fetch install更新源代码。
然后再安装Open Virtual Machine Tools
cd /usr/ports/emulators/open-vm-tools-nox11
make install clean
至少支持了使用VMware Tools关机（这是HA所必不可少的）和vmxnet2网卡。
/usr/target /usr/mount-point nullfs rw,late 0 0
注意必须在Options中添加“late”，否则启动的时候报错。
编辑 /etc/login.conf 增加一个language
enutf8|enutf8 Users Accounts:\
:charset=UTF-8:\
:lang=en_US.UTF-8:\
:tc=default:
vipw 编辑ssh登录的用户，在gid后面增加这个language
yaoge123:$1$***:1001:1001:enutf8:0:0:User &:/home/yaoge123:/bin/sh
创建一个raidz1的ZFS pool
test# zpool create zfspool raidz da1 da2 da3
test# zpool list
NAME SIZE USED AVAIL CAP HEALTH ALTROOT
zfspool 23.9G 192K 23.9G 0% ONLINE -
test# zpool status
pool: zfspool
state: ONLINE
scrub: none requested
config:
NAME STATE READ WRITE CKSUM
zfspool ONLINE 0 0 0
raidz1 ONLINE 0 0 0
da1 ONLINE 0 0 0
da2 ONLINE 0 0 0
da3 ONLINE 0 0 0
errors: No known data errors
用更大的da4、da5、da6替换原来小的da1、da2、da3 (全文…)
创建一个非冗余的ZFS pool
test# zpool create zfspool da1
test# zpool list
NAME SIZE USED AVAIL CAP HEALTH ALTROOT
zfspool 7.94G 110K 7.94G 0% ONLINE -
test# zpool status
pool: zfspool
state: ONLINE
scrub: none requested
config:
NAME STATE READ WRITE CKSUM
zfspool ONLINE 0 0 0
da1 ONLINE 0 0 0
errors: No known data errors
增加一个盘，升级为双路mirror
test# zpool attach zfspool da1 da2
test# zpool list
NAME SIZE USED AVAIL CAP HEALTH ALTROOT
zfspool 7.94G 112K 7.94G 0% ONLINE -
test# zpool status
pool: zfspool
state: ONLINE
scrub: resilver completed with 0 errors on Tue Jul 21 21:24:27 2009
config:
NAME STATE READ WRITE CKSUM
zfspool ONLINE 0 0 0
mirror ONLINE 0 0 0
da1 ONLINE 0 0 0
da2 ONLINE 0 0 0
errors: No known data errors
如果想直接建立一个双路mirror的ZFS pool用
test# zpool create zfspool mirror da1 da2
再增加一个盘，升级为三路mirror (全文…)
HDChina和HDBits上可用的client有Azureus、 BitTornado、 KTorrent、 rtorrent、 Transmission、 uTorrent。uTorrent需要Wine；Azureus(Vuze)和KTorrent需要X，因此都不考虑了。BitTornado 0.3.18 HDChina不认，Transmission 1.61 HDBits不认。rtorrent里凡是没有下载完成的任务，重启后都需要重新hash，Transmission则没有该问题，而且且速度比rtorrent要快。难道用uTorrent + samba？
FreeBSD安装了VMware Tools以后，可以通过VI关闭客户机，但是系统停留在“The operating system has halted.Please press any key to reboot.”不能够自动关闭电源。解决方法为
ee /usr/local/etc/rc.d/vmware-tools.sh
查找vmware_start_guestd()可见
vmware_start_guestd() {
cd "$vmdb_answer_SBINDIR" && "$vmdb_answer_SBINDIR"/vmware-guestd \
--background "$GUESTD_PID_FILE"
}
在vmware-guestd命令下增加参数–halt-command “/sbin/shutdown -p now”，修改为
vmware_start_guestd() {
cd "$vmdb_answer_SBINDIR" && "$vmdb_answer_SBINDIR"/vmware-guestd \
--background "$GUESTD_PID_FILE" --halt-command "/sbin/shutdown -p now"
}
保存退出。执行 /usr/local/etc/rc.d/vmware-tools.sh restart，重启vmware tools即可。
替换前
test# zpool list
NAME SIZE USED AVAIL CAP HEALTH ALTROOT
zfspool 9.94G 1.96G 7.98G 19% ONLINE -
用da2替换da1，都是/dev/下的设备
test# zpool replace zfspool da1 da2
开始替换了
test# zpool status
pool: zfspool
state: ONLINE
status: One or more devices is currently being resilvered. The pool will
continue to function, possibly in a degraded state.
action: Wait for the resilver to complete.
scrub: resilver in progress, 15.24% done, 0h4m to go
config:
NAME STATE READ WRITE CKSUM
zfspool ONLINE 0 0 0
replacing ONLINE 0 0 0
da1 ONLINE 0 0 0
da2 ONLINE 0 0 0
errors: No known data errors
替换完成了
test# zpool status
pool: zfspool
state: ONLINE
scrub: resilver completed with 0 errors on Sat May 9 16:49:35 2009
config:
NAME STATE READ WRITE CKSUM
zfspool ONLINE 0 0 0
da2 ONLINE 0 0 0
errors: No known data errors
容量增加了
test# zpool list
NAME SIZE USED AVAIL CAP HEALTH ALTROOT
zfspool 17.9G 1.96G 16.0G 10% ONLINE -
整个过程中应用没有中断
FreeBSD做服务器，要实现从哪个网卡进来的连接请求，返回时还从哪个网卡出去。这样可以让客户端自己选择线路，无需收集路由表。
ee /etc/rc.conf
pf_enable="YES"
pf_rules="/etc/pf.conf"
//启用PF
defaultrouter="192.168.1.1"
//这个是本机自己发起连接的默认路由
ee /etc/pf.conf
if_cernet="em0"
if_ct="em1"
gw_cernet="192.168.1.1"
gw_ct="192.168.0.1"
block all
pass quick on lo0 all
pass in quick on $if_cernet reply-to ( $if_cernet $gw_cernet ) proto {tcp,udp,icmp} to any keep state
pass in quick on $if_ct reply-to ( $if_ct $gw_ct ) proto {tcp,udp,icmp} to any keep state
pass out keep state
iSCSI-target和NFS Server由一台Raid10(4*2.5′ 10Krpm 146GB)的VMware ESXi 3.5里的FreeBSD服务机提供，在另一台Raid1(2*3.5′ 15Krpm 146GB)的VMware ESXi 3.5里挂载iSCSI和NFS，然后分别以虚拟磁盘添加入FreeBSD测试机中。使用/usr/local/bin/iozone -i 0 -i 1 -i 2 -r 1024 -s 1G -t 2 -C测试。测试结果如下：
iSCSI测试：
Initial write = 5443.42 KB/sec
Rewrite = 4840.85 KB/sec
Read = 19823.13 KB/sec
Re-read = 19298.97 KB/sec
Random read = 44114.65 KB/sec
Random write = 4024.72 KB/sec
NFS测试：
Initial write = 952.76 KB/sec
Rewrite = 975.36 KB/sec
Read = 14782.20 KB/sec
Re-read = 16085.16 KB/sec
Random read = 41878.42 KB/sec
Random write = 794.31 KB/sec
CPU占用率上NFS只有iSCSI的一半，服务机和测试机都差不多。iSCSI时CPU占用率为15%左右，中间还有一段是30%多。NFS时基本都8%左右。两台机器均为2*Intel E5405，分配给虚拟机2个核。
测试机直接加载NFS测试：
Initial write = 2361.99 KB/sec
Rewrite = 2130.92 KB/sec
Read = 17595.85 KB/sec
Re-read = 18904.29 KB/sec
Random read = 13139.79 KB/sec
Random write = 2001.82 KB/sec
测试机本地测试：
Initial write = 8233.32 KB/sec
Rewrite = 12511.68 KB/sec
Read = 34969.73 KB/sec
Re-read = 34179.26 KB/sec
Random read = 82272.52 KB/sec
Random write = 4620.50 KB/sec
服务机本地测试：
Initial write = 6236.64 KB/sec
Rewrite = 9016.30 KB/sec
Read = 47051.42 KB/sec
Re-read = 47444.12 KB/sec
Random read = 27243.86 KB/sec
Random write = 3251.88 KB/sec