—2012/4/16—
iPad 1代，使用iOS5.1后，速度明显变慢，还是iOS4.3比较适合。
—————END————–
—————Thx————–

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163

#coding:utf-8
"""
web.py + ping
www.591rmb.info
ex:
# nohup web_ping.py 9999 &
# curl http://192.168.1.36:9999/192.168.1.86
ping Code，Copy from the following website:
http://www.g-loaded.eu/2009/10/30/python-ping/
"""
import socket, select, time, web, urllib
import os, sys, socket, struct, select, time
 
urls = (
    '/(.*)', 'Ping'
)
 
app = web.application(urls, globals())
 
class Ping:
    def GET(self, ip):
        return verbose_ping(ip, 2, 1)
 
# From /usr/include/linux/icmp.h; your milage may vary.
ICMP_ECHO_REQUEST = 8 # Seems to be the same on Solaris.
 
 
def checksum(source_string):
    """
    I'm not too confident that this is right but testing seems
    to suggest that it gives the same answers as in_cksum in ping.c
    """
    sum = 0
    countTo = (len(source_string)/2)*2
    count = 0
    while count<countTo:
        thisVal = ord(source_string[count + 1])*256 + ord(source_string[count])
        sum = sum + thisVal
        sum = sum & 0xffffffff # Necessary?
        count = count + 2
 
    if countTo<len(source_string):
        sum = sum + ord(source_string[len(source_string) - 1])
        sum = sum & 0xffffffff # Necessary?
 
    sum = (sum >> 16)  +  (sum & 0xffff)
    sum = sum + (sum >> 16)
    answer = ~sum
    answer = answer & 0xffff
 
    # Swap bytes. Bugger me if I know why.
    answer = answer >> 8 | (answer << 8 & 0xff00)
 
    return answer
 
 
def receive_one_ping(my_socket, ID, timeout):
    """
    receive the ping from the socket.
    """
    timeLeft = timeout
    while True:
        startedSelect = time.time()
        whatReady = select.select([my_socket], [], [], timeLeft)
        howLongInSelect = (time.time() - startedSelect)
        if whatReady[0] == []: # Timeout
            return
 
        timeReceived = time.time()
        recPacket, addr = my_socket.recvfrom(1024)
        icmpHeader = recPacket[20:28]
        type, code, checksum, packetID, sequence = struct.unpack(
            "bbHHh", icmpHeader
        )
        if packetID == ID:
            bytesInDouble = struct.calcsize("d")
            timeSent = struct.unpack("d", recPacket[28:28 + bytesInDouble])[0]
            return timeReceived - timeSent
 
        timeLeft = timeLeft - howLongInSelect
        if timeLeft <= 0:
            return
 
 
def send_one_ping(my_socket, dest_addr, ID):
    """
    Send one ping to the given >dest_addr<.
    """
    dest_addr  =  socket.gethostbyname(dest_addr)
 
    # Header is type (8), code (8), checksum (16), id (16), sequence (16)
    my_checksum = 0
 
    # Make a dummy heder with a 0 checksum.
    header = struct.pack("bbHHh", ICMP_ECHO_REQUEST, 0, my_checksum, ID, 1)
    bytesInDouble = struct.calcsize("d")
    data = (192 - bytesInDouble) * "Q"
    data = struct.pack("d", time.time()) + data
 
    # Calculate the checksum on the data and the dummy header.
    my_checksum = checksum(header + data)
 
    # Now that we have the right checksum, we put that in. It's just easier
    # to make up a new header than to stuff it into the dummy.
    header = struct.pack(
        "bbHHh", ICMP_ECHO_REQUEST, 0, socket.htons(my_checksum), ID, 1
    )
    packet = header + data
    my_socket.sendto(packet, (dest_addr, 1)) # Don't know about the 1
 
 
def do_one(dest_addr, timeout):
    """
    Returns either the delay (in seconds) or none on timeout.
    """
    icmp = socket.getprotobyname("icmp")
    try:
        my_socket = socket.socket(socket.AF_INET, socket.SOCK_RAW, icmp)
    except socket.error, (errno, msg):
        if errno == 1:
            # Operation not permitted
            msg = msg + (
                " - Note that ICMP messages can only be sent from processes"
                " running as root."
            )
            raise socket.error(msg)
        raise # raise the original error
 
    my_ID = os.getpid() & 0xFFFF
 
    send_one_ping(my_socket, dest_addr, my_ID)
    delay = receive_one_ping(my_socket, my_ID, timeout)
 
    my_socket.close()
    return delay
 
 
def verbose_ping(dest_addr, timeout = 2, count = 4):
    """
    Send >count< ping to >dest_addr< with the given >timeout< and display
    the result.
    """
    result = ''
    for i in xrange(count):
        result = "ping " + dest_addr
        try:
            delay  =  do_one(dest_addr, timeout)
        except socket.gaierror, e:
            result += " failed. (socket error: '%s')" % e[1]
            break
 
        if delay  ==  None:
            result += " failed. (timeout within %ssec.)" % timeout
        else:
            delay  =  delay * 1000
            result += " => %dms" % delay
    return result
 
 
 
if __name__ == "__main__":
    def void(s): pass
    app.run()

—————END————–
—————Thx————–

在文章开始前，我要感谢BeansDB的开发人员Davies Liu，他非常的热心，我也一直关注BeansDB，早前也有Email往来，当BeansEye发布后，他也第一时间的告诉了我，并且在安装中遇到的问题，也及时的回复我，感激不尽。

服务器环境：
　系统：CentOS
　安装包下载目录：/usr/local/soft
安装步骤：
一、安装hg：
　网上hg的安装教程很多，我这里就不介绍了；
二、安装Go语言的weekly的发布版本：

1
2
3
4
5
6

#  hg clone -u release https://go.googlecode.com/hg/ go
#  cd go/src
#  hg pull
#  hg update weekly
#  ./src/all.bash
#  mv /usr/local/soft/go /usr/local/

将下面的内容追加到/root/.bashrc ：

1
2
3
4
5
6
7

# vi /root/.bashrc
export GOROOT=/usr/local/go
export GOARCH=amd64
export GOOS=linux
export GOBIN=$GOROOT/bin/
export PATH=.:$PATH:$GOBIN
# source /root/.bashrc

三、安装BeansEye：
1、下载BeansEye：https://github.com/douban/beanseye
2、下载goconfig：https://github.com/kless/goconfig/，并保存至src目录。否则会出现以下错误：
src/proxy/proxy.go:19:5: import “github.com/kless/goconfig/config”: cannot find package

开始安装：

1
2
3

# cd beanseye
# go install proxy
# mv /usr/local/soft/beanseye /usr/local/

目录结构：

四、配置：
在conf创建my.ini文件：

介绍N W R：
N:表示写操作时它主动发起的写的节点数
W 表示最少要写成功的次数
R 表示最少要读成功的次数
据Davies Liu介绍，douban设定的是 N=3, W=2，尽量保证数据的可靠性。

五、启动BeansEye:

1

# ./bin/proxy -conf=conf/myconf.ini

六：BeansEye集群管理：

http://localhost:7908/

—————END————–
—————Thx————–

           # wget http://labs.frickle.com/files/ngx_cache_purge-1.3.tar.gz
           # tar zxvf ngx_cache_purge-1.3.tar.gz 

    二、重新编译Nginx（根据实际情况而定）

          # cd nginx-0.8.52
          # ./configure –prefix=/usr/local/nginx –user=acc –group=acc –add-module=../ngx_cache_purge-1.3 –with-http_ssl_module -
-with-http_sub_module –with-http_dav_module –with-http_flv_module –with-http_gzip_static_module –with-http_stub_status_module –
http-proxy-temp-path=/var/nginx/proxy –http-fastcgi-temp-path=/var/nginx/fastcgi –http-client-body-temp-path=/var/nginx/client –w
ith-debug       
         # make
         # mv /usr/local/nginx/sbin/nginx /usr/local/nginx/sbin/nginx-old
         # cp objs/nginx /usr/local/nginx/sbin/

    三、创建文件夹

        # mkdir /tmp/proxy_temp_dir
        # mkdir /tmp/proxy_cache_dir
        # chmod 777 /tmp/proxy_temp_dir /tmp/proxy_cache_dir

    四、修改Nginx配置文件

# vi /usr/local/nginx/conf/nginx.conf
user addcn addcn;
# 8核CPU
worker_processes  8;
worker_cpu_affinity 00000001 00000010 00000100 00001000 00010000 00100000 01000000 10000000;

error_log  logs/error.log;

pid        logs/nginx.pid;

worker_rlimit_nofile 65535;
events {
    use epoll;
    worker_connections  65535;
}

http {
    include       mime.types;
    default_type  application/octet-stream;

    access_log  logs/access.log  main;

    server_names_hash_bucket_size 128;
    client_header_buffer_size 32k;
    large_client_header_buffers 4 32k;
    client_max_body_size 8m;

    sendfile        on;
    tcp_nopush     on;
    server_tokens off;

    keepalive_timeout  60;
    fastcgi_intercept_errors on;

    client_body_buffer_size  512k;
    proxy_connect_timeout    10;
    proxy_read_timeout       60;
    proxy_send_timeout       10;
    proxy_buffer_size        16k;
    proxy_buffers            4 64k;
    proxy_busy_buffers_size 128k;
    proxy_temp_file_write_size 128k;

    # fastcgi
    fastcgi_connect_timeout 300;
    fastcgi_send_timeout 300;
    fastcgi_read_timeout 300;
    fastcgi_buffer_size 64k;
    fastcgi_buffers 4 64k;
    fastcgi_busy_buffers_size 128k;
    fastcgi_temp_file_write_size 128k;

    # gzip
    gzip  on;
    gzip_min_length  1k;
    gzip_buffers     4 16k;
    gzip_http_version 1.0;
    gzip_comp_level 2;
    gzip_types       text/plain application/x-javascript text/css application/xml;
    gzip_vary on;
    # proxy_temp_path和proxy_cache_path指定的路径必须在同一分区
    proxy_temp_path   /tmp/proxy_temp_dir;
    # 设置Web缓存区名称为cache_one，内存缓存空间大小为1000MB，3天清理一次缓存，硬盘缓存空间大小为100GB。
    proxy_cache_path  /tmp/proxy_cache_dir  levels=1:2   keys_zone=cache_one:1000m inactive=3d max_size=100g;
    #  后台请求服务器
    upstream backend_server {
       server   192.168.1.2:8000 weight=1 max_fails=2 fail_timeout=30s;
       server   192.168.1.3:8000 weight=1 max_fails=2 fail_timeout=30s;
    }

server {
    # listen port
    listen       8000;

    server_name  img.591rmb.info
    charset utf-8;

    # root
    include     /usr/local/nginx/conf/module/path_params.conf;
    root   /var/www/
    index  index.html index.htm;

    # blocked
    location ~ .*\.(gif|jpg|png|jpeg|bmp|swf)$ {
        proxy_next_upstream http_502 http_504 error timeout invalid_header;
        proxy_cache cache_one;
        proxy_cache_valid  200 304 30d;
        proxy_cache_key $host$uri$is_args$args;
        proxy_set_header Host  $host;
        proxy_set_header X-Forwarded-For  $remote_addr;
        proxy_pass http://backend_server;

        expires max;
        access_log off;
    }
   #用于清除缓存，假设一个URL为http://192.168.8.42/test.txt，通过访问http://192.168.8.42/purge/test.txt就可以清除该URL的缓存
    location ~ /purge(/.*)
    {
     #设置只允许指定的IP或IP段才可以清除URL缓存。
     allow            127.0.0.1;
     allow            192.168.0.0/16;
     deny            all;
     proxy_cache_purge    cache_one   $host$1$is_args$args;
    } 
}

　五、启动Nginx

    # /usr/local/nginx/sbin/nginx -t
       the configuration file /usr/local/nginx/conf/nginx.conf syntax is ok
       configuration file /usr/local/nginx/conf/nginx.conf test is successful
    # /usr/local/nginx/sbin/nginx

    六、参考文献：
       使用Nginx的proxy_cache缓存功能取代Squid[原创] ( http://blog.s135.com/nginx_cache )

   总结：
            Varnish3最新版还没有试用，后续找时间试一下，使用proxy_cache后，图片访问一切正常^-^。

# wget http://blitiri.com.ar/p/msnlib/files/3.8/msnlib-3.8.tar.gz
# tar zxvf msnlib-3.8.tar.gz
# cd msnlib-3.8
# ./install
# msnsetup
# msn

　　PS: 执行msnsetup后，会请你输入账号与密码，然后执行msn，就自动登录了。
　　二、安装web.py：

# wget http://webpy.org/static/web.py-0.35.tar.gz
# tar zxvf web.py-0.35.tar.gz
# cd web.py-0.35
# python setup.py install

　　三、msn.py
　　　有代码，有真相：

#coding:utf-8
import socket, select, time, msnlib, msncb, web, urllib

urls = (
    '/', 'hello'
)

web.config.debug = False
app = web.application(urls, globals())

class hello:        
    def POST(self):
        i = web.input()
        result = get_friend_list(i.username, i.pwd)
        if result is None:
            return '{status:false}'
        else:
            http_result = '{status:true, body:[';
            for k, v in result.iteritems():
                http_result = http_result + '{name:"' + v.decode("utf-8", "ignore") + '",email:"' + k + '"},'
            http_result = http_result[0:-1] + ']}';

            return http_result

def get_friend_list(email, password):
    m = msnlib.msnd()
    m.cb = msncb.cb()

    m.email = email.strip()
    m.pwd = password.strip()
    m.encoding = 'utf-8'
    try:
        m.login()
        m.sync()
    except:
        return
    m.change_status('invisible')

    begin_time = time.time()

    users = set()
    while True:
        fds = m.pollable()
        infd = fds[0]
        outfd = fds[1]

        fds = select.select(infd, outfd, [], 0)

        for i in fds[0] + fds[1]:
            try:
                m.read(i)
            except ('SocketError', socket.error), err:
                if i != m:
                    m.close(i)
        merge_users = users|set(m.users.keys())
        if len(users) == len(merge_users):
            end_time = time.time()
            if len(users):
                if end_time-begin_time >= 2:
                    break
            elif end_time-begin_time >= 3:
                break
            time.sleep(0.05)
        else:
            users = merge_users
            begin_time = time.time()
    result = {}
    for i in users:
        nick = m.users[i].nick
        i_lower = i.lower()
        if nick.lower() == i_lower:
            nick = nick.split("@", 1)[0]
        result[i_lower] = nick
    m.disconnect();
    return result

if __name__ == "__main__":
    msnlib.debug = msncb.debug = void

    app.run()
 

　　四、执行msn.py程式，并指定端口为8888：

# python msn.py 8888

　　五、参考文献：
     　　 导入MSN联系人（http://kanrs.com/4_0.html）

　　总结：
        　　遇到难点时，可以跳出PHP世界，你就会发现不一样的美。

# wget http://geolite.maxmind.com/download/geoip/database/GeoLiteCity.dat.gz
# gunzip GeoLiteCity.dat.gz
# ls
GeoLiteCity.dat

　　二、下载PHP的GeoIP的类文件和演示文件：
　　　　网址： http://geolite.maxmind.com/download/geoip/api/php/
　　　　本站下载，点我
　　三、修改sample_city.php文件
　　　　将GeoIPCity.dat的文件路径，指向实际的路径，以及获取客户端的IP地址：

$gi = geoip_open(“./GeoLiteCity.dat”,GEOIP_STANDARD);
$record = geoip_record_by_addr($gi, getenv(‘REMOTE_ADDR’));

　　四、完成
　　　　执行sample_city.php，就可以看到你的实际位置了，单击下面的网址查看你的位置吧！
　　　　http://www.591rmb.info/geoip/

　　Baidu一下，发现金山的“爱词霸”也挺不错的，功能也很强，以下我就简单的介绍一下各自的使用方法：

　　一、爱词霸( http://web.iciba.com/partner/jhjy04.shtml )：

　　爱词霸提供了网页插件、浏览器插件、客户端插件，还提供了开放平台API，可以自定义查询，真是很强大。
　　把代码放置footer.php，调用代码如下：

<script type=”text/javascript”>
var iciba_param =new Array();
iciba_param['skin'] = 1;// 皮肤
iciba_param['searchInputDisplay'] = ‘block’;// 搜索框是否显示 默认是” 当为 ‘none’的时候不显示
iciba_param['searchInputWidth'] =210;////输入框长度 当iciba_param['searchInputDisplay'] !=’none’的时候起作用
iciba_param['isCanDraw'] = ’1′;//是否能拖动？0表示不可以拖动 默认为1可拖动
iciba_param['selfDeter'] =’1′;//是否自己划词自己？0 表示不可以
iciba_param['width'] = ’325′; // 宽度
iciba_param['height'] = ’200′;  // 内容高度
iciba_param['isPopIcon'] =’0′;//是否需要 小icon 需要为1 不需要为0
iciba_param['isInputCan'] =’0′;//是否能在文本域输入框内划词 默认不可以 只有’1′的时候才能起作用
iciba_param['context']=[['Dict',''],['Love',''],['Fy',''],['Tf',''],['Dj',''],['Enen',''],['NetDict','']];//词典功能
</script>
<script type=”text/javascript” src=”http://dict-co.iciba.com/ui.php” charset=”UTF-8″></script>

　　二、必应云词典（ http://dict.bing.com.cn/cloudwidget/Default.aspx ）
　　必应云词典 很简洁，还可以针对句子进行翻译，”爱词霸”没有此功能，我很喜欢它显示拼音及音调的功能，调用代码也很简单：

<script type=”text/javascript” src=”http://dict.bing.com.cn/cloudwidget/Scripts/Generated/BingTranslate_Hover_Phrase_Selection_ShowIcon.js”></script>
<script type=”text/javascript”>
BingCW.Init({
AppID:”www.591rmb.info”,      // 换成你博客的网址即可。
MachineTranslation:true,
WebDefinition:true
})</script>

结束语：
　　两款工具各有千秋，根据自己的需求而定，最后我决定支持国产的“爱词霸”^-^

总结：
       初步了解MooseFS后，感觉如果有更好的HA，以及跨机房备份问题，那就非常适合我们了^-^。

global {
  usage-count yes;
}

resource r0 {    
  protocol C;  
  startup {
    wfc-timeout 0;
    degr-wfc-timeout 120;
  }

  disk {
    on-io-error   detach;
    # no-disk-flushes;
    # no-md-flushes
    # size 1G;
   }

  net {
   }

on nfs1 {
   device    /dev/drbd0;
   disk      /dev/sdb1;
   address   192.168.192.129:7789;
   meta-disk internal;          
  }                              

on nfs2 {
   device    /dev/drbd0;
   disk      /dev/sdb1;
   address   192.168.192.131:7789;
   meta-disk internal;
  }
}

5）建立资源r0（两台机执行）：
[root@localhost ~]# drbdadm create-md r0
Writing meta data…
initializing activity log
NOT initialized bitmap
New drbd meta data block successfully created.
6）启动DRBD（两台机执行）
[root@localhost ~]# service drbd start
Starting DRBD resources: [
r0
Found valid meta data in the expected location, 1069248512 bytes into /dev/sdb1.
d(r0) n(r0) ].
7）设定主节点 nfs1 ，nfs2不需要操作：
定义为主节点
# drbdsetup /dev/drbd0 primary -o
格式化
# mkfs.ext3 /dev/drbd0
挂载到/data目录
# mkdir /data
# mount /dev/drbd0 /data 
8）查看状态nfs1：
# cat /proc/drbd
version: 8.3.8 (api:88/proto:86-94)
GIT-hash: d78846e52224fd00562f7c225bcc25b2d422321d build by mockbuild@builder10.centos.org, 2010-06-04 08:04:16
 0: cs:SyncSource ro:Primary/Secondary ds:UpToDate/Inconsistent C r—-
    ns:39028 nr:0 dw:33216 dr:5933 al:13 bm:1 lo:0 pe:0 ua:0 ap:0 ep:1 wo:b oos:1007200
        [>....................] sync'ed:  4.0% (1007200/1044124)K delay_probe: 25141
        finish: 0:41:57 speed: 360 (2,172) K/sec
UpToDate/Inconsistent 代表，数据正在同步中
如果是 UpToDate/Update 则代表，数据已经同步完成。

三、维护
1) DRBD的主备机切换

有时,你需要将DRBD的主备机互换一下.可以执行下面的操作:
在主机上,先要卸载掉DRBD设备:
# umount /data

将主机降级为”备机”.
# drbdadm secondary r0
可看到两台机都是secondary状态
# cat /proc/drbd
drbdadm primary r0
就将从机升级为主机了。

总结：
        目前准备小范围的进行线上测试，如果一切顺利的话，也会用Mysql的HA中，有网友测试，会损耗5%的性能，这个还是可以接受的。
        本文内容来自源于笔记，下次配置DRBD，再把图片等内容补上。

总结：
        可根据自身的需求，而进行选择，如果数据量非常大时，可以采用Excel 2003的XML格式。