PHPor 的Blog – 第69页

关于redis短连接问题

场景：

PHP通过phpredis(https://github.com/phpredis/phpredis)循环执行如下操作：

<?php
while(true) {
        $r = new Redis();
        $r->connect(...);
        $r->auth(...);
        $r->set(...);
        $r->get(...);
}

<?php

while(true) {

$r = new Redis();

$r->connect(...);

$r->auth(...);

$r->set(...);

$r->get(...);

}

发现connect会出现大量超过1s的情况，甚至超过3s；但是，如下图抓包所示：两次请求之间间隔了3s才发送syn请求，并且没有失败，也就是说，并非syn丢包所致，而是因为某种原因导致根本没有发送syn包；但是外部表现却是connect花费了很多时间（3s）；

所以，从此推断，连接超时未必就一定是网络的问题，你们究竟会是什么原因导致client端3s后才发送syn的呢？

阅读源码发现： redis的connect用的是php提供的connect方法：php_stream_xport_create

其实，fsockopen用的也是该方法，如此的话，不放通过如下脚本直接测试connect：

<?php

$errno = 0;
$error = "";
$i = 0;
while(true) {
	$i++;
	$time_start = microtime(1);
	$fp = fsockopen("10.172.90.108", 6379, $errno, $error, 5);
	$time_end = microtime(1);
	$time_use = $time_end - $time_start;
	if ($time_use > 1) {
		echo "timeout: $time_use ($i)\n";
	}
	if ($fp) fclose($fp);
}

<?php

$errno = 0;

$error = "";

$i = 0;

while(true) {

$i++;

$time_start = microtime(1);

$fp = fsockopen("10.172.90.108", 6379, $errno, $error, 5);

$time_end = microtime(1);

$time_use = $time_end - $time_start;

if ($time_use > 1) {

echo "timeout: $time_use ($i)\n";

}

if ($fp) fclose($fp);

}

结果如下：

connect超过1s的还挺不少的
前面300多次的时候连续出了好几次超过1s的，你们怎么会从374到13489一直没出错呢？而且，运行程序的时候也能感觉到，从第374到第13489花费时间很短的，难道这个区间根本没有发生真正的连接？ strace 跟踪统计发现，确实connect了，就是说，一般情况下，connect是很快的
该问题可能发生在PHP中，也可能发生在glibc中，也可能发生在系统层面，排查办法：
1. 写一个C的，如果不出现问题，那么就是PHP的问题；如果依然如此，很可能是系统问题

salt-cp 不能传二进制问题

问题：

salt-cp copy 了一个9MB的字体文件，结果就把内存耗光了，不解；

突然想起哪里又说salt-cp不能处理二进制文件，翻下代码看看：（注意这里的fopen）

/usr/lib/python2.6/site-packages/salt/cli/cp.py

让go get 支持gitlab

参考文章： https://baokun.li/archives/go-get-proxy/

在你的nginx中添加配置：

if ($args ~* "^go-get=1") {
  set $condition goget;
}
if ($uri ~ ^/([a-zA-Z0-9_-]+)/([a-zA-Z0-9_-]+)/.*$) {
  set $condition "{condition}path";   
}
if ($condition = gogetpath) {
  return 200 "<!DOCTYPE html><html><head><meta content='your.domain.com/$1/$2 git http://your.domain.com/$1/$2.git' name='go-import'></head></html>";
}

if ($args ~* "^go-get=1") {

set $condition goget;

}

if ($uri ~ ^/([a-zA-Z0-9_-]+)/([a-zA-Z0-9_-]+)/.*$) {

set $condition "{condition}path";

}

if ($condition = gogetpath) {

return 200 "<!DOCTYPE html><html><head><meta content='your.domain.com/$1/$2 git http://your.domain.com/$1/$2.git' name='go-import'></head></html>";

}

注意：修改你自己的域名

另：

可能你自己的gitlab没有配置https，那么可以给go get 添加 -insecure 选项即可
可能你的gitlab仓库需要输入用户名和密码，那么你可以改成public的就好了

centos 修改时区

cp /usr/share/zoneinfo/Asia/Shanghai /etc/localtime

1	cp /usr/share/zoneinfo/Asia/Shanghai /etc/localtime

rpm打包

参考： https://docs.fedoraproject.org/en-US/Fedora_Draft_Documentation/0.1/html/RPM_Guide/ch-advanced-packaging.html

关于docker

docker create -it 其中 -it选项，如果没有的话，docker exec 进去，top时会报 TERM environment variable not set. 错误 ; 其实没有关系，手动 export TERM=xterm-256color 就可以了;其实， -it 选项和1号进程为 /bin/bash 是密切相关的；如果1号进程不是 /bin/bash 的话，-it选项没有屁用
docker 的CMD 其实是1号进程，一定不能立即就会退出的进程，如： service sshd start 就不行，如下脚本就可以：

#!/bin/bash # this is a start shell for containner service sshd start while :; do sleep 10000 done

1
2
3
4
5
6

#!/bin/bash
# this is a start shell for containner
service sshd start
while :; do
sleep 10000
done
容器可以自杀：容器内进程使用root kill -9 1 默认是杀不死1号进程的，但是：
当strace -p 1 的时候，kill -9 1 是可以杀死1号进程的（难道说这是个bug？）
因为docker stop时是先给1号进程发送sigterm，是在不行才kill -9 ；所以，1号进程最好能合理处理sigterm信号，最后退出；然而，容器内的root进程是有权限给1号进程发送sigterm信号的，所以，容器自杀其实是很方便的
改进版的1号进程
myinit.c

#include <stdlib.h> #include <unistd.h> int main(int ac, char **av) { while(1){ sleep(10000); } return 0; }

1
2
3
4
5
6
7
8
9

#include <stdlib.h>
#include <unistd.h>

int main(int ac, char **av) {
while(1){
sleep(10000);
}
return 0;
}

gcc -o myinit myinit.c
myinit.sh

#!/bin/bash # this is a start shell for containner service sshd start exec /sbin/myinit

1
2
3
4

#!/bin/bash
# this is a start shell for containner
service sshd start
exec /sbin/myinit

其实还不够
你会慢慢发现，ssh连接失败，因为：sshd接到连接请求之后，检查是否session太多了（默认最大10个），超过了就不伺候了，直接断开；为什么会太多了？sshd每次接到连接请求，会fork => fork (两次fork)出来一个子进程处理请求，就因为是两次fork，这个新的sshd进程便“过继”给了1号进程；但是上面的1号进程还不够成熟，不会带孩子，当这个新的sshd进程退出的时候，就需要1号进程来做一些处理（如：清理进程表项），但是1号进程还不会，于是新的sshd进程会处于 <defunct>状态；而sshd的祖先还以为该session还没结束，就不再接受新的请求了，解决办法：添加信号处理，如下：

#include <stdlib.h>
#include <unistd.h>
#include <signal.h>

int main(int ac, char **av) {
	signal(SIGCHLD, SIG_IGN);
	while(1){
		sleep(10000);
	}
	return 0;
}

#include <stdlib.h>

#include <unistd.h>

#include <signal.h>

int main(int ac, char **av) {

signal(SIGCHLD, SIG_IGN);

while(1){

sleep(10000);

}

return 0;

}

参考资料：
https://blog.phusion.nl/2015/01/20/docker-and-the-pid-1-zombie-reaping-problem/
https://github.com/phusion/baseimage-docker/blob/rel-0.9.16/image/bin/my_init
dockerd 提供了init 和init-path的选项，允许配置是否启动一个指定的（默认的）init进程，该进程就是为了转发信号和收割僵尸进程的

上面sshd对于连接请求是两次fork的说法不太对，见下图：

99、134进程并没有立即退出（不过，一旦意外退出，则 101、136将会被1号进程收养）
还有一种情况为：sshd被restart了，这时候30号进程退出，99、134被1号进程收养，如下：

由于我重启了sshd，再次连接就会失败，如下：

为什么呢？
稍后再查，至少可以确定的是，启动容器时候启动的sshd是可以正常服务的，登录进去重启的sshd是不能正常服务的

关于docker的内存限制：

如果每个docker容器限制最大使用4G，则可能会使用超过4G，（似乎是使用了交换分区了）
如果宿主机有64G内存，则允许docker容器内存总和大于64G，（这个不会也和交换分区有关吧）
文档说–memory-swap=-1 就禁止使用交换分区了，但是测试发现还是使用了1倍内存的交换分区

缅怀一下阿里云云引擎

学习一下人家下线产品是咋写的：

阿里云云引擎ACE产品转型下线公告
尊敬的阿里云用户：
您好。
自ACE产品上线以来，我们致力于为用户提供更好的服务以及体验。
今日，我们非常遗憾的通知您，由于产品体系升级，阿里云云引擎ACE产品将于2016年5月12日整体下线，届时ACE产品（包括扩展服务）将全部停止服务（整体下线安排见后）。我们推荐您使用云服务器ECS、弹性web托管、容器服务等其他云计算产品。
在此期间，ACE产品团队将会：
1、 如选择使用ECS或弹性web托管服务，我们将提供相应的代金券；
2、 在用户管理控制台为您提供数据下载的快速通道，您可通过此快速通道进行数据下载\备份以及迁移的工作（如您选择阿里云其他产品的，阿里云并将提供对应的迁移方案）；
3、 针对未到期的包年包月ACE产品，阿里云将启动退款程序（具体方案将通过站内信或邮件发送给您）。
请在ACE服务到期日之前安排好数据，逾期将无法找回。
如您在下线过程中遇到问题，可直接从阿里云用户中心—工单管理—提交工单-云引擎ACE提交工单。
因产品体系升级给您带来不便，我们深表歉意。
再次感谢您对阿里云的支持！
阿里云计算有限公司
2016年4月6日
常用Q&A
1、云引擎ACE为什么要下线？
答：您好，非常抱歉，由于产品体系升级，业务调整影响，云引擎ACE将不再提供服务。
2、已买了云引擎ACE的包年包月某某版本，在2016年4月6日以后服务还未到期，产品下线了怎么办？
答：您好，已购用户一方面我们会根据您购买ACE的不同语言针对性的推荐阿里云其它产品并提供对应的迁移方案，另一方面会启动退款程序，已购用户的邮箱或者站内信、短信4月6日会统一收到通知，请收到的用户及时查到您的手机、邮箱或阿里云账户中心
3、已买了按量付费云引擎ACE，产品下线了怎么办？
答：您好，一方面我们会根据您购买ACE的不同语言针对性的推荐阿里云其它产品并提供对应的迁移方案，另一方面我们已争取替代产品的最大化代金券，已发放至您的账号，您可以直接用来购买新产品
4、产品下线了，我的应用数据怎么办
答：您好，5月12正式下线之前，我们会一直提供数据下载快速通道，您可以随时做好数据下线和备份工作，若您选择阿里云其他产品，我们提供推荐产品的迁移方案，您可直接根据方案迁移数据
5、我的应用正好在你产品下线期间到期了怎么办？
答：您好，为了保证用户体验，在4月6日-5月12日期间到期的客户可以继续使用应用直至5月12日产品正式下线为止。
6、你们产品下线，推荐我购买其他产品有没有优惠？
答：您好，我们已争取其他产品的最大化优惠代金券，已发放至您的账号，您可以直接用来购买新产品，代金券有效期为30天，请及时使用 。
7、下线期限，包年包月超过流量套餐配置外流量还收费吗？
答：您好，4月6日-5月12日针对超套餐的流量系统仍计费，超出部分按照￥0.80/GB收费，基本版和普通版每个月50GB免费公网流出流量，流入流量免费，专业版每个月80GB免费公网流出流量，流入流量免费

阿里云云引擎ACE产品转型下线公告

尊敬的阿里云用户：

您好。

自ACE产品上线以来，我们致力于为用户提供更好的服务以及体验。

今日，我们非常遗憾的通知您，由于产品体系升级，阿里云云引擎ACE产品将于2016年5月12日整体下线，届时ACE产品（包括扩展服务）将全部停止服务（整体下线安排见后）。我们推荐您使用云服务器ECS、弹性web托管、容器服务等其他云计算产品。

在此期间，ACE产品团队将会：

1、如选择使用ECS或弹性web托管服务，我们将提供相应的代金券；

2、在用户管理控制台为您提供数据下载的快速通道，您可通过此快速通道进行数据下载\备份以及迁移的工作（如您选择阿里云其他产品的，阿里云并将提供对应的迁移方案）；

3、针对未到期的包年包月ACE产品，阿里云将启动退款程序（具体方案将通过站内信或邮件发送给您）。

请在ACE服务到期日之前安排好数据，逾期将无法找回。

如您在下线过程中遇到问题，可直接从阿里云用户中心—工单管理—提交工单-云引擎ACE提交工单。

因产品体系升级给您带来不便，我们深表歉意。

再次感谢您对阿里云的支持！

阿里云计算有限公司

2016年4月6日

常用Q&A

1、云引擎ACE为什么要下线？

答：您好，非常抱歉，由于产品体系升级，业务调整影响，云引擎ACE将不再提供服务。

2、已买了云引擎ACE的包年包月某某版本，在2016年4月6日以后服务还未到期，产品下线了怎么办？

答：您好，已购用户一方面我们会根据您购买ACE的不同语言针对性的推荐阿里云其它产品并提供对应的迁移方案，另一方面会启动退款程序，已购用户的邮箱或者站内信、短信4月6日会统一收到通知，请收到的用户及时查到您的手机、邮箱或阿里云账户中心

3、已买了按量付费云引擎ACE，产品下线了怎么办？

答：您好，一方面我们会根据您购买ACE的不同语言针对性的推荐阿里云其它产品并提供对应的迁移方案，另一方面我们已争取替代产品的最大化代金券，已发放至您的账号，您可以直接用来购买新产品

4、产品下线了，我的应用数据怎么办

答：您好，5月12正式下线之前，我们会一直提供数据下载快速通道，您可以随时做好数据下线和备份工作，若您选择阿里云其他产品，我们提供推荐产品的迁移方案，您可直接根据方案迁移数据

5、我的应用正好在你产品下线期间到期了怎么办？

答：您好，为了保证用户体验，在4月6日-5月12日期间到期的客户可以继续使用应用直至5月12日产品正式下线为止。

6、你们产品下线，推荐我购买其他产品有没有优惠？

答：您好，我们已争取其他产品的最大化优惠代金券，已发放至您的账号，您可以直接用来购买新产品，代金券有效期为30天，请及时使用。

7、下线期限，包年包月超过流量套餐配置外流量还收费吗？

答：您好，4月6日-5月12日针对超套餐的流量系统仍计费，超出部分按照￥0.80/GB收费，基本版和普通版每个月50GB免费公网流出流量，流入流量免费，专业版每个月80GB免费公网流出流量，流入流量免费

关于API网关

https://tyk.io/
阿里云API网关： https://help.aliyun.com/product/9194896_apigateway.html
AWS API网关：https://aws.amazon.com/cn/api-gateway/

为什么API网关也能成为一种服务： http://www.d1net.com/cloud/vendors/359931.html

关于httpdns

概要

httpdns就是通过http的方式进行域名解析；阿里云有提供httpdns服务（目前2016-4-9 还处于公测阶段，需要申请公测资格才能用）。

传统dns解析存在的问题：

由于传统dns解析采用udp方式，而且是明文，所以
1. 可能会被篡改
2. 可能会被投毒（什么叫“投毒”？且看下文）

httpdns的实现原理：

通过http的方式进行域名解析，如：

curl http://203.107.1.1/100000/d?host=www.aliyun.com {"host":"www.aliyun.com","ips":["140.205.63.8"],"ttl":274}

1
2

curl http://203.107.1.1/100000/d?host=www.aliyun.com
{"host":"www.aliyun.com","ips":["140.205.63.8"],"ttl":274}
通过设置http host header的方式发起http请求，如：

curl -H"Host: www.aliyun.com" http://140.205.115.67/

1

curl -H"Host: www.aliyun.com" http://140.205.115.67/

问题：

httpdns server的域名如何解析？
httpdns server 不通过域名提供服务，（只能）直接通过一个固定IP提供服务；服务提供商有足够的手段让这个IP不会出现宕机，而且世界各地（或者是全国各地）访问都能很快（就近访问），这个IP在世界上存在多份。
如何保证连接的这个ip是没被劫持呢？
1. 或许可以给这个IP买个https证书，https证书提供商是可以给IP颁发证书的（比较少见），不过可能贵的多

优点：（参考文档： https://www.aliyun.com/product/httpdns?spm=5176.7960203.223922.4.2JohAO）

防劫持
避免dns服务商莫名其妙的cache，可以做到立即修改理解生效

精准调度

对于地址：
http://203.107.1.1/100000/d?host=www.aliyun.com&amp;ip=111.111.11.111
其中：
100000是你的账户ID，host参数是你要解析的域名，ip参数是你的来源IP（用来做精准调度的）；
你可以在你的账户中设置允许解析的域名，以及要解析到的地址，甚至可以根据来源IP解析到不同的IP，
做到精准调度

对于地址：

http://203.107.1.1/100000/d?host=www.aliyun.com&ip=111.111.11.111

其中：

100000是你的账户ID，host参数是你要解析的域名，ip参数是你的来源IP（用来做精准调度的）；

你可以在你的账户中设置允许解析的域名，以及要解析到的地址，甚至可以根据来源IP解析到不同的IP，

做到精准调度

问题：

https请求问题及解决方案： https://help.aliyun.com/document_detail/httpdns/practice/https.html?spm=5176.7947101.220063.5.9DtXGS
1. 这里回避了单个IP上多个https证书的问题，或许也可以通过钩子解决，但是不一定存在这样的钩子； ssl握手的时候，必须告知正确的host信息，否则，服务器无法知道该返回哪个证书，而不仅仅是验证证书的问题
httpdns只适用于app、client的情况，对于web浏览器是不行的，还好现在大家都是在玩app了；
对于app里面适用webview的情况需要特殊处理一下：https://help.aliyun.com/document_detail/httpdns/practice/webview.html?spm=5176.dochttpdns/user-guide/resolve-stats.6.127.VFTz3N；
对于通过react native来开发app的情况不知道是否方便使用

dns投毒：

因为域名解析是udp的，只要我已dns server的ip不断地向你的机器发送域名解析结果响应的话，你可能就会认为确实是dns server响应的结果；当然，有几个关键点需要注意：

用户确实需要解析某域名（一般来讲，解析完了会cache一段时间的）
伪造的响应里面是需要包含用户正在查询的域名的，域名不匹配自然不会被接受
用户发起请求的端口号（2字节）是随机的，响应的数据包需要能匹配上这个端口，否则不会被接受
用户发起的请求有一个随机的请求标识（2字节），响应的数据包需要能匹配上这个请求标识，否则不会被接受
我咋感觉这个难度其实并不算小啊，除非你是个中间人

httpdns示例：

更多参考：

全局精确流量调度新思路-HttpDNS服务详解： http://www.zmke.com/i/8705.html

渗透测试：内网DNS投毒技术劫持会话：http://www.freebuf.com/articles/web/43157.html

http://www.onedns.net/ 需要特殊的dns客户端

加密的dns： https://www.opendns.com/about/innovations/dnscrypt/ 需要特殊的dns客户端

wireshark分析dns协议： http://blog.csdn.net/hunanchenxingyu/article/details/21488291

dns协议格式： http://cjhust.blog.163.com/blog/static/1758271572014111875652363/

关于shell一些小知识点

关于 [[]] 和 [] 的区别

实例1：

说明：

[] 和 [[]] 是不同的语法；

对于 [

事实上， bash 中没有 [] 语法， [ 是一个命令，出于美观考虑，要求有一个 ] 与之配对；所以 [ 的前后是要有空格的

对于 [[

[[]] 是bash的内建语法

[[ expression ]]
              Return a status of 0 or 1 depending on  the  evaluation  of  the  conditional  expression
              expression.   Expressions are composed of the primaries described below under CONDITIONAL
              EXPRESSIONS.  Word splitting and pathname  expansion  are  not  performed  on  the  words
              between  the  [[  and  ]];  tilde expansion, parameter and variable expansion, arithmetic
              expansion, command substitution, process substitution, and quote removal  are  performed.
              Conditional operators such as -f must be unquoted to be recognized as primaries.

              When the == and != operators are used, the string to the right of the operator is consid-
              ered a pattern and matched according to the rules described below under Pattern Matching.
              If  the shell option nocasematch is enabled, the match is performed without regard to the
              case of alphabetic characters.  The return value is 0 if the string matches (==) or  does
              not  match  (!=)  the pattern, and 1 otherwise.  Any part of the pattern may be quoted to
              force it to be matched as a string.

              An additional binary operator, =~, is available, with the same precedence as ==  and  !=.
              When  it is used, the string to the right of the operator is considered an extended regu-
              lar expression and matched accordingly (as in regex(3)).  The return value is  0  if  the
              string  matches the pattern, and 1 otherwise.  If the regular expression is syntactically
              incorrect, the conditional expression's return value is 2.  If the shell  option  nocase-
              match is enabled, the match is performed without regard to the case of alphabetic charac-
              ters.  Substrings matched by parenthesized subexpressions within the  regular  expression
              are  saved  in the array variable BASH_REMATCH.  The element of BASH_REMATCH with index 0
              is the portion of the string matching the entire  regular  expression.   The  element  of
              BASH_REMATCH  with  index  n  is the portion of the string matching the nth parenthesized
              subexpression.

[[ expression ]]

Return a status of 0 or 1 depending on the evaluation of the conditional expression

expression. Expressions are composed of the primaries described below under CONDITIONAL

EXPRESSIONS. Word splitting and pathname expansion are not performed on the words

between the [[ and ]]; tilde expansion, parameter and variable expansion, arithmetic

expansion, command substitution, process substitution, and quote removal are performed.

Conditional operators such as -f must be unquoted to be recognized as primaries.

When the == and != operators are used, the string to the right of the operator is consid-

ered a pattern and matched according to the rules described below under Pattern Matching.

If the shell option nocasematch is enabled, the match is performed without regard to the

case of alphabetic characters. The return value is 0 if the string matches (==) or does

not match (!=) the pattern, and 1 otherwise. Any part of the pattern may be quoted to

force it to be matched as a string.

An additional binary operator, =~, is available, with the same precedence as == and !=.

When it is used, the string to the right of the operator is considered an extended regu-

lar expression and matched accordingly (as in regex(3)). The return value is 0 if the

string matches the pattern, and 1 otherwise. If the regular expression is syntactically

incorrect, the conditional expression's return value is 2. If the shell option nocase-

match is enabled, the match is performed without regard to the case of alphabetic charac-

ters. Substrings matched by parenthesized subexpressions within the regular expression

are saved in the array variable BASH_REMATCH. The element of BASH_REMATCH with index 0

is the portion of the string matching the entire regular expression. The element of

BASH_REMATCH with index n is the portion of the string matching the nth parenthesized

subexpression.

如果是 == ，则右边被视为 Pattern；如果是 =~ ，则右边视为正则；也就是说，左边和右边是不等价的
[[]] 允许内部与、或、非；如： [[ 1 == 1 && 2 == 2 ]]; 而 [] 则不可以，只能： [ 1 == 1 ] && [ 2 == 2 ]

bash基本概念

bash中有保留字(reserved word)和元字符(meta charactor)的区别的，这个很重要
1. 关键字：
  
  ! case do done elif else esac fi for function if in select then until while { } time [[ ]]
  
  1
  
  ! case do done elif else esac fi for function if in select then until while { } time [[ ]]
2. 元字符：
  
  | & ; ( ) < > space tab
  
  1
  
  | & ; ( ) < > space tab
3. xx
对于关键字，和其他非元字符的字符都需要用空白字符分开，否则，关键字就识别不出来了；因为元字符是被特殊对待的，所以，关键字挨着元字符是允许的，所以，如果分不清关键字和元字符的话，就不知道什么时候需要用空白符分隔，什么时候不需要空白符分隔

man bash

关于元字符、保留字

DEFINITIONS
       The following definitions are used throughout the rest of this document.
       blank  A space or tab.
       word   A  sequence  of  characters  considered  as  a single unit by the shell.  Also known as a
              token.
       name   A word consisting only of alphanumeric characters and underscores, and beginning with  an
              alphabetic character or an underscore.  Also referred to as an identifier.
       metacharacter
              A character that, when unquoted, separates words.  One of the following:
              |  & ; ( ) < > space tab
       control operator
              A token that performs a control function.  It is one of the following symbols:
              || & && ; ;; ( ) | <newline>

RESERVED WORDS
       Reserved words are words that have a special meaning to the shell.  The following words are rec-
       ognized as reserved when unquoted and either the first word of a simple command (see SHELL GRAM-
       MAR below) or the third word of a case or for command:

       ! case  do done elif else esac fi for function if in select then until while { } time [[ ]]

DEFINITIONS

The following definitions are used throughout the rest of this document.

blank A space or tab.

word A sequence of characters considered as a single unit by the shell. Also known as a

token.

name A word consisting only of alphanumeric characters and underscores, and beginning with an

alphabetic character or an underscore. Also referred to as an identifier.

metacharacter

A character that, when unquoted, separates words. One of the following:

| & ; ( ) < > space tab

control operator

A token that performs a control function. It is one of the following symbols:

|| & && ; ;; ( ) | <newline>

RESERVED WORDS

Reserved words are words that have a special meaning to the shell. The following words are rec-

ognized as reserved when unquoted and either the first word of a simple command (see SHELL GRAM-

MAR below) or the third word of a case or for command:

! case do done elif else esac fi for function if in select then until while { } time [[ ]]

关于花括弧

Brace Expansion
       Brace expansion is a mechanism by which arbitrary strings may be generated.  This  mechanism  is
       similar to pathname expansion, but the filenames generated need not exist.  Patterns to be brace
       expanded take the form of an optional preamble, followed by either a series  of  comma-separated
       strings  or  a sequence expression between a pair of braces, followed by an optional postscript.
       The preamble is prefixed to each string contained within the braces, and the postscript is  then
       appended to each resulting string, expanding left to right.

       Brace  expansions  may  be  nested.  The results of each expanded string are not sorted; left to
       right order is preserved.  For example, a{d,c,b}e expands into `ade ace abe'.

       A sequence expression takes the form {x..y}, where x and y are either integers or single charac-
       ters.  When integers are supplied, the expression expands to each number between x and y, inclu-
       sive.  When characters are supplied, the expression expands to each character  lexicographically
       between x and y, inclusive.  Note that both x and y must be of the same type.

       Brace  expansion  is  performed before any other expansions, and any characters special to other
       expansions are preserved in the result.  It is strictly textual.  Bash does not apply  any  syn-
       tactic interpretation to the context of the expansion or the text between the braces.

       A  correctly-formed  brace  expansion  must  contain unquoted opening and closing braces, and at
       least one unquoted comma or a valid sequence expression.  Any incorrectly formed brace expansion
       is left unchanged.  A { or , may be quoted with a backslash to prevent its being considered part
       of a brace expression.  To avoid conflicts with parameter expansion, the string ${ is  not  con-
       sidered eligible for brace expansion.

       This construct is typically used as shorthand when the common prefix of the strings to be gener-
       ated is longer than in the above example:

              mkdir /usr/local/src/bash/{old,new,dist,bugs}
       or
              chown root /usr/{ucb/{ex,edit},lib/{ex?.?*,how_ex}}

       Brace expansion introduces a slight incompatibility with historical versions of sh.  sh does not
       treat opening or closing braces specially when they appear as part of a word, and preserves them
       in the output.  Bash removes braces from words as a consequence of brace expansion.   For  exam-
       ple, a word entered to sh as file{1,2} appears identically in the output.  The same word is out-
       put as file1 file2 after expansion by bash.  If strict compatibility with sh is  desired,  start
       bash  with  the  +B option or disable brace expansion with the +B option to the set command (see
       SHELL BUILTIN COMMANDS below).

Brace Expansion

Brace expansion is a mechanism by which arbitrary strings may be generated. This mechanism is

similar to pathname expansion, but the filenames generated need not exist. Patterns to be brace

expanded take the form of an optional preamble, followed by either a series of comma-separated

strings or a sequence expression between a pair of braces, followed by an optional postscript.

The preamble is prefixed to each string contained within the braces, and the postscript is then

appended to each resulting string, expanding left to right.

Brace expansions may be nested. The results of each expanded string are not sorted; left to

right order is preserved. For example, a{d,c,b}e expands into `ade ace abe'.

A sequence expression takes the form {x..y}, where x and y are either integers or single charac-

ters. When integers are supplied, the expression expands to each number between x and y, inclu-

sive. When characters are supplied, the expression expands to each character lexicographically

between x and y, inclusive. Note that both x and y must be of the same type.

Brace expansion is performed before any other expansions, and any characters special to other

expansions are preserved in the result. It is strictly textual. Bash does not apply any syn-

tactic interpretation to the context of the expansion or the text between the braces.

A correctly-formed brace expansion must contain unquoted opening and closing braces, and at

least one unquoted comma or a valid sequence expression. Any incorrectly formed brace expansion

is left unchanged. A { or , may be quoted with a backslash to prevent its being considered part

of a brace expression. To avoid conflicts with parameter expansion, the string ${ is not con-

sidered eligible for brace expansion.

This construct is typically used as shorthand when the common prefix of the strings to be gener-

ated is longer than in the above example:

mkdir /usr/local/src/bash/{old,new,dist,bugs}

chown root /usr/{ucb/{ex,edit},lib/{ex?.?*,how_ex}}

Brace expansion introduces a slight incompatibility with historical versions of sh. sh does not

treat opening or closing braces specially when they appear as part of a word, and preserves them

in the output. Bash removes braces from words as a consequence of brace expansion. For exam-

ple, a word entered to sh as file{1,2} appears identically in the output. The same word is out-

put as file1 file2 after expansion by bash. If strict compatibility with sh is desired, start

bash with the +B option or disable brace expansion with the +B option to the set command (see

SHELL BUILTIN COMMANDS below).

命令替换

 Command Substitution
       Command substitution allows the output of a command to replace the command name.  There are  two
       forms:

              $(command)
       or
              `command`

       Bash performs the expansion by executing command and replacing the command substitution with the
       standard output of the command, with any trailing newlines deleted.  Embedded newlines  are  not
       deleted,  but  they  may be removed during word splitting.  The command substitution $(cat file)
       can be replaced by the equivalent but faster $(< file).

       When the old-style backquote form of substitution is used, backslash retains its literal meaning
       except  when followed by $, `, or \.  The first backquote not preceded by a backslash terminates
       the command substitution.  When using the $(command) form, all characters between the  parenthe-
       ses make up the command; none are treated specially.

       Command  substitutions  may be nested.  To nest when using the backquoted form, escape the inner
       backquotes with backslashes.

       If the substitution appears within double quotes, word splitting and pathname expansion are  not
       performed on the results.

Command Substitution

Command substitution allows the output of a command to replace the command name. There are two

forms:

$(command)

`command`

Bash performs the expansion by executing command and replacing the command substitution with the

standard output of the command, with any trailing newlines deleted. Embedded newlines are not

deleted, but they may be removed during word splitting. The command substitution $(cat file)

can be replaced by the equivalent but faster $(< file).

When the old-style backquote form of substitution is used, backslash retains its literal meaning

except when followed by $, `, or \. The first backquote not preceded by a backslash terminates

the command substitution. When using the $(command) form, all characters between the parenthe-

ses make up the command; none are treated specially.

Command substitutions may be nested. To nest when using the backquoted form, escape the inner

backquotes with backslashes.

If the substitution appears within double quotes, word splitting and pathname expansion are not

performed on the results.