Use this command to get nginx compiler arguments:
nginx -V
Of cause, you should make sure that nginx command is in your PATH variable.
Use this command to get php compiler arguments:
php -i |grep configure
The php command also need in your PATH, if not, use the absolute path.
Use the below way to get mysql compiler arguments:
cat /usr/local/mysql/bin/mysqlbug | grep CONFIGURE_LINE
By the way, you should find where is your mysqlbug ^__^
2013年4月2日星期二
rsync @ERROR: chroot failed
1. Q: 出現以下這個訊息,
是怎麼一回事?
@ERROR: auth failed on module xxxxx
rsync: connection unexpectedly closed (90 bytes read so far)
rsync error: error in rsync protocol data stream (code 12) at
io.c(150)
A: 這是因為密碼設錯了, 無法登入成功, 請再檢查一下 rsyncd.secrets
中的密碼設定, 二端是否一致?
2. Q: 出現以下這個訊息,
是怎麼一回事?
password
file must not be other-accessible
continuing
without password file
Password:
A: 這表示 rsyncd.secrets 的檔案權限屬性不對, 應設為
600
請下 chmod 600 rsyncd.secrets
3. Q: 出現以下這個訊息,
是怎麼一回事?
@ERROR: chroot failed
rsync: connection unexpectedly closed (75 bytes read so far)
rsync error: error in rsync protocol data stream (code 12) at
io.c(150)
A: 這通常是您的 rsyncd.conf 中
的 path 路徑所設的那個目錄並不存在所致.
請先用 mkdir 開設好備份目錄.
的 path 路徑所設的那個目錄並不存在所致.
請先用 mkdir 開設好備份目錄.
PHP 升级后phtml 无法访问一例
nginx 配置包含了phtml
location ~ \.(php|phtml) {
fastcgi_pass fastcgi_upstream_p;
fastcgi_index index.php;
include fastcgi_params;
expires off;
}
访问phtml 文件时出现
Access denied ****
权限没有任何问题；
按照错误提示，看了一下php changelog
查看#55181 ，添加了一个参数 security.limit_extensions = .php （估计.php
是默认支持的，所以加没加无影响）
原来的版本是 5.3.3 新环境版本 5.3.10 不用想都明白了。
2013年4月1日星期一
yum 删除mysql-server 时候出错
错误如下：
error reading information on service mysqld: No such file or
directory
Error in PREUN scriptlet in rpm package m
Erasing :
mysql-connector-odbc 8/9
error: %preun(mysql-server-5.0.95-1.el5_7.1.x86_64) scriptlet failed, exit
status 1
原来装的是yum 下来的rpm 包，后面为了测试osa ，删的时候，先把/etc/ini.d/mysqld 删掉了，后面yum remove 的时候就一直报错。
解决方法： 从别处cp 一个mysqld 文件。（如果没猜错，只是判断这个文件是否存在的话，那就更简单了，新建一个可以搞定，没来得及试）
mosquitto 安装记录
wget http://mosquitto.org/files/source/mosquitto-1.0.3.tar.gz
tar -zxvf mosquitto-1.0.3.tar.gz
cd mosquitto-1.0.3
make WITH_TLS_PSK=no (遇到undefined reference to
`SSL_CTX_set_psk_server_callback' ，添加参数WITH_TLS_PSK=no 禁用SSL PSK 支持)
make install prefix=/home/mosquitto
mkdir /home/mosquitto/etc
mv /etc/mosquitto/* /home/mosquitto/etc/
strip /home/mosquitto/bin/*
strip /homo/mosquitto/sbin/*
strip /homo/mosquitto/lib/*
echo "/home/mosquitto/lib/" >> /etc/ld.so.conf
ldconfig -f /etc/ld.so.conf
修改 /homo/mosquitto/etc/mosquitto.conf 用户：
user nobody
（其它参数目前使用默认值）
启动服务（如有服务相关错误，检查/home/mosquitto/mosquitto.log，端口默认1883）
/home/mosquitto/sbin/mosquitto -d -c /home/mosquitto/etc/mosquitto.conf
> /home/mosquitto/mosquitto.log 2>&1
终端测试：
客户端
mosquitto_sub -h SERVERIP -t test
服务器端执行后
/home/mosquitto/bin/mosquitto_pub -t test -m "123456"
客户端会成功收到"123456"
查看Mqtt订阅者运行状况
mosquitto_sub -v -t \$SYS/#
或者细化为其中一个命令
mosquitto_sub -v -t ‘$SYS/broker/clients/active’
redmine submit issue slowly
Why creating or updating an issue is so slow ?!
回复 (2)
(1-2/2)
找到原因，就要找解决方法：（红色字）
Email Configuration
Configuration Directives
This
page is a work in progress, the following configuration directives are only a
partial list.
authentication
The
type of authentication method expected by your service provider.
Valid
settings:
- :login
- :none
(note:
if you set this to
:none,
you must not include the user_name and password settings)delivery_method
The
mail transport method to be used.
Valid
settings:
- :smtp
- :async_smtp
- :sendmail
- :async_sendmail
Asynchronous delivery_methods
The
:async_smtp and :async_sendmail use asynchronous sends, which means Redmine
does not wait for the email to be sent to display the next page. SeeAsynchronous Email Delivery for more details.
NB : if
you want to use asynchronous sends, be careful there's a regression in Redmine
2.x which implies a modification of the "smtp_settings" key in
your
config/configuration.yml. See #11376 for more informations.Example configuration.yml Configurations
Simple Login Authentication (default settings)
# Outgoing email settings
production:
email_delivery:
delivery_method: :smtp
smtp_settings:
address: smtp.example.net
port: 25
domain: example.net
authentication: :login
user_name: redmine@example.net
password: redmine
development:
email_delivery:
delivery_method: :smtp
smtp_settings:
address: 127.0.0.1
port: 25
domain: example.net
authentication: :login
user_name: redmine@example.net
password: redmine
If
you want to use GMail/Google Apps and other TLS-requiring SMTP servers,
you'll have to add some TLS-related settings :
production:
email_delivery:
delivery_method: :smtp
smtp_settings:
enable_starttls_auto: true
address: "smtp.gmail.com"
port: '587'
domain: "smtp.gmail.com"
authentication: :plain
user_name: "your_email@gmail.com"
password: "your_password"
However,
this will only work with "recent" enough ruby and rails versions (1.8.7 patchset
2xx and 2.3.5).
(See #5814 )
(See #5814 )
No Authentication
Example
for an SMTP service provider with no authentication. Note the colon before
none.
production:
email_delivery:
delivery_method: :smtp
smtp_settings:
address: smtp.knology.net
port: 25
domain: cybersprocket.com
authentication: :none
Using sendmail command
Example
for a unix system that uses the
/usr/sbin/sendmail command.
production:
email_delivery:
delivery_method: :sendmail
More information
将原来的
delivery_method: :smtp
smtp_settings:
改成：
delivery_method: :async_smtp
async_smtp_settings:
redhat entprise 6.2 配置 multipath
[root@xwumpsdb ~]# yum -y install device-mapper-multipath
[root@xwumpsdb ~]# multipath -v2
Jan 09 10:49:42 | /etc/multipath.conf does not exist, blacklisting all
devices.
Jan 09 10:49:42 | A sample multipath.conf file is located at
Jan 09 10:49:42 |
/usr/share/doc/device-mapper-multipath-0.4.9/multipath.conf
Jan 09 10:49:42
| You can run /sbin/mpathconf to create or modify /etc/multipath.conf
复制一个配置文件到 /etc 下
[root@xwumpsdb ~]# modprobe dm-multipath 加载模块
[root@xwumpsdb ~]# service multipathd start 启动服务
[root@xwumpsdb ~]# multipath -v2 查看设备状态
Jan 09 10:53:42 | mpatha: ignoring map
create: mpathb (36782bcb00066e968000008f84f049685) undef DELL,MD36xxf
size=1.0T features='0' hwhandler='0' wp=undef
|-+- policy='round-robin 0' prio=1 status=undef
| `- 1:0:0:0 sdb 8:16 undef ready running
|-+- policy='round-robin 0' prio=0 status=undef
| `- 1:0:1:0 sdc 8:32 undef faulty running
|-+- policy='round-robin 0' prio=1 status=undef
| `- 2:0:1:0 sde 8:64 undef ready running
`-+- policy='round-robin 0' prio=0 status=undef
`- 2:0:0:0 sdd 8:48
undef faulty running
[root@xwumpsdb ~]# fdisk -l
再使用fdisk 看，会出现多个盘符
Disk /dev/mapper/mpathbp1 doesn't contain a valid partition
table
Disk /dev/sda: 146.2 GB, 146163105792 bytes
255 heads, 63 sectors/track, 17769 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x00000080
Device Boot Start End Blocks Id System
/dev/sda1 * 1 26 204800 83 Linux
Partition 1 does not end on cylinder boundary.
/dev/sda2 26 281 2048000 82 Linux swap / Solaris
Partition 2 does not end on cylinder boundary.
/dev/sda3 281 17770 140483584 83 Linux
Disk /dev/sdb: 1099.5 GB, 1099511627776 bytes
255 heads, 63 sectors/track, 133674 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x072cb38e
Device Boot Start End Blocks Id System
/dev/sdb1 1 133674 1073736373+ 83 Linux
Disk /dev/sde: 1099.5 GB, 1099511627776 bytes
255 heads, 63 sectors/track, 133674 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x072cb38e
Device Boot Start End Blocks Id System
/dev/sde1 1 133674 1073736373+ 83 Linux
Disk /dev/mapper/mpathb: 1099.5 GB, 1099511627776 bytes
255 heads, 63 sectors/track, 133674 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x072cb38e
Device Boot Start End Blocks Id System
/dev/mapper/mpathbp1 1 133674 1073736373+ 83 Linux
Disk /dev/mapper/mpathbp1: 1099.5 GB, 1099506046464 bytes
255 heads, 63 sectors/track, 133673 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x00000000
内核参数引发的思考
问题描述：由于步前路，完成剩下的Load Balance 测试，今天玩的是Haproxy。安装和部署
都只是碰到一些小的问题，基本上问题也不是问题了（此类为可通过Google Baidu 轻松解决）。
确认可以正常访问后，分别给两个web server web1 和web2 写了个静态页面，确认可以正常访问后，
在log 服务器上用 ab 做压力测试，按照测试的基准，从100000 请求数的静态页面开始测试
root@log:~# ab -c 10 -n 100000 http://192.168.104.132:80/
This is ApacheBench, Version 2.3 <$Revision: 655654 $>
Copyright 1996 Adam Twiss, Zeus Technology Ltd,
http://www.zeustech.net/
Licensed to The Apache Software Foundation, http://www.apache.org/
Benchmarking 192.168.104.132 (be patient)
apr_poll: The timeout specified has expired
(70007)
Total of 6263 requests
completed
只完成了6263 个请求，这年头，瞎猫子碰死老鼠也不容易，因为这个系统是之前测试过Nginx 的，
理论上跟系统没有关系，不然在测试Nginx 并发的时候，就应该暴露了，检查了Haproxy 的配置文件，把
CPU 的相关东西调大一些，测试，错误依旧；进程数调多一些，测试，错误依旧；把Haproxy 超时时间
降低，错误依旧……
吃饭的时候，也没有点头绪，搞技术怕的就是找不到原因，而不是如何解决，中午休息时间，buffer
一下，睡完了起来接着整吧！
后来再想想，看看是不是TCP 连接数的问题，
root@log:~# ab -c 10 -n 100000 http://192.168.104.132:80/
This is ApacheBench, Version 2.3 <$Revision: 655654 $>
Copyright 1996 Adam Twiss, Zeus Technology Ltd,
http://www.zeustech.net/
Licensed to The Apache Software Foundation, http://www.apache.org/
Benchmarking 192.168.104.132 (be patient)
Completed 10000 requests
apr_poll: The timeout specified has expired
(70007)
Total of 15836 requests
completed
root@dispatch:/usr/local/haproxy# netstat -ant |grep TIME_WAIT | wc -l
8187
root@dispatch:/usr/local/haproxy# netstat -ant |grep TIME_WAIT | wc
-l
30
以上两次统计TIME_WAIT 连接数的情况是：8187 此时 ab 正在跑， 30 此数据 ab 已跑完15836
后退出。
这个情况让我觉得算是正常吧，TIME_WAIT 自动清除了，连接占用的时间也很少，放弃了查这个TIME_WAIT
相关参数。
于是还是围绕着超时的参数去转悠，找了个内核参数 /proc/sys/net/ipv4/tcp_fin_timeout 默认值是60
将其值
改成30 后，测试数据基本与上面相同，原因也不在这。
这个，情何以堪，随便再找找原因吧，看了一下/var/log/syslog 日志文件，妈呀，
dispatch kernel: [ 9984.162370] nf_conntrack: table
full, dropping packet.
大量的以上信息，我怎么才想着看系统日志，这么
，有了这个，很快就找到原来还是连接数的限制，看了
，有了这个，很快就找到原来还是连接数的限制，看了
一下，系统默认的是31732，
root@dispatch:/usr/local/haproxy# cat
/proc/sys/net/ipv4/netfilter/ip_conntrack_max
31732
将其后面加了个0，再跑 ab 测试，这下OK
root@dispatch:/var/log# echo "317320" >
/proc/sys/net/ipv4/netfilter/ip_conntrack_max
耗时 58秒完成ab，结果虽然重要，过程也相当的，如果早看了系统日志，也不至于花上半天的时间摸索。
关于测试结果，这里不是重点，忽略，谨以此作为处理问题的经验罢了。
proftpd 快速增删改用户
以下操作前提是proftp 安装的base 目录为 /home/proftpd
启动proftpd
/home/proftpd/sbin/proftpd -c /home/proftpd/etc/proftpd.conf
或直接
/home/proftpd/sbin/proftpd
新增用户：
/home/proftpd/bin/ftpasswd --name test -file /home/proftpd/etc/ftpd.passwd
--uid 99 --gid 99 --home /var/ftp/test/ --shell /sbin/nologin --passwd
删除用户：
/home/proftpd/bin/ftpasswd --delete-user --name bbb -file
/home/proftpd/etc/ftpd.passwd --uid 600 --gid 600 --home /var/ftp/bb/ --shell
/sbin/nologin --passwd
更改密码：
/home/proftpd/bin/ftpasswd --name bbb --change-password --passwd
proftpd.conf 文件内容修改部分如下：
日志文件路径：
SystemLog /home/proftpd/var/log/proftpd.syslog
TransferLog /home/proftpd/var/log/proftpd.transferlog
密码文件路径：
AuthUserFile /home/proftpd/etc/ftpd.passwd
另外需要确定服务器上的ftp 目录存在和权限。
订阅：
博文 (Atom)
订阅：
博文 (Atom)