技术支持1

一、确认现场故障，将故障分门别类   2012/10/23~2012/10/24

二、针对某一类别的故障，进行处理--并跟踪反馈，已知故障类别：

       1、LABEL设置问题导致系统进不去或者启动不了。  2012/10/23 修正2台，并进行观察

       2、IE无法登陆进去。  2012/10/23 修改mysql的配置，增加日志级别，着重分析什么因素导致数据全乱

 

 

 

-----------------------------------------------------由技术支持来跟踪------------------------------------------

       3、BIOS升级部分，由研华硬件人员来确定问题：

             3.1、找到新上项目，跟踪新上项目是否存在硬盘问题。

 

 

 

针对二：

解决思路如下：每一种类别抽取2台机器进行维修，然后就近替换附近的工控机，

 

 

磁盘故障问题：

问题描述：

/1: ******* FILE SYSTEM WAS MODIFIED ****

/1: ******* REBOOT LINUX *****

/1:106352/2050272 files (0.2% NO-contiguous), 597367/2048287 blocks

 

 

192.168.0.208

what：磁盘故障，系统无法进入  why：磁盘只读，且只读发生在系统分区   how：fsck修复，然后修改fstab的配置

 

192.168.10.206

what：ie登陆不进去，提示 登录人数超过最大设置,请联系管理员修改最大登录数值或稍后登录!

          TSMIS-2.1-1

why：数据库损坏，值被变更。

how：修改日志级别，确认数据库的值为什么被变更。 简洁修复方法用：使用肖东光的升级包

 

     [mysqld]---开启日志

log=/var/log/mysqld_common.log
log-error=/var/log/mysqld_err.log
log-bin=/var/log/mysqld_bin.bin

 

 

碰到问题：

 

'./mysql-bin.index' not found (Errcode: 13) 的解决方法

解决问题：

  

'./mysql-bin.index' not found (Errcode: 13) 的解决方法

分类： Mysql 2011-07-05 22:15  2581人阅读  评论(0)  收藏  举报

【转自】http://blog.haohtml.com/archives/9202

 

今天突然收到消息机房的一台服务器的mysql无法启动了，首先检查了一下mysql的错误日志，发现最后出现以下错误：

020101 00:42:21  mysqld started
/usr/local/mysql/libexec/mysqld: File './mysql-bin.index' not found (Errcode: 13)
020101  0:42:21 [ERROR] Aborting

020101  0:42:21 [Note] /usr/local/mysql/libexec/mysqld: Shutdown complete

提示./mysql-bin.index无法找到（由于mysql开启了bin日志功能），到数据库根目录查看该文件是存在的，可能是文件权限的问题，查看了数据库根目录的权限是700，所有者和用户组都是root，可能是上次转移数据库的时候不小心修改了文件夹的权限。

解决方法：

chgrp -R mysql /usr/local/mysql/data && chown -R mysql /usr/local/mysql/data 

chgrp -R mysql /usr/share/TSMIS/mysql/test && chown -R mysql /usr/share/TSMIS/mysql/test

 

--

 

 

1. 090211  0:23:19 [ERROR] Could not use /var/lib/mysql/testbox-04-slow.log for logging (error 13). Turning logging off for the whole duration of the MySQL server process. To turn it on again: fix the cause, shutdown the MySQL server and restart it.
2. 090211  0:23:20  InnoDB: Started; log sequence number 0 43656

复制代码

mysql对 /var/log权限不够，需要新建一个文件夹如 /var/log/mysqllogdir 然后把所有生成的日志指向这里。

see actical： http://www.electrictoolbox.com/logging-queries-with-mysql/

 

还需要：

--》default-character-set=utf8 注释掉，不然日志没法读取
选项恢复为default-character-set=utf8

 

搞定之后，通过  show variables like 'log_bin'; 确认日志是否打开，打开时有如下提示：

mysql> show variables like 'log_bin';
+---------------+-------+
| Variable_name | Value |
+---------------+-------+
| log_bin       | ON    |
+---------------+-------+

 

然后升级web升级包。

 

 

 

 

--------------------------------解决步骤如下

1.把日志级别都打开，特别是common日志和bin的二进制日志

2.统计出现web故障的机器，10分钟内重启的次数，重启的原因，确认数据库损坏与重启是否有直接关系。

 

 

 

-----------------每一样机器都重新打一个包，方便升级--------------------------------

 

web升级的包：，

 

goldway.war不需要更变，配置相关的表，重新构建

备份数据库中的数据以及日志信息

打开二进制日志，修改my.cnf的配置（新建日志文件夹、修改my.cnf的配置，修改default字符集配置）

 

针对系统无法进入的升级包：

 

 

 

Version: '5.0.77'  socket: '/usr/share/TSMIS/mysql/mysql.sock'  port: 3306  Source distribution
120923 13:06:42 [Note] Found 69 of 70 rows when repairing './test/functionality'
120923 13:06:42 [Note] Found 130 of 173 rows when repairing './test/role_functionality'

 

正常的有173条--默认的数据，但是数据修复的时候，把它修复成130条

 

 

Sep 23 13:06:41 TSMIS avahi-daemon[2550]: Host name conflict, retrying with <TSMIS-5>
Sep 23 13:06:41 TSMIS avahi-daemon[2550]: Registering new address record for fe80::222:46ff:fe14:213 on eth0.
Sep 23 13:06:41 TSMIS avahi-daemon[2550]: Registering new address record for 192.168.10.178 on eth0.
Sep 23 13:06:41 TSMIS avahi-daemon[2550]: Registering new address record for fe80::222:46ff:fe14:5a07 on eth1.
Sep 23 13:06:41 TSMIS avahi-daemon[2550]: Registering new address record for 192.168.6.33 on eth1.
Sep 23 13:06:41 TSMIS avahi-daemon[2550]: Registering HINFO record with values 'I686'/'LINUX'.
Sep 23 13:06:42 TSMIS avahi-daemon[2550]: Withdrawing address record for fe80::222:46ff:fe14:213 on eth0.
Sep 23 13:06:42 TSMIS avahi-daemon[2550]: Withdrawing address record for 192.168.10.178 on eth0.
Sep 23 13:06:42 TSMIS avahi-daemon[2550]: Withdrawing address record for 192.168.6.33 on eth1.
Sep 23 13:06:42 TSMIS avahi-daemon[2550]: Host name conflict, retrying with <TSMIS-6>
Sep 23 13:06:42 TSMIS avahi-daemon[2550]: Registering new address record for fe80::222:46ff:fe14:213 on eth0.
Sep 23 13:06:42 TSMIS avahi-daemon[2550]: Registering new address record for 192.168.10.178 on eth0.
Sep 23 13:06:42 TSMIS avahi-daemon[2550]: Registering new address record for fe80::222:46ff:fe14:5a07 on eth1.
Sep 23 13:06:42 TSMIS avahi-daemon[2550]: Registering new address record for 192.168.6.33 on eth1.
Sep 23 13:06:42 TSMIS avahi-daemon[2550]: Registering HINFO record with values 'I686'/'LINUX'.
Sep 23 13:06:43 TSMIS avahi-daemon[2550]: Withdrawing address record for fe80::222:46ff:fe14:213 on eth0.
Sep 23 13:06:43 TSMIS avahi-daemon[2550]: Withdrawing address record for 192.168.10.178 on eth0.
Sep 23 13:06:43 TSMIS avahi-daemon[2550]: Withdrawing address record for 192.168.6.33 on eth1.
Sep 23 13:06:43 TSMIS avahi-daemon[2550]: Host name conflict, retrying with <TSMIS-7>
Sep 23 13:06:43 TSMIS avahi-daemon[2550]: Registering new address record for fe80::222:46ff:fe14:213 on eth0.
Sep 23 13:06:43 TSMIS avahi-daemon[2550]: Registering new address record for 192.168.10.178 on eth0.
Sep 23 13:06:43 TSMIS avahi-daemon[2550]: Registering new address record for fe80::222:46ff:fe14:5a07 on eth1.
Sep 23 13:06:43 TSMIS avahi-daemon[2550]: Registering new address record for 192.168.6.33 on eth1.
Sep 23 13:06:43 TSMIS avahi-daemon[2550]: Registering HINFO record with values 'I686'/'LINUX'.

 

 

 

相减，等于 2304，接下去83886081，接下去67108863，左右相减是256

 

 

 

---权限问题（登陆不上以及其他相关），归根到底是数据库修复时引发

 

 

privilege_functionality 表

role_functionality 表。---导致缺项

 

 

120922 08：24：05  mysqld started

然后  120922 08:24:10 ~  120922 08:24:11  进行ftp修复，日志信息如下：

found 2  of  6 rows when repairing 表。

 

 

 

a -o -A -- auto-repair  而不是 -r，--r

 

 

 

『----调查最后一条数据在07：58：25，然后重启的原因是ip冲突』

(12591|2946448272)LM_ERROR,Sat Sep 22 2012 07:58:06.274104,CallBack.cpp,3449: 2012-09-22 07:58:05.210 | VN=1081,DT=0,PT=1,PN=1,RN=1,TN=1,VT
=0,Plate=ÎÞ³µÅÆ
(12591|2946448272)LM_ERROR,Sat Sep 22 2012 07:58:26.397562,CallBack.cpp,3449: 2012-09-22 07:58:25.315 | VN=1082,DT=0,PT=1,PN=1,RN=1,TN=1,VT
=0,Plate=ÎÞ³µÅÆ

 

对应的control日志：可以看出不是因为进程导致的重启

1 2012 15:15:45.897007,ProcessControl.cpp,190: SpawnAllChildren() App diskMaintain is already running
(2703|3086395088)LM_DEBUG,Sat Sep 22 2012 08:01:30.494150,../GDW_ConfigManager.cpp,136: NOW Initial
(2703|3086395088)LM_DEBUG,Sat Sep 22 2012 08:01:30.544336,../GDW_ConfigManager.cpp,143: Get the handle of sql
(2703|3086395088)LM_DEBUG,Sat Sep 22 2012 08:01:30.545242,../GDW_ConfigManager.cpp,151: initial sql
(2703|3086395088)LM_DEBUG,Sat Sep 22 2012 08:01:30.545368,../GDW_ConfigManager.cpp,179: Load
(2703|3086395088)LM_DEBUG,Sat Sep 22 2012 08:01:30.545429,../GDW_ConfigManager.cpp,136: NOW Initial
(2703|3086395088)LM_DEBUG,Sat Sep 22 2012 08:01:30.545486,../GDW_ConfigManager.cpp,143: Get the handle of sql

 

 

对应的message日志

o action taken because this happened within the defend interval
Sep 21 16:41:00 TSMIS last message repeated 4 times
Sep 22 07:29:05 TSMIS ipwatchd[2805]: MAC address 0:22:46:14:9d:66 causes IP conflict with address 192.168.10.208 set on interface eth0 - p
assive mode - reply not sent
Sep 22 07:36:41 TSMIS last message repeated 2 times
Sep 22 07:43:46 TSMIS ipwatchd[2805]: MAC address 0:22:46:14:9d:66 causes IP conflict with address 192.168.10.208 set on interface eth0 - p
assive mode - reply not sent
Sep 22 07:52:35 TSMIS ipwatchd[2805]: MAC address 0:22:46:14:9d:66 causes IP conflict with address 192.168.10.208 set on interface eth0 - p
assive mode - reply not sent
Sep 22 07:59:35 TSMIS ipwatchd[2805]: MAC address 0:22:46:14:9d:66 causes IP conflict with address 192.168.10.208 set on interface eth0 - p
assive mode - reply not sent
Sep 22 08:01:04 TSMIS syslogd 1.4.1: restart.

 

 

 

 

 

 

 

机器不断重启，除非load_watchdog.log watchdog_daemon

 

 

 

有机器磁盘只读

 

cmdService=192.168.11.249
dataManage=10.83.59.71

 

一台 缺少etc