本文記錄了一次RedHat AS release4上出現找不到網絡卡和mysql啟動不了解決方案並解決問題的過程

0.機器環境:

[[email protected] ~]# lsb_release -a
LSB Version:    :core-3.0-ia32:core-3.0-noarch:graphics-3.0-ia32:graphics-3.0-noarch
Distributor ID: RedHatEnterpriseAS
Description:    Red Hat Enterprise Linux AS release 4 (Nahant Update 4)
Release:        4
Codename:       NahantUpdate4

[[email protected] ~]# mysql --version
mysql  Ver 14.14 Distrib 5.1.55, for pc-linux-gnu (i686) using  EditLine wrapper

問題描述：

1.開機不能聯網，找不到網絡卡eth0和eth1，只能找到lo網絡卡

service network start:

Bringing up interface eth0: e1000 device eth0 does not seem to be present, delaying initialization.

解決方案：

 在communities.vmware.com看到一個老外說，他在遇到這個問題時說：“What I found was that my modules.dep and modules.alias were empty,which live in /lib/modules/`uname-r`/”。這讓我想起了,下午我在編一個模組時，執行過depmod，這是會改變modules.dep和modules.alias。到/lib/modules/`uname-r`/一看，果然，裡面只有與我的模組有關的依賴資訊。看來，這是由於我以錯誤的方式執行了depmod命令造成的。可能是由於在向安裝好的php中新增新模組的時候編譯了openssl和curl,而這個過程有可能改變了modules.dep

 看來是系統模組的依賴關係資訊丟失了，啟動時網絡卡驅動等模組無法載入。所以，如果想用modprobe向系統加自己的模組時，應將your_module.ko拷入/lib/modules/`uname -r`/，然後depmod -a 。最後，modprobyour_module。

在/lib/modules/核心版本號/ 執行depmod 重新生成modules.dep和modules.alias

2.Mysql啟動不了

2.1Starting MySQL.Manager of pid-file quit without updating file.[FAILED]mysql

檢視錯誤日誌
預設的mysql錯誤日誌是在/usr/local/mysql/data 下面 more ****.err 有下面的資訊
090302 02:27:26 mysqld_safe mysqld from pid file /opt/mysql/web2.pid ended
090302 02:30:03 mysqld_safe Starting mysqld daemon with databases from /opt/mysql
090302 2:30:03 [ERROR] /usr/local/mysql/libexec/mysqld: unknown option'--skip-federated'
090302 2:30:03 [ERROR] Aborting
解決方法：
只要將/etc/my.cnf裡面的 skip federated註釋掉即可。

2.2

mysql不能啟動 Manager of pid-file quit without updating file

啟動mysql失敗，提示Manager of pid-file quit without updating file.錯誤。從網上直接搜尋錯誤資訊，感覺出錯原因及解決方案都不適用於本機的情況。

本著出錯要查出其原因的原則，查看了mysql的出錯日誌（出錯日誌所在路徑，請根據您自己機子的實際情況查詢），發現錯誤資訊：File ‘./mysql-bin.index’ not found (Errcode: 13)，檢視資料庫檔案，mysql-bin.index確實存在。

檢視資料庫檔案的許可權，發現全都變成了 drwx—— 2 1001 1001。突然想起之前重建過mysql使用者，造成資料庫檔案的許可權使用者資訊丟失。重新授權即可。

網上有不少這個原因的解釋，但是都不是我想說的。我要說的原因其實很白痴：data/mysql-bin.index沒有刪除，data/mysql-bin.index是存放日誌檔案索引的檔案，只刪除了日誌檔案而沒有對日誌的索引檔案做處理顯然是不行的。

2.3 errcode 122 waiting for someone to free sapce

Running Centos 6.5

We have been getting these errors for a few days when mysql tries to write to the datafiles in /var/lib/mysql tree. The file names vary. There is 120Gb free in the root partion. Every thing is in root.

Filesystem 1K-blocks Used Available Use% Mounted on
/dev/mapper/vg_myserver-lv_root 150245636 14936520 127681188 11% /
tmpfs 4030892 0 4030892 0% /dev/shm
/dev/sda1 495844 116407 353837 25% /boot

140506 16:11:28 [Warning] Disk is full writing './george_wpress/ella_commentmeta.TMD' (Errcode: 122). Waiting for someone to free space... (Expect up to 60 secs delay for server to continue after freeing disk space)

As a test I ( using dd command ) created a 5G testfile in the /var/directory without problem.

The total size of the mysql directory is around 500Mb

The fault was resovled by switching off the quota mechanism using the
quotaoff command

2.4 MySQL is running but PID file is not found

今天vps訪問速度很慢，我用/root/lnmp restart 命令給lnmp重啟了一下，重啟之後發覺mysql連結不上了。 ssh裡面執行 service mysql restrat 出現 MySQL is running but PID file could not be found 錯誤。

用mysql -u root -p 命令執行mysql時出現 Can't connect to MySQL through socket mysql.sock 錯誤。

後來網上找到一篇不錯的文章 mysql PID file not found and Can't connect to MySQL through socket mysql.sock雖然後來我是直接重啟vps(reboot)解決這個問題，不過上面文章裡的很多內容還是挺有用的。

記錄幾個以前不知道的知識：

1、grep 是查詢命令 可做檔案內的字串查詢，使用如： ps aux | grep mysql

2、ps 命令 顯示所有程序 跟windows系統的 tasklist 命令相似 aux 顯示所有包含其他使用者的行程 比如我執行 ps aux | grep mysql 命令後能夠找到

–datadir=/usr/local/mysql/var–pid-file=/usr/local/mysql/var/li220-124.pid

3、kill -9 [PID] 終止程序命令 通過ps 找到程序pid 然後用kill -9 [PID]終止程序(-9表示強迫程序立即停止)

4、hostname 命令顯示當前主機名 uname -a 命令顯示當前系統核心版本

5、service命令

通過 service mysql status 可以檢視 mysql的狀態。但是其他服務名稱如何查詢？

發現通過 ls /etc/init.d/ 目錄下面可以找到各個服務名稱。

6、top命令 檢視記憶體和cup使用情況

< shift+m 按記憶體使用排序>

<1 顯示cpu使用情況>

7、chown 更改屬主和屬主組

chown [OPTION]… [OWNER][:[GROUP]] FILE…

如：chown mysql:mysql redhat.pid 更改redhat.pid 的ownner為mysql 所屬組為mysql
2.5 MySQL: mysql is not running but lock exists

A MySQL crash on Red Hat system resulted with:

/etc/init.d/mysql status

coming back with:

mysql is not running but lock exists

Solved by removing the lock file:

rm /var/lock/subsys/mysql

If this happens again it may also be necessary to remove the pid file from /var/lib/mysql
2.6
MySQL: write failed, user block limit reached

this mean that you don't have enouph space to create this backup file. check your qouta and file size

2.7
 sda5: write failed, user block limit reached

1.

da5: write failed, user block limit reached ???

sda5 : write failed, user block limit reached ???

--------------------------------------------------------------------------------

I got write failed, user block limit reached in my server shell ....
I can not start apache and mysql because if I do that server load average goes up ti 100! in seconds..!!  

Already runed /scripts/fixquotas but the problem stills.

sda5 is 50% free space....

I use RH4.

Last edited by sh4ka; 01-31-2006 at 06:05 PM.

Reply With QuoteReply With Quote

2.

01-31-2006, 06:14 PM

nyjimbo nyjimbo is offline

Registered Member

nyjimbo's Avatar
Join DateJan 2003LocationNew YorkPosts1,123

Default

did you try to google or yahoo search:

write failed, user block limit reached

Sounds like you ran out of some kind of room either due to quotas or some other limits.

Is this a virtual server ?

"A dog has raised it’s hind leg on the age of nevermore !"
-- Rolf

Reply With QuoteReply With Quote

3.

01-31-2006, 06:22 PM

sh4ka sh4ka is offline

Registered Member

sh4ka's Avatar
Join DateMay 2005Posts435

Default

Yes is a virtual server...
I already searched google and know the problem is with the disk quota maybe it was bad configured at the setup.. I really don't know, that's why I am asking.

On sda5 I have the / partition... including home, and the system directories. /tmp /boot are the other partitions..

[email protected] [~]# quota
Disk quotas for user root (uid 0): none

[email protected] [~]# df -h
Filesystem Size Used Avail Use% Mounted on
/dev/sda5 71G 35G 32G 53% /
/dev/sda1 99M 11M 83M 12% /boot
/dev/sda3 1012M 36M 925M 4% /tmp
none 506M 0 506M 0% /dev/shm
/dev/sdb1 74G 20G 50G 29% /backup/cpbackup

Also the fstab is well configured for this partition:

[email protected] [~]# cat /etc/fstab
/dev/sda5 / ext3 defaults,usrquota 1 1

Any ideas, pleaseeeee ?

thkz!

Reply With QuoteReply With Quote

4.

02-01-2006, 01:48 AM

madaboutlinux madaboutlinux is offline

Registered Member

Join DateJan 2005LocationEarthPosts1,052

Default

If its a VPS, the hosting company do set a Quota limit for each VPS on the main VPS server. There are Disk quotas set for each VPS to control the number of inodes a vps can use. You will have to ask your hosting company to change the quota limit for you. It seems you have reached your Hard limit. You will not be able to sort this issue at your end.

Hope this will sort out your issue.

Reply With QuoteReply With Quote

5.

02-01-2006, 05:32 AM

xerophyte xerophyte is offline

Registered Member

Join DateMar 2003LocationCanadaPosts217

Default

That error happens when you are over the quota, and you try to write to the disk.

In your case looks like one (many) of your hosted site over the quota, and when they try to write to the disk you get that error. Check site(s) which are over the quota.

hope that helps

相關資訊截圖：

1.mysql log

2.檢視 mysql程序

3.mysql失敗

4.mysql版本

5.RedHat版本

6.ulimit設定