Solution to the failure of MySQL service startup under CentOS 7

Today, starting the MySQL server failed as follows:

[root@spark01 ~]# /etc/init.d/mysqld start
Starting mysqld (via systemctl): Job for mysqld.service failed because the control process exited with error code. See "systemctl status mysqld.service" and "journalctl -xe" for details.
                                                           [FAILED]
According to the prompts, use systemctl status mysqld.service and journalctl -xe respectively to check the reason for the failure to start the service

[root@spark01 ~]# systemctl status mysqld. serviceCopy

code
?.mysqld.service - SYSV: MySQL database server. Loaded
   : loaded (/etc/rc.d/init.d/mysqld)
   Active: failed (Result: exit-code) since Wed 2016-01-20 18 :26:57 CST; 40s ago
     Docs: man:systemd-sysv-generator(8)
  Process: 2979 ExecStart=/etc/rc.d/init.d/mysqld start (code=exited, status=1/FAILURE)

Jan 20 18:26:56 spark01 systemd[1]: Starting SYSV: MySQL database server....
Jan 20 18:26:57 spark01 mysqld[2979]: MySQL Daemon failed to start.
Jan 20 18:26:57 spark01 mysqld[2979]: Starting mysqld:  [FAILED]
Jan 20 18:26:57 spark01 systemd[1]: mysqld.service: control process exited, code=exited status=1
Jan 20 18:26:57 spark01 systemd[1]: Failed to start SYSV: MySQL database server..
Jan 20 18:26:57 spark01 systemd[1]: Unit mysqld.service entered failed state.
Jan 20 18:26:57 spark01 systemd[1]: mysqld.service failed.
复制代码
[root@spark01 ~]# journalctl -xe

复制代码
--
-- Unit session-2.scope has begun starting up.
Jan 20 18:26:48 spark01 sshd[2916]: pam_unix(sshd:session): session opened for user spark by (uid=0)
Jan 20 18:26:52 spark01 su[2944]: (to root) spark on pts/1
Jan 20 18:26:52 spark01 su[2944]: pam_unix(su-l:session): session opened for user root by spark(uid=1000)
Jan 20 18:26:56 spark01 polkitd[909]: Registered Authentication Agent for unix-process:2974:117137 (system bus name :1.25
Jan 20 18:26:56 spark01 systemd[1]: Starting SYSV: MySQL database server....
-- Subject: Unit mysqld.service has begun start-up
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
--
-- Unit mysqld.service has begun starting up.
Jan 20 18:26:57 spark01 mysqld[2979]: MySQL Daemon failed to start.
Jan 20 18:26:57 spark01 mysqld[2979]: Starting mysqld:  [FAILED]
Jan 20 18:26:57 spark01 systemd[1]: mysqld.service: control process exited, code=exited status=1
Jan 20 18:26:57 spark01 systemd[1]: Failed to start SYSV: MySQL database server..
-- Subject: Unit mysqld.service has failed
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
--
-- Unit mysqld.service has failed.
--
-- The result is failed.
Jan 20 18:26:57 spark01 systemd[1]: Unit mysqld.service entered failed state.
Jan 20 18:26:57 spark01 systemd[1]: mysqld.service failed.
Jan 20 18:26:57 spark01 polkitd[909]: Unregistered Authentication Agent for unix-process:2974:117137 (system bus name :
1.copy
However , unfortunately, these information cannot provide the real reason for the failure of the service to start.

At this time, you might as well open the alarm log of MySQL. After all, as long as the MySQL service starts, the alarm log will have output information. Sure enough

2016-01-20T10 :00:19.935771Z 0 [ERROR] /usr/sbin/mysqld: Can't create/write to file '/var/run/mysqld/mysqld.pid' (Errcode: 2 - No such file or directory)
2016-01 -20T10:00:19.935795Z 0 [ERROR] Can't start server: can't create PID file: No such file or directory
160120 18:00:20 mysqld_safe mysqld from pid file /var/run/mysqld/mysqld.pid ended
The pid file cannot be created when the MySQL service is started. Check

whether the directory exists in the terminal. Sure enough, it does not exist.

So, the /var/run/mysqld/ directory was created and the MySQL service was restarted

[root@spark01 ~]# mkdir -p /var/run/mysqld/

[root@spark01 ~]# /etc/init.d/mysqld start

Starting mysqld (via systemctl):  Job for mysqld.service failed because the control process exited with error code. See "systemctl status mysqld.service" and "journalctl -xe" for details.
                                                           [FAILED]
依旧报错，重新查看告警日志，有以下输出

2016-01-20T10:28:37.183387Z 0 [ERROR] /usr/sbin/mysqld: Can't create/write to file '/var/run/mysqld/mysqld.pid' (Errcode: 13 - Permission denied)
2016-01-20T10:28:37.183431Z 0 [ERROR] Can't start server: can't create PID file: Permission denied
160120 18:28:37 mysqld_safe mysqld from pid file /var/run/mysqld/mysqld.pid ended
160120 18:32:06 mysqld_safe Starting mysqld daemon with databases from /var/lib/mysql
Originally, the owner and group of /var/run/mysqld/ are still root, mysql cannot create files in it, and then modify the directory Owner and group, start OK.

[root@spark01 ~]# ls -ld /var/run/mysqld/
drwxr-xr-x 2 root root 40 Jan 20 18:28 /var/run/mysqld/
[root@spark01 ~]# chown mysql.mysql / var/run/mysqld/
[root@spark01 ~]# /etc/init.d/mysqld start
Starting mysqld (via systemctl): [ OK ]


Summary:

When playing Kubernetes, I often encountered startup failures. According to The systemctl prompt, through the systemctl status mysqld.service and journalctl -xe commands to check the cause of the service startup failure is often not satisfactory, but gives a wrong hint, thinking that this is related to the system. In fact, by viewing the log of the service, it is often more clear why the service fails to start.