3

I have a VPS with ChicagoVPS that's running Centos 6. Official version of the build is:

Linux RedHat 6.5(Final 2.6.32-042stab078.26 x86_64)

I followed this guide to add the repository and installed MySQL 5.6.15.

It installed with no errors. The following error is the one that I get whenever I try to access the database:

[root@sacroiliac log]# mysql
ERROR 2002 (HY000): Can't connect to local MySQL server through socket '/var/lib/mysql/mysql.sock' (2)

These are the commands I'm running. (I'm also adding the mysql portion because I read somewhere the daemon was changed from mysqld to mysql, but it's not being found on my system.)

[root@sacroiliac log]# service mysqld start
MySQL Daemon failed to start.
Starting mysqld:                                           [FAILED]
[root@sacroiliac log]# service mysql start
mysql: unrecognized service
[root@sacroiliac log]# /etc/init.d/mysql start
-bash: /etc/init.d/mysql: No such file or directory
[root@sacroiliac log]# /etc/init.d/mysqld start
MySQL Daemon failed to start.
Starting mysqld:                                           [FAILED]

There is no error file in /var/lib/mysql/ and /var/log/mysqld.log says the following:

131210 17:34:04 mysqld_safe Starting mysqld daemon with databases from /var/lib/mysql
131210 17:34:04 mysqld_safe mysqld from pid file /var/run/mysqld/mysqld.pid ended
131210 17:34:20 mysqld_safe Starting mysqld daemon with databases from /var/lib/mysql
131210 17:34:21 mysqld_safe mysqld from pid file /var/run/mysqld/mysqld.pid ended

There is MySQL 5.1 on the default repository, but I'd like 5.6 for full-text search of InnoDB databases, and the new(er) SHA2 algorithms.

Edit
Also, I wanted to note that I have rebuilt the VM a few times (eliminate the issue of having a corrupt download), and Googled the issue. None of the results I've found have been helpful at all. Most of them were actually unanswered questions similarly to mine.

Edit 2
/var/log/messages contains the following:

Dec 10 17:32:10 sacroiliac yum[1327]: Installed: mysql-community-release-el6-5.noarch
Dec 10 17:32:45 sacroiliac yum[1328]: Installed: mysql-community-common-5.6.15-1.el6.x86_64
Dec 10 17:32:46 sacroiliac yum[1328]: Installed: mysql-community-libs-5.6.15-1.el6.x86_64
Dec 10 17:32:51 sacroiliac yum[1328]: Installed: mysql-community-client-5.6.15-1.el6.x86_64
Dec 10 17:32:51 sacroiliac yum[1328]: Installed: libaio-0.3.107-10.el6.x86_64
Dec 10 17:33:04 sacroiliac yum[1328]: Installed: mysql-community-server-5.6.15-1.el6.x86_64
Dec 10 17:34:04 sacroiliac kernel: [627203.702793] OOM killed process 1657 (mysqld) vm:385760kB, rss:238984kB, swap:0kB
Dec 10 17:34:21 sacroiliac kernel: [627220.208520] OOM killed process 1919 (mysqld) vm:385760kB, rss:238300kB, swap:0kB
Dec 10 17:50:03 sacroiliac kernel: [628160.758387] OOM killed process 2228 (mysqld) vm:385760kB, rss:238988kB, swap:0kB
Dec 10 17:51:33 sacroiliac kernel: [628250.613146] OOM killed process 2486 (mysqld) vm:385760kB, rss:237220kB, swap:0kB
Dec 10 17:51:49 sacroiliac kernel: [628266.621825] OOM killed process 2744 (mysqld) vm:385760kB, rss:238412kB, swap:0kB

Edit 3 This is the result of trying to run mysqld_safe

131210 21:59:17 mysqld_safe Logging to '/var/log/mysqld.log'.
131210 21:59:17 mysqld_safe Starting mysqld daemon with databases from /var/lib/mysql
/usr/bin/mysqld_safe: line 166:  2677 Killed                  nohup /usr/sbin/mysqld --     basedir=/usr --datadir=/var/lib/mysql --plugin-dir=/usr/lib/mysql/plugin --user=mysql --  log-error=/var/log/mysqld.log --pid-file=/var/run/mysqld/mysqld.pid -- socket=/var/lib/mysql/mysql.sock < /dev/null >> /var/log/mysqld.log 2>&1
131210 21:59:18 mysqld_safe mysqld from pid file /var/run/mysqld/mysqld.pid ended

Edit 4

I never found a fix for my problem; I did get a solution. One of my issues was that I never had enough memory. A VM on my local machine was using 300MB or so, which was more than my VPS had. I tried a swap file, but my VPS runs on an OpenVZ kernel and they don't support swap files.

The solution ended up being a KVM VPS. It ended up costing a few bucks more a month with double the memory, storage, and bandwidth. MySQL 5.6 on Centos 6.5 runs just fine now.

7
  • Have you checked that the mysql user actually owns and can write into /var/lib/mysql? sc. Dec 10, 2013 at 14:51
  • Yep. The mysql user owns that folder. [root@sacroiliac lib]# ls -l | grep mysql drwxr-xr-x 4 mysql mysql 4096 Dec 10 17:34 mysql
    – Cameron
    Dec 10, 2013 at 14:55
  • Ok - i'd also check /var/log/messages and/or /var/log/mysqld.log (wherever your install has it located). If all all fails, you can: strace -aef /path/to/safe_mysqld Dec 10, 2013 at 14:59
  • Please see my next edit.
    – Cameron
    Dec 10, 2013 at 15:03
  • Reading your log (messages), it shows a memory problem: mysqld is consuming all the available memory, and the OOM process kills it before it manages the whole system to hang...
    – AndrewQ
    Dec 10, 2013 at 15:53

1 Answer 1

2

The problem/reason of this is that you do not have enough free memory to run MySQL on your VPS. It is in /var/log/messages:

Dec 10 17:34:04 sacroiliac kernel: [627203.702793] OOM killed process 1657 (mysqld) vm:385760kB, rss:238984kB, swap:0kB
Dec 10 17:34:21 sacroiliac kernel: [627220.208520] OOM killed process 1919 (mysqld) vm:385760kB, rss:238300kB, swap:0kB
Dec 10 17:50:03 sacroiliac kernel: [628160.758387] OOM killed process 2228 (mysqld) vm:385760kB, rss:238988kB, swap:0kB
Dec 10 17:51:33 sacroiliac kernel: [628250.613146] OOM killed process 2486 (mysqld) vm:385760kB, rss:237220kB, swap:0kB
Dec 10 17:51:49 sacroiliac kernel: [628266.621825] OOM killed process 2744 (mysqld) vm:385760kB, rss:238412kB, swap:0kB

OOM is an acronym for Out of Memory.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .