I was having this problem since last few days. My mysql server was getting killed on its own on aws/ec2 micro instance. So far I was restarting the whole instance to fix this problem.
When I looked at the log (/var/log/mysqld.log) on my centos linux instance, here is what I saw:
130304 23:55:47 InnoDB: Initializing buffer pool, size = 128.0M InnoDB: mmap(137363456 bytes) failed; errno 12 130304 23:55:47 InnoDB: Completed initialization of buffer pool 130304 23:55:47 InnoDB: Fatal error: cannot allocate memory for the buffer pool 130304 23:55:47 [ERROR] Plugin 'InnoDB' init function returned error. 130304 23:55:47 [ERROR] Plugin 'InnoDB' registration as a STORAGE ENGINE failed.
After reducing innodb_buffer_pool_size (default 256M) to 10M in /etc/my.cnf the problem seems to have gone away. Here is the value I changed in /etc/my.cnf (On Ubuntu the location is /etc/mysql/my.cnf)
[mysqld] innodb_buffer_pool_size = 10M
This worked on both CentOS and Ubuntu Linux. I took a very low value for now as I don’t have too much data and traffic. I’ll watch and see if this needs to be increased in next few days.