LNMPA mysql服务运行几个小时后总是自动挂掉，系统重做好多次了，总是搞不定，求救！

notair

Ubuntu 14.04.2 LTS (GNU/Linux 3.13.0-32-generic x86_64)512M内存，
只装了LNMPAmysql启用innoDB，更改了MySQL的数据库存放目录，innodb_data_home_dir 和innodb_log_group_home_dir地址
同步修改清理并禁用mysql-bin.00001日志文件，
my.cnf文件[mysqld]下增加下面配置innodb_file_per_table=1
一开始一起运行正常，过几个小时后mysql就会自动挂掉，
重启之后正常运行，过几个小时后mysql又自动挂掉
mysql出错日志如下，请高手帮忙指点一下

151215 23:25:06 mysqld_safe mysqld from pid file /mnt/data/AY1404082155398856bcZ.pid ended
151215 23:25:08 mysqld_safe Starting mysqld daemon with databases from /mnt/data
151215 23:25:08 InnoDB: The InnoDB memory heap is disabled
151215 23:25:08 InnoDB: Mutexes and rw_locks use GCC atomic builtins
151215 23:25:08 InnoDB: Compressed tables use zlib 1.2.8
151215 23:25:08 InnoDB: Initializing buffer pool, size = 16.0M
151215 23:25:08 InnoDB: Completed initialization of buffer pool
151215 23:25:08 InnoDB: highest supported file format is Barracuda.
151215 23:25:08  InnoDB: Waiting for the background threads to start
151215 23:25:09 InnoDB: 5.5.42 started; log sequence number 696396357
151215 23:25:09 [Note] Server hostname (bind-address): '0.0.0.0'; port: 3306
151215 23:25:09 [Note]   - '0.0.0.0' resolves to '0.0.0.0';
151215 23:25:09 [Note] Server socket created on IP: '0.0.0.0'.
151215 23:25:09 [Note] Event Scheduler: Loaded 0 events
151215 23:25:09 [Note] /usr/local/mysql/bin/mysqld: ready for connections.
Version: '5.5.42'  socket: '/tmp/mysql.sock'  port: 3306  Source distribution
Killed
151216 02:31:45 mysqld_safe Number of processes running now: 0
151216 02:31:45 mysqld_safe mysqld restarted
151216  2:31:45 InnoDB: The InnoDB memory heap is disabled
151216  2:31:45 InnoDB: Mutexes and rw_locks use GCC atomic builtins
151216  2:31:45 InnoDB: Compressed tables use zlib 1.2.8
151216  2:31:45 InnoDB: Initializing buffer pool, size = 16.0M
151216  2:31:45 InnoDB: Completed initialization of buffer pool
151216  2:31:45 InnoDB: highest supported file format is Barracuda.
InnoDB: The log sequence number in ibdata files does not match
InnoDB: the log sequence number in the ib_logfiles!
151216  2:31:45  InnoDB: Database was not shut down normally!
InnoDB: Starting crash recovery.
InnoDB: Reading tablespace information from the .ibd files...
InnoDB: Restoring possible half-written data pages from the doublewrite
InnoDB: buffer...
151216  2:31:46  InnoDB: Waiting for the background threads to start
151216  2:31:47 InnoDB: 5.5.42 started; log sequence number 697241756
151216  2:31:47 [Note] Server hostname (bind-address): '0.0.0.0'; port: 3306
151216  2:31:47 [Note]   - '0.0.0.0' resolves to '0.0.0.0';
151216  2:31:47 [Note] Server socket created on IP: '0.0.0.0'.
151216  2:31:48 [Note] Event Scheduler: Loaded 0 events
151216  2:31:48 [Note] /usr/local/mysql/bin/mysqld: ready for connections.
Version: '5.5.42'  socket: '/tmp/mysql.sock'  port: 3306  Source distribution
Killed
151216 02:33:44 mysqld_safe Number of processes running now: 0
151216 02:33:44 mysqld_safe mysqld restarted
151216  2:33:44 InnoDB: The InnoDB memory heap is disabled
151216  2:33:44 InnoDB: Mutexes and rw_locks use GCC atomic builtins
151216  2:33:44 InnoDB: Compressed tables use zlib 1.2.8
151216  2:33:44 InnoDB: Initializing buffer pool, size = 16.0M
151216  2:33:44 InnoDB: Completed initialization of buffer pool
151216  2:33:44 InnoDB: highest supported file format is Barracuda.
InnoDB: The log sequence number in ibdata files does not match
InnoDB: the log sequence number in the ib_logfiles!
151216  2:33:44  InnoDB: Database was not shut down normally!
InnoDB: Starting crash recovery.
InnoDB: Reading tablespace information from the .ibd files...
InnoDB: Restoring possible half-written data pages from the doublewrite
InnoDB: buffer...
151216  2:33:45  InnoDB: Waiting for the background threads to start
151216  2:33:46 InnoDB: 5.5.42 started; log sequence number 697241766
151216  2:33:46 [Note] Server hostname (bind-address): '0.0.0.0'; port: 3306
151216  2:33:46 [Note]   - '0.0.0.0' resolves to '0.0.0.0';
151216  2:33:48 [Note] Server socket created on IP: '0.0.0.0'.
151216  2:35:20 [Note] Event Scheduler: Loaded 0 events
151216  2:35:20 [Note] /usr/local/mysql/bin/mysqld: ready for connections.
Version: '5.5.42'  socket: '/tmp/mysql.sock'  port: 3306  Source distribution
Killed
151216 02:35:40 mysqld_safe Number of processes running now: 0
151216 02:35:40 mysqld_safe mysqld restarted
151216  2:35:40 InnoDB: The InnoDB memory heap is disabled
151216  2:35:40 InnoDB: Mutexes and rw_locks use GCC atomic builtins
151216  2:35:40 InnoDB: Compressed tables use zlib 1.2.8
151216  2:35:40 InnoDB: Initializing buffer pool, size = 16.0M
151216  2:35:40 InnoDB: Completed initialization of buffer pool
151216  2:35:40 InnoDB: highest supported file format is Barracuda.
InnoDB: The log sequence number in ibdata files does not match
InnoDB: the log sequence number in the ib_logfiles!
151216  2:35:40  InnoDB: Database was not shut down normally!
InnoDB: Starting crash recovery.
InnoDB: Reading tablespace information from the .ibd files...
InnoDB: Restoring possible half-written data pages from the doublewrite
InnoDB: buffer...
151216  2:35:43  InnoDB: Waiting for the background threads to start
151216  2:35:44 InnoDB: 5.5.42 started; log sequence number 697241776
151216  2:35:44 [Note] Server hostname (bind-address): '0.0.0.0'; port: 3306
151216  2:35:44 [Note]   - '0.0.0.0' resolves to '0.0.0.0';
151216  2:35:44 [Note] Server socket created on IP: '0.0.0.0'.
151216  2:35:44 [Note] Event Scheduler: Loaded 0 events
151216  2:35:44 [Note] /usr/local/mysql/bin/mysqld: ready for connections.
Version: '5.5.42'  socket: '/tmp/mysql.sock'  port: 3306  Source distribution
Killed
151216 02:59:39 mysqld_safe Number of processes running now: 0
151216 02:59:39 mysqld_safe mysqld restarted
151216  2:59:39 InnoDB: The InnoDB memory heap is disabled
151216  2:59:39 InnoDB: Mutexes and rw_locks use GCC atomic builtins
151216  2:59:39 InnoDB: Compressed tables use zlib 1.2.8
151216  2:59:39 InnoDB: Initializing buffer pool, size = 16.0M
InnoDB: mmap(17170432 bytes) failed; errno 12
151216  2:59:39 InnoDB: Completed initialization of buffer pool
151216  2:59:39 InnoDB: Fatal error: cannot allocate memory for the buffer pool
151216  2:59:39 [ERROR] Plugin 'InnoDB' init function returned error.
151216  2:59:39 [ERROR] Plugin 'InnoDB' registration as a STORAGE ENGINE failed.
151216  2:59:39 [ERROR] Unknown/unsupported storage engine: InnoDB
151216  2:59:39 [ERROR] Aborting

151216  2:59:39 [Note] /usr/local/mysql/bin/mysqld: Shutdown complete

151216 02:59:39 mysqld_safe mysqld from pid file /mnt/data/AY1404082155398856bcZ.pid ended

licess

前面日志里只有Killed，应该是内存不足被系统kill了

InnoDB: mmap(17170432 bytes) failed; errno 12
151216  2:59:39 InnoDB: Completed initialization of buffer pool
151216  2:59:39 InnoDB: Fatal error: cannot allocate memory for the buffer pool

后面日志也有无法分配内存的提示

notair

为什么会这样？ 难道512M内存不能支持LNMPA吗

licess

512MB运行lnmpa确实内存挺紧张
最好是1G以上
本身apache就挺能吃内存的

notair

果然是，换成lnmp一切正常，伤心呀！用不了lnmpa

smartweb

我也有512的，很稳定