apache用不了多久就把mysql搞挂了，why？

okwiner

1. 130528  0:23:53 [Note] Plugin 'FEDERATED' is disabled.
   
2. 130528  0:23:53 InnoDB: The InnoDB memory heap is disabled
   
3. 130528  0:23:53 InnoDB: Mutexes and rw_locks use GCC atomic builtins
   
4. 130528  0:23:53 InnoDB: Compressed tables use zlib 1.2.3.4
   
5. 130528  0:23:53 InnoDB: Initializing buffer pool, size = 128.0M
   
6. 130528  0:23:53 InnoDB: Completed initialization of buffer pool
   
7. 130528  0:23:53 InnoDB: highest supported file format is Barracuda.
   
8. InnoDB: The log sequence number in ibdata files does not match
   
9. InnoDB: the log sequence number in the ib_logfiles!
   
10. 130528  0:23:53  InnoDB: Database was not shut down normally!
    
11. InnoDB: Starting crash recovery.
    
12. InnoDB: Reading tablespace information from the .ibd files...
    
13. InnoDB: Restoring possible half-written data pages from the doublewrite
    
14. InnoDB: buffer...
    
15. InnoDB: Last MySQL binlog file position 0 588594, file name /var/log/mysql/mysql-bin.000006
    
16. 130528  0:23:54  InnoDB: Waiting for the background threads to start
    
17. 130528  0:23:55 InnoDB: 1.1.8 started; log sequence number 17212102
    
18. 130528  0:23:55 [Note] Server hostname (bind-address): '127.0.0.1'; port: 3306
    
19. 130528  0:23:55 [Note]   - '127.0.0.1' resolves to '127.0.0.1';
    
20. 130528  0:23:55 [Note] Server socket created on IP: '127.0.0.1'.
    
21. 130528  0:23:55 [Note] Event Scheduler: Loaded 0 events
    
22. 130528  0:23:55 [Note] /usr/sbin/mysqld: ready for connections.
    
23. Version: '5.5.28-0ubuntu0.12.04.3'  socket: '/var/run/mysqld/mysqld.sock'  port: 3306  (Ubuntu)
    
24. 130528  0:24:04 [ERROR] /usr/sbin/mysqld: Table './everitsm/wp_options' is marked as crashed and should be repaired
    
25. 130528  0:24:04 [Warning] Checking table:   './everitsm/wp_options'
    
26. 130528  0:27:57 [Note] Plugin 'FEDERATED' is disabled.
    
27. 130528  0:27:57 InnoDB: The InnoDB memory heap is disabled
    
28. 130528  0:27:57 InnoDB: Mutexes and rw_locks use GCC atomic builtins
    
29. 130528  0:27:57 InnoDB: Compressed tables use zlib 1.2.3.4
    
30. 130528  0:27:57 InnoDB: Initializing buffer pool, size = 128.0M
    
31. InnoDB: mmap(137363456 bytes) failed; errno 12
    
32. 130528  0:27:57 InnoDB: Completed initialization of buffer pool
    
33. 130528  0:27:57 InnoDB: Fatal error: cannot allocate memory for the buffer pool
    
34. 130528  0:27:57 [ERROR] Plugin 'InnoDB' init function returned error.
    
35. 130528  0:27:57 [ERROR] Plugin 'InnoDB' registration as a STORAGE ENGINE failed.
    
36. 130528  0:27:57 [ERROR] Unknown/unsupported storage engine: InnoDB
    
37. 130528  0:27:57 [ERROR] Aborting
    
38. 
    
39. 130528  0:27:57 [Note] /usr/sbin/mysqld: Shutdown complete
    
40. 
    
41. 130528  0:27:57 [Note] Plugin 'FEDERATED' is disabled.
    
42. 130528  0:27:57 InnoDB: The InnoDB memory heap is disabled
    
43. 130528  0:27:57 InnoDB: Mutexes and rw_locks use GCC atomic builtins
    
44. 130528  0:27:57 InnoDB: Compressed tables use zlib 1.2.3.4
    
45. 130528  0:27:57 InnoDB: Initializing buffer pool, size = 128.0M
    
46. InnoDB: mmap(137363456 bytes) failed; errno 12
    
47. 130528  0:27:57 InnoDB: Completed initialization of buffer pool
    
48. 130528  0:27:57 InnoDB: Fatal error: cannot allocate memory for the buffer pool
    
49. 130528  0:27:57 [ERROR] Plugin 'InnoDB' init function returned error.
    
50. 130528  0:27:57 [ERROR] Plugin 'InnoDB' registration as a STORAGE ENGINE failed.
    
51. 130528  0:27:57 [ERROR] Unknown/unsupported storage engine: InnoDB
    
52. 130528  0:27:57 [ERROR] Aborting
    
53. 
    
54. 130528  0:27:57 [Note] /usr/sbin/mysqld: Shutdown complete

复制代码

用这个命令
ab -n 100 -c 10 http://xxxxxx.com/
用不了几次，mysql就挂了，出现的错误都一样，网站是wordpress的。

求助，请问怎么解决？

action08

mysql是不是用的默认配置啊，编译安装还是yum安装的呀？？

action08

上面的错了



48行，
130528  0:27:57 InnoDB: Fatal error: cannot allocate memory for the buffer pool

内存不够，贴free -m

okwiner

mysql 是apt-get安装的，不是编译安装的。

是内存不足，但是怎么限制apache对内存的消耗和mysql自己对内存的消耗？

Shaquile

对于ab：并发数越多，内存使用量越大
对于mysql：支持的线程数越多，内存使用量越大

所以，小内存没法进行性能测试，因为内存首先是瓶颈，无法测到mysql的瓶颈

RogerZhuo

#
nnoDB: mmap(137363456 bytes) failed; errno 12
130528  0:27:57 InnoDB: Completed initialization of buffer pool
130528  0:27:57 InnoDB: Fatal error: cannot allocate memory for the buffer pool
#
在你启动MySQL的时候，innodb就会根据innodb_buffer_size指定的大小为innodb预分配内存， 但你此时OS的内存已不足指定的128M，所以innodb bp无法初始化，MySQL也就启动不起来了。
[root@DB ~]# perror 12
OS error code  12:  Cannot allocate memory
解决：1）把你apache的内存配置调小，2）用free -m看一下，还有多少内存，然后设置innodb_buffer_pool_size,不过128M用于测试都太小了。
如果用的VM，建议去扩一下内存。

ruochen

哈哈

action08

楼上版主宝宝已经五岁了呀

maochanglu

用 nginx 替代apache