shmmax

shmmax内核参数定义是系统允许的单个共享内存段的最大值
现在模拟设置shmmax很小：
[root@localhost kernel]# cat /etc/redhat-release
Red Hat Enterprise Linux Server release 5.5 (Tikanga)

SQL> select * from v$version where rownum=1;

BANNER
--------------------------------------------------------------------------------
Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - Production


[root@localhost ~]# cd /proc/sys/kernel | echo 1000 > shmmax                  => 注意：该种修改会在os重启后失效

[oracle@localhost ~]$ sqlplus /  as  sysdba


SQL*Plus: Release 11.2.0.1.0 Production on 星期日 5月 10 23:27:36 2015


Copyright (c) 1982, 2009, Oracle.  All rights reserved.


ERROR:
ORA-12547: TNS: 丢失连接


请输入用户名:  

告警日志：
......
......
Sun May 10 23:27:42 2015
Process J000 died, see its trace file
kkjcre1p: unable to spawn jobq slave process 
Errors in file /u01/app/oracle/diag/rdbms/orcl3939/orcl3939/trace/orcl3939_cjq0_10511.trc:
Process J000 died, see its trace file
kkjcre1p: unable to spawn jobq slave process 
Errors in file /u01/app/oracle/diag/rdbms/orcl3939/orcl3939/trace/orcl3939_cjq0_10511.trc:

/u01/app/oracle/diag/rdbms/orcl3939/orcl3939/trace/orcl3939_cjq0_10511.trc:

Trace file /u01/app/oracle/diag/rdbms/orcl3939/orcl3939/trace/orcl3939_cjq0_10511.trc
Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options
ORACLE_HOME = /u01/app/oracle/product/11.2.0/dbhome_1
System name: Linux
Node name: localhost.localdomain
Release: 2.6.18-194.el5
Version: #1 SMP Mon Mar 29 20:06:41 EDT 2010
Machine: i686
Instance name: orcl3939
Redo thread mounted by this instance: 1
Oracle process number: 25
Unix process pid: 10511, image: [email protected] (CJQ0)




*** 2015-05-10 23:08:21.485
*** SESSION ID:(133.9) 2015-05-10 23:08:21.485
*** CLIENT ID:() 2015-05-10 23:08:21.485
*** SERVICE NAME:(SYS$BACKGROUND) 2015-05-10 23:08:21.485
*** MODULE NAME:() 2015-05-10 23:08:21.485
*** ACTION NAME:() 2015-05-10 23:08:21.485
 
Setting Resource Manager plan SCHEDULER[0x3009]:DEFAULT_MAINTENANCE_PLAN via scheduler window
Setting Resource Manager plan DEFAULT_MAINTENANCE_PLAN via parameter


*** 2015-05-10 23:27:42.805
Process J000 is dead (pid=11060 req_ver=3 cur_ver=3 state=KSOSP_SPAWNED).


*** 2015-05-10 23:27:43.876
Process J000 is dead (pid=11063 req_ver=16 cur_ver=16 state=KSOSP_SPAWNED).


*** 2015-05-10 23:28:43.621
Process J000 is dead (pid=11106 req_ver=5 cur_ver=5 state=KSOSP_SPAWNED).


*** 2015-05-10 23:28:44.743
Process J000 is dead (pid=11109 req_ver=17 cur_ver=17 state=KSOSP_SPAWNED).


*** 2015-05-10 23:34:43.874
Process J000 is dead (pid=11244 req_ver=6 cur_ver=6 state=KSOSP_SPAWNED).


*** 2015-05-10 23:34:44.888
Process J000 is dead (pid=11247 req_ver=18 cur_ver=18 state=KSOSP_SPAWNED).


*** 2015-05-10 23:35:42.550
Process J000 is dead (pid=11280 req_ver=7 cur_ver=7 state=KSOSP_SPAWNED).


*** 2015-05-10 23:35:43.565
Process J000 is dead (pid=11285 req_ver=19 cur_ver=19 state=KSOSP_SPAWNED).

[oracle@localhost ~]$ ipcs -a


------ Shared Memory Segments --------
key        shmid      owner      perms      bytes      nattch     status      
0x00000000 3276801    oracle    600        393216     2          dest         
0x00000000 3309571    oracle    600        393216     2          dest         
0x00000000 3342340    oracle    600        393216     2          dest         
0x00000000 3375109    oracle    600        393216     2          dest         
0x00000000 3407878    oracle    600        393216     2          dest         
0x00000000 3440647    oracle    600        393216     2          dest         
0x00000000 3473416    oracle    600        393216     2          dest         
0x00000000 3506185    oracle    600        393216     2          dest         
0x00000000 3538954    oracle    600        393216     2          dest         
0x00000000 3571723    oracle    600        393216     2          dest         
0x05a35eec 4194316    oracle    660        4096       0                       
0x038880a0 3735567    oracle    755        1079228    0                       


------ Semaphore Arrays --------
key        semid      owner      perms      nsems     
0xed251740 622592     oracle    660        154       


------ Message Queues --------
key        msqid      owner      perms      used-bytes   messages    


SQL> select * from v$sgainfo;


NAME                                  BYTES RES
-------------------------------- ---------- ---
Fixed SGA Size                      1336960 No
Redo Buffers                        6094848 No
Buffer Cache Size                  79691776 Yes
Shared Pool Size                  176160768 Yes
Large Pool Size                     4194304 Yes
Java Pool Size                      4194304 Yes
Streams Pool Size                   4194304 Yes
Shared IO Pool Size                       0 Yes
Granule Size                        4194304 No
Maximum SGA Size                  422670336 No
Startup overhead in Shared Pool    58720256 No

NAME                                  BYTES RES
-------------------------------- ---------- ---
Free SGA Memory Available         146800640

如果shmmax的大小小于SGA设置，那么sga任然可以创建成功，但是会被分配多个共享内存段，但是这个shmmax的大小要有底线，
否则 OS资源被耗尽，Process J000 is dead
通常建议shmmax设置大些，把sga限制在一个共享内存段中。

官方文档：
Memory Resources

The resources provided by the UNIX default kernels are often inadequate for a medium or large instance of Oracle Database. The maximum size of a shared memory segment (SHMMAX) and maximum number of semaphores available (SEMMNS) may be too low for Oracle Database recommendations.
由UNIX默认内核提供的资源往往是不足的中型或大型的数据库实例。最大的共享内存段的大小（SHMMAX）和信号量可用的最大数目（SEMMNS）可能太低对Oracle数据库的建议。

On Windows, fewer resources are needed for interprocess communication (IPC), because the Oracle Database relational database management system is thread-based and not process-based. These resources, including shared memory and semaphores, are not adjustable by the user.

在Windows中，更少的资源用于进程间通信（IPC）所需要的，因为Oracle数据库的关系型数据库管理系统是基于线程的而不是进程。这些资源，包括共享内存和信号量，都是用户不可修改的。