ORA-00845: MEMORY_TARGET not supported on this system

Hi,
As we know MEMORY_TARGET is a new dynamic parameter added in Oracle 11g for automatic SGA and PGA management.Working with LINUX OS you might face error
ORA-00845.

Error:
ORA-00845: MEMORY_TARGET not supported on this system

The ORA-00845:can arises for the following two reasons on linux system.

1)If the shared memory which is mapped to /dev/shm directory is less than the size of MEMORY_TARGET or MEMORY_MAX_TARGET.
or,
2)If the shared memory is not mapped to /dev/shm directory.

For me the first case had happen i.e MEMORY_TARGET=1G,So I decided to increase to 2G,This is because of the below factor:
•The AMM (Automatic Memory Management) now in 11g manages both SGA and PGA together by MMAN process so we need extra memory or size of this parameter must be high,If you have 2 Databases running in your machine you can keep MEMORY_TARGET=2G.For 3 or 4 Database MEMORY_TARGET=2.5g should
be fine.

Resolving ORA-00845:
As a root user
mount -t tmpfs shmfs -o size=2g /dev/shm

In order to make the settings persistence so that it will affect after
restarting machine add an entry in /etc/fstab similar to the following:
vi /etc/fstab

shmfs /dev/shm tmpfs size=12g 0

[root@node1 /]# mount -t tmpfs shmfs -o size=2g /dev/shm

:wq
Save and quit

Note 1:In the above example I've created the shmfs(shared memory file system) with a size of 2G as that is the size of the buffer cache I am planning to use.
The other elements of the SGA are placed in regular memory,not this shared memory file system, so they should not be included when deciding on the size of the shmfs.
It is recommended to size this slightly bigger than the actual size needed,
Here I've used a 2G shmfs for a 2G buffer cache.

Note 2: The size of memory_target parameter we can grow upto memory_max_target and not beyond memory_max_target.

SQL> show parameter %MEMORY%

NAME TYPE VALUE
------------------------------------ ----------- ------------------------------
hi_shared_memory_address integer 0
memory_max_target big integer 1G
memory_target big integer 1G


[root@node1 /]# reboot

Broadcast message from root (pts/3) (Thu May 19 19:28:02 2011):

The system is going down for reboot NOW

Check after reboot:
-------------------

[root@node1 ~]# df -k
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/sda6 5952252 4117872 1527140 73% /
/dev/sda7 5080796 4292008 526532 90% /u01
/dev/sda3 6605824 1807356 4457492 29% /u03
/dev/sda2 35550448 26956160 6759264 80% /u02
/dev/sda1 497829 16695 455432 4% /boot
tmpfs 2097152 0 2097152 0% /dev/shm
/dev/sdb1 39381744 35555696 1825556 96% /u04
/dev/sdc1 51605436 1233976 47750056 3% /u05
shmfs 2097152 0 2097152 0% /dev/shm

Check after starting database:
-------------------------------
[oracle@node1 ~]$ df -h
Filesystem Size Used Avail Use% Mounted on
/dev/sda6 5.7G 4.0G 1.5G 73% /
/dev/sda7 4.9G 4.1G 515M 90% /u01
/dev/sda3 6.3G 1.8G 4.3G 29% /u03
/dev/sda2 34G 26G 6.5G 80% /u02
/dev/sda1 487M 17M 445M 4% /boot
tmpfs 2.0G 1.4G 714M 66% /dev/shm
/dev/sdb1 38G 34G 1.8G 96% /u04
/dev/sdc1 50G 1.2G 46G 3% /u05
shmfs 2.0G 1.4G 714M 66% /dev/shm

Hope it helps.


Best regards,

Rafi.