0

i usring Oracle Enterprise Linux 6.5 and Oracle Database 11g when i execute "top" show this to me 

top - 14:08:02 up 5 days, 18:34,  2 users,  load average: 0.00, 0.01, 0.05
Tasks: 340 total,   1 running, 331 sleeping,   8 stopped,   0 zombie
Cpu(s):  0.0%us,  0.0%sy,  0.0%ni, 99.9%id,  0.0%wa,  0.0%hi,  0.0%si,  0.0%st
Mem:  132038392k total, 100871400k used, 31166992k free,    16912k buffers
Swap: 32767996k total,        0k used, 32767996k free, 99462140k cached

i dedicate 70GB memory to my Database so when i start my database show following error 

ORA-00845: MEMORY_TARGET not supported on this system

then i execute "vmstat -s" show following output

132038392  total memory
100928456  used memory
   278536  active memory
 99495304  inactive memory

i realize 96GB of my used memory is INACTIVE memory then i used following command to free up my INACTIVE memory 

sync && echo 3 > /proc/sys/vm/drop_caches

nothing happened and still i cant start my database unless i reboot my OS How can i resolve this problem?

Improve this question
2
  • I think this is talking about temp shared memory filesystem (shmfs/tmpfs). Can you check your df and see what you have for tmpfs?
    – Tux_DEV_NULL
    Mar 25, 2019 at 12:23
  • Yes. Thanks for answering to me. i checked that and this is exact 95GB used of tmpfs. so how can i release that. still i can't start my database
    – Soheil Mamdouhi
    Mar 26, 2019 at 3:46

2 Answers 2

Reset to default
0

I think to solve this problem with the below string of commands:

first of all Writing to this will cause the kernel to drop clean caches Once dropped, their memory becomes free.

To free pagecache:

echo 1 > /proc/sys/vm/drop_caches

echo 2 > /proc/sys/vm/drop_caches

echo 3 > /proc/sys/vm/drop_caches

if this problem wasnot resolve then you should to do it:

Now change the value by modifying /etc/sysctl.conf 
kernel.shmmax 
kernel.shmall 
then force the system to activate this setting file without a reboot.
$ sysctl -p
Improve this answer
0

Can you try this?

[1] Below command to mount is manually

mount -t tmpfs shmfs -o size=70g /dev/shm

[2] Then try

SQL> startup nomount If all works add below line to fstab to make it persistent.

"tmpfs /dev/shm tmpfs size=72g 0 0"

I hope server having enough resources for db and other process.

Improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .