Hi,

I have redhat 5.3 64bit. it has kernel 2.6.18-308.13.1.el5.

it has showing all memory used in chache.
[abc]$ free -m
total used free shared buffers cached
Mem: 24103 23965 137 0 16 2641
-/+ buffers/cache: 21308 2794
Swap: 8001 4176 3824

.

please help me to get the memory in use.

it IS "in use". If running applications need the space, they'll reclaim it. There is nothing to do.

you CAN do what you asked ... "sysctrl -w vm.drop_caches=3" BUT if you don't actually understand this already, don't do it. In a similar vein you can lower the "swappiness" value but these things work as they do by default for a reason, and most people have no valid reason to change them, including, I suspect, you.

Wellll ...
Looking at that, out of 24 Gig RAM, more than 88% is in use (i.e. not buffer/cache).
Maybe the OP has a reasonable request - what's using it.

What does "top" show for a start ?.

Top output is normal...

top - 13:13:12 up 5 days, 14:27, 2 users, load average: 2.45, 1.55, 1.01
Tasks: 461 total, 2 running, 459 sleeping, 0 stopped, 0 zombie
Cpu(s): 5.1%us, 0.4%sy, 0.0%ni, 92.0%id, 2.4%wa, 0.0%hi, 0.1%si, 0.0%st
Mem: 24681492k total, 24582896k used, 98596k free, 15576k buffers
Swap: 8193140k total, 4787228k used, 3405912k free, 2882564k cached

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
11724 oracle 18 0 3380m 105m 35m R 81.9 0.4 0:27.65 oracle
5489 oracle 16 0 3316m 17m 16m S 3.3 0.1 2:02.03 oracle
28181 oracle 18 0 3316m 10m 6972 S 2.3 0.0 27:07.90 oracle
18859 oracle 18 0 4348m 62m 53m S 0.7 0.3 32:48.52 oracle


Acid--- Can I use this command on running server without downtime?

Please reply.

dropping caches shouldn't incur downtime, but from personal experience I've known it lock a box up on at least one occasion.