Enterprise Manager

Hi, My grid control is sending the following warning for all 4 nodes from the cluster -

Memory Utilization is 99.7%, crossed warning (99) or critical ( ) threshold.

When looked at the "Target->Host->Performance", It shows almost 100% memory utilization on all the nodes. I guess this is for the host memory status and not just any oracle instance.

I have 16Gigs memory on each node. Here is the current status -

$ free
total used free shared buffers cached
Mem: 16412416 16220100 192316 0 361732 14227204
-/+ buffers/cache: 1631164 14781252
Swap: 16777208 285676 16491532

Looks like it still has 14gigs free.

$ cat /etc/sysctl.conf
net.ipv4.ip_forward = 0
net.ipv4.conf.default.rp_filter = 1
net.ipv4.conf.default.accept_source_route = 0
kernel.core_uses_pid = 1
fs.file-max = 327679
fs.aio-max-nr = 3145728
kernel.msgmni = 2878
kernel.msgmax = 8192
kernel.msgmnb = 65536
kernel.sem = 250 32000 100 142
kernel.shmall = 3279547
kernel.shmmax = 8589934592
kernel.shmmni = 4096
kernel.sysrq = 1
net.ipv4.ip_local_port_range = 1024 65000
net.core.rmem_max = 2097152
net.core.rmem_default = 1048576
net.core.wmem_default = 262144
net.core.wmem_max = 262144



I have 2 databases running on all 4 nodes with folowwing parameter -
sga_max_size big integer 1776M
sga_target big integer 1776M

Any idea why 100% memory utilization? Any recommendation/best practices for memory tuning? Following is more info -

Dell2950/Linux AS 4 - 4 Node Cluster with Oracle 10.2.0.3 with ASM.


Thanks