cjb123
asked on
Sudden High LINUX CPU Utilization...
Hello:
Starting the other day, we saw a CPU utilization on our production LINUX server go from a "normal" 10% average utilization to approx 45 - 60% utilization. It's been "pegged" at about
45 - 60% utilization since then.
When I do a "top" command, I get these results below. Can you help me decipher these data?
If "top" is not the best command to use, please advise which command I should run and I'll provide supplement data in a follow-up post. Thank you.
top - 16:43:43 up 75 days, 13:21, 36 users, load average: 2.31, 2.36, 2.33
Tasks: 671 total, 5 running, 666 sleeping, 0 stopped, 0 zombie
Cpu(s): 45.9% us, 9.0% sy, 0.0% ni, 45.0% id, 0.1% wa, 0.0% hi, 0.0% si
Mem: 33264168k total, 17173904k used, 16090264k free, 359788k buffers
Swap: 25165792k total, 234908k used, 24930884k free, 13276400k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
1381 oracle 25 0 2022m 107m 89m R 100 0.3 21522:51 oracle
4350 oracle 16 0 2001m 91m 89m R 55 0.3 9789:05 oracle
4348 oracle 16 0 22620 9168 5568 S 47 0.0 8024:24 DP800
14660 oracle 16 0 2002m 179m 176m S 4 0.6 0:01.62 oracle
16254 oracle 16 0 2002m 222m 218m S 2 0.7 0:04.14 oracle
8674 hallsftp 16 0 3656 1296 740 R 2 0.0 0:05.74 top
11299 oracle 23 0 18368 6928 4628 S 1 0.0 0:00.03 DP700
11303 oracle 18 0 1999m 23m 21m S 1 0.1 0:00.03 oracle
14658 rfhalls 15 0 25104 19m 5276 S 1 0.1 0:01.04 f60run
16252 rfhalls 16 0 27852 22m 5960 S 1 0.1 0:02.71 f60run
9349 oracle 17 0 67932 24m 12m S 1 0.1 0:10.65 frmweb
21503 rfhalls 16 0 28128 22m 5856 S 1 0.1 0:01.83 f60run
Starting the other day, we saw a CPU utilization on our production LINUX server go from a "normal" 10% average utilization to approx 45 - 60% utilization. It's been "pegged" at about
45 - 60% utilization since then.
When I do a "top" command, I get these results below. Can you help me decipher these data?
If "top" is not the best command to use, please advise which command I should run and I'll provide supplement data in a follow-up post. Thank you.
top - 16:43:43 up 75 days, 13:21, 36 users, load average: 2.31, 2.36, 2.33
Tasks: 671 total, 5 running, 666 sleeping, 0 stopped, 0 zombie
Cpu(s): 45.9% us, 9.0% sy, 0.0% ni, 45.0% id, 0.1% wa, 0.0% hi, 0.0% si
Mem: 33264168k total, 17173904k used, 16090264k free, 359788k buffers
Swap: 25165792k total, 234908k used, 24930884k free, 13276400k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
1381 oracle 25 0 2022m 107m 89m R 100 0.3 21522:51 oracle
4350 oracle 16 0 2001m 91m 89m R 55 0.3 9789:05 oracle
4348 oracle 16 0 22620 9168 5568 S 47 0.0 8024:24 DP800
14660 oracle 16 0 2002m 179m 176m S 4 0.6 0:01.62 oracle
16254 oracle 16 0 2002m 222m 218m S 2 0.7 0:04.14 oracle
8674 hallsftp 16 0 3656 1296 740 R 2 0.0 0:05.74 top
11299 oracle 23 0 18368 6928 4628 S 1 0.0 0:00.03 DP700
11303 oracle 18 0 1999m 23m 21m S 1 0.1 0:00.03 oracle
14658 rfhalls 15 0 25104 19m 5276 S 1 0.1 0:01.04 f60run
16252 rfhalls 16 0 27852 22m 5960 S 1 0.1 0:02.71 f60run
9349 oracle 17 0 67932 24m 12m S 1 0.1 0:10.65 frmweb
21503 rfhalls 16 0 28128 22m 5856 S 1 0.1 0:01.83 f60run
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
Thanks guys...
45.9% us,
which meant , linux is spending 45.9% cpu time for User level activities which i guess is oracle
is spending 9.0% sy, so system level activities only 9% which is normal
so that means, its the oracle process eating too much processor speed
also, your system is using Swap memory, which i would not say a good thing about server performance, you should avoid to use Swap memory