| View previous topic :: View next topic |
| Author |
Message |
richis
n00b
Joined: 10 Feb 2004
Posts: 20
|
 Posted: Wed Jan 05, 2005 7:16 pm Post subject: Strange performance problem |
 |
|
Hi,
I going to loose my mind. I cant understand what is slowing down my system.
All the day it is running quite ok, LA=1-4, but then at evening LA goes up to sky without any reason. Here is my TOP at this moment. I dont see any big problems why it happens.
top - 21:15:17 up 17 days, 8:14, 2 users, load average: 68.47, 53.37, 34.17
Tasks: 207 total, 61 running, 142 sleeping, 3 stopped, 1 zombie
Cpu(s): 53.3% us, 42.5% sy, 0.0% ni, 0.0% id, 0.0% wa, 4.3% hi, 0.0% si
Mem: 1036940k total, 1023240k used, 13700k free, 4732k buffers
Swap: 498004k total, 16140k used, 481864k free, 368684k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
27902 squid 15 0 184m 54m 1416 S 76.7 5.4 11:56.89 squid
27995 apache 16 0 28780 10m 4684 S 14.8 1.1 0:12.76 apache
25996 apache 16 0 28924 11m 4768 S 13.9 1.1 0:23.85 apache
8925 dchub 15 0 16512 13m 2504 S 7.1 1.3 318:17.91 verlihub
29763 apache 16 0 28656 10m 4696 D 7.1 1.1 0:04.25 apache
30153 apache 16 0 28548 10m 4680 S 4.5 1.1 0:04.73 apache
30130 apache 16 0 28028 10m 4676 S 3.5 1.0 0:02.12 apache
28008 apache 15 0 30492 11m 4916 S 2.9 1.1 0:09.29 apache
28461 apache 17 0 28776 11m 4720 R 1.9 1.1 0:15.31 apache
21546 apache 17 0 28800 11m 4732 R 1.6 1.1 0:44.87 apache
30891 apache 16 0 26404 8232 4124 S 1.6 0.8 0:00.56 apache
2999 apache 17 0 29148 11m 4736 R 1.3 1.1 1:38.39 apache
27570 apache 15 0 30216 11m 5012 S 1.3 1.1 0:12.54 apache
30965 apache 17 0 28624 10m 4632 R 1.3 1.1 0:01.05 apache
6114 dchub 15 0 8332 5056 2280 S 1.0 0.5 55:09.76 verlihub
11694 apache 17 0 28588 11m 4920 R 1.0 1.1 1:54.14 apache
17056 apache 15 0 28824 11m 4924 S 1.0 1.1 0:57.09 apache
25993 apache 16 0 28616 11m 4916 S 1.0 1.1 0:24.75 apache
27999 apache 16 0 30804 11m 4908 S 1.0 1.1 0:10.02 apache
do you have any ideas? |
|
| Back to top |
|
 |
SerfurJ
l33t
Joined: 10 Apr 2004
Posts: 824
Location: Texas
|
| Posted: Wed Jan 05, 2005 7:21 pm Post subject: |
|
|
paste your top results using sorted for mem usage and cpu usage.
is your DMA on your hard drive turned on? |
|
| Back to top |
|
|
richis
n00b
Joined: 10 Feb 2004
Posts: 20
|
| Posted: Wed Jan 05, 2005 7:26 pm Post subject: |
|
|
This is sorted by CPU:
| Code: | top - 21:24:24 up 17 days, 8:23, 2 users, load average: 10.54, 24.43, 29.60
Tasks: 172 total, 3 running, 162 sleeping, 3 stopped, 4 zombie
Cpu(s): 27.4% us, 13.1% sy, 0.0% ni, 11.0% id, 46.0% wa, 2.5% hi, 0.0% si
Mem: 1036940k total, 963276k used, 73664k free, 8380k buffers
Swap: 498004k total, 16140k used, 481864k free, 395336k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
27902 squid 15 0 198m 69m 1416 R 32.6 6.8 15:41.19 squid
27999 apache 16 0 29064 11m 4680 S 27.1 1.1 0:12.74 apache
25998 apache 16 0 29024 11m 4884 S 12.7 1.1 0:22.99 apache
8925 dchub 15 0 16712 12m 2504 S 7.2 1.3 318:50.92 verlihub
31950 root 16 0 2128 984 720 R 3.6 0.1 0:00.02 top
6114 dchub 15 0 8224 4932 2280 R 1.8 0.5 55:18.06 verlihub
1 root 16 0 1512 460 412 S 0.0 0.0 0:03.60 init
2 root 34 19 0 0 0 S 0.0 0.0 0:02.56 ksoftirqd/0
3 root 5 -10 0 0 0 S 0.0 0.0 0:01.04 events/0
4 root 5 -10 0 0 0 S 0.0 0.0 0:00.00 khelper
5 root 5 -10 0 0 0 S 0.0 0.0 7:07.23 kblockd/0
9 root 15 -10 0 0 0 S 0.0 0.0 0:00.00 aio/0
10 root 25 0 0 0 0 S 0.0 0.0 0:00.00 kseriod
11 root 15 0 0 0 0 S 0.0 0.0 0:00.29 kjournald
75 root 15 0 1920 948 688 S 0.0 0.1 0:00.29 devfsd
4292 root 20 0 0 0 0 S 0.0 0.0 0:00.00 kjournald
4293 root 15 0 0 0 0 S 0.0 0.0 0:07.97 kjournald
4294 root 15 0 0 0 0 S 0.0 0.0 0:59.06 kjournald
4295 root 15 0 0 0 0 S 0.0 0.0 0:03.56 kjournald |
This is by MEM:
| Code: | top - 21:25:33 up 17 days, 8:24, 2 users, load average: 16.97, 23.23, 28.79
Tasks: 188 total, 22 running, 160 sleeping, 3 stopped, 3 zombie
Cpu(s): 52.8% us, 42.1% sy, 0.0% ni, 0.0% id, 0.0% wa, 5.0% hi, 0.0% si
Mem: 1036940k total, 1011348k used, 25592k free, 8892k buffers
Swap: 498004k total, 16140k used, 481864k free, 414384k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
27902 squid 16 0 198m 69m 1416 S 45.5 6.8 16:12.37 squid
5399 mysql 15 0 123m 34m 2840 S 0.0 3.4 1:26.72 mysqld
28468 apache 19 0 33332 14m 4532 S 0.0 1.4 0:00.44 apache
8925 dchub 15 0 16988 13m 2504 S 8.0 1.3 318:55.08 verlihub
25531 apache 16 0 31096 13m 6444 S 0.0 1.3 0:15.34 apache
23335 apache 16 0 30600 12m 6628 S 0.0 1.3 1:04.07 apache
6468 apache 16 0 31152 12m 5704 S 0.0 1.2 1:31.65 apache
30153 apache 16 0 30560 12m 6252 S 0.0 1.2 0:07.80 apache
3587 apache 16 0 29820 12m 4924 S 0.0 1.2 1:36.07 apache
28451 apache 16 0 30692 12m 5920 S 0.0 1.2 0:24.32 apache
27994 apache 16 0 30620 12m 5808 S 0.0 1.2 0:22.07 apache
23282 apache 15 0 30964 11m 5264 S 0.0 1.2 0:33.92 apache
17056 apache 15 0 30836 11m 5340 S 0.0 1.2 1:01.97 apache
29761 apache 18 0 30416 11m 4672 S 0.0 1.1 0:02.09 apache
32523 apache 17 0 29056 11m 4876 S 0.0 1.1 2:15.52 apache
7467 apache 16 0 29036 11m 4892 S 0.0 1.1 1:15.71 apache
25998 apache 16 0 29024 11m 4884 S 0.0 1.1 0:23.36 apache
2999 apache 17 0 29148 11m 4736 R 2.0 1.1 1:41.22 apache
29749 apache 16 0 29184 11m 4684 S 0.0 1.1 0:08.70 apache
25996 apache 15 0 29068 11m 4768 S 0.0 1.1 0:29.49 apache
25993 apache 15 0 28900 11m 4916 S 0.0 1.1 0:27.51 apache
30126 apache 15 0 30816 11m 4924 S 0.0 1.1 0:10.23 apache
23291 apache 16 0 28856 11m 4916 S 0.0 1.1 0:28.47 apache
23336 apache 16 0 28860 11m 4912 S 0.0 1.1 0:36.45 apache
27991 apache 17 0 29028 11m 4720 R 1.7 1.1 0:18.47 apache
27999 apache 16 0 29064 11m 4680 S 0.0 1.1 0:13.34 apache
24435 apache 16 0 30544 11m 5136 S 0.0 1.1 0:27.81 apache |
Yes all HDD have DMA turned on:
/sbin/hdparm -q -A1 -a8 -c1 -m16 -W1 -u1 /dev/hda
/sbin/hdparm -q -A1 -a16 -c1 -m16 -W1 -u1 /dev/hdb
/sbin/hdparm -q -A1 -a4 -c1 -m16 -W1 -u1 /dev/hde
/sbin/hdparm -q -A1 -a16 -c1 -m16 -W1 -u1 /dev/hdd
I even use squid in as revers-proxy in other case apache dies imediatly =(
Some times I think imaybe there is some problem with network connections, but I dont know where I can see what is maximum allowed and how much is used. |
|
| Back to top |
|
|
SerfurJ
l33t
Joined: 10 Apr 2004
Posts: 824
Location: Texas
|
| Posted: Wed Jan 05, 2005 8:07 pm Post subject: |
|
|
something to do with squid is using a lot of system cpu. i'm guessing squid is misconfigured. i don't use squid so i can't help you there.
you can use iptraf to see network activity. |
|
| Back to top |
|
|
planetsheinker
Guru
Joined: 26 Feb 2004
Posts: 403
Location: Israel
|
| Posted: Wed Jan 05, 2005 8:10 pm Post subject: |
|
|
| richis wrote: | This is sorted by CPU:
| Code: | top - 21:24:24 up 17 days, 8:23, 2 users, load average: 10.54, 24.43, 29.60
Tasks: 172 total, 3 running, 162 sleeping, 3 stopped, 4 zombie
Cpu(s): 27.4% us, 13.1% sy, 0.0% ni, 11.0% id, 46.0% wa, 2.5% hi, 0.0% si
Mem: 1036940k total, 963276k used, 73664k free, 8380k buffers
Swap: 498004k total, 16140k used, 481864k free, 395336k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
27902 squid 15 0 198m 69m 1416 R 32.6 6.8 15:41.19 squid
27999 apache 16 0 29064 11m 4680 S 27.1 1.1 0:12.74 apache
25998 apache 16 0 29024 11m 4884 S 12.7 1.1 0:22.99 apache
8925 dchub 15 0 16712 12m 2504 S 7.2 1.3 318:50.92 verlihub
31950 root 16 0 2128 984 720 R 3.6 0.1 0:00.02 top
6114 dchub 15 0 8224 4932 2280 R 1.8 0.5 55:18.06 verlihub
1 root 16 0 1512 460 412 S 0.0 0.0 0:03.60 init
2 root 34 19 0 0 0 S 0.0 0.0 0:02.56 ksoftirqd/0
3 root 5 -10 0 0 0 S 0.0 0.0 0:01.04 events/0
4 root 5 -10 0 0 0 S 0.0 0.0 0:00.00 khelper
5 root 5 -10 0 0 0 S 0.0 0.0 7:07.23 kblockd/0
9 root 15 -10 0 0 0 S 0.0 0.0 0:00.00 aio/0
10 root 25 0 0 0 0 S 0.0 0.0 0:00.00 kseriod
11 root 15 0 0 0 0 S 0.0 0.0 0:00.29 kjournald
75 root 15 0 1920 948 688 S 0.0 0.1 0:00.29 devfsd
4292 root 20 0 0 0 0 S 0.0 0.0 0:00.00 kjournald
4293 root 15 0 0 0 0 S 0.0 0.0 0:07.97 kjournald
4294 root 15 0 0 0 0 S 0.0 0.0 0:59.06 kjournald
4295 root 15 0 0 0 0 S 0.0 0.0 0:03.56 kjournald |
This is by MEM:
| Code: | top - 21:25:33 up 17 days, 8:24, 2 users, load average: 16.97, 23.23, 28.79
Tasks: 188 total, 22 running, 160 sleeping, 3 stopped, 3 zombie
Cpu(s): 52.8% us, 42.1% sy, 0.0% ni, 0.0% id, 0.0% wa, 5.0% hi, 0.0% si
Mem: 1036940k total, 1011348k used, 25592k free, 8892k buffers
Swap: 498004k total, 16140k used, 481864k free, 414384k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
27902 squid 16 0 198m 69m 1416 S 45.5 6.8 16:12.37 squid
5399 mysql 15 0 123m 34m 2840 S 0.0 3.4 1:26.72 mysqld
28468 apache 19 0 33332 14m 4532 S 0.0 1.4 0:00.44 apache
8925 dchub 15 0 16988 13m 2504 S 8.0 1.3 318:55.08 verlihub
25531 apache 16 0 31096 13m 6444 S 0.0 1.3 0:15.34 apache
23335 apache 16 0 30600 12m 6628 S 0.0 1.3 1:04.07 apache
6468 apache 16 0 31152 12m 5704 S 0.0 1.2 1:31.65 apache
30153 apache 16 0 30560 12m 6252 S 0.0 1.2 0:07.80 apache
3587 apache 16 0 29820 12m 4924 S 0.0 1.2 1:36.07 apache
28451 apache 16 0 30692 12m 5920 S 0.0 1.2 0:24.32 apache
27994 apache 16 0 30620 12m 5808 S 0.0 1.2 0:22.07 apache
23282 apache 15 0 30964 11m 5264 S 0.0 1.2 0:33.92 apache
17056 apache 15 0 30836 11m 5340 S 0.0 1.2 1:01.97 apache
29761 apache 18 0 30416 11m 4672 S 0.0 1.1 0:02.09 apache
32523 apache 17 0 29056 11m 4876 S 0.0 1.1 2:15.52 apache
7467 apache 16 0 29036 11m 4892 S 0.0 1.1 1:15.71 apache
25998 apache 16 0 29024 11m 4884 S 0.0 1.1 0:23.36 apache
2999 apache 17 0 29148 11m 4736 R 2.0 1.1 1:41.22 apache
29749 apache 16 0 29184 11m 4684 S 0.0 1.1 0:08.70 apache
25996 apache 15 0 29068 11m 4768 S 0.0 1.1 0:29.49 apache
25993 apache 15 0 28900 11m 4916 S 0.0 1.1 0:27.51 apache
30126 apache 15 0 30816 11m 4924 S 0.0 1.1 0:10.23 apache
23291 apache 16 0 28856 11m 4916 S 0.0 1.1 0:28.47 apache
23336 apache 16 0 28860 11m 4912 S 0.0 1.1 0:36.45 apache
27991 apache 17 0 29028 11m 4720 R 1.7 1.1 0:18.47 apache
27999 apache 16 0 29064 11m 4680 S 0.0 1.1 0:13.34 apache
24435 apache 16 0 30544 11m 5136 S 0.0 1.1 0:27.81 apache |
Yes all HDD have DMA turned on:
/sbin/hdparm -q -A1 -a8 -c1 -m16 -W1 -u1 /dev/hda
/sbin/hdparm -q -A1 -a16 -c1 -m16 -W1 -u1 /dev/hdb
/sbin/hdparm -q -A1 -a4 -c1 -m16 -W1 -u1 /dev/hde
/sbin/hdparm -q -A1 -a16 -c1 -m16 -W1 -u1 /dev/hdd
I even use squid in as revers-proxy in other case apache dies imediatly =(
Some times I think imaybe there is some problem with network connections, but I dont know where I can see what is maximum allowed and how much is used. |
I have been reading the "man hdparm" manpages, and as far as I understand, you forgot the "-d1" in order to enable DMA:
/sbin/hdparm -d1 -q -A1 -a8 -c1 -m16 -W1 -u1 /dev/hda |
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
Gentoo Chat
