Mageia forum

by **DentAr** » Jun 28th, '11, 11:40

Hi all,

i noticed on my mageia box (64bit, single core cpu) that the load average is nearly constantly at 1.0/1.01/1.02 BUT - the cpu load monitored using top or htop does not really change from 0.5% to 2%.
This machine sits normally idle being used for mageia tests or, its main purpose, serving as a squid proxy.

Accessing the machine through requests to squid gives the mentioned 2% cpu load, just monitoring it using top/htop in a terminal window (ssh) gives a load below 2% (ususally around 0.7%)

As far as i understand, the values of the load average should then be somewhere around 0.something, but never around 1 (for the one minute/5 minute/15 minute load average values) unless the machine is really under heavy load.

Even the values at
cat /proc/loadavg
displays
1.19 1.06 1.04 2/196 20759

Any ideas why?
Or is my assumption with the load average wrong?

by **martinw** » Jun 28th, '11, 20:58

Your reported load average definitely looks wrong. I get:

Code: Select all: % cat /proc/loadavg 0.02 0.04 0.05 2/285 7957

The only thing I can think of is that there is a process that is being counted as runnable, but is blocked in some way. Try

Code: Select all: ps agux

and check the STAT column for anything unusual.

by **DentAr** » Jun 29th, '11, 09:40

Hi,

i checked the ps output, for me, there is nothing unusual.
Maybe you could also take a look at the following ps output?

Could this be a bug in mageia? I did have the same configuration in mandriva before, the load average was 0.something as expected and in no way similar to the mageia output...

Thnx.

Code: Select all: USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND root 1 0.0 0.0 3984 640 ? Ss Jun25 0:02 init [5] root 2 0.0 0.0 0 0 ? S Jun25 0:00 [kthreadd] root 3 0.0 0.0 0 0 ? S Jun25 0:00 [ksoftirqd/0] root 5 0.0 0.0 0 0 ? S Jun25 0:00 [kworker/u:0] root 6 0.0 0.0 0 0 ? S Jun25 0:00 [migration/0] root 7 0.0 0.0 0 0 ? S< Jun25 0:00 [cpuset] root 8 0.0 0.0 0 0 ? S< Jun25 0:00 [khelper] root 9 0.0 0.0 0 0 ? S< Jun25 0:00 [netns] root 10 0.0 0.0 0 0 ? S Jun25 0:00 [sync_supers] root 11 0.0 0.0 0 0 ? S Jun25 0:00 [bdi-default] root 12 0.0 0.0 0 0 ? S< Jun25 0:00 [kintegrityd] root 13 0.0 0.0 0 0 ? S< Jun25 0:00 [kblockd] root 14 0.0 0.0 0 0 ? S< Jun25 0:00 [kacpid] root 15 0.0 0.0 0 0 ? S< Jun25 0:00 [kacpi_notify] root 16 0.0 0.0 0 0 ? S< Jun25 0:00 [kacpi_hotplug] root 17 0.0 0.0 0 0 ? S< Jun25 0:00 [md] root 20 0.0 0.0 0 0 ? S Jun25 0:00 [khungtaskd] root 21 0.0 0.0 0 0 ? S Jun25 0:00 [kswapd0] root 22 0.0 0.0 0 0 ? SN Jun25 0:00 [ksmd] root 23 0.0 0.0 0 0 ? SN Jun25 0:03 [khugepaged] root 24 0.0 0.0 0 0 ? S Jun25 0:00 [fsnotify_mark] root 25 0.0 0.0 0 0 ? S< Jun25 0:00 [aio] root 26 0.0 0.0 0 0 ? S< Jun25 0:00 [crypto] root 29 0.0 0.0 0 0 ? S< Jun25 0:00 [kthrotld] root 31 0.0 0.0 0 0 ? S< Jun25 0:00 [kpsmoused] root 48 0.0 0.0 0 0 ? S< Jun25 0:00 [ata_sff] root 51 0.0 0.0 0 0 ? S Jun25 0:14 [scsi_eh_0] root 52 0.0 0.0 0 0 ? S Jun25 0:00 [scsi_eh_1] root 53 0.0 0.0 0 0 ? S Jun25 0:00 [kworker/u:1] root 66 0.0 0.0 0 0 ? S Jun25 0:00 [scsi_eh_2] root 67 0.0 0.0 0 0 ? S Jun25 0:00 [scsi_eh_3] root 68 0.0 0.0 0 0 ? S Jun25 0:00 [jbd2/sda5-8] root 69 0.0 0.0 0 0 ? S< Jun25 0:00 [ext4-dio-unwrit] root 100 0.0 0.0 0 0 ? S Jun25 0:00 [khubd] root 130 0.0 0.0 10856 1172 ? S<s Jun25 0:00 /sbin/udevd -d root 243 0.0 0.0 0 0 ? S< Jun25 0:00 [edac-poller] root 841 0.0 0.0 0 0 ? S< Jun25 0:00 [kconservative] root 842 0.0 0.0 0 0 ? S< Jun25 0:00 [kondemand] root 933 0.0 0.0 0 0 ? S Jun25 0:00 [jbd2/sda1-8] root 934 0.0 0.0 0 0 ? S< Jun25 0:00 [ext4-dio-unwrit] root 935 0.0 0.0 0 0 ? S Jun25 0:03 [jbd2/sdb1-8] root 936 0.0 0.0 0 0 ? S< Jun25 0:00 [ext4-dio-unwrit] root 937 0.0 0.0 0 0 ? S Jun25 0:01 [jbd2/sda9-8] root 938 0.0 0.0 0 0 ? S< Jun25 0:00 [ext4-dio-unwrit] root 939 0.0 0.0 0 0 ? S Jun25 0:00 [jbd2/sda7-8] root 940 0.0 0.0 0 0 ? S< Jun25 0:00 [ext4-dio-unwrit] root 941 0.0 0.0 0 0 ? S Jun25 0:00 [jbd2/sda8-8] root 942 0.0 0.0 0 0 ? S< Jun25 0:00 [ext4-dio-unwrit] root 962 0.0 0.0 10852 1196 ? S< Jun25 0:00 /sbin/udevd -d root 963 0.0 0.0 10852 1168 ? S< Jun25 0:00 /sbin/udevd -d root 1357 0.0 0.0 4376 336 ? Ss Jun25 0:00 mdadm --monitor --scan -f --pid-file=/var/run/mdadm/mdadm.pid daemon 1371 0.0 0.0 12384 344 ? Ss Jun25 0:00 /usr/sbin/atd -l 0.8 root 1402 0.0 0.0 244516 1316 ? Sl Jun25 0:00 rsyslogd -c 4 root 1502 0.0 0.0 14780 1204 ? Ss Jun25 0:01 crond 499 1544 0.0 0.0 13512 1588 ? Ss Jun25 0:01 dbus-daemon --system 494 1597 0.0 0.2 43332 5492 ? Ssl Jun25 0:02 hald root 1599 0.0 0.1 583744 3676 ? Sl Jun25 0:00 /usr/sbin/console-kit-daemon --no-daemon root 1691 0.0 0.1 186348 3736 ? Sl Jun25 0:00 /usr/lib64/polkit-1/polkitd root 1704 0.0 0.0 22464 1396 ? S Jun25 0:00 hald-runner root 1784 0.0 0.0 24584 1176 ? S Jun25 0:00 hald-addon-input: Listening on /dev/input/event0 /dev/input/event1 /dev/input/event2 root 1802 0.0 0.0 24584 1176 ? S Jun25 0:06 hald-addon-storage: polling /dev/sr0 (every 16 sec) root 1804 0.0 0.0 24592 1176 ? S Jun25 0:00 /usr/lib64/hal/hald-addon-cpufreq 494 1808 0.0 0.0 20048 1136 ? S Jun25 0:04 /usr/lib64/hal/hald-addon-acpi root 1865 0.0 0.0 28708 1252 ? S Jun25 0:00 /usr/bin/kdm -nodaemon root 1958 0.0 0.0 0 0 ? S Jun25 0:00 [kauditd] avahi 2258 0.0 0.0 25664 1660 ? S Jun25 0:00 avahi-daemon: running [amd900.local] avahi 2259 0.0 0.0 25532 436 ? S Jun25 0:00 avahi-daemon: chroot helper root 2269 0.0 0.2 9792 4796 ? S Jun25 0:00 /usr/sbin/hddtemp -d -l 127.0.0.1 /dev/hd? /dev/sda /dev/sdb root 2279 0.0 0.1 39324 3456 ? Ss Jun25 0:00 squid squid 2282 0.0 7.3 184384 151908 ? S Jun25 1:10 (squid) rpc 2283 0.0 0.0 14624 896 ? Ss Jun25 0:00 rpcbind -i squid 2297 0.0 0.0 11628 952 ? S Jun25 0:00 (unlinkd) root 2304 0.0 0.0 20972 1100 ? Ss Jun25 0:00 rpc.statd --port 4001 root 2312 0.0 0.0 0 0 ? S< Jun25 0:00 [rpciod] root 2314 0.0 0.0 0 0 ? S< Jun25 0:00 [nfsiod] root 2322 0.0 0.0 18864 784 ? Ss Jun25 0:00 rpc.idmapd root 2399 0.0 0.0 43584 1260 ? Ss Jun25 0:00 /usr/sbin/sshd root 2405 0.0 0.0 0 0 ? S Jun25 0:00 [lockd] root 2408 0.0 0.0 0 0 ? S< Jun25 0:00 [nfsd4] root 2409 0.0 0.0 0 0 ? S< Jun25 0:00 [nfsd4_callbacks] root 2410 0.0 0.0 0 0 ? S Jun25 0:00 [nfsd] root 2411 0.0 0.0 0 0 ? S Jun25 0:00 [nfsd] root 2412 0.0 0.0 0 0 ? S Jun25 0:00 [nfsd] root 2413 0.0 0.0 0 0 ? S Jun25 0:01 [nfsd] root 2414 0.0 0.0 0 0 ? S Jun25 0:00 [nfsd] root 2415 0.0 0.0 0 0 ? S Jun25 0:00 [nfsd] root 2416 0.0 0.0 0 0 ? S Jun25 0:00 [nfsd] root 2417 0.0 0.0 0 0 ? S Jun25 0:00 [nfsd] root 2443 0.0 0.0 21328 1420 ? Ss Jun25 0:00 rpc.mountd --port 4003 root 2497 0.0 0.0 14616 448 ? Ss Jun25 0:05 gpg-agent --keep-display --daemon --write-env-file //.gnupg/gpg-agent-info ntp 2667 0.0 0.0 26344 1488 ? Ss Jun25 0:00 ntpd -u ntp:ntp -p /var/run/ntpd.pid root 2853 0.0 0.0 10552 140 ? Ss Jun25 0:00 /usr/sbin/mandi -d root 2882 0.0 0.0 3964 548 tty1 Ss+ Jun25 0:00 /sbin/mingetty tty1 root 2883 0.0 0.0 3964 548 tty2 Ss+ Jun25 0:00 /sbin/mingetty tty2 root 2884 0.0 0.0 3964 548 tty3 Ss+ Jun25 0:00 /sbin/mingetty tty3 root 2885 0.0 0.0 3964 548 tty4 Ss+ Jun25 0:00 /sbin/mingetty tty4 root 2886 0.0 0.0 3964 552 tty5 Ss+ Jun25 0:00 /sbin/mingetty tty5 root 2887 0.0 0.0 3964 552 tty6 Ss+ Jun25 0:00 /sbin/mingetty tty6 userXX 2942 0.0 0.0 17412 416 ? Ss Jun25 0:05 gpg-agent --keep-display --daemon --write-env-file /home/userXX/.gnupg/gpg-agent-info root 3089 0.0 0.1 121444 3388 ? Sl Jun25 0:00 /usr/lib64/upowerd root 3147 0.0 0.1 125084 3616 ? Sl Jun25 0:01 /usr/lib64/udisks-daemon root 3149 0.0 0.0 42820 768 ? S Jun25 0:22 udisks-daemon: polling /dev/sr0 rtkit 3311 0.0 0.0 162200 1176 ? SNl Jun25 0:01 /usr/lib64/rtkit-daemon root 3935 0.0 0.0 6056 412 ? S Jun25 0:09 /sbin/ifplugd -I -b -i eth0 root 4130 0.0 0.1 9888 2460 ? Ss Jun25 0:00 dhclient -1 -q -lf /var/lib/dhcp/dhclient-eth0.leases -pf /var/run/dhclient-eth0.pid -cf /etc/dhclient-eth0.conf eth0 root 8723 0.0 0.0 0 0 ? S Jun25 0:00 [nfsv4.0-svc] root 10278 0.0 0.0 0 0 ? S Jun25 0:37 [kworker/0:1] root 10370 0.0 1.2 106508 25560 tty7 Ss+ Jun25 2:19 /etc/X11/X :0 vt7 -nr -nolisten tcp -auth /var/run/xauth/A:0-iRBUJa root 10374 0.0 0.0 35076 1456 ? S Jun25 0:00 -:0 root 10377 0.1 2.4 382596 49736 ? Sl Jun25 8:53 /usr/lib64/kde4/libexec/kdm_greet root 19802 0.0 0.0 39924 1640 ? S Jun28 0:00 CROND root 19803 0.0 0.0 3960 556 ? SNs Jun28 0:00 run-parts --report /etc/cron.daily root 19907 0.0 0.0 13260 1428 ? SN Jun28 0:00 /bin/bash /etc/cron.daily/msec root 20509 0.0 0.3 64272 7428 ? DN Jun28 0:00 /usr/bin/python -O /usr/share/msec/msecperms.py -q -e /var/log/security/open_port.daily.today /var/log/security/open_port.daily.yesterday root 21655 0.0 0.0 0 0 ? S 09:15 0:00 [flush-8:0] root 21676 0.0 0.0 0 0 ? S 09:28 0:00 [kworker/0:0] root 21787 0.0 0.0 0 0 ? S 09:33 0:00 [kworker/0:2] root 21788 0.0 0.1 54428 2928 ? Ss 09:36 0:00 sshd: userXX [priv] userXX 21790 0.0 0.0 54428 1424 ? S 09:36 0:00 sshd: userXX@pts/0 userXX 21791 0.1 0.0 15368 1584 pts/0 Ss 09:36 0:00 -bash root 21831 0.1 0.0 45936 1396 pts/0 S 09:36 0:00 su root 21832 0.1 0.0 15380 1704 pts/0 S 09:36 0:00 bash root 21862 0.0 0.0 12836 1032 pts/0 R+ 09:36 0:00 ps agux

by **martinw** » Jun 30th, '11, 00:53

The only thing I can see that might possibly be the problem is

Code: Select all: root 20509 0.0 0.3 64272 7428 ? DN Jun28 0:00 /usr/bin/python -O /usr/share/msec/msecperms.py -q -e /var/log/security/open_port.daily.today /var/log/security/open_port.daily.yesterday

The "D" in the STAT column means the process is in an uninterruptible wait state - so it might still be counted as a runnable process. Is that process still there?

by **DentAr** » Jun 30th, '11, 11:13

Hi,

yes, this process is always there, even if i kill it, it will be respawn by the system immediately.
I thought this was part of the msec suite running in the background,
as this was also true for mandriva, i did not tune anything regarding to the periodic scan... (and mandriva did not show the 1.0 load average)

by **martinw** » Jul 1st, '11, 00:47

I stand ready to be corrected on this, but my understanding is that the msec process is started by cron on a daily basis, so should not be there permanently. Certainly it is not showing up when I run 'ps' at the moment on my machine.

Does the msec process always have the "DN" status when you run 'ps'? If so, this would suggest that it is getting hung up when trying to access a device. Are there any messages from msec in /var/log/syslog?

by **DentAr** » Jul 1st, '11, 20:23

Hmm, strange...
For several days, this python process was active (or after killing with -9) reactivated by the system.
Always in the 'DN' state.
Now, after leaving the machine alone (only accessing the squid proxy from time to time) this python process is gone, and, even better, the load average is as it should be: 0.00, 0.01, 0.05

I am not really sure what happened, but this problem seems to be solved then...

Good to know, that processes having the DN state are taken into account when calculating the load average?
Anyway,
thanks for your help!!!

by **martinw** » Jul 2nd, '11, 22:13

As you say, strange... But good to hear that it's sorted itself out.

Mageia forum

Idle machine with load average at 1.0/1.01/1.02?

Idle machine with load average at 1.0/1.01/1.02?

Re: Idle machine with load average at 1.0/1.01/1.02?

Re: Idle machine with load average at 1.0/1.01/1.02?

Re: Idle machine with load average at 1.0/1.01/1.02?

Re: Idle machine with load average at 1.0/1.01/1.02?

Re: Idle machine with load average at 1.0/1.01/1.02?

Re: Idle machine with load average at 1.0/1.01/1.02

Re: Idle machine with load average at 1.0/1.01/1.02?

Who is online