Hello,

I have the following problem, my actual ubunt 12.04 linux became during one year extremely slowly. Everything is slow, booting ... . Now firefox is not coming up. the iostat command shows, that the CPUs are 99% idle:

vg-cpu: %user %nice %system %iowait %steal %idle
0,76 0,00 0,00 0,00 0,00 99,24

Device: rrqm/s wrqm/s r/s w/s rkB/s wkB/s avgrq-sz avgqu-sz await r_await w_await svctm %util
sda 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00
sdb 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00


Linux 3.2.0-54-generic 06.01.2014 _x86_64_ (4 CPU)

avg-cpu: %user %nice %system %iowait %steal %idle
3,10 0,09 0,76 8,35 0,00 87,70

Device: tps kB_read/s kB_wrtn/s kB_read kB_wrtn
sda 4,40 123,32 21,78 645746 114056
sdb 0,09 0,57 0,00 2974 8


The hard disc sda is a SSD-disk: OCZ agility 2

Does anybody has an idea?

Best Regards,
Lutz

Are you maybe out of disk space?

What does df -T say?

Hello,

there seems no problem with the disk use

Dateisystem Typ 1K-Blöcke Benutzt Verfügbar Verw% Eingehängt auf
/dev/sda3 ext3 59620892 10094016 46498224 18% /
udev devtmpfs 4076920 4 4076916 1% /dev
tmpfs tmpfs 1635520 828 1634692 1% /run
none tmpfs 5120 0 5120 0% /run/lock
none tmpfs 4088796 168 4088628 1% /run/shm
/dev/sdb1 ext3 1922858352 759388648 1065794104 42% /media/data
10.96.135.3:/export/home nfs 49468416 4316416 42639104 10% /home

sda3 is the SSD boot disk with ubuntu
the home partition is mounted from a server
sdb1 is a data disk

Additionally, I performed manually trim as root:
fstrim -v /
Trim was successful finished.

Best Regards,
Lutz

Have you used a system monitor to ID the serivce(s) that is(are) eating up your resources?

You didn't mention how much ram you have, nor if you have a swap partition.

Are you out of memory and swapping?

Use the "top" command to identify the process(s) that are using the most memory. When it's running, hit "M" to sort by memory usage. Alternately, hit "G", then hit "3" to show the memory grouping page.

p.s. - Those need to uppercase "M" and "G" above, lowercase are different commands.

Hello,

please find attachted a typical top output. The computer is equipped with 4 cores in two CPUs, 8 GB memory and 8 GB swap. It does not seem that the hardware resources are insufficient.

top - 08:30:00 up 23:16, 6 users, load average: 10.10, 10.13, 9.75
Tasks: 185 total, 1 running, 183 sleeping, 0 stopped, 1 zombie
Cpu(s): 1.0%us, 0.2%sy, 0.0%ni, 74.5%id, 24.4%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 8177592k total, 6784808k used, 1392784k free, 218652k buffers
Swap: 7815616k total, 0k used, 7815616k free, 5655300k cached

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
3157 root 20 0 227m 80m 21m S 4 1.0 36:00.09 Xorg
3888 luedeman 20 0 582m 29m 12m S 1 0.4 5:22.61 gnome-terminal
6157 luedeman 20 0 17340 1372 960 R 0 0.0 0:00.17 top
1 root 20 0 25052 2952 1300 S 0 0.0 0:01.62 init
2 root 20 0 0 0 0 S 0 0.0 0:00.03 kthreadd
3 root 20 0 0 0 0 S 0 0.0 0:02.34 ksoftirqd/0
6 root RT 0 0 0 0 S 0 0.0 0:01.50 migration/0
7 root RT 0 0 0 0 S 0 0.0 0:00.38 watchdog/0
8 root RT 0 0 0 0 S 0 0.0 0:00.47 migration/1
10 root 20 0 0 0 0 S 0 0.0 0:01.30 ksoftirqd/1
12 root RT 0 0 0 0 S 0 0.0 0:00.27 watchdog/1
13 root RT 0 0 0 0 S 0 0.0 0:00.85 migration/2
15 root 20 0 0 0 0 S 0 0.0 0:02.29 ksoftirqd/2
16 root RT 0 0 0 0 S 0 0.0 0:00.32 watchdog/2
17 root RT 0 0 0 0 S 0 0.0 0:00.31 migration/3
19 root 20 0 0 0 0 S 0 0.0 0:01.16 ksoftirqd/3
20 root RT 0 0 0 0 S 0 0.0 0:00.25 watchdog/3

Here the top output sorted by memory use "M":

top - 08:35:51 up 23:22, 6 users, load average: 8.51, 9.14, 9.44
Tasks: 185 total, 1 running, 183 sleeping, 0 stopped, 1 zombie
Cpu(s): 2.2%us, 0.5%sy, 0.0%ni, 80.3%id, 17.1%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 8177592k total, 6787404k used, 1390188k free, 220268k buffers
Swap: 7815616k total, 0k used, 7815616k free, 5655276k cached

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
3729 luedeman 20 0 484m 109m 18m S 0 1.4 4:08.38 ubuntuone-syncd
4052 luedeman 20 0 662m 87m 40m D 0 1.1 0:04.87 firefox
3157 root 20 0 227m 80m 21m S 5 1.0 36:10.46 Xorg
2388 luedeman 20 0 567m 45m 10m D 0 0.6 0:02.35 dropbox
2717 luedeman 20 0 567m 45m 10m D 0 0.6 0:01.84 dropbox
3554 luedeman 20 0 567m 45m 10m D 0 0.6 0:02.54 dropbox
3836 luedeman 20 0 567m 45m 10m D 0 0.6 0:01.92 dropbox
3519 luedeman 20 0 608m 31m 16m S 0 0.4 0:21.64 gnome-panel
3516 luedeman 20 0 591m 31m 12m S 0 0.4 0:03.05 metacity
3532 luedeman 20 0 1400m 30m 19m S 0 0.4 0:29.61 nautilus
3888 luedeman 20 0 582m 30m 12m S 1 0.4 5:24.73 gnome-terminal
3497 luedeman 20 0 538m 28m 14m S 1 0.4 3:21.70 python
3486 luedeman 20 0 667m 26m 13m S 0 0.3 0:05.83 gnome-settings-
3604 luedeman 20 0 577m 20m 11m S 1 0.3 1:53.52 indicator-apple
4281 root 20 0 165m 17m 11m S 1 0.2 0:04.97 OCZToolbox
3540 luedeman 20 0 686m 17m 12m S 0 0.2 0:00.45 nm-applet
3528 luedeman 20 0 480m 16m 11m S 0 0.2 0:00.60 polkit-gnome-au

And the top output using memory grouping, "G3":

top - 08:39:08 up 23:25, 6 users, load average: 7.29, 8.45, 9.13
Tasks: 185 total, 1 running, 183 sleeping, 0 stopped, 1 zombie
Cpu(s): 1.1%us, 0.2%sy, 0.0%ni, 98.7%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 8177592k total, 6786676k used, 1390916k free, 220688k buffers
Swap: 7815616k total, 0k used, 7815616k free, 5655352k cached

PID %MEM VIRT SWAP RES CODE DATA SHR nFLT nDRT S PR NI %CPU COMMAND
3729 1.4 484m 375m 109m 2496 306m 18m 0 0 S 20 0 0 ubuntuone-syncd
4052 1.1 662m 575m 87m 92 211m 40m 117 0 D 20 0 0 firefox
3157 1.0 227m 146m 80m 1984 52m 21m 0 0 S 20 0 5 Xorg
2388 0.6 567m 521m 45m 3524 322m 10m 123 0 D 20 0 0 dropbox
2717 0.6 567m 521m 45m 3524 322m 10m 0 0 D 20 0 0 dropbox
3554 0.6 567m 521m 45m 3524 322m 10m 0 0 D 20 0 0 dropbox
3836 0.6 567m 521m 45m 3524 322m 10m 0 0 D 20 0 0 dropbox
3519 0.4 608m 576m 31m 556 310m 16m 44 0 S 20 0 0 gnome-panel
3516 0.4 591m 560m 31m 544 234m 12m 1 0 S 20 0 0 metacity
3888 0.4 582m 552m 30m 284 304m 12m 0 0 S 20 0 2 gnome-terminal
3532 0.4 1400m 1.3g 30m 1428 897m 19m 0 0 S 20 0 0 nautilus
3497 0.4 538m 509m 28m 2496 228m 14m 0 0 S 20 0 0 python
3486 0.3 667m 640m 26m 32 299m 13m 0 0 S 20 0 0 gnome-settings-
3604 0.3 577m 557m 20m 32 292m 11m 0 0 S 20 0 0 indicator-apple
3540 0.2 686m 668m 17m 352 292m 12m 0 0 S 20 0 0 nm-applet
4281 0.2 165m 148m 16m 12m 77m 11m 3 0 S 20 0 0 OCZToolbox
3528 0.2 480m 464m 16m 40 219m 11m 0 0 S 20 0 0 polkit-gnome-au
3702 0.2 409m 394m 14m 124 147m 10m 0 0 S 20 0 0 gnome-screensav
3769 0.2 394m 380m 14m 68 219m 10m 0 0 S 20 0 0 update-notifier
1043 0.2 490m 477m 13m 120 222m 8860 71 0 S 20 0 0 colord
3570 0.2 390m 377m 13m 32 145m 10m 0 0 S 20 0 0 goa-daemon
3527 0.2 483m 470m 12m 36 219m 9040 1 0 S 20 0 0 bluetooth-apple
3713 0.2 380m 368m 12m 68 146m 9396 0 0 S 20 0 0 evolution-alarm
3537 0.2 417m 405m 12m 280 218m 9220 0 0 S 20 0 0 vino-server

What does "sudo hdparm -tT /dev/sda" outputs (its a basic read/write speed test, safe to do)?
Also try copying large files (around 500mb or larger) in some graphical application that reports write speeds.

hdparm provides for the SSD disk (sda) and hard disk (sdb) the following values:

hdparm -Tt /dev/sda

/dev/sda:
Timing cached reads: 1630 MB in 2.00 seconds = 815.34 MB/sec
Timing buffered disk reads: 366 MB in 3.01 seconds = 121.44 MB/sec


hdparm -Tt /dev/sdb

/dev/sdb:
Timing cached reads: 1606 MB in 2.00 seconds = 802.59 MB/sec
Timing buffered disk reads: 184 MB in 3.01 seconds = 61.13 MB/sec

what I do not understand:
Hava a look at the memory use:
Mem: 8177592k total, 6787404k used, 1390188k free,

There are no applications running and already 6,7 GB of 8 GB memory are in use. But the memory output does not tell me the processes using the memory:
No process uses more than 2% of the memory!


Here the top output sorted by memory use "M":

top - 08:35:51 up 23:22, 6 users, load average: 8.51, 9.14, 9.44
Tasks: 185 total, 1 running, 183 sleeping, 0 stopped, 1 zombie
Cpu(s): 2.2%us, 0.5%sy, 0.0%ni, 80.3%id, 17.1%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 8177592k total, 6787404k used, 1390188k free, 220268k buffers
Swap: 7815616k total, 0k used, 7815616k free, 5655276k cached

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
3729 luedeman 20 0 484m 109m 18m S 0 1.4 4:08.38 ubuntuone-syncd
4052 luedeman 20 0 662m 87m 40m D 0 1.1 0:04.87 firefox
3157 root 20 0 227m 80m 21m S 5 1.0 36:10.46 Xorg
2388 luedeman 20 0 567m 45m 10m D 0 0.6 0:02.35 dropbox
2717 luedeman 20 0 567m 45m 10m D 0 0.6 0:01.84 dropbox
3554 luedeman 20 0 567m 45m 10m D 0 0.6 0:02.54 dropbox
3836 luedeman 20 0 567m 45m 10m D 0 0.6 0:01.92 dropbox
3519 luedeman 20 0 608m 31m 16m S 0 0.4 0:21.64 gnome-panel
3516 luedeman 20 0 591m 31m 12m S 0 0.4 0:03.05 metacity
3532 luedeman 20 0 1400m 30m 19m S 0 0.4 0:29.61 nautilus
3888 luedeman 20 0 582m 30m 12m S 1 0.4 5:24.73 gnome-terminal
3497 luedeman 20 0 538m 28m 14m S 1 0.4 3:21.70 python
3486 luedeman 20 0 667m 26m 13m S 0 0.3 0:05.83 gnome-settings-
3604 luedeman 20 0 577m 20m 11m S 1 0.3 1:53.52 indicator-apple
4281 root 20 0 165m 17m 11m S 1 0.2 0:04.97 OCZToolbox
3540 luedeman 20 0 686m 17m 12m S 0 0.2 0:00.45 nm-applet
3528 luedeman 20 0 480m 16m 11m S 0 0.2 0:00.60 polkit-gnome-au

Have a read of this - working as designed.

Maybe you can look at latencytop to see what might be contributing.

1 members found this post helpful.

Thanks, I started now the grapical interface for system processes. They show that only 680 MB of 7.8GB memory is used by ubuntu. The four cores are use only approximately 10% of there resources, each. Never the less, I am not able to start any larger application, e.g. firefox.

I installed latencytop and got something like the following output:

Waiting for event (poll): 4.5msec 2.1%
waiting for event (select): 4.3msec 97.9%

Process Xorg Total 610 msec
Waiting for avent (select): 4.3msec 100%

What does it mean?

What happens (output) when you start firefox in a terminal?

I started firefox in a terminal and nothing happened over hours. No error massages.
Then I started firefox using the option -jsconsole to open the error console, also nothing happend.

With respect to LatencyTop:
When I started LatencyTop in an alpha-numeric Terminal (e.g. Str Alt F1), it starts immediately. Using the classical gnome graphical interface, it took hours. If an application is running using the gnome surface e.g. system monitoring the output is alright.

The problems seem to have something to do with the window manager. I used the new ubuntu and classical gnome but the problem is independent of the window manager.

Can you ssh into the slow computer from another computer (thus bypassing any window manager or GUI stuff), and see what kind of performance you are getting over that ssh connection?