arch boots quick enough, but after login hdd light is on constantly

pdincubus · 2009-03-18 16:57:39

Hi all, not sure what info to post to give the best overview on the problem.

Arch seems to boot reasonably quickly, but after login my system seems to grind to a near halt for a few minutes. if i leave it for a while and come back to it, everything is much more responsive. it seems that the longer i leave it, the quicker it feels.

Here is a copy of top after a few minutes, and while the hdd light is still on constant.

top - 16:50:03 up 15 min,  2 users,  load average: 0.14, 0.41, 0.34
Tasks: 123 total,   1 running, 122 sleeping,   0 stopped,   0 zombie
Cpu(s):  2.4%us,  1.4%sy,  0.0%ni, 90.4%id,  4.1%wa,  0.1%hi,  1.5%si,  0.0%st
Mem:   3113992k total,   580316k used,  2533676k free,    19040k buffers
Swap:  4194296k total,        0k used,  4194296k free,   235128k cached

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
 4087 root      20   0  408m  32m 8696 S    5  1.1   0:14.33 X
 4264 pd        20   0  279m  51m  25m S    4  1.7   0:03.83 plasma
 4257 pd        20   0  120m  56m  36m S    4  1.9   1:06.87 kwin
 2003 root      15  -5     0    0    0 S    3  0.0   0:26.87 kcopyd
 4308 pd        20   0  239m 115m  23m S    2  3.8   0:16.27 firefox
 4303 pd        20   0 73448  19m  13m S    1  0.6   0:02.11 konsole
 4269 pd        20   0 51036 9300 7648 S    0  0.3   0:00.04 kaccess
    1 root      20   0  1688  576  512 S    0  0.0   0:00.98 init
    2 root      15  -5     0    0    0 S    0  0.0   0:00.00 kthreadd
    3 root      RT  -5     0    0    0 S    0  0.0   0:00.00 migration/0
    4 root      15  -5     0    0    0 S    0  0.0   0:00.01 ksoftirqd/0
    5 root      RT  -5     0    0    0 S    0  0.0   0:00.00 watchdog/0
    6 root      RT  -5     0    0    0 S    0  0.0   0:00.00 migration/1
    7 root      15  -5     0    0    0 S    0  0.0   0:00.09 ksoftirqd/1
    8 root      RT  -5     0    0    0 S    0  0.0   0:00.00 watchdog/1
    9 root      RT  -5     0    0    0 S    0  0.0   0:00.00 migration/2
   10 root      15  -5     0    0    0 S    0  0.0   0:00.04 ksoftirqd/2
   11 root      RT  -5     0    0    0 S    0  0.0   0:00.00 watchdog/2
   12 root      RT  -5     0    0    0 S    0  0.0   0:00.00 migration/3
   13 root      15  -5     0    0    0 S    0  0.0   0:00.12 ksoftirqd/3
   14 root      RT  -5     0    0    0 S    0  0.0   0:00.00 watchdog/3
   15 root      15  -5     0    0    0 S    0  0.0   0:00.00 events/0
   16 root      15  -5     0    0    0 S    0  0.0   0:00.03 events/1
   17 root      15  -5     0    0    0 S    0  0.0   0:00.00 events/2
   18 root      15  -5     0    0    0 S    0  0.0   0:00.00 events/3
   19 root      15  -5     0    0    0 S    0  0.0   0:00.00 khelper
   20 root      RT  -5     0    0    0 S    0  0.0   0:00.00 kstop/0
   21 root      RT  -5     0    0    0 S    0  0.0   0:00.00 kstop/1
   22 root      RT  -5     0    0    0 S    0  0.0   0:00.00 kstop/2
   23 root      RT  -5     0    0    0 S    0  0.0   0:00.00 kstop/3
   24 root      15  -5     0    0    0 S    0  0.0   0:00.06 kblockd/0
   25 root      15  -5     0    0    0 S    0  0.0   0:00.06 kblockd/1
   26 root      15  -5     0    0    0 S    0  0.0   0:00.07 kblockd/2
   27 root      15  -5     0    0    0 S    0  0.0   0:00.06 kblockd/3
   28 root      15  -5     0    0    0 S    0  0.0   0:00.00 kacpid
   29 root      15  -5     0    0    0 S    0  0.0   0:00.00 kacpi_notify
   30 root      15  -5     0    0    0 S    0  0.0   0:00.00 kseriod
   31 root      20   0     0    0    0 S    0  0.0   0:00.00 pdflush
   32 root      20   0     0    0    0 S    0  0.0   0:00.00 pdflush
   33 root      15  -5     0    0    0 S    0  0.0   0:00.00 kswapd0
   34 root      15  -5     0    0    0 S    0  0.0   0:00.00 aio/0
   35 root      15  -5     0    0    0 S    0  0.0   0:00.00 aio/1
   36 root      15  -5     0    0    0 S    0  0.0   0:00.00 aio/2
   37 root      15  -5     0    0    0 S    0  0.0   0:00.00 aio/3
   48 root      15  -5     0    0    0 S    0  0.0   0:00.01 ata/0
   49 root      15  -5     0    0    0 S    0  0.0   0:00.00 ata/1
   50 root      15  -5     0    0    0 S    0  0.0   0:00.00 ata/2
   51 root      15  -5     0    0    0 S    0  0.0   0:00.01 ata/3
   52 root      15  -5     0    0    0 S    0  0.0   0:00.00 ata_aux
   56 root      15  -5     0    0    0 S    0  0.0   0:00.00 scsi_eh_0
   57 root      15  -5     0    0    0 S    0  0.0   0:00.01 scsi_eh_1
  409 root      15  -5     0    0    0 S    0  0.0   0:00.00 ksuspend_usbd
  445 root      15  -5     0    0    0 S    0  0.0   0:00.06 khubd
  562 root      15  -5     0    0    0 S    0  0.0   0:00.00 kstriped

As far as i can see, the cpu isn't being hammered and memory usage is pretty low.

Any ideas what to check? i,m also using openbox and see the same issues in there too, so i'm pretty sure it's not a kde specific problem. At first i thought it might be all that nepomuk stuff, but i believe that is now all disabled.

Thanks in advance

byte · 2009-03-19 18:05:40

My guess would be Nepomuk/Strigi, check the systray and "Desktop Search" in the system settings.
Another app known for its harddisk abuse is Firefox when you have the "Tell me if the site..." security options enabled.

lukrop · 2009-03-19 20:28:00

maybe you check wich program is causing that with iotop

mingus · 2009-03-20 04:41:42

To reiterate lukrop's suggestion, I think iotop will be more useful to you rather than top to figure out what's going wrong. I was experiencing the same HD lockup/system freeze as you, the culprit being pdflush/kjournald. Sounds like you may need to tweak your vm.dirty_background_ratio and vm.dirty_ratio settings via /etc/sysctl.conf.

Here's a fairly straightforward description of the problem and a suggested solution: Kernel Tuning (Scroll down to section "Kernel Virtual Memory Management") Please don't take my word for it, though. I'm still a newbie trying to find my own ass in the dark using a candle, so hopefully a guru can weigh in on the issue and provide better guidance.

pdincubus · 2009-03-22 22:27:43

Thanks for the replies, i've pasted below from iotop a few seconds after logging in. it looks like it could be kcopy that is causing the problem. i've been on for a few minutes now and it's kcopy still going for it. Any ideas why it's running or what it might be doing?

2008 root       72.68 M/s       0 B/s  0.00 %  0.00 % [kcopyd]
    1 root           0 B/s       0 B/s  0.00 %  0.00 % init [3]
    2 root           0 B/s       0 B/s  0.00 %  0.00 % [kthreadd]
    3 root           0 B/s       0 B/s  0.00 %  0.00 % [migration/0]
    4 root           0 B/s       0 B/s  0.00 %  0.00 % [ksoftirqd/0]
    5 root           0 B/s       0 B/s  0.00 %  0.00 % [watchdog/0]
    6 root           0 B/s       0 B/s  0.00 %  0.00 % [migration/1]
    7 root           0 B/s       0 B/s  0.00 %  0.00 % [ksoftirqd/1]
    8 root           0 B/s       0 B/s  0.00 %  0.00 % [watchdog/1]
    9 root           0 B/s       0 B/s  0.00 %  0.00 % [migration/2]
   10 root           0 B/s       0 B/s  0.00 %  0.00 % [ksoftirqd/2]
   11 root           0 B/s       0 B/s  0.00 %  0.00 % [watchdog/2]
   12 root           0 B/s       0 B/s  0.00 %  0.00 % [migration/3]
   13 root           0 B/s       0 B/s  0.00 %  0.00 % [ksoftirqd/3]
   14 root           0 B/s       0 B/s  0.00 %  0.00 % [watchdog/3]
   15 root           0 B/s       0 B/s  0.00 %  0.00 % [events/0]
   16 root           0 B/s       0 B/s  0.00 %  0.00 % [events/1]
   17 root           0 B/s       0 B/s  0.00 %  0.00 % [events/2]
   18 root           0 B/s       0 B/s  0.00 %  0.00 % [events/3]
   19 root           0 B/s       0 B/s  0.00 %  0.00 % [khelper]
   20 root           0 B/s       0 B/s  0.00 %  0.00 % [kstop/0]
   21 root           0 B/s       0 B/s  0.00 %  0.00 % [kstop/1]
   22 root           0 B/s       0 B/s  0.00 %  0.00 % [kstop/2]
   23 root           0 B/s       0 B/s  0.00 %  0.00 % [kstop/3]
   24 root           0 B/s       0 B/s  0.00 %  0.00 % [kblockd/0]
   25 root           0 B/s       0 B/s  0.00 %  0.00 % [kblockd/1]
   26 root           0 B/s       0 B/s  0.00 %  0.00 % [kblockd/2]
   27 root           0 B/s       0 B/s  0.00 %  0.00 % [kblockd/3]
   28 root           0 B/s       0 B/s  0.00 %  0.00 % [kacpid]

edit: i had already stopped strigi from running a while back as i didn't use it's features.

Last edited by pdincubus (2009-03-22 22:32:40)

pdincubus · 2009-03-25 08:37:19

I hate to *bump*, but I don't want this to get lost in the ether. Anybody any ideas about the problem above?

Thanks in advance!

gattschardo · 2009-03-25 08:59:08

Well, to see what it's "doing" you can of course call [b]ps aux |grep kcopyd[b], maybe one of the parameters that it was called with might give a hint. Although, seeing that it seems to be a daemon, maybe it won't - in this case there should exist some configuration somewhere, something might be wrong in there.

I'm not a kde user though, so I really don't know what kcopyd is supposed to do, amybe someone else can answer this?

dcrabs · 2009-03-25 09:08:09

kcopyd doesn't belong to KDE it is rather kernel related; something with device mapper.
@pdincubus
It it is probably related to your LVM2 setup.

pdincubus · 2009-03-25 11:28:10

thanks for the replies. i did a bit of a lookup on Google and it seemed that Kcopy was definitely to do with the kernel. I do have an LVM setup:

2 x 250 GB drives as physical volumes, one volume group on there, with a few bits in there as logical volumes. At least one of those is mirrored: could this be causing this problem?

I don't see why it would need to copy so much data across on boot, unless i've misunderstood somewhere i thought the LVM mirroring was kept up to date all the time, so shouldn't need to copy a load of stuff across all the time? Can anyone shed any further light on this? As far as I can tell, the mirrored bits are in two: one copy on one drive, one on the other. Do you know of any way to check what is being copied and between which two points?

It seems we might be finally getting to the root of the problem.

Thanks for you help!

pdincubus · 2009-03-29 09:32:20

*bump*
Anyone with LVM that can shed any light on this for me? I have one of my logical volumes set to mirror so I guess that is the data kcopy is copying around, but how come it waits til the first login and then copies everythin instead of just copying what's changed when it changes?
Would a snapshot work better for me? Really what I want is a copy of my important stuff on a seperate hard drive, so mirrored logical volumes is not absolutely necessary I don't think.

Any ideas?

Odysseus · 2009-03-30 00:36:05

Sounds like a misconfiguration. LVM thinks it must "update" one of your drives to reflect the other.

And yes, I think you want a separate backup of your important data, not a mirror. If you accidentally (or otherwise) delete some important data, It will be gone from the mirror too.

pdincubus · 2009-03-30 09:45:35

Any ideas how to check for a misconfiguration. I've followed instructions to get it working, so unsure as to what could be the problem with it?
I think i might change LVM to have no mirrors, and reboot. See if it is still a problem. If not, I'll know for definite that it's the mirroring that's the issue.

Any pointers for a simple, easy backup solution? Only needs to probably backup one partition/drive to another within the same box.

Thanks

Odysseus · 2009-03-31 01:02:17

dd if=/dev/sda of=/dev/sdb

Does a low level copy from hard drive 1 to hard drive 2.

pdincubus · 2009-04-02 13:02:52

i turned off mirroring on the lvm setup, and rebooted.

the kcopy problem seems to have gone for now, so i'll need to set up that copy command as a cron job or something to see how it goes and see where it leaves me.

on the plus side, i've just freed up nearly 200GB to add into lvm and use

sacarde · 2009-04-09 12:42:36

I have similar problem

I view in iotop a kjournald proces, every 3/4 seconds, and access to HD

I have 2 ext3 partition mounted

what suggest to me?

sacarde · 2009-04-09 18:35:27

I read:

kjournald:
It is for the journaling feature of the EXT3 file system. Kjournald is the journaling kernel thread for ext3 filesystem. One for each mounted ext3 filesystem.

then, possible solutions I found are:

- add mount parameter: noatime

- add mount parameter: norelatime

- install laptop-mode-tools

help

Arch Linux

#1 2009-03-18 16:57:39

arch boots quick enough, but after login hdd light is on constantly

#2 2009-03-19 18:05:40

Re: arch boots quick enough, but after login hdd light is on constantly

#3 2009-03-19 20:28:00

Re: arch boots quick enough, but after login hdd light is on constantly

#4 2009-03-20 04:41:42

Re: arch boots quick enough, but after login hdd light is on constantly

#5 2009-03-22 22:27:43

Re: arch boots quick enough, but after login hdd light is on constantly

#6 2009-03-25 08:37:19

Re: arch boots quick enough, but after login hdd light is on constantly

#7 2009-03-25 08:59:08

Re: arch boots quick enough, but after login hdd light is on constantly

#8 2009-03-25 09:08:09

Re: arch boots quick enough, but after login hdd light is on constantly

#9 2009-03-25 11:28:10

Re: arch boots quick enough, but after login hdd light is on constantly

#10 2009-03-29 09:32:20

Re: arch boots quick enough, but after login hdd light is on constantly

#11 2009-03-30 00:36:05

Re: arch boots quick enough, but after login hdd light is on constantly

#12 2009-03-30 09:45:35

Re: arch boots quick enough, but after login hdd light is on constantly

#13 2009-03-31 01:02:17

Re: arch boots quick enough, but after login hdd light is on constantly

#14 2009-04-02 13:02:52

Re: arch boots quick enough, but after login hdd light is on constantly

#15 2009-04-09 12:42:36

Re: arch boots quick enough, but after login hdd light is on constantly

#16 2009-04-09 18:35:27

Re: arch boots quick enough, but after login hdd light is on constantly

Board footer