System getting slow / freezing

Guys,

I have opensuse 12.3 installed on my laptop. Since some days ago, this system gets very slow to boot. It hangs at “Starting tell plymouth to write out runtime data” and then it delays some five minutes to continue boot process then starts X server. Also, after boot, on normal use, it got some freezing too, when switching between apps or windows or event switching between browser tabs. The hdd led stays on all the time, so I think may be there’s a problem with filesystem or some other process is keeping the disk busy, but I don’t know how to check it. Can you give me some advices?

On 2013-09-08 14:16, edersc wrote:
>
> Guys,
>
> I have opensuse 12.3 installed on my laptop. Since some days ago, this
> system gets very slow to boot. It hangs at “Starting tell plymouth to
> write out runtime data” and then it delays some five minutes to continue
> boot process then starts X server. Also, after boot, on normal use, it
> got some freezing too, when switching between apps or windows or event
> switching between browser tabs. The hdd led stays on all the time, so I
> think may be there’s a problem with filesystem or some other process is
> keeping the disk busy, but I don’t know how to check it. Can you give me
> some advices?

I would suspect hard disk problem. If this is the case, there would be
telltale entries in the syslog - sorry, I know what they are when I see
them, but I don’t remember the words to type them here.

You could run the long test in smartctl, or download the seatools ISO
from seagate; it is a boot CD or usb stick, designed for testing disks.

If you need more details, please ask.


Cheers / Saludos,

Carlos E. R.
(from 12.3 x86_64 “Dartmouth” at Telcontar)

[HR][/HR] Well, I had this same problem some months ago. Since I hasn’t another machine for search for solution, I decided to do a fresh install, and then the system get fine again.
But since I supposed to do not do that (this sounds like windows universal solution: format c: ), this time I decided to investigate why it happens. The harddisk it pretty new, And then I doubt it has phisical defects.
Anyway, I’ll put some rescue disk on and then try to check filesystem.

On 2013-09-08 15:46, edersc wrote:
>
> The harddisk it pretty new, And then I doubt it has phisical
> defects.

But sometimes they do. Please verify.

> Anyway, I’ll put some rescue disk on and then try to check filesystem.

Not the filesytem, the hard disk itself is what you need to verify.

Just run this:


smartctl -a /dev/sda

and post it here.


Cheers / Saludos,

Carlos E. R.
(from 12.3 x86_64 “Dartmouth” at Telcontar)

Here is
smartctl 6.0 2012-10-10 r3643 [x86_64-linux-3.7.10-1.16-desktop] (SUSE RPM) Cop - Pastebin.com

Well… I think the problem should be on some system service (one of that started by systemd), because boots always stalls at the same stage, and because after some minutes of usage, the system runs normally.

You could install systemd-analyze and run

systemd-analyze blame

That command will show you what’s slowing down the boot process.

Thanks! With your help I think I found the origins of the problem:


222472ms systemd-tmpfiles-setup.service
  8410ms wicd.service
  4210ms avahi-daemon.service
  4023ms systemd-logind.service
  3628ms xdm.service
  3206ms systemd-vconsole-setup.service
  3206ms postfix.service
  2458ms cycle.service
  2244ms vboxdrv.service
  2138ms systemd-udev-root-symlink.service
  1568ms systemd-remount-fs.service
  1257ms polkit.service
  1055ms systemd-modules-load.service
   880ms dev-mqueue.mount
   876ms dev-hugepages.mount
   656ms rsyslog.service
   556ms win.mount
   525ms network.service
   517ms bluetooth.service
   444ms nfs.service
   419ms systemd-udevd.service
   344ms systemd-user-sessions.service
   312ms upower.service
   234ms fbset.service
   171ms rc-local.service
   157ms systemd-readahead-replay.service
   145ms sshd.service
    95ms udisks2.service
    87ms home.mount
    36ms rtkit-daemon.service
    33ms systemd-udev-trigger.service
    32ms systemd-readahead-collect.service
    10ms systemd-sysctl.service
    10ms var-lock.mount
     5ms sys-kernel-debug.mount
     3ms var-run.mount


Now i just need to know how the f**k I fix it.

On 2013-09-09 18:29, edersc wrote:
>
> Here is
> ‘smartctl 6.0 2012-10-10 r3643 [x86_64-linux-3.7.10-1.16-desktop] (SUSE
> RPM) Cop - Pastebin.com’ (http://pastebin.com/UQBM0Fct)

Disk is good.

Important values are these:


>   5 Reallocated_Sector_Ct   0x0033   252   252   010    Pre-fail  Always       -       0
>
> 198 Offline_Uncorrectable   0x0030   252   252   000    Old_age   Offline      -       0
> 199 UDMA_CRC_Error_Count    0x0036   200   200   000    Old_age   Always       -       0

It has at one time run hot, though (59 degrees).

An important thing you could do is run the long test, as none has been
ever run:


SMART Self-test log structure revision number 1
No self-tests have been logged.  [To run self-tests, use: smartctl -t]

You should, I think, run the long test thus:


smartctl --test=long /dev/sda

and after the specified time (you may continue working with the computer
at the same time, just that it will be slower), check the results with:


smartctl -a /dev/sda

it will be something like this:


> SMART Self-test log structure revision number 1
> Num  Test_Description    Status                  Remaining  LifeTime(hours)  LBA_of_first_error
> # 1  Extended offline    Completed without error       00%         1         -

About checking the time boot takes, I like the graphical info you obtain
thus:


Telcontar:~ # systemd-analyze plot > plot.svg
Telcontar:~ #

You can display the graphic with “display”, but it is probably better to
do it with libreoffice.


Cheers / Saludos,

Carlos E. R.
(from 12.3 x86_64 “Dartmouth” at Telcontar)

Any/All errors which reference Plymouth almost certainly have something to do with the graphic driver, it’s when the system switches from the driver used by GRUB (usually VESA) to whatever you’ve configured in openSUSE.

See the release notes on enabling/disabling KMS.
Consider installing an alternate GPU driver, that would depend on what your GPU is.

TSU

Booting is slow even when starting on console mode only. As we see above on output of systemd-analyse
222472ms systemd-tmpfiles-setup.service

is getting around 3’40" to complete. So now we know where resides the problem, just have to search about this particular service, and how to fix it.

On 2013-09-12 13:46, edersc wrote:
>
> Booting is slow even when starting on console mode only. As we see above
> on output of systemd-analyse
> 222472ms systemd-tmpfiles-setup.service
>
> is getting around 3’40" to complete. So now we know where resides the
> problem, just have to search about this particular service, and how to
> fix it.


systemctl status systemd-tmpfiles-setup.service

and see what it says…


Cheers / Saludos,

Carlos E. R.
(from 12.3 x86_64 “Dartmouth” at Telcontar)

systemd-tmpfiles-setup.service - Recreate Volatile Files and Directories
          Loaded: loaded (/usr/lib/systemd/system/systemd-tmpfiles-setup.service; static)
          Active: active (exited) since Thu, 2013-09-12 23:42:39 BRT; 20h ago
            Docs: man:tmpfiles.d(5)
         Process: 528 ExecStart=/usr/bin/systemd-tmpfiles --create --remove (code=exited, status=0/SUCCESS)
          CGroup: name=systemd:/system/systemd-tmpfiles-setup.service


On 2013-09-14 01:36, edersc wrote:
>
> Code:
> --------------------
> systemd-tmpfiles-setup.service - Recreate Volatile Files and Directories
> Loaded: loaded (/usr/lib/systemd/system/systemd-tmpfiles-setup.service; static)
> Active: active (exited) since Thu, 2013-09-12 23:42:39 BRT; 20h ago
> Docs: man:tmpfiles.d(5)
> Process: 528 ExecStart=/usr/bin/systemd-tmpfiles --create --remove (code=exited, status=0/SUCCESS)
> CGroup: name=systemd:/system/systemd-tmpfiles-setup.service
>
>
> --------------------

Nothing there… all normal, I think.


Cheers / Saludos,

Carlos E. R.
(from 12.3 x86_64 “Dartmouth” at Telcontar)