Author Topic: Process named udevd in a CPU loop [SOLVED]  (Read 3639 times)

Offline Software Monkey

  • Jr. Member
  • **
  • Posts: 21
    • Software Monkey's Web Site
Process named udevd in a CPU loop [SOLVED]
« on: February 09, 2012, 06:31:49 PM »
Just moved from Linux Mint 11 to PCLOS 2011 a few days ago, and we have noticed that my (dual core) CPU is constantly showing about 50% load; looking the the process manager I can see a process named udevd, is consuming CPU constantly, about 100M in memory and was started with the command line "/sbin/udevd -d".  What is this process, and why is it constantly consuming resources?

Code: [Select]
Process 110 - udevd

Summary

The process udevd (with pid 110) is using approximately 99.0 MB of memory.
It is using 98.9 MB privately, and a further 712.0 KB that is, or could be, shared with other programs.
Dividing up the shared memory between all the processes sharing that memory we get a reduced shared memory usage of 54.0 KB. Adding that to the private usage, we get the above mentioned total memory footprint of 99.0 MB.
Library Usage

The memory usage of a process is found by adding up the memory usage of each of its libraries, plus the process's own heap, stack and any other mappings.
Private
more
101236 KB [heap]
40 KB [stack]
16 KB /sbin/udevd
4 KB /lib/i686/libpthread-2.13.so
4 KB /lib/i686/libc-2.13.so
Shared
more
328 KB /lib/i686/libc-2.13.so
236 KB [heap]
72 KB /sbin/udevd
32 KB /lib/ld-2.13.so
16 KB /lib/i686/libpthread-2.13.so
Totals

Private 101304 KB (= 12 KB clean + 101292 KB dirty)
Shared 712 KB (= 436 KB clean + 276 KB dirty)
Rss 102016 KB (= Private + Shared)
Pss 101358 KB (= Private + Shared/Number of Processes)
Swap 0 KB
Full Details

Information about the complete virtual space for the process is available, with sortable columns. An empty filename means that it is an anonymous mapping.
Both the MMU page size and the kernel page size are 4 KB.
Address Perm Size Rss Pss Shared Clean Shared Dirty Private Clean Private Dirty Referenced Anonymous Swap Locked Filename
08048000-0806b000 r-xp 140 KB 84 KB 24 KB 72 KB 0 KB 12 KB 0 KB 84 KB 0 KB 0 KB 0 KB /sbin/udevd
0806b000-0806c000 rw-p 4 KB 4 KB 4 KB 0 KB 0 KB 0 KB 4 KB 4 KB 4 KB 0 KB 0 KB /sbin/udevd
08be5000-0eed6000 rw-p 101316 KB 101244 KB 101220 KB 0 KB 28 KB 0 KB 101216 KB 101220 KB 101244 KB 0 KB 0 KB [heap]
b757e000-b75af000 rw-p 196 KB 196 KB 32 KB 0 KB 196 KB 0 KB 0 KB 196 KB 196 KB 0 KB 0 KB
b7610000-b7612000 rw-p 8 KB 8 KB 4 KB 0 KB 4 KB 0 KB 4 KB 8 KB 8 KB 0 KB 0 KB
b7612000-b7627000 r-xp 84 KB 12 KB 0 KB 12 KB 0 KB 0 KB 0 KB 12 KB 0 KB 0 KB 0 KB /lib/i686/libpthread-2.13.so
b7627000-b7628000 r--p 4 KB 4 KB 0 KB 0 KB 4 KB 0 KB 0 KB 0 KB 4 KB 0 KB 0 KB /lib/i686/libpthread-2.13.so
b7628000-b7629000 rw-p 4 KB 4 KB 4 KB 0 KB 0 KB 0 KB 4 KB 4 KB 4 KB 0 KB 0 KB /lib/i686/libpthread-2.13.so
b7629000-b762b000 rw-p 8 KB 4 KB 4 KB 0 KB 0 KB 0 KB 4 KB 4 KB 4 KB 0 KB 0 KB
b762b000-b778c000 r-xp 1412 KB 320 KB 4 KB 320 KB 0 KB 0 KB 0 KB 320 KB 0 KB 0 KB 0 KB /lib/i686/libc-2.13.so
b778c000-b778d000 ---p 4 KB 0 KB 0 KB 0 KB 0 KB 0 KB 0 KB 0 KB 0 KB 0 KB 0 KB /lib/i686/libc-2.13.so
b778d000-b778f000 r--p 8 KB 8 KB 1 KB 0 KB 8 KB 0 KB 0 KB 8 KB 8 KB 0 KB 0 KB /lib/i686/libc-2.13.so
b778f000-b7790000 rw-p 4 KB 4 KB 4 KB 0 KB 0 KB 0 KB 4 KB 4 KB 4 KB 0 KB 0 KB /lib/i686/libc-2.13.so
b7790000-b7793000 rw-p 12 KB 12 KB 8 KB 0 KB 4 KB 0 KB 8 KB 12 KB 12 KB 0 KB 0 KB
b7793000-b779a000 r-xp 28 KB 4 KB 0 KB 4 KB 0 KB 0 KB 0 KB 4 KB 0 KB 0 KB 0 KB /lib/i686/librt-2.13.so
b779a000-b779b000 r--p 4 KB 4 KB 0 KB 0 KB 4 KB 0 KB 0 KB 4 KB 4 KB 0 KB 0 KB /lib/i686/librt-2.13.so
b779b000-b779c000 rw-p 4 KB 4 KB 0 KB 0 KB 4 KB 0 KB 0 KB 0 KB 4 KB 0 KB 0 KB /lib/i686/librt-2.13.so
b77be000-b77c8000 r-xp 40 KB 0 KB 0 KB 0 KB 0 KB 0 KB 0 KB 0 KB 0 KB 0 KB 0 KB /lib/libnss_files-2.13.so
b77c8000-b77c9000 r--p 4 KB 4 KB 0 KB 0 KB 4 KB 0 KB 0 KB 0 KB 4 KB 0 KB 0 KB /lib/libnss_files-2.13.so
b77c9000-b77ca000 rw-p 4 KB 4 KB 0 KB 0 KB 4 KB 0 KB 0 KB 0 KB 4 KB 0 KB 0 KB /lib/libnss_files-2.13.so
b77ca000-b77cb000 rw-p 4 KB 4 KB 4 KB 0 KB 0 KB 0 KB 4 KB 4 KB 4 KB 0 KB 0 KB
b77cb000-b77e5000 r-xp 104 KB 28 KB 0 KB 28 KB 0 KB 0 KB 0 KB 28 KB 0 KB 0 KB 0 KB /lib/ld-2.13.so
b77e5000-b77e6000 rw-p 4 KB 4 KB 0 KB 0 KB 4 KB 0 KB 0 KB 4 KB 4 KB 0 KB 0 KB
b77e6000-b77e7000 r--p 4 KB 4 KB 0 KB 0 KB 4 KB 0 KB 0 KB 4 KB 4 KB 0 KB 0 KB /lib/ld-2.13.so
b77e7000-b77e8000 rw-p 4 KB 4 KB 4 KB 0 KB 0 KB 0 KB 4 KB 4 KB 4 KB 0 KB 0 KB /lib/ld-2.13.so
bfd29000-bfd4a000 rw-p 136 KB 48 KB 41 KB 0 KB 8 KB 0 KB 40 KB 44 KB 48 KB 0 KB 0 KB [stack]
ffffe000-fffff000 r-xp 4 KB 0 KB 0 KB 0 KB 0 KB 0 KB 0 KB 0 KB 0 KB 0 KB 0 KB [vdso]
« Last Edit: February 12, 2012, 12:13:01 AM by Software Monkey »

Offline AS

  • Hero Member
  • *****
  • Posts: 4111
  • Have a nice ... night!
Re: Process named udevd in a CPU loop
« Reply #1 on: February 09, 2012, 07:17:44 PM »
Just moved from Linux Mint 11 to PCLOS 2011 a few days ago, and we have noticed that my (dual core) CPU is constantly showing about 50% load; looking the the process manager I can see a process named udevd, is consuming CPU constantly, about 100M in memory and was started with the command line "/sbin/udevd -d".  What is this process, and why is it constantly consuming resources?


udevd is program responsible for monitor "events", usually coming from kernel, upon event it will take appropriate actions depending on the the event, i.e. when you plug-in a USB stick it will drive the creation of the related entry in /dev ...

Usually is not an heavy process, in your case there could be some issue ...

try, from a terminal:
Code: [Select]
udevadm monitor
very probably you will find a lot of messages ... and possibly some hint about what is going wrong ...

AS

Offline Software Monkey

  • Jr. Member
  • **
  • Posts: 21
    • Software Monkey's Web Site
Re: Process named udevd in a CPU loop
« Reply #2 on: February 09, 2012, 07:34:58 PM »
The output from udevadm monitor is a hard loop of:

Code: [Select]
KERNEL[1328840672.590525] change   /devices/pci0000:00/0000:00:0d.0/ide1/1.0/block/hdc (block)
KERNEL[1328840672.601523] change   /devices/pci0000:00/0000:00:0d.0/ide1/1.0/block/hdc (block)
KERNEL[1328840672.613524] change   /devices/pci0000:00/0000:00:0d.0/ide1/1.0/block/hdc (block)
UDEV  [1328840672.621052] change   /devices/pci0000:00/0000:00:0d.0/ide1/1.0/block/hdc (block)
KERNEL[1328840672.653048] change   /devices/pci0000:00/0000:00:0d.0/ide1/1.0/block/hdc (block)
KERNEL[1328840672.671547] change   /devices/pci0000:00/0000:00:0d.0/ide1/1.0/block/hdc (block)
KERNEL[1328840672.683522] change   /devices/pci0000:00/0000:00:0d.0/ide1/1.0/block/hdc (block)
UDEV  [1328840672.695588] change   /devices/pci0000:00/0000:00:0d.0/ide1/1.0/block/hdc (block)

That would seem to be my DVD burner, which is working, although whenever I open the drive it immediately closes by itself:

Code: [Select]
Mount point: /media/cdrom
Device: hdc
Name: SONY DVD RW DW-Q30A
Type: auto
Options: umask=0,users,iocharset=utf8,noauto,ro,exec

Offline AS

  • Hero Member
  • *****
  • Posts: 4111
  • Have a nice ... night!
Re: Process named udevd in a CPU loop
« Reply #3 on: February 09, 2012, 07:43:19 PM »
The output from udevadm monitor is a hard loop of:

Code: [Select]
KERNEL[1328840672.590525] change   /devices/pci0000:00/0000:00:0d.0/ide1/1.0/block/hdc (block)
KERNEL[1328840672.601523] change   /devices/pci0000:00/0000:00:0d.0/ide1/1.0/block/hdc (block)
KERNEL[1328840672.613524] change   /devices/pci0000:00/0000:00:0d.0/ide1/1.0/block/hdc (block)
UDEV  [1328840672.621052] change   /devices/pci0000:00/0000:00:0d.0/ide1/1.0/block/hdc (block)
KERNEL[1328840672.653048] change   /devices/pci0000:00/0000:00:0d.0/ide1/1.0/block/hdc (block)
KERNEL[1328840672.671547] change   /devices/pci0000:00/0000:00:0d.0/ide1/1.0/block/hdc (block)
KERNEL[1328840672.683522] change   /devices/pci0000:00/0000:00:0d.0/ide1/1.0/block/hdc (block)
UDEV  [1328840672.695588] change   /devices/pci0000:00/0000:00:0d.0/ide1/1.0/block/hdc (block)

That would seem to be my DVD burner, which is working, although whenever I open the drive it immediately closes by itself:

Code: [Select]
Mount point: /media/cdrom
Device: hdc
Name: SONY DVD RW DW-Q30A
Type: auto
Options: umask=0,users,iocharset=utf8,noauto,ro,exec


If fact your DVD (or may be the related driver... doesn't change much) it is sending 'change' messages like if the media was changed ...

Try the solution posted here and report back: (you will need to reboot your system)
http://www.pclinuxos.com/forum/index.php/topic,100800.0.html


Online Just17

  • PCLinuxOS Tester
  • Super Villain
  • *******
  • Posts: 10606
  • MLUs Forever!
Re: Process named udevd in a CPU loop
« Reply #4 on: February 09, 2012, 08:58:29 PM »
I will add a little here .......  I have had this problem for some time, or one similar to it.
I have two optical drives in this PC.
One is a DVD Multiplayer (/dev/sr0) while the other is a Superwritemaster burner (not 'seen')
Both are connected by IDE cable.

According to PCC Hardware

/dev/sr0 is using pata_amd module. This drive used to close immediately it was opened ......  some update seems to have corrected that carry on.

The second drive (burner) does not register at all apparently. (I have been using an USB connected optical drive for a while, and just never got around to looking at this problem.)

It is 4 am so will catch up on this tomorrow .....  when I can do some tests etc.

EDIT:
           Probably should not have posted at all.
On checking this am it seems that the second optical drive which is not seen at all is a SATA drive ......  and the reason it ain't seen now is that the SATA cable has bee disconnected!

It sure looks as if that was my "temporary" 'fix' until I got around to looking at this again.
Without the external burner that might have been a long time ago  :D

 
« Last Edit: February 10, 2012, 05:55:02 AM by Just18 »
MLUs rule the roost!

Linux XPS 3.2.18-pclos2.pae.bfs  32 bit
Intel Core2 Quad CPU Q9450 @ 2.66GHz
4 GB RAM
MCP51 High Def Audio
GeForce GTX 550 Ti
PHILIPS  ‎DVD+-RW DVD8701
‎Logitech ‎BT Mini-Receiver
Afatech DTT

Offline AS

  • Hero Member
  • *****
  • Posts: 4111
  • Have a nice ... night!
Re: Process named udevd in a CPU loop
« Reply #5 on: February 10, 2012, 06:03:00 AM »
Software Monkey,

please, post the content of the file:  /etc/udev/rules.d/70-persistent-cd.rules

Offline Software Monkey

  • Jr. Member
  • **
  • Posts: 21
    • Software Monkey's Web Site
Re: Process named udevd in a CPU loop
« Reply #6 on: February 10, 2012, 05:19:23 PM »
AS:

There is no rules file for the CD of the name you specified:

Code: [Select]
[root@localhost rules.d]# ls
10-blackberry.rules          45-vmc-novatel.rules                   60-libmtp.rules          69-printers_lp_user_fix.rules  97-bluetooth-serial.rules
10-vboxdrv.rules             45-vmc-option.rules                    60-libsane.rules         70-gpsd.rules                  alsa.rules
40-hplip.rules               45-vmc-zte.rules                       60-pcmcia.rules          70-persistent-net.rules        kino.rules
41-odvr.rules                49_hso-udev.rules                      60-raw.rules             90-tpb.rules
42-midisport-firmware.rules  55-alsa-tascam-firmware-loaders.rules  61-x11-input.rules       91-drm-modeset.rules
45-hpdjconsole.rules         56-hpmud_support.rules                 67-xorg-wizardpen.rules  97-bluetooth-hid2hci.rules
45-vmc-huawei.rules          60-dynamic.rules                       69-floppy-acl.rules      97-bluetooth.rules

Offline djohnston

  • PCLinuxOS Tester
  • Hero Member
  • *******
  • Posts: 6227
  • I don't do Windows
Re: Process named udevd in a CPU loop
« Reply #7 on: February 10, 2012, 05:24:50 PM »
AS:

There is no rules file for the CD of the name you specified:

There should be. It should look something like the following.

Code: [Select]
# This file was automatically generated by the /lib/udev/write_cd_rules
# program, run by the cd-aliases-generator.rules rules file.
#
# You can modify it, as long as you keep each rule on a single
# line, and set the $GENERATED variable.

# ATAPI_DVD_A_DH20A4P (pci-0000:00:14.1-scsi-0:0:1:0)
SUBSYSTEM=="block", ENV{ID_CDROM}=="?*", ENV{ID_PATH}=="pci-0000:00:14.1-scsi-0:0:1:0", SYMLINK+="cdrom", ENV{GENERATED}="1"
SUBSYSTEM=="block", ENV{ID_CDROM}=="?*", ENV{ID_PATH}=="pci-0000:00:14.1-scsi-0:0:1:0", SYMLINK+="cdrw", ENV{GENERATED}="1"
SUBSYSTEM=="block", ENV{ID_CDROM}=="?*", ENV{ID_PATH}=="pci-0000:00:14.1-scsi-0:0:1:0", SYMLINK+="dvd", ENV{GENERATED}="1"
SUBSYSTEM=="block", ENV{ID_CDROM}=="?*", ENV{ID_PATH}=="pci-0000:00:14.1-scsi-0:0:1:0", SYMLINK+="dvdrw", ENV{GENERATED}="1"
Bare metal                           VBox
AMD Athlon 7750 Dual-Core    Single core
4GiB RAM                              1GiB RAM
nVidia GeForce FX 5200          64MB video
LXDE 32bit                            KDE 64bit

Registered Linux User #416378

Offline AS

  • Hero Member
  • *****
  • Posts: 4111
  • Have a nice ... night!
Re: Process named udevd in a CPU loop
« Reply #8 on: February 10, 2012, 05:39:21 PM »
AS:

There is no rules file for the CD of the name you specified:

Code: [Select]
[root@localhost rules.d]# ls
10-blackberry.rules          45-vmc-novatel.rules                   60-libmtp.rules          69-printers_lp_user_fix.rules  97-bluetooth-serial.rules
10-vboxdrv.rules             45-vmc-option.rules                    60-libsane.rules         70-gpsd.rules                  alsa.rules
40-hplip.rules               45-vmc-zte.rules                       60-pcmcia.rules          70-persistent-net.rules        kino.rules
41-odvr.rules                49_hso-udev.rules                      60-raw.rules             90-tpb.rules
42-midisport-firmware.rules  55-alsa-tascam-firmware-loaders.rules  61-x11-input.rules       91-drm-modeset.rules
45-hpdjconsole.rules         56-hpmud_support.rules                 67-xorg-wizardpen.rules  97-bluetooth-hid2hci.rules
45-vmc-huawei.rules          60-dynamic.rules                       69-floppy-acl.rules      97-bluetooth.rules


Yes, I see. And I think you have just found a bug ... where udev actually fail to recognize ide CD/DVD ...

described here:
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=635107

on my system (and probably on yours too) the content of the file /lib/udev/rules.d/60-cdrom_id.rules is:
Quote
# do not edit this file, it will be overwritten on update

ACTION=="remove", GOTO="cdrom_end"
SUBSYSTEM!="block", GOTO="cdrom_end"
KERNEL!="sr[0-9]*|xvd*", GOTO="cdrom_end"
ENV{DEVTYPE}!="disk", GOTO="cdrom_end"

KERNEL=="sr[0-9]*", ENV{ID_CDROM}="1"
IMPORT{program}="cdrom_id --export $tempnode"

LABEL="cdrom_end"


following the fix suggested in the thread above, should be:
Quote
# do not edit this file, it will be overwritten on update

ACTION=="remove", GOTO="cdrom_end"
SUBSYSTEM!="block", GOTO="cdrom_end"
KERNEL!="sr[0-9]*|xvd*", GOTO="cdrom_end"
ENV{DEVTYPE}!="disk", GOTO="cdrom_end"

KERNEL=="sr[0-9]*", ENV{ID_CDROM}="1"
IMPORT{program}="cdrom_id --export $tempnode"

KERNEL=="hd[a-z]*", ENV{ID_CDROM}="1"
IMPORT{program}="cdrom_id --export $tempnode"


LABEL="cdrom_end"


If you want to try to apply the modif, we will be glad to know the result ... ide drives are quite rare lately ...
(don't forget to reboot after you apply the changes)
« Last Edit: February 10, 2012, 05:51:50 PM by AS »

Offline Software Monkey

  • Jr. Member
  • **
  • Posts: 21
    • Software Monkey's Web Site
Re: Process named udevd in a CPU loop
« Reply #9 on: February 10, 2012, 06:01:49 PM »
AS & djohnston:

There are two sets of rules each of you has mentioned in two different directories; I am uncertain as to what changes if any to make.

Here is the contents of both directories, and the contents of the file mentioned by AS:

Code: [Select]
[root@localhost rules.d]# ls /lib/udev/rules.d/
10-console.rules          60-persistent-input.rules         70-hid2hci.rules                   85-regulatory.rules
10-dm.rules               60-persistent-serial.rules        70-hotplug_map.rules               85-usbmuxd.rules
11-dm-lvm.rules           60-persistent-storage.rules       70-infrared.rules*                 88-clock.rules
13-dm-disk.rules          60-persistent-storage-tape.rules  70-printers.rules                  90-alsa-restore.rules
40-libgphoto2.rules       60-persistent-v4l.rules           75-cd-aliases-generator.rules      90-hal.rules
40-usb_modeswitch.rules   61-mobile-action.rules            75-net-description.rules           90-libgpod.rules
42-qemu-usb.rules         61-mobile-zte-drakx-net.rules     75-persistent-net-generator.rules  92-pcscd-ccid.rules
50-firmware.rules         61-persistent-storage-edd.rules   75-probe_mtd.rules                 95-dm-notify.rules
50-udev-default.rules     64-md-raid.rules                  75-tty-description.rules           95-keyboard-force-release.rules
51-udev-pclos.rules       65-xorg-wacom.rules               76-net.rules                       95-keymap.rules
60-cdrom_id.rules         69-printeracl.rules               78-sound-card.rules                95-udev-late.rules
60-floppy.rules           69-xorg-vmmouse.rules             80-drivers.rules
60-persistent-alsa.rules  70-acl.rules                      80-udisks.rules
[root@localhost rules.d]# ls /etc/udev/rules.d/
10-blackberry.rules          45-vmc-novatel.rules                   60-libmtp.rules          69-printers_lp_user_fix.rules  97-bluetooth-serial.rules
10-vboxdrv.rules             45-vmc-option.rules                    60-libsane.rules         70-gpsd.rules                  alsa.rules
40-hplip.rules               45-vmc-zte.rules                       60-pcmcia.rules          70-persistent-net.rules        kino.rules
41-odvr.rules                49_hso-udev.rules                      60-raw.rules             90-tpb.rules
42-midisport-firmware.rules  55-alsa-tascam-firmware-loaders.rules  61-x11-input.rules       91-drm-modeset.rules
45-hpdjconsole.rules         56-hpmud_support.rules                 67-xorg-wizardpen.rules  97-bluetooth-hid2hci.rules
45-vmc-huawei.rules          60-dynamic.rules                       69-floppy-acl.rules      97-bluetooth.rules
[root@localhost rules.d]# cat /lib/udev/rules.d/60-cdrom_id.rules
# do not edit this file, it will be overwritten on update

ACTION=="remove", GOTO="cdrom_end"
SUBSYSTEM!="block", GOTO="cdrom_end"
KERNEL!="sr[0-9]*|xvd*", GOTO="cdrom_end"
ENV{DEVTYPE}!="disk", GOTO="cdrom_end"

KERNEL=="sr[0-9]*", ENV{ID_CDROM}="1"
IMPORT{program}="cdrom_id --export $tempnode"

LABEL="cdrom_end"

BTW, I do see devices for the DVD writer in /dev:

Code: [Select]
[root@localhost rules.d]# ls /dev
adsp             fb@   hpet        loop7               null       ram11   rawctl      shm/      tty13  tty26  tty39  tty51  tty7        vcs12   vcsa5
audio            fb0   hugepages/  lp0                 nvidia0    ram12   root        snapshot  tty14  tty27  tty4   tty52  tty8        vcs2    vcsa6
block/           fd@   initctl|    lp1                 nvidiactl  ram13   rtc@        snd/      tty15  tty28  tty40  tty53  tty9        vcs3    vcsa7
bsg/             fd0   input/      lp2                 oldmem     ram14   rtc0        st0       tty16  tty29  tty41  tty54  ttyS0       vcs4    vcsa8
bus/             fd1   kmem        lp3                 parport0   ram15   sda         st1       tty17  tty3   tty42  tty55  ttyS1       vcs5    vga_arbiter
cdrom@           full  kmsg        mapper/             port       ram2    sdb         stderr@   tty18  tty30  tty43  tty56  ttyS2       vcs6    video0
char/            fuse  log=        mcelog              ppp        ram3    sdc         stdin@    tty19  tty31  tty44  tty57  ttyS3       vcs7    watchdog
console          hda   loop0       mem                 psaux      ram4    sdd         stdout@   tty2   tty32  tty45  tty58  urandom     vcs8    zero
core@            hda1  loop1       mice                ptmx       ram5    sequencer   tty       tty20  tty33  tty46  tty59  v4l/        vcsa
cpu/             hda2  loop2       mixer               pts/       ram6    sequencer2  tty0      tty21  tty34  tty47  tty6   vboxdrv     vcsa1
cpu_dma_latency  hda5  loop3       mouse0              ram@       ram7    sg0         tty1      tty22  tty35  tty48  tty60  vboxnetctl  vcsa12
disk/            hdb   loop4       net/                ram0       ram8    sg1         tty10     tty23  tty36  tty49  tty61  vboxusb/    vcsa2
dsp              hdb1  loop5       network_latency     ram1       ram9    sg2         tty11     tty24  tty37  tty5   tty62  vcs         vcsa3
dvd@             hdc   loop6       network_throughput  ram10      random  sg3         tty12     tty25  tty38  tty50  tty63  vcs1        vcsa4

But they are flashing reverse-image red.
« Last Edit: February 10, 2012, 06:06:07 PM by Software Monkey »

Offline AS

  • Hero Member
  • *****
  • Posts: 4111
  • Have a nice ... night!
Re: Process named udevd in a CPU loop
« Reply #10 on: February 10, 2012, 06:11:48 PM »
AS & djohnston:

There are two sets of rules each of you has mentioned in two different directories; I am uncertain as to what changes if any to make.

the changes should be applied in /lib/udev/rules.d/60-cdrom_id.rules
as a result you should find a new file in /etc/udev/rules.d ...

Quote
BTW, I do see devices for the DVD writer in /dev:
...
But they are flashing reverse-image red.

please post the output of:
Code: [Select]
ls -l /dev/hd*

Offline Software Monkey

  • Jr. Member
  • **
  • Posts: 21
    • Software Monkey's Web Site
Re: Process named udevd in a CPU loop
« Reply #11 on: February 10, 2012, 06:15:04 PM »
AS:

Here's the output you requested, plus sd* (and the entire /dev was in my earlier post):

Code: [Select]
[root@localhost rules.d]# ls -l /dev/hd*
brw-rw---- 1 root disk  3,  0 Feb 10 10:48 /dev/hda
brw-rw---- 1 root disk  3,  1 Feb 10 10:48 /dev/hda1
brw-rw---- 1 root disk  3,  2 Feb 10 10:48 /dev/hda2
brw-rw---- 1 root disk  3,  5 Feb 10 10:48 /dev/hda5
brw-rw---- 1 root disk  3, 64 Feb 10 10:48 /dev/hdb
brw-rw---- 1 root disk  3, 65 Feb 10 10:48 /dev/hdb1
brw-rw---- 1 root disk 22,  0 Feb 10 17:16 /dev/hdc
Code: [Select]
[root@localhost rules.d]# ls -l /dev/sd*
brw-rw---- 1 root disk 8,  0 Feb 10 10:48 /dev/sda
brw-rw---- 1 root disk 8, 16 Feb 10 10:48 /dev/sdb
brw-rw---- 1 root disk 8, 32 Feb 10 10:48 /dev/sdc
brw-rw---- 1 root disk 8, 48 Feb 10 10:48 /dev/sdd
« Last Edit: February 10, 2012, 06:17:16 PM by Software Monkey »

Offline AS

  • Hero Member
  • *****
  • Posts: 4111
  • Have a nice ... night!
Re: Process named udevd in a CPU loop
« Reply #12 on: February 10, 2012, 06:23:19 PM »
AS:

Here's the output you requested, plus sd* (and the entire /dev was in my earlier post):


Look fine ... did you applied the suggested changes ?

Offline Software Monkey

  • Jr. Member
  • **
  • Posts: 21
    • Software Monkey's Web Site
Re: Process named udevd in a CPU loop
« Reply #13 on: February 10, 2012, 06:56:59 PM »
AS:

Yes, I added the changes and rebooted:

lib/udev... is now:
Code: [Select]
[root@localhost ~]# ls /lib/udev/rules.d/
10-console.rules          60-persistent-input.rules         70-hid2hci.rules                   85-regulatory.rules
10-dm.rules               60-persistent-serial.rules        70-hotplug_map.rules               85-usbmuxd.rules
11-dm-lvm.rules           60-persistent-storage.rules       70-infrared.rules*                 88-clock.rules
13-dm-disk.rules          60-persistent-storage-tape.rules  70-printers.rules                  90-alsa-restore.rules
40-libgphoto2.rules       60-persistent-v4l.rules           75-cd-aliases-generator.rules      90-hal.rules
40-usb_modeswitch.rules   61-mobile-action.rules            75-net-description.rules           90-libgpod.rules
42-qemu-usb.rules         61-mobile-zte-drakx-net.rules     75-persistent-net-generator.rules  92-pcscd-ccid.rules
50-firmware.rules         61-persistent-storage-edd.rules   75-probe_mtd.rules                 95-dm-notify.rules
50-udev-default.rules     64-md-raid.rules                  75-tty-description.rules           95-keyboard-force-release.rules
51-udev-pclos.rules       65-xorg-wacom.rules               76-net.rules                       95-keymap.rules
60-cdrom_id.rules         69-printeracl.rules               78-sound-card.rules                95-udev-late.rules
60-floppy.rules           69-xorg-vmmouse.rules             80-drivers.rules
60-persistent-alsa.rules  70-acl.rules                      80-udisks.rules

etc/udev... is now:
Code: [Select]
[root@localhost ~]# ls /etc/udev/rules.d/
10-blackberry.rules          45-vmc-novatel.rules                   60-libmtp.rules          69-printers_lp_user_fix.rules  97-bluetooth-serial.rules
10-vboxdrv.rules             45-vmc-option.rules                    60-libsane.rules         70-gpsd.rules                  alsa.rules
40-hplip.rules               45-vmc-zte.rules                       60-pcmcia.rules          70-persistent-net.rules        kino.rules
41-odvr.rules                49_hso-udev.rules                      60-raw.rules             90-tpb.rules
42-midisport-firmware.rules  55-alsa-tascam-firmware-loaders.rules  61-x11-input.rules       91-drm-modeset.rules
45-hpdjconsole.rules         56-hpmud_support.rules                 67-xorg-wizardpen.rules  97-bluetooth-hid2hci.rules
45-vmc-huawei.rules          60-dynamic.rules                       69-floppy-acl.rules      97-bluetooth.rules

and the contents of 60-cdrom_id.rules is now:
Code: [Select]
[root@localhost ~]# cat /lib/udev/rules.d/60-cdrom_id.rules
# do not edit this file, it will be overwritten on update

ACTION=="remove", GOTO="cdrom_end"
SUBSYSTEM!="block", GOTO="cdrom_end"
KERNEL!="sr[0-9]*|xvd*", GOTO="cdrom_end"
ENV{DEVTYPE}!="disk", GOTO="cdrom_end"

# this is only a button press event
ENV{DISK_EJECT_REQUEST}=="?*", GOTO="cdrom_end"

KERNEL=="sr[0-9]*", ENV{ID_CDROM}="1"
IMPORT{program}="cdrom_id --export $tempnode"

KERNEL=="hd[a-z]*", ENV{ID_CDROM}="1"
IMPORT{program}="cdrom_id $tempnode"

LABEL="cdrom_end"

EDIT: If I simply disconnect the DVD writer's power the notifications immediately stop, so it's definitely the DVD drive causing them. Almost like the kernel notification causes udevd to do something which immediately trips more kernel notifications, in a loop.
« Last Edit: February 10, 2012, 07:11:56 PM by Software Monkey »

Offline AS

  • Hero Member
  • *****
  • Posts: 4111
  • Have a nice ... night!
Re: Process named udevd in a CPU loop
« Reply #14 on: February 10, 2012, 07:13:04 PM »
EDIT: If I simply disconnect the DVD writer's power the notifications immediately stop, so it's definitely the DVD drive causing them. Almost like the kernel notification cause udevd to do something which immediately trips more kernel notifications, in a loop.


Of course ... I'm perplexed because you reported (it seems ...) it was working on Mint ...
Do not forget to revert back the changes experimented so far ... out of ideas right now ...  :(