Communitygate: Kernel Oops - KNX-User-Forum

henfri · #1 21.12.2012, 21:26

Hallo,

mein CG lief einige Zeit lang wunderbar in einer VM. Der Host war ein Ubuntu 10.04.
Jetzt habe ich auf 12.04 aktualisiert und die Konfiguration der VM lief so erstmal nicht mehr (ich erinnere mich nicht genau, aber die Syntax war nicht mehr ok so). Ich habe daraufhin eine neue Konfiguration (libvirt) erstellt, mit der das CG dann startet und auch lange wunderbar läuft.

Nach einiger Zeit gibt es dann aber Probleme:
BUG: unable to handle kernel paging request at a5000000

Code:

Message from syslogd@wiregate at Dec 21 21:21:25 ...
 kernel:[299416.943949] Oops: 0000 [#56] SMP

Message from syslogd@wiregate at Dec 21 21:21:25 ...
 kernel:[299416.944025] last sysfs file: /sys/devices/pci0000:00/0000:00:01.3/i2c-0/name

Message from syslogd@wiregate at Dec 21 21:21:25 ...
 kernel:[299416.944025] Process uptime (pid: 2303, ti=c001a000 task=c0eb1dc0 task.ti=c001a000)

Message from syslogd@wiregate at Dec 21 21:21:25 ...
 kernel:[299416.944025] Stack:

Message from syslogd@wiregate at Dec 21 21:21:25 ...
 kernel:[299416.944025] Call Trace:

Message from syslogd@wiregate at Dec 21 21:21:25 ...
 kernel:[299416.944025] Code: 9c 58 8d 74 26 00 89 c7 fa 90 8d 74 26 00 64 a1 54 79 41 c1 8b 94 85 84 00 00 00 8b 42 10 89 44 24 10 8b 1a 85 db 74 0a 8b 42 0c <8b> 04 83 89 02 eb 15 52 83 c9 ff ff 74 24 0c 89 f2 89 e8 e8 3d

Message from syslogd@wiregate at Dec 21 21:21:25 ...
 kernel:[299416.944025] EIP: [<c10aed2c>] kmem_cache_alloc+0x55/0xe5 SS:ESP 0068:c001be40

Message from syslogd@wiregate at Dec 21 21:21:25 ...
 kernel:[299416.944025] CR2: 00000000a5000000

Hat jemand eine Idee, woran das liegen kann?

Unten noch die Config

Gruß,
Hendrik

Code:

cat /var/lib/zentyal/machines/WG/domain.xml
<domain type='kvm'>
  <name>WG</name>
  <uuid>d4fd6ae5-31cb-a7fb-73a4-2ea911cf7017</uuid>
  <memory>262144</memory>

  <vcpu>1</vcpu>
  <os>
    <type arch='i686' machine='pc'>hvm</type>
    <boot dev='hd'/>
    <bootmenu enable='yes'/>
  </os>
  <features><acpi/></features>
  <on_poweroff>destroy</on_poweroff>
  <on_reboot>restart</on_reboot>
  <on_crash>restart</on_crash>
  <devices>
    <emulator>/usr/bin/kvm</emulator>
    <interface type='bridge'>
      <source bridge='br1'/>
      <mac address='52:54:00:2b:85:3d'/>
      <target dev='vnet0'/>
      <model type='virtio'/>
      <alias name='net0'/>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/>
    </interface>
    <disk type='file' device='disk'>
      <driver name='qemu' type='qcow2'/>
      <target dev='sda' bus='sata'/>
      <source file='/var/lib/zentyal/machines/WG/WG-Test.qcow2'/>

      <alias name='sata0-0-0'/>
      <address type='drive' controller='0' bus='0' unit='0'/>
    </disk>
    <controller type='sata' index='0'>
      <alias name='sata0'/>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x05' function='0x0'/>
    </controller>

    <console type='pty' tty='/dev/pts/12'>
      <source path='/dev/pts/12'/>
      <target type='serial' port='0'/>
      <alias name='serial0'/>
    </console>




    <hostdev mode='subsystem' type='usb'>
      <source>
        <address bus='0x1' device='0x4'/>
        </source>

    </hostdev>
    <hostdev mode='subsystem' type='usb'>
      <source>
         <address bus='0x1' device='0x5'/>
     </source>

    </hostdev>

    <video>
      <model type='cirrus' vram='9216' heads='1'/>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x02' function='0x0'/>
    </video>
         <graphics type='vnc' listen='0.0.0.0' port='5900' passwd='till1234' keymap='de'/>
  </devices>

</domain>

Und der Beginn des Troubles:

Code:

Dec 21 20:36:25 wiregate kernel: [296716.685615] BUG: unable to handle kernel paging request at a5000000
Dec 21 20:36:25 wiregate kernel: [296716.685890] IP: [<c10aed2c>] kmem_cache_alloc+0x55/0xe5
Dec 21 20:36:25 wiregate kernel: [296716.686107] *pde = 00000000
Dec 21 20:36:25 wiregate kernel: [296716.686454] Oops: 0000 [#1] SMP
Dec 21 20:36:25 wiregate kernel: [296716.686861] last sysfs file: /sys/devices/pci0000:00/0000:00:01.3/i2c-0/name
Dec 21 20:36:25 wiregate kernel: [296716.687327] Modules linked in: loop snd_pcm snd_timer snd soundcore snd_page_alloc i2c_piix4 pcspkr i2c_core processor psmouse butt
on evdev virtio_balloon serio_raw ext3 jbd mbcache dm_mod sd_mod crc_t10dif ata_generic ata_piix ahci virtio_net thermal uhci_hcd ehci_hcd floppy thermal_sys libata vir
tio_pci usbcore nls_base scsi_mod [last unloaded: scsi_wait_scan]
Dec 21 20:36:25 wiregate kernel: [296716.688019]
Dec 21 20:36:25 wiregate kernel: [296716.688019] Pid: 11779, comm: sed Not tainted (2.6.32-5-686 #1) Bochs
Dec 21 20:36:25 wiregate kernel: [296716.688019] EIP: 0060:[<c10aed2c>] EFLAGS: 00210086 CPU: 0
Dec 21 20:36:25 wiregate kernel: [296716.688019] EIP is at kmem_cache_alloc+0x55/0xe5
Dec 21 20:36:25 wiregate kernel: [296716.688019] EAX: 00000000 EBX: a5000000 ECX: 00000000 EDX: c1804f9c
Dec 21 20:36:25 wiregate kernel: [296716.688019] ESI: 000000d0 EDI: 00200246 EBP: c1387ed0 ESP: cbaede40
Dec 21 20:36:25 wiregate kernel: [296716.688019]  DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068
Dec 21 20:36:25 wiregate kernel: [296716.688019] Process sed (pid: 11779, ti=cbaec000 task=cbb59980 task.ti=cbaec000)
Dec 21 20:36:25 wiregate kernel: [296716.688019] Stack:
Dec 21 20:36:25 wiregate kernel: [296716.688019]  2c9c4a2d 000000d0 c10c7862 c1f72aa0 00000010 00000000 ce253a80 cbbe05a0
Dec 21 20:36:25 wiregate kernel: [296716.688019] <0> cb847880 c10c7862 00000000 c10c43ea 00000000 ce253a80 cbbe05a0 ce253ac8
Dec 21 20:36:25 wiregate kernel: [296716.688019] <0> c10e64dc cb847880 c1c5d018 c10c6d86 00000000 cb847880 00000000 c1c5d018
Dec 21 20:36:25 wiregate kernel: [296716.688019] Call Trace:
Dec 21 20:36:25 wiregate kernel: [296716.688019]  [<c10c7862>] ? single_open+0x1f/0xa9
Dec 21 20:36:25 wiregate kernel: [296716.688019]  [<c10c7862>] ? single_open+0x1f/0xa9
Dec 21 20:36:25 wiregate kernel: [296716.688019]  [<c10c43ea>] ? filesystems_proc_show+0x0/0x51
Dec 21 20:36:25 wiregate kernel: [296716.688019]  [<c10e64dc>] ? proc_reg_open+0xac/0x117
Dec 21 20:36:25 wiregate kernel: [296716.688019]  [<c10c6d86>] ? single_release+0x0/0x1b
Dec 21 20:36:25 wiregate kernel: [296716.688019]  [<c10b2303>] ? __dentry_open+0x156/0x246
Dec 21 20:36:25 wiregate kernel: [296716.688019]  [<c10b2484>] ? nameidata_to_filp+0x29/0x3c
Dec 21 20:36:25 wiregate kernel: [296716.688019]  [<c10e6430>] ? proc_reg_open+0x0/0x117
Dec 21 20:36:25 wiregate kernel: [296716.688019]  [<c10bc51c>] ? do_filp_open+0x43f/0x802
Dec 21 20:36:25 wiregate kernel: [296716.688019]  [<c109c076>] ? handle_mm_fault+0x48f/0x959
Dec 21 20:36:25 wiregate kernel: [296716.688019]  [<c10c4241>] ? alloc_fd+0x52/0xb7
Dec 21 20:36:25 wiregate kernel: [296716.688019]  [<c10b20bb>] ? do_sys_open+0x4c/0xdf
Dec 21 20:36:25 wiregate kernel: [296716.688019]  [<c10b2192>] ? sys_open+0x1e/0x23
Dec 21 20:36:25 wiregate kernel: [296716.688019]  [<c10030fb>] ? sysenter_do_call+0x12/0x28
Dec 21 20:36:25 wiregate kernel: [296716.688019] Code: 9c 58 8d 74 26 00 89 c7 fa 90 8d 74 26 00 64 a1 54 79 41 c1 8b 94 85 84 00 00 00 8b 42 10 89 44 24 10 8b 1a 85 db
 74 0a 8b 42 0c <8b> 04 83 89 02 eb 15 52 83 c9 ff ff 74 24 0c 89 f2 89 e8 e8 3d
Dec 21 20:36:25 wiregate kernel: [296716.688019] EIP: [<c10aed2c>] kmem_cache_alloc+0x55/0xe5 SS:ESP 0068:cbaede40
Dec 21 20:36:25 wiregate kernel: [296716.688019] CR2: 00000000a5000000
Dec 21 20:36:25 wiregate kernel: [296716.688019] ---[ end trace 16192d79aa82026a ]---
Dec 21 20:36:25 wiregate kernel: [296716.703117] BUG: unable to handle kernel paging request at a5000000
Dec 21 20:36:25 wiregate kernel: [296716.703551] IP: [<c10af472>] __kmalloc+0xb9/0x128
Dec 21 20:36:25 wiregate kernel: [296716.703909] *pde = 00000000
Dec 21 20:36:25 wiregate kernel: [296716.704219] Oops: 0000 [#2] SMP
Dec 21 20:36:25 wiregate kernel: [296716.704590] last sysfs file: /sys/devices/pci0000:00/0000:00:01.3/i2c-0/name
Dec 21 20:36:25 wiregate kernel: [296716.704960] Modules linked in: loop snd_pcm snd_timer snd soundcore snd_page_alloc i2c_piix4 pcspkr i2c_core processor psmouse butt
on evdev virtio_balloon serio_raw ext3 jbd mbcache dm_mod sd_mod crc_t10dif ata_generic ata_piix ahci virtio_net thermal uhci_hcd ehci_hcd floppy thermal_sys libata vir
tio_pci usbcore nls_base scsi_mod [last unloaded: scsi_wait_scan]
Dec 21 20:36:25 wiregate kernel: [296716.707028]
Dec 21 20:36:25 wiregate kernel: [296716.707028] Pid: 11775, comm: owserver Tainted: G      D    (2.6.32-5-686 #1) Bochs
Dec 21 20:36:25 wiregate kernel: [296716.707028] EIP: 0060:[<c10af472>] EFLAGS: 00210086 CPU: 0
Dec 21 20:36:25 wiregate kernel: [296716.707028] EIP is at __kmalloc+0xb9/0x128
Dec 21 20:36:25 wiregate kernel: [296716.707028] EAX: 00000000 EBX: a5000000 ECX: 00000010 EDX: c1804f9c
Dec 21 20:36:25 wiregate kernel: [296716.707028] ESI: 000000d0 EDI: 00200246 EBP: c1387ed0 ESP: cc0fdd4c
Dec 21 20:36:25 wiregate kernel: [296716.707028]  DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068
Dec 21 20:36:25 wiregate kernel: [296716.707028] Process owserver (pid: 11775, ti=cc0fc000 task=cbb5a640 task.ti=cc0fc000)
Dec 21 20:36:25 wiregate kernel: [296716.707028] Stack:
Dec 21 20:36:25 wiregate kernel: [296716.707028]  00000074 000000d0 00000009 d08ba700 00000010 cb847180 cc0fdde0 00000000
Dec 21 20:36:25 wiregate kernel: [296716.707028] <0> cc18f4c0 d08ba700 cc0fdde0 cb9a1420 00000000 cc2a36ec 00000000 cc195f00
Dec 21 20:36:25 wiregate kernel: [296716.707028] <0> 00000000 00000000 cc0fdde0 cc291800 b5d51cf8 ffffffff d08bb7de b5d51cf8
Dec 21 20:36:25 wiregate kernel: [296716.707028] Call Trace:
Dec 21 20:36:25 wiregate kernel: [296716.707028]  [<d08ba700>] ? proc_do_submiturb+0x3e9/0x798 [usbcore]
Dec 21 20:36:25 wiregate kernel: [296716.707028]  [<d08ba700>] ? proc_do_submiturb+0x3e9/0x798 [usbcore]
Dec 21 20:36:25 wiregate kernel: [296716.707028]  [<d08bb7de>] ? usbdev_ioctl+0xae1/0x111e [usbcore]
Dec 21 20:36:25 wiregate kernel: [296716.707028]  [<c1051912>] ? do_futex+0xcf/0xae7
Dec 21 20:36:25 wiregate kernel: [296716.707028]  [<c11fc0d2>] ? ip_finish_output2+0x192/0x1cd
Dec 21 20:36:25 wiregate kernel: [296716.707028]  [<c108b43d>] ? zone_watermark_ok+0x16/0x99
Dec 21 20:36:25 wiregate kernel: [296716.707028]  [<c10250c0>] ? update_curr+0x106/0x1b3
Dec 21 20:36:25 wiregate kernel: [296716.707028]  [<c10beec4>] ? set_fd_set+0x18/0x2d
Dec 21 20:36:25 wiregate kernel: [296716.707028]  [<c10bfa89>] ? core_sys_select+0x19d/0x1c2
Dec 21 20:36:25 wiregate kernel: [296716.707028]  [<c102b862>] ? finish_task_switch+0x34/0x95
Dec 21 20:36:25 wiregate kernel: [296716.707028]  [<c126dfa6>] ? schedule+0x78f/0x7dc
Dec 21 20:36:25 wiregate kernel: [296716.707028]  [<c10471c5>] ? __hrtimer_start_range_ns+0x24c/0x258
Dec 21 20:36:25 wiregate kernel: [296716.707028]  [<c101b29b>] ? pvclock_clocksource_read+0xf9/0x10f
Dec 21 20:36:25 wiregate kernel: [296716.707028]  [<c10be1dd>] ? vfs_ioctl+0x49/0x5f
Dec 21 20:36:25 wiregate kernel: [296716.707028]  [<c10be744>] ? do_vfs_ioctl+0x4aa/0x4e5
Dec 21 20:36:25 wiregate kernel: [296716.707028]  [<c104b5c9>] ? getnstimeofday+0x4d/0xc9
Dec 21 20:36:25 wiregate kernel: [296716.707028]  [<c104b67f>] ? do_gettimeofday+0xf/0x2c
Dec 21 20:36:25 wiregate kernel: [296716.707028]  [<c10be7c0>] ? sys_ioctl+0x41/0x58
Dec 21 20:36:25 wiregate kernel: [296716.707028]  [<c10030fb>] ? sysenter_do_call+0x12/0x28
Dec 21 20:36:25 wiregate kernel: [296716.707028] Code: 9c 58 8d 74 26 00 89 c7 fa 90 8d 74 26 00 64 a1 54 79 41 c1 8b 94 85 84 00 00 00 8b 4a 10 89 4c 24 10 8b 1a 85 db
 74 0a 8b 42 0c <8b> 04 83 89 02 eb 15 52 83 c9 ff ff 74 24 10 89 f2 89 e8 e8 f7

makki · #2 22.12.2012, 15:32

Schätzung aus der hohlen Hüfte: Speicherleck bzw. speicher ausgegangen..

Makki

henfri · #3 23.12.2012, 11:00

Danke für den Tipp.
Ich hab jetzt mal ein
@hourly date >> /var/log/freemem.txt && free -m |head -n 2 |tail -n1 >> /var/log/freemem.txt
Im Crontab.

Mal schauen.

Du hattest mir mal den Tipp gegeben, weniger RRDs pro Plugin zu speichern. Das habe ich noch nicht umgesetzt. Daran kann es aber nicht liegen?!

Gruß&frohes Fest,
Hendrik

P.S: Ist ein Communitygate, aber die Sensorik wird -ohne Preisvergleich- bei euch bestellt. So gerade passiert ;-)

henfri · #4 26.12.2012, 08:55

Hallo,

gestern hatte ich wieder einen. Der Speicher sah da so aus:

Di 25. Dez 16:00:01 CET 2012
total used free shared buffers cached
Mem: 248 230 18 0 61 113

Kann ich somit einen zu vollen Speicher ausschließen?

Auch diesmal übrigens wieder:

last sysfs file: /sys/devices/pci0000:00/0000:00:01.3/i2c-0/name

lspci sagt mir:

00:01.3 Bridge: Intel Corporation 82371AB/EB/MB PIIX4 ACPI (rev 03)

Ist das der Übeltäter?

Gruß,
Hendrik

greentux · #5 26.12.2012, 10:44

Am I2C steckt was genau dran?

henfri · #6 26.12.2012, 11:12

Nix. Und nix kann ich deshalb sagen, weil ich keine Hardware außer den Busmastern (USB) an die VM weitergebe.

Übrigens hier mal die Unterschiede zwischen meiner ursprünglichen VM-Konfiguration und der aktuellen. Hinzugefügt wurden:

<uuid>d4fd6ae5-31cb-a7fb-73a4-2ea911cf7017</uuid>

Und bei der Netzwerkkarte:
<target dev='vnet0'/>
<model type='virtio'/>
<alias name='net0'/>
<address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/>

Sowie:

<console type='pty' tty='/dev/pts/12'>
<source path='/dev/pts/12'/>
<target type='serial' port='0'/>
<alias name='serial0'/>
</console>

Aber dass kann es doch nicht sein, oder?

Naja, korrellation ist nicht Kausalität. Es muss ja nicht an der VM liegen.

Gruß,
Hendrik

greentux · #7 26.12.2012, 11:24

Welche Virtualisierungslösung? Falls VirtualBox, wie sieht die Config aus?

henfri · #8 26.12.2012, 12:07

libvirt/qemu:

Code:

<domain type='kvm'>
  <name>WG</name>
  <uuid>d4fd6ae5-31cb-a7fb-73a4-2ea911cf7017</uuid>
  <memory>262144</memory>

  <vcpu>1</vcpu>
  <os>
    <type arch='i686' machine='pc'>hvm</type>
    <boot dev='hd'/>
    <bootmenu enable='yes'/>
  </os>
  <features><acpi/></features>
  <on_poweroff>destroy</on_poweroff>
  <on_reboot>restart</on_reboot>
  <on_crash>restart</on_crash>
  <devices>
    <emulator>/usr/bin/kvm</emulator>
    <interface type='bridge'>
      <source bridge='br1'/>
      <mac address='52:54:00:2b:85:3d'/>
      <target dev='vnet0'/>
      <model type='virtio'/>
      <alias name='net0'/>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/>
    </interface>
    <disk type='file' device='disk'>
      <driver name='qemu' type='qcow2'/>
      <target dev='sda' bus='sata'/>
      <source file='/var/lib/zentyal/machines/WG/WG-Test.qcow2'/>

      <alias name='sata0-0-0'/>
      <address type='drive' controller='0' bus='0' unit='0'/>
    </disk>
    <controller type='sata' index='0'>
      <alias name='sata0'/>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x05' function='0x0'/>
    </controller>

    <console type='pty' tty='/dev/pts/12'>
      <source path='/dev/pts/12'/>
      <target type='serial' port='0'/>
      <alias name='serial0'/>
    </console>




    <hostdev mode='subsystem' type='usb'>
      <source>
        <address bus='0x1' device='0x4'/>
        </source>

    </hostdev>
    <hostdev mode='subsystem' type='usb'>
      <source>
         <address bus='0x1' device='0x5'/>
     </source>

    </hostdev>

    <video>
      <model type='cirrus' vram='9216' heads='1'/>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x02' function='0x0'/>
    </video>
         <graphics type='vnc' listen='0.0.0.0' port='5900' passwd='till1234' keymap='de'/>
  </devices>

</domain>

Gruß,
Hendrik

greentux · #9 26.12.2012, 12:54

KVM kenne ich zu wenig. Ggf. ein SMP Problem oder sowas?

henfri · #10 26.12.2012, 13:11

Wie finde ich das raus?

Gruß,
Hendrik