Have you checked to see it isn't a power issue? Does the server has ample power to support the five camera, even though they may not be taking power from the server, it might still be affecting the server - I had a server with random crashes, turned out I had too many self-powered USB devices attached, so even though they had there own power supply, the server just was freaking out with all the additional devices.


On 1 May 2010 19:47, Moses <moses@neonixie.com> wrote:
Hi Everybody,

I've been dealing with my motion server constantly crashing.. every week
or two, and I've posted to linux-media but no response. The machine
crashed today and I noticed a similarity in the logs.. all the crashes
have "last sysfs file: /devices/pci0000:00/0000:00:00.0/irq" in common.
The same app doesn't even crash, sometimes motion will trigger it
sometimes sh will, or some other random process, but the last sysfs file
is always the same (pci / irq).

I have 5 cameras now on a 8 chip bt878 kodicom clone card. I'm wondering
if I'm somehow filling up the PCI bus and crashing the machine? The
crashes seem somewhat random, but can sometimes be triggered with a lot
of disk IO. If saturating the bus is the problem.. is there anyway I can
calculate how much of the bus I'm using? I have the following line in
the log if it helps..

Using palette YUYV (720x480) bytesperlines 1440 sizeimage 691200
colorspace 00000000

I'm capturing 720x480, should I be using something else other then YUYV?
Is my thinking correct, is this a PCI bus problem I'm hitting?

Machine is centos on 2.6.18-164.15.1.el5, and I've swapped machines
completely and reinstalled, same problem. Even tried another capture
card (same model).

Thank you for any help!


BUG: unable to handle kernel NULL pointer dereference at virtual address
 printing eip:
*pde = 00000000
Oops: 0002 [#1]
last sysfs file: /devices/pci0000:00/0000:00:00.0/irq
ypto_api lp snd_hda_intel snd_seq_dummy snd_seq_oss snd_seq_midi_event
snd_seq snd_seq_device snd_pcm_oss snd_mixer_oss snd_pcm tuner snd_timer
snd_page_alloc snd_hwdep bttv snd video_buf ir_common soundcore
compat_ioctl32 i2c_i801 i2c_algo_bit pcspkr btcx_risc tveeprom i2c_core
videodev v4l1_compat parport_pc v4l2_common r8169 serio_raw parport mii
dm_raid45 dm_message dm_region_hash dm_log dm_mod dm_mem_cache
usb_storage ata_piix libata sd_mod scsi_mod raid1 ext3 jbd uhci_hcd
ohci_hcd ehci_hcd
CPU:    1
EIP:    0060:[<c0618d22>]    Not tainted VLI
EFLAGS: 00210246   (2.6.18-164.15.1.el5 #1)
EIP is at mutex_lock+0xd/0x19
eax: 00000000   ebx: 00000000   ecx: f7d76a80   edx: d8921000
esi: c99d7920   edi: 00000000   ebp: f78a8200   esp: d8921e30
ds: 007b   es: 007b   ss: 0068
Process motion (pid: 1214, ti=d8921000 task=f7f8baa0 task.ti=d8921000)
Stack: f7448b54 00000000 f8a5746f ec7c4374 ee9ac7b8 ec7c4374 c0466646
       f68f1740 c0466734 00008a4e c97fd644 f68f1740 f68f178c c06404ab
       f68f1740 00000080 c047f19f e0413c00 00000000 f7bab540 00000000
Call Trace:
 [<f8a5746f>] videobuf_vm_close+0x52/0x9b [video_buf]
 [<c0466646>] remove_vma+0x1e/0x3b
 [<c0466734>] exit_mmap+0xd1/0xee
 [<c042344c>] mmput+0x25/0x69
 [<c047f19f>] flush_old_exec+0x5f7/0x87d
 [<c047e323>] kernel_read+0x32/0x43
 [<c049ca42>] load_elf_binary+0x486/0x1552
 [<c0464cf3>] get_user_pages+0x284/0x306
 [<c047e615>] get_arg_page+0x30/0x7c
 [<c04eefc5>] copy_from_user+0x31/0x5d
 [<c047e86d>] copy_strings+0x169/0x173
 [<c047e143>] search_binary_handler+0x99/0x219
 [<c047fef2>] do_execve+0x13b/0x1b1
 [<c0403200>] sys_execve+0x2a/0x4a
 [<c0404f17>] syscall_call+0x7/0xb
Code: 94 c0 c3 53 89 c3 e8 77 f7 ff ff 83 c8 ff f0 0f c1 03 48 79 05 89
d8 5b eb 1d 5b 31 c0 c3 53 89 c3 83 ec 04 e8 59 f7 ff ff 89 d8 <f0> ff
08 0f 88 53 01 00 00 58 5b c3 57 56 89 c6 53 83 ec 0c 89
EIP: [<c0618d22>] mutex_lock+0xd/0x19 SS:ESP 0068:d8921e30

Motion-user mailing list