My system has been experiencing kernel panics every morning at around 0345 (3:45am). I don't know what triggers this. I don't have any cron jobs at that time, and mythfilldatabase runs at some off times, usually in the early evening. The system runs fine while recording, playing back or just sitting idle. This all started back on Nov 23 but I hadn't done any changes. Usually I have the system off most of the time except when recording or playing back in the evening which is why this hasn't been that big of an issue. This system is used as a backend although it does have a monitor and keyboard connected and can function as a frontend if needed.

Does anyone know what might be starting up at around 0345?

Or can I install auditd from the regular Arch repo? I don't see it in the LinHES repo anywhere. I was thinking it might be able to tell me what process gets started before the panic.

System:
Via C7 processor on all-in-one motherboard VIA PC2500E
1GB RAM
1 TB Seagate ST31000528AS (SATA)
Hauppauge PVR-150 capture card
CDROM drive (IDE)
LinHES 6.04

I managed to capture the kernel panic by hooking up a serial console and adding a parameter to the kernel command line (console=ttyS0,115200) as documented in the kernel source Documentation/serial-console.txt. Here's the beginning of the panic:

BUG: unable to handle kernel paging request at 3b9aca00
IP: [<c01484f8>] ktime_get+0x28/0x40
Oops: 0000 [#1] PREEMPT SMP
last sysfs file: /sys/block/sr0/capability
Modules linked in: netconsole configfs nfsd exportfs ipv6 snd_seq_oss snd_seq_midi_event snd_seq snd_pcm_oss snd_mixer_oss ppdev snd_via82xx gameport uhci_hcd snd_ac97_codec lp ac97_bus parport_pc ppp_generic wm8775 slhc parport snd_pcm cx25840 snd_mpu401_uart serio_raw snd_timer snd_page_alloc snd_rawmidi snd_seq_device tuner tea5767 tda8290 tda18271 tda827x ehci_hcd mt20xx tea5761 psmouse snd soundcore pcspkr usbcore i2c_viapro ivtv compat_ioctl32 shpchp pci_hotplug cx2341x v4l2_common via_agp agpgart sg thermal evdev processor fan button battery ac cx88_vp3054_i2c mt352 cx8802 cx88xx videodev v4l1_compat ir_common i2c_algo_bit tveeprom btcx_risc tuner_xc2028 s5h1409 xc5000 dvb_pll s5h1411 or51132 cx24116 videobuf_dvb videobuf_dma_sg videobuf_core nxt200x isl6421 stb6000 tda9887 tuner_simple tuner_types zl10353 cx24123 lgdt330x stv0299 dvb_core stv0288 cx22702 i2c_core rtc_cmos rtc_core rtc_lib ext3 jbd mbcache sd_mod sr_mod cdrom sata_via pata_via pata_acpi libata scsi_mod

Pid: 0, comm: swapper Not tainted (2.6.28-LinHES #1) ID-PCM7E PC2500
EIP: 0060:[<c01484f8>] EFLAGS: 00010092 CPU: 0
EIP is at ktime_get+0x28/0x40
EAX: 000042f9 EBX: 11cb2e99 ECX: 3b9aca00 EDX: 000042f9
ESI: 2d4225fb EDI: 00000000 EBP: c17865a0 ESP: c0575efc
DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068
Process swapper (pid: 0, ti=c0574000 task=c0505340 task.ti=c0574000)
Stack:
000042f9 2d4225fb 11cb2e99 00000f98 0ee49ab6 c0150012 c0147a00 c1787d10
c1787ce0 c040e7f1 11cb2e99 00000f98 0ee49ab6 00000f98 c015012a 00000000
f5ab5b2c c1787d10 c01486b5 00000000 11cb2e99 00000f98 0ee49ab6 00000f98
Call Trace:
[<c0150012>] tick_dev_program_event+0x12/0xb0
[snip]

After messing around with the system for a few days I found that the problem was something being done by mythbackend. I couldn't get any further than that so I wiped the install partition and did an upgrade install to preserve my data. There were a few annoyances getting everything back as it was but it has fixed the problem. Apparently something had gotten corrupted and the reinstall fixed it.