Screen freezes and there is a spike in interrupt request. What could cause it?

MikeyCarter · 09-17-2010, 09:56 AM

I have a computer that freezes for about 41 seconds freqently. There's no logs other than a rtkit complaining about a starving bird.

Then I started running vmstat, below is when the glitch happens. Any ideas on what could be causing the very high "in: The number of interrupts per second, including the clock." all of a sudden?

Code:

procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu------ ---timestamp---
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa st
 1  0   4704 117188 141196 1279312    0    0     0     1  110  195  1  2 97  0  0       2010-09-17 10:37:29 EDT
45  1   4704 117188 141200 1279312    0    0     0    16 9390   79  0 91  9  0  0       2010-09-17 10:38:09 EDT

xeleema · 09-17-2010, 10:32 AM

Greetingz!

I would suspect it's a peice of hardware. Probably an I/O timeout of somesort.
If you don't have any Fibre attached storage, then I'd check and make sure all of your drives are okay.

Aside from that, check the crontabs and make sure there isn't something big running every minute or so.

Hope this helps!

MikeyCarter · 09-17-2010, 10:50 AM

I figure it's something waking up. The times are erratic when it happens. There's no corresponding cron job or anything else in the logs. Just the ~10,000 in requests.

Any way to track down what's making those requests?

xeleema · 09-17-2010, 01:35 PM

I'm not entirely sure. I want to start asking vauge questions like "Are you running a hardcore database, like Oracle?" or "change kernel schedulers", but I would just be talking outta my posterier at that point...
Are you always getting spikes like that? Even after a reboot?

H_TeXMeX_H · 09-17-2010, 02:02 PM

Well, can you post 'cat /proc/interrupts', it might be useful.

MikeyCarter · 09-17-2010, 02:27 PM

Code:

            CPU0       
   0:    9390956   IO-APIC-edge      timer
   1:          4   IO-APIC-edge      i8042
   3:         11   IO-APIC-edge    
   4:         12   IO-APIC-edge    
   7:          0   IO-APIC-edge      parport0
   8:          1   IO-APIC-edge      rtc0
   9:          0   IO-APIC-fasteoi   acpi
  12:          1   IO-APIC-edge    
  14:    2689246   IO-APIC-edge      ata_piix
  15:          0   IO-APIC-edge      ata_piix
  16:     179014   IO-APIC-fasteoi   uhci_hcd:usb5
  18:    3897474   IO-APIC-fasteoi   uhci_hcd:usb4
  19:     890802   IO-APIC-fasteoi   ata_piix, uhci_hcd:usb3
  23:    1884482   IO-APIC-fasteoi   ehci_hcd:usb1, uhci_hcd:usb2
  26:    1180967   PCI-MSI-edge      i915
  27:    3942417   PCI-MSI-edge      eth0
  28:      13454   PCI-MSI-edge      hda_intel
 NMI:          0   Non-maskable interrupts
 LOC:   38761815   Local timer interrupts
 SPU:          0   Spurious interrupts
 PMI:          0   Performance monitoring interrupts
 PND:          0   Performance pending work
 RES:          0   Rescheduling interrupts
 CAL:          0   Function call interrupts
 TLB:          0   TLB shootdowns
 TRM:          0   Thermal event interrupts
 THR:          0   Threshold APIC interrupts
 MCE:          0   Machine check exceptions
 MCP:        909   Machine check polls
 ERR:          0
 MIS:          0

I've been tracking it by a canary (which seems to be a symptom):

Sep 17 06:11:08 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 06:17:14 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 06:21:48 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 06:32:43 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 06:34:42 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 06:38:42 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 06:39:28 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 06:43:08 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 06:46:32 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 07:01:22 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 07:08:34 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 07:34:20 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 07:47:23 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 07:48:35 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 08:35:12 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 09:29:11 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 10:38:09 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 10:45:56 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 10:47:15 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 10:57:34 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 11:21:05 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 13:52:37 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 13:53:35 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.
Sep 17 15:18:42 silver-dragon rtkit-daemon[23904]: The canary thread is apparently starving. Taking action.

I've had my dad unplug most of his USB devices to see if it was one of them. (as he has hubs, upon hubs plugged into each other)