Bug 540550

Summary:	ksoftirqd causes massive disk access
Product:	[openSUSE] openSUSE 11.2	Reporter:	Elmar Stellnberger <estellnb>
Component:	Kernel	Assignee:	E-mail List <kernel-maintainers>
Status:	RESOLVED DUPLICATE	QA Contact:	E-mail List <qa-bugs>
Severity:	Critical
Priority:	P5 - None	Flags:	coolo: SHIP_STOPPER-
Version:	Milestone 7
Target Milestone:	---
Hardware:	x86-64
OS:	SUSE Other
Whiteboard:
Found By:	---	Services Priority:
Business Priority:		Blocker:	---
Marketing QA Status:	---	IT Deployment:	---
Attachments:	/proc/interrupts & /var/log/messages & top same info for 100% cpu no disk acces ksoftirqd-hang

Description Elmar Stellnberger 2009-09-20 09:46:35 UTC

User-Agent:       Mozilla/5.0 (compatible; Konqueror/3.5; Linux) KHTML/3.5.10 (like Gecko) SUSE

  The ksoftirqd problem has emerged again. Resolution by letting lay things far off has not worked. However now it has emerged in a somehow different flavour causing massive disk acccess instead of 100% cpu load.
  see for the old os11.1-related bug: Bug 465039.
Rebooting is the only choice. It is not possible to work seriously with a system like this!

Reproducible: Always

Comment 1 Elmar Stellnberger 2009-09-20 09:55:41 UTC

Created attachment 318975 [details]
/proc/interrupts & /var/log/messages & top

Comment 2 Elmar Stellnberger 2009-09-20 10:01:35 UTC

  I don`t know whether this is related to the ksoftirqd problem, but my external keyboard could not be recognized after the ksoftirqd problem had emerged although I have plugged it a dozen of times on multiple USB boards. i.e. The disk load problem had already been gone, but the USB keyboard was not recognized afterwards any more. Problems with the recognition of my USB-mouse or keyboard can normally be resolved by plugging them in once another time.

Comment 3 Elmar Stellnberger 2009-09-20 10:22:50 UTC

Comment #2: invalid; accidentially plugged in wrong device.

Comment 4 Elmar Stellnberger 2009-09-21 19:33:13 UTC

Created attachment 319136 [details]
same info for 100% cpu no disk acces ksoftirqd-hang

  This time there was no disk access but a 100% CPU-load. In a fact nothing seems having changed since the times of OpenSuSE 11.1.

Comment 5 Stephan Kulow 2009-09-22 18:26:20 UTC

THe 11.1 one hasn't fixed yet, so I don't see a reason to have 2 bugs.

*** This bug has been marked as a duplicate of bug 465039 ***

Comment 6 Elmar Stellnberger 2009-09-25 08:57:16 UTC

  It was possibly already fixed in the meanwhile since I have never experienced this with kernel 2.6.30.