Bug 186821

Summary:	pm-suspend doesn't finish after suspend, eats 100% of CPU
Product:	[Fedora] Fedora	Reporter:	Yaron Minsky <yminsky>
Component:	pm-utils	Assignee:	Phil Knirsch <pknirsch>
Status:	CLOSED WORKSFORME	QA Contact:
Severity:	medium	Docs Contact:
Priority:	medium
Version:	5	CC:	ncunning, rvokal
Target Milestone:	---
Target Release:	---
Hardware:	All
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2007-01-23 14:53:47 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Yaron Minsky 2006-03-26 21:54:24 UTC

Description of problem:

I'm running on a T22 Thinkpad.  If I suspend (using pm-suspend) and then
recover, I don't get back to X, I just stay on the console that I went to after
the suspend.  I can switch back to X using Ctl-Alt-F7.  At that point,
pm-suspend still hasn't returned, and is now eating the entire CPU.  This is
running with what I believe are the latest kernel and pm-utils:

kernel-2.6.15-1.2054_FC5
pm-utils-0.15-1


How reproducible:

100%

Steps to Reproduce:
1. Suspend using pm-suspend
2. Press a key to recover from the suspend
3. Switch back to X
  
Actual results:


Expected results:


Additional info:

Comment 1 Yaron Minsky 2006-03-26 22:03:43 UTC

Another potentially intersting detail --- after the suspend, the half-moon icon
on the thinkpad is left flashing.  This suggests that it's in some weird
suspension-related state.

Comment 2 Yaron Minsky 2006-03-27 02:12:31 UTC

A slight refinement.  pm-suspend doesn't take over the CPU until you kill it. 
It looks initially like this in ps:

root      6264  0.0  0.5   4448  1388 ?        S    19:49   0:01 /bin/sh
/usr/sbin/pm-suspend

And after you kill it, it looks like this:

root      6264  0.0  0.5   4448  1388 ?        R    19:49   0:04 /bin/sh
/usr/sbin/pm-suspend

And now it eats the entire CPU.  It can't be killed even with kill -9.

Comment 3 Phil Knirsch 2006-09-05 14:38:58 UTC

That sounds like suspend seems to be hanging somehow in the kernel.

Have you tried newer FC5 kernels? Or would you be even able to test the latest
devel kernels and pm-utils from there?

Read ya, Phil

Comment 4 Phil Knirsch 2007-01-23 14:53:47 UTC

Closing due to inactivity.

Read ya, Phil