Bug 155074 – hotplug gets stuck.

Bugzilla will be upgraded to version 5.0 on a still to be determined date in the near future. The original upgrade date has been delayed.

Bug 155074 - hotplug gets stuck.

Summary:	hotplug gets stuck.

Status:	CLOSED RAWHIDE

Aliases:	None

Product:	Fedora
Classification:	Fedora
Component:	hotplug (Show other bugs)
Sub Component:
Version:	4
Hardware:	All Linux

Priority	medium Severity medium
Target Milestone:	---
Target Release:	---
Assigned To:	Bill Nottingham
QA Contact:	Brock Organ
Docs Contact:

URL:
Whiteboard:
Keywords:

Depends On:
Blocks:
	Show dependency tree / graph

Reported:	2005-04-15 19:48 EDT by Dave Jones
Modified:	2015-01-04 17:18 EST (History)
CC List:	3 users (show)

See Also:
Fixed In Version:
Doc Type:	Bug Fix
Doc Text:
Story Points:	---
Clone Of:
Environment:
Last Closed:	2005-10-03 19:06:35 EDT
Type:	---
Regression:	---
Mount Type:	---
Documentation:	---
CRM:
Verified Versions:
Category:	---
oVirt Team:	---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---

Attachments	(Terms of Use)
Add an attachment (proposed patch, testcase, etc.)

Groups: None (edit)

Description Dave Jones 2005-04-15 19:48:56 EDT

I did an upgrade of an FC3 to rawhide, and on logging in, found that everything
was really sluggish. Firing up top showed that the CPUs were both at 100% !

These two..

 3811 root      20  -5  5268  404  328 R 92.3  0.0  24:03.47 05-pam_console.
 3997 root      20  -5  5268  404  328 R 90.6  0.0  23:19.10 default.hotplug

Take it in turns to be scheduled, and do.. bugger all.
strace of those processes shows no output whatsoever.

sysrq-t shows..
05-pam_consol R  running task       0  3811   7552                     (NOTLB)
bash          S ffff81002e4e0170     0  7571   4634  7620    7768  7449 (NOTLB)
ffff81006fea3eb8 0000000000000086 ffff81002e4e0170 0000000000000002
       ffffffff80543720 0000000000000239 00000074000001e3 ffff81002e4e0380
       ffff81002e4e0170 ffff81002e1770f0
Call Trace:<ffffffff8013d51e>{do_wait+2798}
<ffffffff80134440>{default_wake_function+0}
       <ffffffff8019d59c>{sys_ioctl+108} <ffffffff8010ea72>{system_call+126}

default.hotpl R  running task       0  3997   7674                     (NOTLB)
bash          S ffff81002d6ad880     0  7768   4634  7813          7571 (NOTLB)
ffff81002de2beb8 0000000000000082 ffff81002d6ad880 0000000000000002
       ffffffff80543720 00000000000000f3 00000074000001e3 ffff81002d6ada90
       ffff81002d6ad880 ffff8100302a00f0
Call Trace:<ffffffff8013d51e>{do_wait+2798}
<ffffffff80134440>{default_wake_function+0}
       <ffffffff8019d59c>{sys_ioctl+108} <ffffffff8010ea72>{system_call+126}


What they're waiting for is a mystery to me.

more oddness.. something is calling sys_swapon once a second.
in 30 mins of uptime, its been called over 10,000 times.

Comment 1 Dave Jones 2005-04-15 19:54:59 EDT

killing these processes brings back a speedy desktop. The calls to swapon still
occur though. Wieeeerd.

Comment 2 Dave Jones 2005-04-15 19:57:42 EDT

false alarm on the sys_swapon -- wrong System.map in use. (its slab debug :-)

Why hotplug went nuts is still a mystery though.

Comment 3 Bill Nottingham 2005-04-15 22:42:59 EDT

I'd love to know what that ioctl was..... if you can reproduce it, what files do
they have open?

Comment 4 Bill Nottingham 2005-10-03 18:16:49 EDT

You still seeing this? (Woo, I get to do this this time. :) )

Comment 5 Dave Jones 2005-10-03 19:06:35 EDT

nope, only ever happened once.  *shrug* rawhide oddness of the day I suppose.

Note You need to log in before you can comment on or make changes to this bug.