[hardy] network interfaces not restarted after resume

Bug #214523 reported by P
2
Affects Status Importance Assigned to Milestone
acpi-support (Ubuntu)
Invalid
Medium
Daniel Hahler

Bug Description

Binary package hint: acpi-support

I have an thinkpad X40. After suspend to RAM and resume, my wired interface is not restored correctly.

I have to /etc/init.d/networking restart to get it operational.

I have also the same problem after hibernate/resume on another computer. The eth0 network interface is not operational.

Both computers use dhcp to get an address.

$ lsb_release -rd
Description: Ubuntu hardy (development branch)
Release: 8.04
$ apt-cache policy acpi-support
acpi-support:
  Installé : 0.108
  Candidat : 0.108
 Table de version :
 *** 0.108 0
        500 http://archive.ubuntu.com hardy/main Packages
        100 /var/lib/dpkg/status

Revision history for this message
P (p92) wrote :
Revision history for this message
P (p92) wrote :
Revision history for this message
P (p92) wrote :
Revision history for this message
P (p92) wrote :
Revision history for this message
P (p92) wrote :
Revision history for this message
P (p92) wrote :
Revision history for this message
P (p92) wrote :
Revision history for this message
P (p92) wrote :
Revision history for this message
P (p92) wrote :

doing /etc/acpi/suspend.d/55-down-interfaces.sh
manually works OK and I can see that the dhcpclient3 process has been killed

# /etc/acpi/suspend.d/55-down-interfaces.sh
+ pccardctl eject
+ killall dhclient dhclient3
+ /sbin/ifconfig
+ awk /^[^ ]+/ {print $1}
+ INTERFACES=eth0
eth0:0
eth0:1
eth1
lo
+ ifdown eth0
There is already a pid file /var/run/dhclient.eth0.pid with pid 17093
removed stale PID file
Internet Systems Consortium DHCP Client V3.0.6
Copyright 2004-2007 Internet Systems Consortium.
All rights reserved.
For info, please visit http://www.isc.org/sw/dhcp/

Listening on LPF/eth0/00:0a:e4:24:7b:a8
Sending on LPF/eth0/00:0a:e4:24:7b:a8
Sending on Socket/fallback
DHCPRELEASE on eth0 to 192.168.0.254 port 67
+ ifconfig eth0 down
+ ifdown eth0:0
RTNETLINK answers: No such process
+ ifconfig eth0:0 down
SIOCSIFFLAGS: Cannot assign requested address
+ ifdown eth0:1
RTNETLINK answers: No such process
+ ifconfig eth0:1 down
SIOCSIFFLAGS: Cannot assign requested address
+ ifdown eth1
ifdown: interface eth1 not configured
+ ifconfig eth1 down
+ ifdown lo
+ ifconfig lo down

BUT when I do a suspend to ram / resume
I can still see the same dhcpclient3 process ID active.

In fact it is as if /etc/acpi/suspend.d/55-down-interfaces.sh is never called upon suspend

Also I have some trouble to understand how the variable INTERFACES is populated in the
following script resume.d script :

# cat /etc/acpi/resume.d/62-ifup.sh
#!/bin/sh

# Bring up the interfaces (this should probably be left up to some policy
# manager, but at the moment we just bring back whatever we ifdowned)
for x in $INTERFACES; do
    ifup $x &
done

Revision history for this message
P (p92) wrote :

I am using KDE 3 with knetworkmanager :

$ dpkg -l network-manager-kde
Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Installed/Config-f/Unpacked/Failed-cfg/Half-inst/t-aWait/T-pend
|/ Err?=(none)/Hold/Reinst-required/X=both-problems (Status,Err: uppercase=bad)
||/ Nom Version Description
+++-==================-==================-====================================================
ii network-manager-kd 1:0.2.2-1ubuntu1 KDE systray applet for controlling NetworkManager

Revision history for this message
Daniel Hahler (blueyed) wrote :

Thank you for your bug report and all the information.

acpi-support does not get used for suspend/hibernation in Hardy anymore, but pm-utils.

This bug is caused by a fix for bug 162654, which unloads all networking modules before hibernate/suspend and caused this regression.

I'm marking it as duplicate. Please try the proposed patch over there.

Changed in acpi-support:
assignee: nobody → blueyed
importance: Undecided → Medium
status: New → Invalid
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.