Bug 254412

Summary:	Network is sometimes not correctly restarted after resume
Product:	[openSUSE] openSUSE 10.3	Reporter:	Matthias Hopf <mhopf>
Component:	Network	Assignee:	Tambet Ingo <tambet>
Status:	RESOLVED DUPLICATE	QA Contact:	E-mail List <qa-bugs>
Severity:	Critical
Priority:	P5 - None	CC:	behlert
Version:	Alpha 1plus
Target Milestone:	---
Hardware:	Other
OS:	Other
Whiteboard:
Found By:	Other	Services Priority:
Business Priority:		Blocker:	---
Marketing QA Status:	---	IT Deployment:	---
Attachments:	Logfile of NetworkManager

Description Matthias Hopf 2007-03-14 11:57:44 UTC

On this D410, the network doesn't seem to be correctly restarted after resume.

Sometimes (approx. 1 out of 4 tries) the networkmanager tries to get an IP address, and turns up with 169.254.x.x (local ad-hoc network) after a timeout.

Explicitly asking the networkmanager to connect to wired network again solves the issue, but this is especially annoying if the screensaver has popped up, because it won't allow you to log in in order to restart network detection, if you are not working on a local account, because YP doesn't work (no network ;), and the screensaver doesn't allow you to log in with the root password (some desktop guys apparently consider this a security feature...).

What logs do you need when this happens again?

Comment 1 Magnus Boman 2007-03-14 12:01:12 UTC

Log for nm is /var/log/NetworkManager

Comment 3 Forgotten User ZhJd0F0L3x 2007-03-14 13:09:09 UTC

this is not a Mobile devices bug, it is a Network Manager bug.
We tell NM to suspend and to resume, however, NM does usually do something completely bogus.
E.g. starting the wired interface although there is no cable plugged in and other weird stuff.
I guess that this bug falls into the same category.

Comment 4 Matthias Hopf 2007-03-14 13:38:49 UTC

Agreed. Assigning to networkmanager maintainer.

Comment 5 Matthias Hopf 2007-03-14 13:58:31 UTC

Created attachment 124331 [details]
Logfile of NetworkManager

This logfile contains the last two resumes. After the first resume, NetworkManager got a valid IP, after the second it didn't.

Especially noteworthy is the line

Couldn't send DHCP 'up' message because: name 'com.redhat.dhcp.OperationInProgress', message 'interface eth0 is being released. Please try again later.'.

This sounds like a race condition.

Comment 6 Matthias Hopf 2007-04-05 09:37:34 UTC

Raising severity, as this doesn't seems to be looked at.

Note that sometimes even on boot network isn't correctly detected. This is especially bad for NIS configurations (noone except root can log in to restart network detection).

Comment 7 Bryan Perry 2007-04-18 23:37:14 UTC

Isn't this the same as bug 246966?

Comment 8 Matthias Hopf 2007-04-19 10:05:35 UTC

Seems to be. I'll try the rpms there.

*** This bug has been marked as a duplicate of bug 246966 ***