After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 542916 - Network-manager shows wired interface as disconnected when started (under VirtualPC and Hyper-V)
Network-manager shows wired interface as disconnected when started (under Vir...
Status: RESOLVED NOTGNOME
Product: NetworkManager
Classification: Platform
Component: general
0.6.6
Other All
: Normal normal
: ---
Assigned To: Dan Williams
Dan Williams
Depends on:
Blocks:
 
 
Reported: 2008-07-14 12:40 UTC by Franklin PIAT
Modified: 2008-07-15 21:07 UTC
See Also:
GNOME target: ---
GNOME version: ---


Attachments
/var/log/syslog (110.62 KB, text/plain)
2008-07-14 12:56 UTC, Franklin PIAT
Details
dmesg output (19.74 KB, text/plain)
2008-07-14 12:57 UTC, Franklin PIAT
Details

Description Franklin PIAT 2008-07-14 12:40:11 UTC
Please describe the problem:
When I boot Debian under Microsoft VirtualPC 2007 or WindowsServer Hyper-V, the network is "disconnected".

The network interface is has PCI id =1011:0009
"Digital Equipment Corporation DECchip 21140 [FasterNet]" 
The Linux kernel module to handle it is "tulip".

Steps to reproduce:
1. Boot Debian (stable or Testing) under VirtualPC 2007 or Hyper-V
2a. Open a session in a console, type "ifconfig" : The interface isn't configured
2b. Open a session under gnome : the NM applet says the link is down.
    (click on the icon, the "Wired Networking" isn't ticked in !).
3a. Type "ifup eth0" : the interface get configured.
3b. Click on the NM applet, tick-in "Wired Networking" : the interface get configured.


Actual results:
We notice that the interface is down.

Expected results:
The interface should be up and configured.

Does this happen every time?
Yes.
Under Debian stable (kernel 2.6.18 + NM 0.6.4) and Debian Testing (kernel 2.6.25 + NM 0.6.6)

Other information:
I also made the following test :
1. boot into debian, and remove the files /etc/rc2.d/*network-manager*
2. reboot again and login the console as root.
3. Run ifconfig : notice eth0 is up and running
4. Run the command : "NetworkManager --no-daemon"
### sample output : start ###
NetworkManager: <info>  starting...
NetworkManager: <info>  eth0: Driver 'tulip' does not support carrier detection.        You must switch to it manually.
NetworkManager: <info>  nm_device_init(): waiting for device's worker thread to
start
NetworkManager: <info>  nm_device_init(): device's worker thread started, continuing.
NetworkManager: <info>  Now managing wired Ethernet (802.3) device 'eth0'.
NetworkManager: <info>  Deactivating device eth0.
### sample output : end ###
5. Run ifconfig again. Notice that the interface isn't configured anymore.
Comment 1 Franklin PIAT 2008-07-14 12:56:35 UTC
Created attachment 114521 [details]
/var/log/syslog
Comment 2 Franklin PIAT 2008-07-14 12:57:10 UTC
Created attachment 114522 [details]
dmesg output
Comment 3 Dan Williams 2008-07-15 17:44:12 UTC
NetworkManager is not aware of the state of devices that are not configured by NetworkManager.  Thus typing 'ifup eth0' has no effect becuase you have not told NetworkManager to bring the device up (by picking it from the list).

The real problem here is that the device does not support carrier detection via ethtool or mii-tool.  If the driver supported carrier detection, NetworkManager would bring it up automatically.

NetworkManager: <info>  eth0: Driver 'tulip' does not support carrier
detection.        You must switch to it manually.

That should be fixed in the driver.
Comment 4 Franklin PIAT 2008-07-15 18:52:11 UTC
Couldn't NetworkManager try to bring up those devices on boot ? Especially when the computer has no other network interface.

Thanks.
Comment 5 Franklin PIAT 2008-07-15 21:07:41 UTC
I don't know If this problem can be solved on this application side only, but it would be great if we could find a solution so Windows admins don't get mad when they try Linux ;)

I've opened a similar bug again linux's tulip driver :
http://bugzilla.kernel.org/show_bug.cgi?id=11092

Your help would be appreciated.

Franklin