On Mon, 2008-07-14 at 15:34 +0200, Michael Biebl wrote: > It seems as if the tulip driver or the emulated hardware doesn't > support link beat detection. That's why NM doesn't activate the > connection. > > This is not a bug in NM itself but rather a driver problem.
I don't understand what's the exact problem. Is it : A.The driver sends inaccurate information about the link detection capability and/or the current link state. In which case the driver should be fixed (i.e forward this bug the kernel). or B.The driver sends information that accurately describe the link detection capability and link state. But NM don't handle it properly (for example if NM don't handle the case where the card can't do link detection). I digged a little bit with "nm-tool" and in /sys/class/net/eth0/ but I couldn't find the documentation on how to interpret the results... ### BEFORE clicking on "Wired connection" ### $nm-tools state NetworkManager Tool State: disconnected - Device: eth0 ---------------------------------------------------------------- NM Path: /org/freedesktop/NetworkManager/Devices/eth0 Type: Wired Driver: tulip Active: no HW Address: 40:00:50:00:40:50 Capabilities: Supported: yes Wired Settings Hardware Link: yes ### AFTER clicking on "Wired connection" ### $nm-tools state NetworkManager Tool State: connected - Device: eth0 ---------------------------------------------------------------- NM Path: /org/freedesktop/NetworkManager/Devices/eth0 Type: Wired Driver: tulip Active: yes HW Address: 40:00:50:00:40:50 Capabilities: Supported: yes Wired Settings Hardware Link: yes IP Settings: IP Address: 192.168.0.144 Subnet Mask: 255.255.255.0 Broadcast: 192.168.0.255 Gateway: 192.168.0.1 Primary DNS: 192.168.0.33 Secondary DNS: 0.0.0.0 > If you prefer, I can reassign the bug report to the kernel, otherwise > I'm simply going to close this bug report. If it is really a driver bug, yes please reassign. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]