I can confirm this bug on 7.04 with kernel 2.6.20-15-generic. One difference I notice from previous posts is that on boot the network gets up on 1000Mb/s and works fine until ifdown. After ifup the link is at 100Mb/s.
On forcing eth0 to 1000 with ethtool makes these messages appear in /var/log/messages: Jun 28 13:59:40 husk kernel: [19933.630813] e1000: eth0: e1000_watchdog: NIC Link is Up 1000 Mbps Full Duplex Jun 28 13:59:40 husk kernel: [19933.630915] ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready Jun 28 13:59:43 husk kernel: [19936.197642] e1000: eth0: e1000_watchdog: NIC Link is Down Jun 28 13:59:46 husk kernel: [19939.276850] e1000: eth0: e1000_watchdog: NIC Link is Up 100 Mbps Full Duplex Jun 28 13:59:46 husk kernel: [19939.276855] e1000: eth0: e1000_watchdog: 10/100 speed: disabling TSO lspci for ethernet controller: 00:19.0 Ethernet controller: Intel Corporation 82566DM Gigabit Network Connection (rev 02) The affected system is a HP Compaq dc7700 desktop. A identical system with Fedora Core 6 with kernel 2.6.19 is also experiencing this bug. I guess this means this bug is in upstream. Rasjid's sollution doesn't seem to help. -- Intel Gigabit network adapter e1000 fails to connect at gigabit speed, is "geared down" to only 100 Mbps https://bugs.launchpad.net/bugs/86437 You received this bug notification because you are a member of Ubuntu Bugs, which is the bug contact for Ubuntu. -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
