Slackware repeatedly drops connection and then reconnects

storckm · 05-15-2014, 12:11 PM

My internet connection seems to work fine with puppy and mint, but in slackware, the connection is dropped every minute or two, and then reconnects after a few seconds.

Ifconfig gives me this:

Code:

eth0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet6 fe80::227:13ff:fe64:4ac5  prefixlen 64  scopeid 0x20<link>
        inet6 fd97:518:b147:0:227:13ff:fe64:4ac5  prefixlen 64  scopeid 0x0<global>
        ether 00:27:13:64:4a:c5  txqueuelen 1000  (Ethernet)
        RX packets 243607  bytes 286860288 (273.5 MiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 199251  bytes 26550836 (25.3 MiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
        device interrupt 20  memory 0xfa400000-fa420000  

lo: flags=73<UP,LOOPBACK,RUNNING>  mtu 65536
        inet 127.0.0.1  netmask 255.0.0.0
        inet6 ::1  prefixlen 128  scopeid 0x10<host>
        loop  txqueuelen 0  (Local Loopback)
        RX packets 10684  bytes 919960 (898.3 KiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 10684  bytes 919960 (898.3 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

lspci | grep Ethernet gives me this:

Code:

00:19.0 Ethernet controller: Intel Corporation 82567LM Gigabit Network Connection (rev 03)

And dmesg |grep -i eth0 gives me this:

Code:

[  257.067086] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
[  287.543459] IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready
[  289.045031] e1000e: eth0 NIC Link is Up 100 Mbps Full Duplex, Flow Control: Rx/Tx
[  289.045154] e1000e 0000:00:19.0 eth0: 10/100 speed: disabling TSO
[  289.045222] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
[  319.543430] IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready
[  321.189030] e1000e: eth0 NIC Link is Up 100 Mbps Full Duplex, Flow Control: Rx/Tx
[  321.189152] e1000e 0000:00:19.0 eth0: 10/100 speed: disabling TSO
[  321.189220] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
[  351.562517] IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready
[  353.218034] e1000e: eth0 NIC Link is Up 100 Mbps Full Duplex, Flow Control: Rx/Tx
etc.,
etc.

Other information:
Thinkpad T500, Linksys E900 router, cable internet, Modem Motorola SBG901 (in bridge mode).

metaschima · 05-15-2014, 02:44 PM

The e1000e driver is iffy from my experience. Try installing the latest:
http://sourceforge.net/projects/e100...?source=navbar

storckm · 05-16-2014, 11:03 PM

O.K. So I compiled the latest driver version

Code:

make install

. There are now two files, /lib/modules/3.14.3/kernel/drivers/net/ethernet/intel/e1000e/e1000e.ko and /lib/modules/3.14.4/kernel/drivers/net/ethernet/intel/e1000e/e1000e.ko. How do I designate which one the system loads?

metaschima · 05-17-2014, 11:04 AM

It always loads the same version as the kernel version you are running (uname -r). In other words the kernel looks in the module directory of its own version.

storckm · 05-19-2014, 10:03 AM

I've booted from the same kernel as the most recent version of the e1000e module, but the problem is not resolved. What should I try next?

metaschima · 05-19-2014, 10:48 AM

So are you using network manager or wicd, try each.

storckm · 05-19-2014, 08:35 PM

Aha! Forcing wicd to use ethtool solved the problem.