LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Slackware (https://www.linuxquestions.org/questions/slackware-14/)
-   -   e1000e eth0 NIC and kernel 2.6.35.3 under Slackware64 (https://www.linuxquestions.org/questions/slackware-14/e1000e-eth0-nic-and-kernel-2-6-35-3-under-slackware64-827741/)

dimm0k 08-22-2010 01:00 AM

e1000e eth0 NIC and kernel 2.6.35.3 under Slackware64
 
No networking issues when I first installed Slackware 64 13.1, as well as when I compiled a 2.6.34.1 kernel to replace the stock. Now that I have compiled and am running 2.6.35.3, my ethernet connection won't go and obtain an IP address from DHCP as it had done so previously. Did I miss something in this particular kernel release? The funny thing is my wifi connection works fine... isn't it usually the wired that is easier to get working than wireless! I did not change anything pertaining to e1000e and from the looks of it, it is listed under lspci -v and lsmod. Any suggestions?

business_kid 08-22-2010 03:09 AM

Is it listed under ifconfig? I would also check rc.inet1.conf in /etc/rc.d

onebuck 08-22-2010 07:53 AM

Hi,

Try 'ifconfig -a'.

guanx 08-22-2010 08:05 AM

Since it's in lsmod, can you try "dhcpcd eth0" (presuming you are using dhcp and your e1000e is eth0) or just "ifconfig eth0 YOUR_IP_ADDR up" (if you use a static ip address).

dimm0k 08-22-2010 09:22 AM

ifconfig lists eth0, but with no IP address. I just issued "dhcpcd eth0" and it proceeded by acquiring an IP address from the DHCP server and everything is working from that. However if I take down eth0 with "ifconfig eth0 down" and stop /etc/rc.d/rc.inet1, I get the following when I start /etc/rc.d/rc.inet1

Code:

Polling for DHCP server on interface eth0:
dhcpcd: version 5.2.2 starting
dhcpcd: eth0: waiting for carrier
dhcpcd: timed out
dhcpcd: allowing 8 seconds for IPv4LL timeout
dhcpcd: timed outPolling for DHCP server on interface eth0:
dhcpcd: version 5.2.2 starting
dhcpcd: eth0: waiting for carrier
dhcpcd: timed out
dhcpcd: allowing 8 seconds for IPv4LL timeout
dhcpcd: timed out

After this if I issue "dhcpcd eth0" again, it comes up with an IP address. What gives??

GazL 08-22-2010 09:49 AM

Yep. I'm getting the same thing.

When I run a "/etc/rc.d/rc.inet1 restart" I still get the no-carrier/timeout message. I haven't tried using dhcpcd directly from the command line yet. My work around was to use "dhclient -4 eth0" instead of dhcpcd.

I'm wondering if there's something under /sys they've removed in .35 that the slackware version of dhcpcd doesn't like.

update: when I run dhcpcd directly from the command line i get:
Code:

root@nix:~# dhcpcd
dhcpcd: version 5.2.2 starting
dhcpcd: no interfaces have a carrier
dhcpcd: forking to background

so mine is at least being consistent.

GazL 08-22-2010 10:29 AM

found this:
http://roy.marples.name/archives/dhc...2010/0218.html

So, update is needed for .35

dimm0k 08-22-2010 11:03 AM

Quote:

Originally Posted by GazL (Post 4074177)
found this:
http://roy.marples.name/archives/dhc...2010/0218.html

So, update is needed for .35

Good eye/research! Got the 5.2.7 source and the SlackBuild script and like that, it's working without a hitch again! Had to recreate the little diff patch, but other than this is SOLVED! Thanks all, GazL!

GazL 08-22-2010 11:25 AM

Just done the same mate. :)

What did you change in the diff? from what I saw the 50-ntp.conf shipped with 5.2.7 now includes the slackware ntp restart clause in the if, so unless I'm missing something then it shouldn't be necessary to patch, though I suppose it doesn't hurt to make it slackware specific.

Code:

if type rc-service >/dev/null 2>&1 && rc-service --exists ntpd; then
        ntpd_restart_cmd="rc-service ntpd -- -Ds restart"
elif type invoke-rc.d >/dev/null 2>&1; then
        if [ -x /etc/init.d/ntp ]; then
                ntpd_restart_cmd="invoke-rc.d ntp try-restart"
        fi
        # Debian as a seperate file for DHCP config to avoid stamping on
        # the master.
        [ -e /var/lib/ntp ] || mkdir /var/lib/ntp
        : ${NTP_DHCP_CONF:=/var/lib/ntp/ntp.conf.dhcp}
elif [ -x /etc/rc.d/ntpd ]; then
        ntpd_restart_cmd="/etc/rc.d/ntpd status >/dev/null 2>&1 && /etc/rc.d/ntp
d restart"

elif [ -x /usr/local/etc/rc.d/ntpd ]; then
        ntpd_restart_cmd="/usr/local/etc/rc.d/ntpd status >/dev/null 2>&1 && /us
r/local/etc/rc.d/ntpd restart"
fi


dimm0k 08-22-2010 08:51 PM

GazL, while I removed that whole section, if you look at that lines you have highlighted in red, you will see that it still needs a little editing. There is no /etc/rc.d/ntpd in Slackware, but instead it is /etc/rc.d/rc.ntpd.

GazL 08-23-2010 04:49 AM

D'oh! I completely missed that!

Well, if it needs changing anyway might as well remove those unneeded if clauses too while I'm about it.

Thanks for putting me right.


All times are GMT -5. The time now is 03:19 AM.