Network Dies every 5 - 10 minutes, Start/Stop Fixes

Waggoneer · #1 15th November 2006, 04:53 PM

I am on a HP laptop at work. Every 5 - 10 minutes my network connection stops working and if I do deactivate/reactivate it will start working again.

Here is what is in the log file, during that time period.

Quote:

Nov 15 09:46:59 LLX1002481 NET[4511]: /sbin/dhclient-script : updated /etc/resolv.conf
Nov 15 09:46:59 LLX1002481 avahi-daemon[2281]: Withdrawing address record for 10.81.1.81 on eth0.
Nov 15 09:46:59 LLX1002481 avahi-daemon[2281]: Leaving mDNS multicast group on interface eth0.IPv4 with address 10.81.1.81.
Nov 15 09:46:59 LLX1002481 avahi-daemon[2281]: iface.c: interface_mdns_mcast_join() called but no local address available.
Nov 15 09:46:59 LLX1002481 avahi-daemon[2281]: Interface eth0.IPv4 no longer relevant for mDNS.
Nov 15 09:46:59 LLX1002481 avahi-daemon[2281]: Withdrawing address record for fe80::208:2ff:fe6c:16ad on eth0.
Nov 15 09:46:59 LLX1002481 avahi-daemon[2281]: Leaving mDNS multicast group on interface eth0.IPv6 with address fe80::208:2ff:fe6c:16ad.
Nov 15 09:46:59 LLX1002481 avahi-daemon[2281]: iface.c: interface_mdns_mcast_join() called but no local address available.
Nov 15 09:46:59 LLX1002481 avahi-daemon[2281]: Interface eth0.IPv6 no longer relevant for mDNS.
Nov 15 09:47:01 LLX1002481 kernel: ADDRCONF(NETDEV_UP): eth0: link is not ready
Nov 15 09:47:01 LLX1002481 kernel: e100: eth0: e100_watchdog: link up, 100Mbps, full-duplex
Nov 15 09:47:01 LLX1002481 kernel: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
Nov 15 09:47:01 LLX1002481 dhclient: DHCPREQUEST on eth0 to 255.255.255.255 port 67
Nov 15 09:47:01 LLX1002481 dhclient: DHCPACK from 10.81.1.27
Nov 15 09:47:02 LLX1002481 avahi-daemon[2281]: New relevant interface eth0.IPv4 for mDNS.
Nov 15 09:47:02 LLX1002481 avahi-daemon[2281]: Joining mDNS multicast group on interface eth0.IPv4 with address 10.81.1.81.
Nov 15 09:47:02 LLX1002481 avahi-daemon[2281]: Registering new address record for 10.81.1.81 on eth0.
Nov 15 09:47:02 LLX1002481 avahi-daemon[2281]: Withdrawing address record for 10.81.1.81 on eth0.
Nov 15 09:47:02 LLX1002481 avahi-daemon[2281]: Leaving mDNS multicast group on interface eth0.IPv4 with address 10.81.1.81.
Nov 15 09:47:02 LLX1002481 avahi-daemon[2281]: iface.c: interface_mdns_mcast_join() called but no local address available.
Nov 15 09:47:02 LLX1002481 avahi-daemon[2281]: Interface eth0.IPv4 no longer relevant for mDNS.
Nov 15 09:47:02 LLX1002481 avahi-daemon[2281]: New relevant interface eth0.IPv4 for mDNS.
Nov 15 09:47:02 LLX1002481 avahi-daemon[2281]: Joining mDNS multicast group on interface eth0.IPv4 with address 10.81.1.81.
Nov 15 09:47:02 LLX1002481 avahi-daemon[2281]: Registering new address record for 10.81.1.81 on eth0.
Nov 15 09:47:02 LLX1002481 NET[4661]: /sbin/dhclient-script : updated /etc/resolv.conf
Nov 15 09:47:02 LLX1002481 dhclient: bound to 10.81.1.81 -- renewal in 328459 seconds.
Nov 15 09:47:03 LLX1002481 avahi-daemon[2281]: New relevant interface eth0.IPv6 for mDNS.
Nov 15 09:47:03 LLX1002481 avahi-daemon[2281]: Joining mDNS multicast group on interface eth0.IPv6 with address fe80::208:2ff:fe6c:16ad.
Nov 15 09:47:03 LLX1002481 avahi-daemon[2281]: Registering new address record for fe80::208:2ff:fe6c:16ad on eth0.

What could be the issue? I asked some of our sysadmin's if I needed to do anything to get fedora working in our domain. They use unubuntu with no problems.

tomcat · #2 15th November 2006, 05:16 PM

Disable IPV6 on your box* and add valid DNS entries from your ISP to your /etc/resolv.conf file.

*= add the following lines to /etc/modprobe.conf:
alias net-pf-10 off
alias ipv6 off

Waggoneer · #3 15th November 2006, 05:35 PM

Ok, I made the changes to modprobe.conf.
My resolv.conf had the correct DNS entries, already.
Then I rebooted. Seemed to boot faster, too.
I will surf around a little and see if the connection is stable.

Thanks for the reply.

Waggoneer · #4 15th November 2006, 05:39 PM

Problem persists. Here is the relevent log info....

During start-up

Quote:

Nov 15 10:28:08 LLX1002481 avahi-daemon[2277]: Joining mDNS multicast group on interface eth0.IPv4 with address 10.81.1.81.
Nov 15 10:28:08 LLX1002481 avahi-daemon[2277]: Network interface enumeration completed.
Nov 15 10:28:08 LLX1002481 avahi-daemon[2277]: Registering new address record for 10.81.1.81 on eth0.
Nov 15 10:28:08 LLX1002481 avahi-daemon[2277]: Registering HINFO record with values 'I686'/'LINUX'.
Nov 15 10:28:09 LLX1002481 avahi-daemon[2277]: Server startup complete. Host name is LLX1002481.local. Local service cookie is 591013232.

during deactivate/activate

Quote:

Nov 15 10:34:45 LLX1002481 NET[3076]: /sbin/dhclient-script : updated /etc/resolv.conf
Nov 15 10:34:45 LLX1002481 avahi-daemon[2277]: Withdrawing address record for 10.81.1.81 on eth0.
Nov 15 10:34:45 LLX1002481 avahi-daemon[2277]: Leaving mDNS multicast group on interface eth0.IPv4 with address 10.81.1.81.
Nov 15 10:34:45 LLX1002481 avahi-daemon[2277]: iface.c: interface_mdns_mcast_join() called but no local address available.
Nov 15 10:34:45 LLX1002481 avahi-daemon[2277]: Interface eth0.IPv4 no longer relevant for mDNS.
Nov 15 10:34:47 LLX1002481 kernel: e100: eth0: e100_watchdog: link up, 100Mbps, full-duplex
Nov 15 10:34:47 LLX1002481 dhclient: DHCPREQUEST on eth0 to 255.255.255.255 port 67
Nov 15 10:34:47 LLX1002481 dhclient: DHCPACK from 10.81.1.27
Nov 15 10:34:47 LLX1002481 avahi-daemon[2277]: New relevant interface eth0.IPv4 for mDNS.
Nov 15 10:34:47 LLX1002481 avahi-daemon[2277]: Joining mDNS multicast group on interface eth0.IPv4 with address 10.81.1.81.
Nov 15 10:34:47 LLX1002481 avahi-daemon[2277]: Registering new address record for 10.81.1.81 on eth0.
Nov 15 10:34:47 LLX1002481 avahi-daemon[2277]: Withdrawing address record for 10.81.1.81 on eth0.
Nov 15 10:34:47 LLX1002481 avahi-daemon[2277]: Leaving mDNS multicast group on interface eth0.IPv4 with address 10.81.1.81.
Nov 15 10:34:47 LLX1002481 avahi-daemon[2277]: iface.c: interface_mdns_mcast_join() called but no local address available.
Nov 15 10:34:47 LLX1002481 avahi-daemon[2277]: Interface eth0.IPv4 no longer relevant for mDNS.
Nov 15 10:34:47 LLX1002481 avahi-daemon[2277]: New relevant interface eth0.IPv4 for mDNS.
Nov 15 10:34:47 LLX1002481 avahi-daemon[2277]: Joining mDNS multicast group on interface eth0.IPv4 with address 10.81.1.81.
Nov 15 10:34:47 LLX1002481 avahi-daemon[2277]: Registering new address record for 10.81.1.81 on eth0.
Nov 15 10:34:47 LLX1002481 NET[3225]: /sbin/dhclient-script : updated /etc/resolv.conf
Nov 15 10:34:47 LLX1002481 dhclient: bound to 10.81.1.81 -- renewal in 272533 seconds.

Anything else I should check?

tomcat · #5 15th November 2006, 09:22 PM

Please post the output of

cat /etc/resolv.conf

stevea · #6 16th November 2006, 10:14 AM

Do the network link LEDs blink off when you lose the connection ? I suspect so. Normally loss/restore of connectivity (link) causes these messages and similar.
Nov 15 09:47:01 LLX1002481 kernel: ADDRCONF(NETDEV_UP): eth0: link is not ready
Nov 15 09:47:01 LLX1002481 kernel: e100: eth0: e100_watchdog: link up, 100Mbps, full-duplex
Nov 15 09:47:01 LLX1002481 kernel: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
This could mean a bad cable or a bad slot i nthe corporate switch.

It is EXTREMELY unlikely that /etc/resolv.conf is related to the problem as it is an upper protocol and you are losing connectivity. Find an absolute IP you can always ping (like ping 4.2.2.2) and *if* you can still ping it when you are having problems then *maybe resolv.conf is the problem. Most likely not.

Waggoneer · #7 17th November 2006, 03:42 PM

Here is the requested output. One thing is, when my network is down, I can't ping via ip addy. I don't think it is a resolution issue.
Also, I have been using this laptop, cable, port for a while and had no problems. The issue happened as soon as I switched from XP to FC6.

Quote:

[root@LLX1002481 Jason]# cat /etc/resolv.conf
; generated by /sbin/dhclient-script
search unisource.corp
nameserver 10.81.1.38
nameserver 10.128.10.64

Waggoneer · #8 17th November 2006, 08:05 PM

==SOLVED==
Stevea was the closest to the right answer. I just found out that somehow my laptop and a video conference device were using the same IP addy.
Every five minutes the device would knock me off, and then when I would disable/enable the port it would kill the video conference.
They are researching how I got that IP in the first place, since my client is set to DHCP.
Thanks for the help!