5

I'm one of the many confused by the new several minute delay when booting an upgraded Ubuntu 11.10 system. I have two NICs, both with static IP addresses, plus lo and nothing else (that I know of). So (a) what is the system waiting for? and (b) why does it think it failed and warn that is it is going to proceed with "broken" network configuration? This is EXTREMELY disconcerting even though every network thing I've known to check appears to be working fine. FWIW, here's my interfaces file (actual addresses obscured):

# The loopback network interface
auto lo
iface lo inet loopback

# The primary network interface
auto eth1
iface eth1 inet static
        address w.x.y.z
        netmask 255.255.255.248
        gateway w.x.y.z

# The extra network interface
auto eth0
iface eth0 inet static
        address w.x.y.z
        netmask 255.255.255.0
        gateway w.x.y.z

Oh, I'm not running under VMware, this is a dedicated PC, but I've followed the instructions about changing /var/run and /var/lock into symlinks anyway, but of course that didn't help.

3
  • No answer yet but I have the same problem. I was noticing that Oneiric took much much longer to boot to desktop, but eventually decided it must be waiting to finishing network settings. Oct 21, 2011 at 13:02
  • FWIW, I thought this might have something to do with IPv6 and, since IPv6 isn't yet available through our provider, I turned it all off. Unfortunately, that didn't help at all. And on a related note, I must manually add the default gateway for eth1 every time the machine boots, too. For some reason, that one is lost on every boot, though the eth0 gateway is always correct. I really wish 11.10 hadn't broken the networking so badly. Sigh. Oct 24, 2011 at 16:04
  • It's still pretty fresh out, so hopefully updates help with that. Oct 25, 2011 at 0:19

2 Answers 2

1

Remove the "auto" lines (comment them out if you're not confident). This is a bug, either #839595, “failsafe.conf's 30 second time out is too low” or #845914, “Failsafe boot delay causes a real delay on every boot” (now marked as a dupe of #847782, “installer writes a permanent ethernet entry in interfaces file”). That edit fixes it.

0

Your boot error may have nothing to do with networking at all (and may not even be a boot error!) For example, my similar symptoms were due to https://bugs.launchpad.net/bugs/858122. (example only!)

Look through your dmesg and boot syslog. Find those big pauses, and post the couple lines before each pause...or Google them.

7
  • In dmesg the pause comes just after the two network cards are activated and ssh terminated (don't know why that would be): (in next comment because too few characters are allowed in each) Dec 7, 2011 at 20:15
  • [ 21.539811] 8139too 0000:00:09.0: eth0: link up, 100Mbps, full-duplex, lpa 0x45E1 [ 21.543118] via-rhine 0000:00:12.0: eth1: link up, 10Mbps, half-duplex, lpa 0x0021 [ 24.780925] init: ssh main process (564) terminated with status 255 [ 142.228746] audit_printk_skb: 6 callbacks suppressed [ 142.228754] type=1400 audit(1322484952.766:14): apparmor="STATUS" operation="profile_replace" name="/sbin/dhclient" pid=728 comm="apparm$ Dec 7, 2011 at 20:15
  • There seems to be no way to have code in comments, or at least the backticks and four leading spaces, and two trailing spaces things in the docs aren't working for me. The previous message is the requested few lines from dmesg. Dec 7, 2011 at 20:19
  • boot.log doesn't contain any time tags (or anything indicating possible errors) so it was no help and syslog doesn't go back far enough. When I am next able to boot the machine, I'll try to remember to take a look to see if there is anything there. Dec 7, 2011 at 20:20
  • Are you running ssh or sshd? If the latter, check /var/auth/log for sshd errors to understand why it died.
    – user535733
    Dec 14, 2011 at 20:11

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .