Yet another recent strangeness that started sometime last week...
My system has a single, wired NIC (eth0) configured to be started at
boot. It is also configured to use DHCP and to get its hostname from
the DHCP server.
For the last few days, when I boot, the interface comes up, but the
hostname is set to localhost.localdomain, and there are no messages in
syslog indicating that any DHCP requests are being sent. I get to the
network just fine, but ifconfig shows:
eth0 Link encap:Ethernet HWaddr 00:0C:41:EE:E2:E1
inet6 addr: fe80::20c:41ff:feee:e2e1/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:25770 errors:0 dropped:0 overruns:0 frame:0
TX packets:24117 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:8936889 (8.5 MiB) TX bytes:3339309 (3.1 MiB)
Interrupt:21 Base address:0xe800
i. e. eth0 has no ipv4 IP address. However, syslog *does* contain the line:
Jan 24 11:10:55 localhost kernel: eth0: no IPv6 routers present
So to whom is he talking ? Moreover, looking at ifcfg-eth0:
DEVICE=eth0
BOOTPROTO=dhcp
ONBOOT=yes
METRIC=10
MII_NOT_SUPPORTED=no
USERCTL=no
RESOLV_MODS=no
IPV6INIT=no
IPV6TO4INIT=no
ACCOUNTING=no
NM_CONTROLLED=no
DHCP_CLIENT=dhclient
NEEDHOSTNAME=yes
PEERDNS=yes
PEERYP=yes
PEERNTPD=no
IPv6 seems to be disabled completely.
If I ifdown/ifup eth0, things go back to normal:
eth0 Link encap:Ethernet HWaddr 00:0C:41:EE:E2:E1
inet addr:192.168.3.102 Bcast:192.168.3.255 Mask:255.255.255.0
inet6 addr: fe80::20c:41ff:feee:e2e1/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:25979 errors:0 dropped:0 overruns:0 frame:0
TX packets:24333 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:8971923 (8.5 MiB) TX bytes:3392939 (3.2 MiB)
Interrupt:21 Base address:0xe800
So what has changed in the boot-time interface starting procedure ?
