bind() fd 6, family 28, port 123 at boot time

Michael Proto mike at jellydonut.org
Thu Apr 6 11:57:05 UTC 2006


Miguel wrote:
> Hi, looking at the messages i see errors with ntp everytime i reboot my 
> server:
> 
> shiva2# tail /var/log/messages
> Apr  5 15:32:46 shiva2 kernel: Trying to mount root from ufs:/dev/da0s1a
> Apr  5 15:32:46 shiva2 ntpd[385]: ntpd 4.2.0-a Tue Mar 14 04:43:54 UTC 
> 2006 (1)
> Apr  5 15:32:46 shiva2 ntpd[385]: bind() fd 6, family 28, port 123, addr 
> fe80:1::20b:cdff:fe42:3d63, in6_is_addr_multicast=0 flags=0 fails: Can't 
> assign requested address
> Apr  5 15:32:46 shiva2 ntpd[385]: bind() fd 7, family 28, port 123, addr 
> fe80:2::20b:cdff:fe42:3d62, in6_is_addr_multicast=0 flags=0 fails: Can't 
> assign requested address
> Apr  5 15:32:47 shiva2 kernel: bge0: link state changed to UP
> Apr  5 15:32:48 shiva2 kernel: bge1: link state changed to UP
> Apr  5 15:33:51 shiva2 ntpd[385]: sendto(200.138.151.150): No route to
host
> Apr  5 15:36:03 shiva2 ntpd[385]: kernel time sync disabled 2041
> Apr  5 15:37:08 shiva2 ntpd[385]: kernel time sync enabled 2001
> 
> 
> someone suggested in the archives that disabling ipv6 could resolv the 
> problem, i already commented out options inet6 and recompiled the 
> kernel, i think the time is syncronized thoug.
> what can i check?
> ---
> Miguel
> _______________________________________________
> freebsd-stable at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-stable
> To unsubscribe, send any mail to "freebsd-stable-unsubscribe at freebsd.org"
> 

I've seen the "sendto... no route to host" error myself, although I'm
not sure if you are seeing the same behavior I was. In my case, my
external interface was a Netgraph tunnel interface (ng0) ports/net/mpd
which would sometimes die at the hands of my Internet provider. When the
tunnel would die and then get rebuilt, ntpd would give this error
because it was still bound to the old tunnel. My fix was to create an
mpd startup script that would restart ntpd whenever ng0 was recreated.

In any case, perhaps it is due to your Internet connection not being
available when ntpd starts? You could verify that your time is still
being synced by running ntpdc at the command line, then running "peers"
at the prompt.


-Proto


More information about the freebsd-stable mailing list