Date: Tue, 19 Nov 2002 21:47:36 +0100 (CET) From: BOUWSMA Beery <freebsd-misuser@ipv6.netscum.dyndns.dk> To: freebsd-net@freebsd.org Subject: Question about rt_llinfo panic in netinet6 Message-ID: <200211192047.gAJKlaQ00550@MAIL.NetScum.DynDNS.dK>
next in thread | raw e-mail | index | archive | help
[IPv6-only address above; strip the obvious for IPv4-only mail]
Howdy,
I'm able to readily reproduce a panic with -stable that seems to
be caused by my IPv6 script interacting with dhclient, that normally
happens after the IPv4 address granted by the dhcp server expires.
An infrequent occurrence, but I can simulate it pretty easily.
The panic is in a sanity check in the src/sys/netinet6/nd6.c file:
456 /* sanity check */
[...]
459 if (rt->rt_llinfo && (struct llinfo_nd6 *)rt->rt_llinfo
459 != ln)
460 panic("rt_llinfo(%p) is not equal to ln(%p)\n",
461 rt->rt_llinfo, ln);
I'm unclear, just what is this check, and in what cases is it
likely to happen and induce this panic?
I haven't figured out exactly what my stf(4)-based IPv6 script
that also plays with faith and aliases might be doing to result
in this panic, but I'll keep poking at it.
What happens when the dhcp lease expires is that my interface IP
address is set to 0. Then when I get connectivity to the dhcp
server again and re-obtain the old IP address (or, apparently, a
new one as well), that is enough to invoke this panic. Thus I
can manually set the interface IP to 0.0.0.0 and restart dhclient
and get the same panic. The intent of my script is to assign
initial IPv6 addresses, or update them in the event the IPv4
address changes out from under me, but my initial debuggery makes
it seem the panic happens later, when I try to get the info about
the change to update my DNS zone files, or something.
thanks,
barry bouwsma
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-net" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200211192047.gAJKlaQ00550>