On Feb 10, 2014, at 1:17 PM, Richard PALO <[email protected]> wrote:
> This is the first time I've come across this:
>> panic[cpu1]/thread=ffffff02e79e7880:
>> assertion failed: !IN6_IS_ADDR_V4MAPPED(dst_addr), file:
>> ../../common/inet/ip/ip6.c, line: 1984
<SNIP!>
Richard provided me a coredump, and I've filed bug 4596:
https://www.illumos.org/issues/4596
against it.
What happened was his program (firefox) was using not-quite-fully-initialized
sockaddr_in6 structures, and his __sin6_src_id field was filled in with a
non-zero value which triggered the aforementioned bug. The proper fix is for
us to be more robust in these situations.
BTW, on a NON-debug kernel, bits black-hole for localhost, and WILL produce
v4-mapped IPv6 source addresses for packets destined for IPv6 systems. Our own
input code should guard against these as malicious and drop 'em, but I'm not so
sure about it for other hosts, save MacOS 10.9 (which also drops). The test
program attached to the above bug can and should be used to find out more.
Dan
-------------------------------------------
illumos-discuss
Archives: https://www.listbox.com/member/archive/182180/=now
RSS Feed: https://www.listbox.com/member/archive/rss/182180/21175430-2e6923be
Modify Your Subscription:
https://www.listbox.com/member/?member_id=21175430&id_secret=21175430-6a77cda4
Powered by Listbox: http://www.listbox.com
