Tianran Zhou wrote:
> What if we ask for symmetric request and response? I.e., carry reserved
> bytes in the request packet.
That would make me happier actually.
--
Michael Richardson. o O ( IPv6 IøT consulting )
Sandelman Software Works Inc, Ottawa and Worldwide
si
Michael,
Please note the following sentence in the security considerations
section: "the amplification effect in this case is similar to ICMPv6
error message, and specifically similar to Traceroute."
Sending a Loopback causes the exact same amplification as invoking the
last packet of Traceroute,
Hi Michael,
What if we ask for symmetric request and response? I.e., carry reserved bytes
in the request packet.
Tianran
-Original Message-
From: ipv6 [mailto:ipv6-boun...@ietf.org] On Behalf Of Michael Richardson
Sent: Thursday, June 8, 2023 12:59 AM
To: int-area@ietf.org; i...@ietf.o
Despite what the Security Considerations suggests, this still looks ripe for
use as an amplication attack to me.
--
Michael Richardson. o O ( IPv6 IøT consulting )
Sandelman Software Works Inc, Ottawa and Worldwide
signature.asc
Description: PGP signature
On 2023-06-07 11:42 UTC, Luigi IANNONE
wrote:
> For FreeBSD it looks like code MUST be 0 or the packet is dropped:
>
this is coming from KAME so everything that derives from there has
this. E.g. it's also present in OpenBSD & NetBSD.
> (file netinet6/icmp6.c line 550:
> http://fxr.watson.org/f
)
goto badcode;
Ciao
L.
> -Original Message-
> From: Int-area On Behalf Of Justin Iurman
> Sent: Wednesday, 7 June 2023 13:24
> To: Erik Kline ; Tal Mizrahi
> Cc: int-area@ietf.org; i...@ietf.org
> Subject: Re: [Int-area] [IPv6] New Draft - ICMPv6 Loopback
&g
On 6/7/23 08:06, Erik Kline wrote:
Poking around the Linux kernel source, my reading of net/ipv6/icmp.c's
icmpv6_rcv() is that it checks the type byte before dispatching to
icmpv6_echo_reply(), and inside icmpv6_echo_reply() I'm not seeing any
checking of the code byte, so I'd assume (without tes
Poking around the Linux kernel source, my reading of net/ipv6/icmp.c's
icmpv6_rcv() is that it checks the type byte before dispatching to
icmpv6_echo_reply(), and inside icmpv6_echo_reply() I'm not seeing any
checking of the code byte, so I'd assume (without testing) that it
just constructs a norma
