Re: Suresh Krishnan's Discuss on draft-ietf-l2tpext-keyed-ipv6-tunnel-07: (with DISCUSS)

Mark Townsley <[email protected]> Wed, 8 Feb 2017 19:20:05 +0100
Newsgroups gmane.ietf.l2tpext
Message-ID <[email protected]>
--===============1867236896858231854==
Content-Type: multipart/alternative;
 boundary=Apple-Mail-7E4A99E1-2D93-4B1A-AA82-C810F0A1DBF3
Content-Transfer-Encoding: 7bit


--Apple-Mail-7E4A99E1-2D93-4B1A-AA82-C810F0A1DBF3
Content-Type: text/plain;
	charset=utf-8
Content-Transfer-Encoding: quoted-printable


RFC 4623, section 5.1 essentially says "Try not to be in a situation where y=
ou must fragment. If you must, try to use native PW fragmentation and reasse=
mbly if available. IP frag as a last resort." - which I think is the spirit o=
f what you are trying to say here, though the all caps in #3 may be coming a=
cross as overkill to the reviewer (as if v6 host frag is somehow evil or bro=
ken - it's not, it's just hard for network gear that isn't used to this kind=
 of thing, and we'd rather avoid it).

- Mark=20


> On Feb 8, 2017, at 6:07 PM, Giles Heron <[email protected]> wrote:
>=20
> Sorry for the delay in responding.
>=20
> At any rate I don=E2=80=99t think this is an issue.  As far as I can tell t=
he mention of RFC2473 in section 4.1.4 of RFC3931 pertains to an IPv6 payloa=
d, not to IPv6 transport.  Our draft doesn=E2=80=99t discuss payload fragmen=
tation - as we=E2=80=99ve generally assumed systems are acting as LACs rathe=
r than as LNSes (i.e. just forwarding at layer 2, not routing between a L2 c=
ircuit and a =E2=80=9Chome network=E2=80=9D), though there=E2=80=99s nothing=
 to stop an implementation that routes into a keyed IPv6 tunnel from fragmen=
ting IPv4 packets before forwarding them into the tunnel, or performing L2TP=
v3 fragmentation for IPv4 or IPv6 packets.
>=20
> In terms of preference the draft is pretty clear that it=E2=80=99s:
> (1) ensure MTU is sufficient
> (2) use L2TPv3 fragmentation
> (3) NOT RECOMMENDED - use IPv6 fragmentation.
>=20
> In case 3 (as in case 2) the ingress router will fragment and the egress r=
outer will reassemble.
>=20
> or have I missed something?
>=20
> Giles
>=20
>=20
>> On 3 Nov 2016, at 04:04, Suresh Krishnan <[email protected]> w=
rote:
>>=20
>> Suresh Krishnan has entered the following ballot position for
>> draft-ietf-l2tpext-keyed-ipv6-tunnel-07: Discuss
>>=20
>> When responding, please keep the subject line intact and reply to all
>> email addresses included in the To and CC lines. (Feel free to cut this
>> introductory paragraph, however.)
>>=20
>>=20
>> Please refer to https://www.ietf.org/iesg/statement/discuss-criteria.html=

>> for more information about IESG DISCUSS and COMMENT positions.
>>=20
>>=20
>> The document, along with other ballot positions, can be found here:
>> https://datatracker.ietf.org/doc/draft-ietf-l2tpext-keyed-ipv6-tunnel/
>>=20
>>=20
>>=20
>> ----------------------------------------------------------------------
>> DISCUSS:
>> ----------------------------------------------------------------------
>>=20
>> * Section 5
>> I am having a hard time seeing how fragmentation is expected to work=20
>>=20
>>   It is NOT RECOMMENDED for routers implementing this specification to
>>   enable IPv6 fragmentation (as defined in section 4.5 of RFC2460) for
>>   keyed IP tunnels.  IP fragmentation issues for L2TPv3 are discussed
>>   in section 4.1.4 of RFC3931.
>>=20
>> And that specific section of RFC3931 recommends using RFC2473 to tunnel
>> the packets which again ends up using the RFC2460 fragment header that
>> this draft is trying to forbid.
>>=20
>> So, can you please clarify exactly what happens when the size of the
>> packet to be tunneled exceeds the MTU?
>>=20
>>=20
>>=20
>>=20
>> _______________________________________________
>> L2tpext mailing list
>> [email protected]
>> https://www.ietf.org/mailman/listinfo/l2tpext
>=20
> _______________________________________________
> L2tpext mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/l2tpext

--Apple-Mail-7E4A99E1-2D93-4B1A-AA82-C810F0A1DBF3
Content-Type: text/html;
	charset=utf-8
Content-Transfer-Encoding: quoted-printable

<html><head><meta http-equiv=3D"content-type" content=3D"text/html; charset=3D=
utf-8"></head><body dir=3D"auto"><div></div><div><br></div><div>RFC 4623, se=
ction 5.1 essentially says "Try not to be in a situation where you must frag=
ment. If you must, try to use native PW fragmentation and reassembly if avai=
lable. IP frag as a last resort." - which I think is the spirit of what you a=
re trying to say here, though the all caps in #3 may be coming across as ove=
rkill to the reviewer (as if v6 host frag is somehow evil or broken - it's n=
ot, it's just hard for network gear that isn't used to this kind of thing, a=
nd we'd rather avoid it).</div><div><br></div><div>- Mark&nbsp;</div><div><b=
r></div><div><br>On Feb 8, 2017, at 6:07 PM, Giles Heron &lt;<a href=3D"mail=
to:[email protected]">[email protected]</a>&gt; wrote:<br><br></div>=
<blockquote type=3D"cite"><div><meta http-equiv=3D"Content-Type" content=3D"=
text/html charset=3Dutf-8"><font face=3D"Courier" class=3D"">Sorry for the d=
elay in responding.</font><div class=3D""><div class=3D""><font face=3D"Cour=
ier" class=3D""><br class=3D""></font></div><div class=3D""><font face=3D"Co=
urier" class=3D"">At any rate I don=E2=80=99t think this is an issue. &nbsp;=
As far as I can tell the mention of RFC2473 in section 4.1.4 of RFC3931 pert=
ains to an IPv6 payload, not to IPv6 transport. &nbsp;Our draft&nbsp;doesn=E2=
=80=99t discuss payload&nbsp;fragmentation - as we=E2=80=99ve&nbsp;generally=
 assumed&nbsp;systems are acting as LACs rather than as LNSes (i.e. just&nbs=
p;forwarding at layer 2, not routing between a L2 circuit and a&nbsp;=E2=80=9C=
home network=E2=80=9D), though there=E2=80=99s nothing to stop an implementa=
tion that routes into a keyed IPv6 tunnel from fragmenting IPv4 packets befo=
re forwarding them into the tunnel, or performing L2TPv3 fragmentation for I=
Pv4 or IPv6 packets.</font></div><div class=3D""><font face=3D"Courier" clas=
s=3D""><br class=3D""></font></div><div class=3D""><font face=3D"Courier" cl=
ass=3D"">In terms of preference the draft is pretty clear that it=E2=80=99s:=
</font></div><div class=3D""><font face=3D"Courier" class=3D"">(1) ensure MT=
U is sufficient</font></div><div class=3D""><font face=3D"Courier" class=3D"=
">(2) use L2TPv3 fragmentation</font></div><div class=3D""><font face=3D"Cou=
rier" class=3D"">(3) NOT RECOMMENDED - use IPv6 fragmentation.</font></div><=
div class=3D""><font face=3D"Courier" class=3D""><br class=3D""></font></div=
><div class=3D""><font face=3D"Courier" class=3D"">In case 3 (as in case 2) t=
he ingress router will fragment and the egress router will reassemble.</font=
></div><div class=3D""><font face=3D"Courier" class=3D""><br class=3D""></fo=
nt></div><div class=3D""><font face=3D"Courier" class=3D"">or have I missed s=
omething?</font></div><div class=3D""><font face=3D"Courier" class=3D""><br c=
lass=3D""></font></div><div class=3D""><font face=3D"Courier" class=3D"">Gil=
es</font></div><div class=3D""><br class=3D""></div><div class=3D""><br clas=
s=3D""><div><blockquote type=3D"cite" class=3D""><div class=3D"">On 3 Nov 20=
16, at 04:04, Suresh Krishnan &lt;<a href=3D"mailto:suresh.krishnan@ericsson=
.com" class=3D"">[email protected]</a>&gt; wrote:</div><br class=3D=
"Apple-interchange-newline"><div class=3D""><div class=3D"">Suresh Krishnan h=
as entered the following ballot position for<br class=3D"">draft-ietf-l2tpex=
t-keyed-ipv6-tunnel-07: Discuss<br class=3D""><br class=3D"">When responding=
, please keep the subject line intact and reply to all<br class=3D"">email a=
ddresses included in the To and CC lines. (Feel free to cut this<br class=3D=
"">introductory paragraph, however.)<br class=3D""><br class=3D""><br class=3D=
"">Please refer to <a href=3D"https://www.ietf.org/iesg/statement/discuss-cr=
iteria.html" class=3D"">https://www.ietf.org/iesg/statement/discuss-criteria=
.html</a><br class=3D"">for more information about IESG DISCUSS and COMMENT p=
ositions.<br class=3D""><br class=3D""><br class=3D"">The document, along wi=
th other ballot positions, can be found here:<br class=3D""><a href=3D"https=
://datatracker.ietf.org/doc/draft-ietf-l2tpext-keyed-ipv6-tunnel/" class=3D"=
">https://datatracker.ietf.org/doc/draft-ietf-l2tpext-keyed-ipv6-tunnel/</a>=
<br class=3D""><br class=3D""><br class=3D""><br class=3D"">----------------=
------------------------------------------------------<br class=3D"">DISCUSS=
:<br class=3D"">------------------------------------------------------------=
----------<br class=3D""><br class=3D"">* Section 5<br class=3D"">I am havin=
g a hard time seeing how fragmentation is expected to work <br class=3D""><b=
r class=3D""> &nbsp;&nbsp;It is NOT RECOMMENDED for routers implementing thi=
s specification to<br class=3D""> &nbsp;&nbsp;enable IPv6 fragmentation (as d=
efined in section 4.5 of RFC2460) for<br class=3D""> &nbsp;&nbsp;keyed IP tu=
nnels. &nbsp;IP fragmentation issues for L2TPv3 are discussed<br class=3D"">=
 &nbsp;&nbsp;in section 4.1.4 of RFC3931.<br class=3D""><br class=3D"">And t=
hat specific section of RFC3931 recommends using RFC2473 to tunnel<br class=3D=
"">the packets which again ends up using the RFC2460 fragment header that<br=
 class=3D"">this draft is trying to forbid.<br class=3D""><br class=3D"">So,=
 can you please clarify exactly what happens when the size of the<br class=3D=
"">packet to be tunneled exceeds the MTU?<br class=3D""><br class=3D""><br c=
lass=3D""><br class=3D""><br class=3D"">____________________________________=
___________<br class=3D"">L2tpext mailing list<br class=3D""><a href=3D"mail=
to:[email protected]">[email protected]</a><br class=3D""><a href=3D"https://w=
ww.ietf.org/mailman/listinfo/l2tpext">https://www.ietf.org/mailman/listinfo/=
l2tpext</a><br class=3D""></div></div></blockquote></div><br class=3D""></di=
v></div></div></blockquote><blockquote type=3D"cite"><div><span>____________=
___________________________________</span><br><span>L2tpext mailing list</sp=
an><br><span><a href=3D"mailto:[email protected]">[email protected]</a></span>=
<br><span><a href=3D"https://www.ietf.org/mailman/listinfo/l2tpext">https://=
www.ietf.org/mailman/listinfo/l2tpext</a></span><br></div></blockquote></bod=
y></html>=

--Apple-Mail-7E4A99E1-2D93-4B1A-AA82-C810F0A1DBF3--


--===============1867236896858231854==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
L2tpext mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/l2tpext

--===============1867236896858231854==--