| Re: Issue 151: MTU discovery doesn't look right | <– Date –> <– Thread –> |
|
From: Pat Calhoun (pacalhou) (pcalhoun |
|
| Date: Thu, 14 Aug 2008 10:37:38 -0700 (PDT) | |
Given the lack of comments, or objections, to the proposed text since
July 28th, I will consider this issue closed. Fixed in
draft-ietf-capwap-protocol-specification-12.txt
PatC
-----Original Message-----
From: Pat Calhoun (pacalhou)
Sent: Monday, July 28, 2008 12:43 PM
To: capwap [at] frascone.com
Cc: Pasi.Eronen [at] nokia.com
Subject: [Capwap] Issue 151: MTU discovery doesn't look right
Pasi's comment was:
3.5 seems to assume that after the WTP has discovered the AC it
wishes to
communicate with, RFC 1191/1981/4821 would provide it with the MTU,
and then
the WTP would establish the CAPWAP session. But those RFCs don't
specify e.g.
what packets would be used for probing for the MTU.
I have taken a look at the RFCs, and he is correct. It looks like we had
removed from functionality from the CAPWAP protocol prematurely some
time ago. So I re-instated the MTU Discovery Padding message element
(glad we had only commented out the text :).
Here is the new text:
<text>
3.5. MTU Discovery
Once a WTP has discovered the AC it wishes to establish a CAPWAP
session with, it SHOULD perform a Path MTU (PMTU) discovery. One
recommendation for performing PMTU discovery is to have the WTP
transmit Discovery Request (see Section 5.1) messages, and include
the MTU Discovery Padding message element (see Section 4.6.33). The
actual procedures used for PMTU discovery are described in [RFC1191],
for IPv4, or for IPv6 [RFC1981] SHOULD be used. Alternatively,
implementers MAY use the procedures defined in [RFC4821]. The WTP
SHOULD also periodically re-evaluate the PMTU using the guidelines
provided in these two RFCs, using the Primary Discovery Request (see
Section 5.3) along with the MTU Discovery Padding message element
(see Section 4.6.33). When the MTU is initially known, or updated in
the case where an existing session already exists, the discovered
PMTU is used to configure the DTLS component (see Section 2.3.2.1),
while non-DTLS frames need to be fragmented to fit the MTU, defined
in Section 3.4.
4.6. CAPWAP Protocol Message Elements
[...]
MTU Discovery Padding 52
4.6.33. MTU Discovery Padding
The MTU Discovery Padding message element is used as padding to
perform MTU discovery, and MUST contain octets of value 0xFF, of any
length.
0
0 1 2 3 4 5 6 7
+-+-+-+-+-+-+-+-+
| Padding...
+-+-+-+-+-+-+-+-
Type: 52 for MTU Discovery Padding
Length: variable
Pad: A variable length pad, filled with the value 0xFF.
5.1. Discovery Request Message
[...]
The following message elements MAY be included in the Discovery
Request message:
o MTU Discovery Padding, see Section 4.6.33
5.3. Primary Discovery Request Message
The Primary Discovery Request message is sent by the WTP to determine
whether its preferred (or primary) AC is available or to performa a
Path MTU Discovery (see section Section 3.5.
[...]
The following message elements MAY be included in the Primary
Discovery Request message:
o MTU Discovery Padding, see Section 4.6.33 </text>
PatC
_________________________________________________________________
To unsubscribe or modify your subscription options, please visit:
http://lists.frascone.com/mailman/listinfo/capwap
Archives: http://lists.frascone.com/pipermail/capwap
-
Issue 151: MTU discovery doesn't look right Pat Calhoun (pacalhou), July 28 2008
- Re: Issue 151: MTU discovery doesn't look right Pat Calhoun (pacalhou), August 14 2008
Results generated by Tiger Technologies using MHonArc.