The MPLS WG Archive
[Date Prev][Date Next][Thread Prev][Thread Next] [Date Index][Thread Index][Author Index][Subject Index]
Doubts in "draft-ietf-mpls-lsp-query-00.txt"
-
From: "Antonela Paraschiv" <antonela@nortelnetworks.com>
-
Date: Wed, 25 Oct 2000 11:23:15 -0400
-
CC: mpls@UU.NET
Thank you for comments. Please find my answers below.
Best regards,
Antonela Paraschiv
Manikantan S wrote:
Hello Paraschiv
I read the draft "draft-ietf-mpls-lsp-query-00.txt".
I have a few doubts and can you please clarify them?
Thanks.
1) Label TLV is used in the Query message and the Query Label
TLV as Optional parameters in the Query Reply message.
Both the TLVs contain a set of Generalized Label TLVs.
I am not clear as why do we require Query Label TLV and
Label TLV i.e., two TLVs when the contents are the same?
You are right. The only reason I created 2 different TLVs was to
make it clearer . I could certainly remove one of them.
2) ER TLV has been specified as Optional parameters in
Query Message and in Query-reply message. I understand
that this TLV's usage is to keep track of the hops.
Path Vector TLV (in LDP), RRO in (RSVP-TE)helps in
providing this information too. What is the significance
of using ER TLV?
The ER TLV provides the proper format to encode a list of interface addresses.
Its only purpose is to carry the list of hops. It is not used for loop
detection as the Path Vector TLV.
3) In Section 5.2 Page 9, we have
--------------------------------------------------------------------
Upon receiving a Query Message, an LSR decodes the label
to identify
which LSP is queried. If it cannot find the LSP which
is using the
label, it sends back a Notification message.
--------------------------------------------------------------------
What is the status value that should be indicated in this
case?
4) In Section 6.2 Page 12, we have
-------------------------------------------------------------------
A Query-Reply message is initiated by an egress node which
receives a
Query message, if the egress is able to identify the queried
LSP. If
not, the egress replies with a Notification message.
--------------------------------------------------------------------
What is the status value that should be indicated in this
case?
3 and 4 : I will add a new Status code: "No Lsp to query" with status data
= next available.
Should this be the same one as for question 3?
I noticed Following minor typos/missing info
1) Section 6.3 Page 13
we have
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|0| Query-Reply (0x0411)
| Message Length
|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
should be
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|0| Partial-Query-Reply (0x0411)|
Message Length
|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2) Section 7.2 Page 16
for "The format for the Query Label TLV " read "The format
for the Merge
Flags TLV"
3) Section 7.3 Page 16
for "The format for the Query Label TLV " read "The format
for the Label
TLV"
Thanks in advance
with best regards
mani
-----------------------------------------
S.Manikantan
Future Software Limited
480-481, Anna Salai,
Nandanam, Chennai, India.
Zip (PIN CODE) : 600 035
Phone : 91-44-4330550
Fax
: 91-44-4344157
email : manis@future.futsoft.com
-----------------------------------------
--
Antonela Paraschiv
Carrier Packet Division, Nortel Networks
Esn : 978-288-6136
E-mail: antonela@nortelnetworks.com
|