RFC 5818
Data Channel Status Confirmation Extensions for the Link Management Protocol
RFC 5818 📜🦋
Data Channel Status Confirmation Extensions for the Link Management Protocol 💖
- Number: RFC 5818 ✨
- Authors: D. Li and H. Xu (both Huawei), S. Bardalai (Fujitsu), J. Meuric (France Telecom), and D. Caviglia (Ericsson) 🫶
- Category: Standards Track 👑
- ISSN: 2070-1721 🌈
- Date: April 2010 🍓
- Obsoletes / Updates: The title block doesn't list anything this one obsoletes or updates. 🙈
The IETF is the group that writes the Internet's rulebook, so this is an official, real-deal standard. 💅
Abstract 🌸
okay so this doc adds a few simple extras to the Link Management Protocol (LMP) 💜. The goal is a control plane tool that helps find stranded resources (channels that are stuck in limbo 😭). Adjacent Label-Switching Routers (LSRs) can confirm data channel statuses with each other, and if they spot a mismatch, they can trigger a heads-up to the management plane 🔥. LMP already checks that the data plane is connected, so it's a totally natural fit for this. it's giving "of course it goes here" 🦋
Status of This Memo 🎀
This is an Internet Standards Track document. 💖
This document is a product of the Internet Engineering Task Force (IETF) 🌈. It represents the consensus of the IETF community, has had public review, and was approved for publication by the Internet Engineering Steering Group (IESG) 👑. You can read more about Internet Standards in Section 2 of RFC 5741 ✨.
Info about the current status of this document, any errata, and how to send feedback is over at http://www.rfc-editor.org/info/rfc5818 🧚
Copyright Notice 💅
Copyright (c) 2010 IETF Trust and the persons identified as the document authors. All rights reserved. 🫶
This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (http://trustee.ietf.org/license-info), as in effect on the date of publication. Please read these carefully, because they spell out your rights and restrictions for this document 📜. Code Components extracted from this document must include Simplified BSD License text, as described in Section 4.e of the Trust Legal Provisions. They're provided without warranty, as described in the Simplified BSD License 🌸.
Table of Contents 🗂️
1. Introduction ....................................................3
2. Specification of Requirements ...................................4
3. Problem Explanation .............................................4
3.1. Mismatch Caused by Manual Configuration ....................4
3.2. Mismatch Caused by LSP Deletion ............................5
3.3. Failed Resources ...........................................6
4. Motivation ......................................................6
5. Extensions to LMP ...............................................7
5.1. Confirm Data Channel Status Messages .......................7
5.1.1. ConfirmDataChannelStatus Messages ...................8
5.1.2. ConfirmDataChannelStatusAck Messages ................8
5.1.3. ConfirmDataChannelStatusNack Messages ...............8
5.2. Data Channel Status Subobject ..............................9
5.3. Message Construction ......................................10
5.4. Backward Compatibility ....................................10
6. Procedures .....................................................11
7. Security Considerations ........................................12
8. IANA Considerations ............................................12
8.1. LMP Message Types .........................................12
8.2. LMP Data Link Object Subobject ............................13
8.3. LMP Error_Code Class Type .................................13
9. Acknowledgments ................................................13
10. References ....................................................13
10.1. Normative References .....................................13
10.2. Informative References ...................................14
Contributor's Address .............................................141. Introduction 🌸
okay so GMPLS networks are built out of Traffic Engineering (TE) links, and those TE links connect Label Switching Routers (LSRs) 🧚. Each TE link is made from a set of data channels. In this context, a data channel is a resource label in a non-packet technology, like a timeslot or a lambda ✨.
Here's the core problem 😭. A data channel mismatch happens when the LSR at one end of a TE link thinks the data channel is assigned to carry data, but the LSR at the other end doesn't. The phrase "ready to carry data" means the channel is cross-connected or bound to an end-point, so it can receive or deliver data 🎀.
Data channel mismatches can't be spotted from the TE info that routing protocols advertise [RFC4203], [RFC5307] 🙈. Some of them might show up as a mismatch in the advertised bandwidths, but only where bidirectional TE links and bidirectional services are in use. If unidirectional services exist, or if several data channel mismatches happen at once, routing advertisements just can't catch them. And honestly, there's no mechanism to isolate the mismatches by figuring out which data channels are at fault 💀.
If a data channel mismatch exists, any attempt to use that channel for a new Label Switched Path (LSP) will fail 😭. One end of the TE link might try to assign the TE link for use, but the other end reports the data channel as unavailable whenever the control plane or management plane tries to assign it to an LSP 💅.
Yes, the Acceptable Label Set object in GMPLS signaling [RFC3473] can fix this, but it's super inefficient 🙄. It may need an extra signaling exchange for every LSP that gets set up. When lots of LSPs are being set up and there are lots of mismatches, those inefficiencies pile up fast. So it's desirable to avoid the extra signaling overhead and to report the problems to the management plane, so they can get fixed and LSP setup gets more efficient 🔥.
A mismatch like this can also cause misconnections in the data plane, especially when LSPs are set up through management plane operations 😬.
Mismatched resources are often called "stranded resources" 🍓. They aren't in use by any LSP, but they can't be assigned to a new LSP because they look like they're already in use. In theory, management plane applications could audit every network resource to find stranded resources and release them. In practice, that almost never happens 🙈. Coordinating different Element Management Systems (EMSs) is a pain, and there's a real risk of accidentally releasing resources that are actually in use. So a control plane mechanism that detects and reports stranded resources is what everybody wants 👑.
Stranded resources are sneaky. They look free to nobody and busy to everybody, which is exactly why a control plane check helps so much. 🦋
This document defines simple additions to the Link Management Protocol (LMP) [RFC4204]. These additions give the control plane a tool that helps locate stranded resources, by letting adjacent LSRs confirm data channel statuses and trigger notifications to the management plane whenever they find discrepancies 🧚. And since LMP already verifies data plane connectivity, it's considered an appropriate candidate for this feature 💖.
2. Specification of Requirements 🎀
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in [RFC2119]. 📜✨
3. Problem Explanation 🌈
Here are three scenarios that show what data channel mismatches look like 👀.
In every one of these scenarios, the specific channel resource of a data link becomes unavailable because of the data channel status mismatch, which means that channel resource gets wasted 😭. On top of that, a mismatch can make it less likely that an LSP gets set up successfully, because it can cause the LSP establishment to fail 🙈.
So it's really important to confirm the data channel statuses as early as possible 🫶.
3.1. Mismatch Caused by Manual Configuration 🎀
Picture this: an operator configured a cross-connect at only one end of a TE link, using an EMS 🧑💻. The resource at one end of the data channel is allocated, but the matching resource at the other end is still available. So from one end, the data channel looks free to the control plane. From the other end, though, it's considered unavailable 😭. Or, the available end might get cross-connected by the management plane, and then a misconnection can happen, because the other end is already cross-connected 💀.
Figure 1 shows a data channel between nodes A and B. The resource at A's end of the TE link is allocated through manual configuration, while the resource at B's end is available. That means the data channel status is mismatched 💜.
allocated available
+-+------------+-+
A |x| | | B
+-+------------+-+
data channel
Figure 1. Mismatch Caused by Manual Configuration3.2. Mismatch Caused by LSP Deletion 🗑️
okay so the channel status of a data link can get mismatched during LSP deletion 😭 If the deletion gets aborted halfway through (maybe a temporary control plane glitch), the cross-connect at the upstream node can be removed while the downstream node still keeps its cross-connect, as long as the source node started the deletion. 💔
Check out Figure 2. An LSP goes through nodes A, B, and C, and node B resets abnormally while the LSP is being deleted. Now the cross-connects at A and C are gone, but B's cross-connect is still being used. 🙈 That means the data channel statuses between A and B, and between B and C, are both mismatched. Messy!
<---------LSP--------->
+-+-------+-+-------+-+
| | |X| | |
+-+-------+-+-------+-+
A B C
Figure 2. Mismatch Caused by LSP DeletionNow for the backstory 📚 In [RFC2205] and [RFC3209], a "soft state" mechanism was defined to stop state discrepancies between LSRs, which is super important. Then the Resource ReSerVation Protocol-Traffic Engineering (RSVP-TE) restart processes ([RFC3473], [RFC5063]) came along too. With those, adjacent LSRs can resynchronize their control plane state so they can reinstate info about LSPs that have stuck around in the data plane. ✨ Both mechanisms are all about keeping state consistent across nodes and letting LSRs spot mismatched data plane states. 🫶
What to do about a mismatched data plane state is a local policy decision. Some deployments might automatically clean up the data plane so it matches the control plane. Others might raise an alert to the management plane and leave the data plane alone, just in case somebody is still using it. Both are valid, honestly the vibe depends on your network 💅
Either way, data channel mismatches can show up after a restart and might not get cleared up by the restart procedures. Not ideal, but that's life 🥲
3.3. Failed Resources 🔧
It's not common, but it can happen that a termination point of a TE link looks failed from one end while it looks totally OK from the other. 👀 This can happen because of a hardware failure or a problem with how the termination point's status gets detected.
This mismatch in termination point status can cause a failure when you're setting up a bidirectional LSP. Yikes! 😱
Good Failed
+-+------------+-+
A | | |X| B
+-+------------+-+
data channel
Path Message with Upstream Label---->
Figure 3. Mismatch Caused by Resource FailureHere's what goes down in this case. The upstream node picks termination point A to receive traffic from the downstream node. From the upstream node's point of view, the resource looks available and therefore usable 👍 But over at the downstream node, the matching termination point (resource B) is broken. Result: setup failure. 💀
4. Motivation 🌸
The requirement doesn't come from some gap in the GMPLS specifications themselves. It comes from operational concerns, because in most cases GMPLS-controlled networks have to live side by side with legacy networks and legacy procedures. Real world problems, you know? 🌍
The protocol extensions in this document are meant to catch data plane problems caused by misuse or misconfiguration from user error, or by a failure to clean up the data plane after the control plane disconnects. The authors expect human mistakes to be the biggest source of errors here, which, relatable 🙃
This document is not intended to provide a protocol mechanism for dealing with broken implementations. So if the software itself is buggy, that's a different problem! 🛠️
The procedures in this document are designed to run on a periodic or on-demand basis. They are NOT RECOMMENDED for continuous, online monitoring. Don't go overboard with it, bestie 🙅♀️
LMP is already used to verify data plane connectivity, so it's considered a good candidate to support this feature. It's giving "obviously the right tool for the job" 💜
5. Extensions to LMP 🦋
A control plane tool to detect and isolate data channel mismatches is provided in this document as simple additions to the Link Management Protocol (LMP) [RFC4204]. It helps track down stranded resources by letting adjacent LSRs confirm their data channel statuses. Iconic 👑
Here are the outline procedures. More detailed procedures are in Section 6.
The message formats in the subsections below use Backus-Naur Form (BNF) encoding, as defined in [RFC5511]. 📝
5.1. Confirm Data Channel Status Messages ✅
Here are the extensions to LMP for confirming a data channel status. To confirm a data channel status, the new LMP messages get sent between adjacent nodes. They can go out periodically or be triggered by some event, like an operator command, a configurable timer, or an LSP setup message getting rejected because a resource is unavailable. 🚀 The new LMP messages run over the control channel, encapsulated in UDP with an LMP port number and IP addressing, as defined in "Link Management Protocol (LMP)" [RFC4204].
There are three new messages for checking data channel status: ConfirmDataChannelStatus, ConfirmDataChannelStatusAck, and ConfirmDataChannelStatusNack. Each one gets its own detailed subsection below. The Message Type numbers are in Section 8.1. 🎀
5.1.1. ConfirmDataChannelStatus Messages 💌
The ConfirmDataChannelStatus message gives the remote end of the data channel the status of the local end, and asks the remote end to report its own data channel. It can cover more than one data channel at once, so it's pretty efficient 🧚
<ConfirmDataChannelStatus Message> ::= <Common Header>
<LOCAL_LINK_ID>
<MESSAGE_ID>
<DATA_LINK>[<DATA_LINK>...]When a node gets a ConfirmDataChannelStatus message, and it supports the data channel status confirmation procedure, the node compares its own data channel statuses with all the data channel statuses the remote end sent in that message. If it finds a mismatch, the result is expected to get reported to the management plane for further action. What happens in the management plane is outside the scope of this document, so that's a whole other mood 🙃
Also, if the message is a Confirm Data Channel Status message and its MESSAGE_ID value is less than the largest MESSAGE_ID value previously received from the sender for the specified TE link, then the message SHOULD be treated as out-of-order. ⏪ Out-of-order messages are a no-no, babe.
5.1.2. ConfirmDataChannelStatusAck Messages 💖
The ConfirmDataChannelStatusAck message goes back to the node that originated the ConfirmDataChannelStatus message, to return the data channel statuses that were requested. 📬
When a ConfirmDataChannelStatusAck message is received, the node compares the data channel statuses it got from the remote end with the ones at its own local end. This is the same comparison the receiver of the ConfirmDataChannelStatus message does. If a mismatch shows up, the result is expected to get reported to the management plane for further action. 🔍
<ConfirmDataChannelStatusAck Message> ::= <Common Header>
<MESSAGE_ID_ACK>
<DATA_LINK>[<DATA_LINK>...]The contents of the MESSAGE_ID_ACK objects MUST be obtained from the ConfirmDataChannelStatus message being acknowledged. No making stuff up here, it has to match 🫡
Also, the ConfirmDataChannelStatusAck message gets used whether the data channel statuses match or don't match. Both are valid replies, so don't panic 💅
5.1.3. ConfirmDataChannelStatusNack Messages 🙅♀️
When a node gets a ConfirmDataChannelStatus message, and the data channel status confirmation procedure is not supported (but the message is recognized), a ConfirmDataChannelStatusNack message MUST be sent. It has to contain an ERROR_CODE indicating "Channel Status Confirmation Procedure not supported". 🥲
If the data channel status confirmation procedure is supported but the node can't begin it, a ConfirmDataChannelStatusNack message MUST be sent with an ERROR_CODE indicating "Unwilling to Confirm". When the originating node gets a ConfirmDataChannelStatusNack carrying that ERROR_CODE, it MAY schedule a retransmission of the ConfirmDataChannelStatus message after a configured time. The document suggests a default of 10 minutes for this timer, which is a nice little starting point ⏰
<ConfirmDataChannelStatusNack Message> ::= <Common Header>
[<LOCAL_LINK_ID>]
<MESSAGE_ID_ACK>
<ERROR_CODE>The contents of the MESSAGE_ID_ACK objects MUST be obtained from the ConfirmDataChannelStatus message being rejected. Same rule as before, the IDs have to match 🎯
The ERROR_CODE object in this message has a new Class Type (see Section 8.3), but it's formed the same way as the ERROR_CODE object defined in [RFC4204]. These Error Codes are defined:
0x01 = Channel Status Confirmation Procedure not supported
0x02 = Unwilling to Confirm5.2. Data Channel Status Subobject 🍓
A new Data Channel Status subobject type is added to the DATA LINK object, to hold the Data Channel Status and the Data Channel ID. ✨
The Subobject Type value is in Section 8.2.
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type | Length | Data Channel Status |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| |
// Data Channel ID //
| |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+Data Channel Status: This is a series of bit flags that indicate the status of the data channel. These values are defined:
0x0000 : The channel is available/free.
0x0001 : The channel is unavailable/in-use.Data Channel ID: This identifies the data channel. The length of this field can be deduced from the Length field in the subobject. Heads up: all subobjects have to be padded to a four-byte boundary with trailing zeros. 🧸
If padding is needed, the Length field MUST indicate the length of the subobject up to, but not including, the first byte of padding. So the amount of padding is worked out on the fly and never gets written into the Length field. Neat trick 🪄
Also, the Data Channel ID is given in the context of the sender of the ConfirmChannelStatus message. 💭
The Data Channel ID must be encoded as a label value. The encoding method depends on the type of signal (for example, Synchronous Optical Network/Synchronous Digital Hierarchy (SONET/SDH), Lambda, and so on). For SONET/SDH, the label value is encoded as per [RFC4606]. 🌈
5.3. Message Construction
Okay so, the Data_Link Class (defined in Section 13.12 of [RFC4204]) gets included in both the ConfirmDataChannelStatus and ConfirmDataChannelStatusAck messages 🎀
The status of the TE link end MUST be carried by the Data Channel Status subobject, which is defined in Section 5.2 of this document. 💅 The new subobject MUST be part of Data_Link Class. No skipping this, bestie 🫶
In the case of SONET/SDH, the Data Channel ID in the new subobject SHOULD be used to identify each timeslot of the data link. ✨ That's the recommended move for SONET/SDH, so lowkey it's the way to go 🌸
5.4. Backward Compatibility
Real talk, some nodes out there might only support the LMP Message Types that are already defined in [RFC4204]. 😭 The three new types of LMP messages defined in this document are totally invisible to these nodes, because they can't recognize them at all 💀 The behavior of an LMP node that receives an unknown message is not specified in [RFC4204] and will be clarified in a separate document. 📚
Since the behavior of legacy nodes has to be assumed unknown, this document assumes that a deployment meant to support the function described here will consist completely of nodes that support the protocol extensions also described in this document. 🧚 Full squad or nothing, periodt 👑
In the future, it may be the case that LMP will be extended to allow function support to be detected. 🔮 If that happens, it may become possible to deploy this function in a mixed environment. 🌈
6. Procedures 🦋
Adjacent nodes MAY send data channel status confirmation-related LMP messages. ✨ Periodical timers or some other events requesting the confirmation of channel status for the data link may trigger these messages. ⏰ It's a local policy decision to start the data channel status confirmation process 💅 The procedure goes like this:
- Initially, the SENDER constructs a ConfirmDataChannelStatus message that MUST contain one or more DATA_LINK objects. The DATA_LINK object is defined in [RFC4204]. Each DATA_LINK object MUST contain one or more Data Channel Status subobjects. The Data Channel ID field in the Data Channel Status subobject MUST indicate which data channel needs to be confirmed, and MUST report the data channel status at the SENDER. Then the ConfirmDataChannelStatus message is sent to the RECEIVER. 💌
- Upon receipt of a ConfirmDataChannelStatus message, the RECEIVER MUST extract the data channel statuses from the ConfirmDataChannelStatus message and SHOULD compare these with its data channel statuses for the reported data channels. If a data channel status mismatch is found, the mismatch result SHOULD be reported to the management plane for further action. 🚨 The RECEIVER also SHOULD send the ConfirmDataChannelStatusAck message, which MUST carry all the local end statuses of the requested data channels to the SENDER. 📬
- If the RECEIVER is not able to support or to begin the confirmation procedure, the RECEIVER MUST send a ConfirmDataChannelStatusNack message containing the ERROR_CODE that indicates the reason for rejection. 🙅♀️
- Upon receipt of a ConfirmDataChannelStatusAck message, the SENDER MUST compare the received data channel statuses at the remote end with the data channel statuses at the local end. If a data channel status mismatch is found, the mismatch result SHOULD be reported to the management plane for further action. 🔍
The data channel status mismatch issue identified by LMP may be automatically resolved by RSVP restart. 🔄 For example, the restarting node may also have damaged its data plane, and that leaves the data channels mismatched. 🙈 However, RSVP restart will re-install the data plane state in the restarting node, so it's giving redemption arc 🎬 The issue may also be resolved via RSVP soft state timeout. ⏳
If the ConfirmDataChannelStatus message is not recognized by the RECEIVER, the RECEIVER ignores this message and will not send out an acknowledgment message to the SENDER. 🙅♀️
Due to the message loss problem, the SENDER may not be able to receive the acknowledgment message. 😵
ConfirmDataChannelStatus SHOULD be sent using LMP [RFC4204] reliable transmission mechanisms. 🛡️ If, after the retry limit is reached, a ConfirmDataChannelStatusAck message or a ConfirmDataChannelStatusNack message is not received by the SENDER, the SENDER SHOULD terminate the data channel confirmation procedure and SHOULD raise an alert to the management plane. 🚨
7. Security Considerations 🔐
[RFC4204] describes how LMP messages between peers can be secured, and these measures are equally applicable to the new messages defined in this document. 🔒
Okay this section is serious bestie 😭
The operation of the procedures described in this document does not of itself constitute a security risk because it does not cause any change in network state. 🙌 It would be possible, if the messages were intercepted or spoofed, to cause bogus alerts in the management plane, and so the use of LMP security measures described in [RFC4204] is RECOMMENDED. 💀 Lock it down!
Note that performing the procedures described in this document may provide a useful additional security measure to verify that data channels have not been illicitly modified. 🕵️♀️✨
8. IANA Considerations 📋
8.1. LMP Message Types
IANA maintains the "Link Management Protocol (LMP)" registry, which has a subregistry called "LMP Message Type". 🗂️ IANA has made the following three new allocations from this registry.
Value Description
------ ---------------------------------
32 ConfirmDataChannelStatus
33 ConfirmDataChannelStatusAck
34 ConfirmDataChannelStatusNack8.2. LMP Data Link Object Subobject
IANA maintains the "Link Management Protocol (LMP)" registry, which has a subregistry called "LMP Object Class name space and Class type (C-Type)". 🗃️ This subregistry has an entry for the DATA_LINK object, and there is a further embedded registry called "DATA_LINK Sub-object Class name space". IANA has made the following allocation from this embedded registry.
Value Description
------ ---------------------------------
9 Data Channel Status8.3. LMP Error_Code Class Type
IANA maintains the "Link Management Protocol (LMP)" registry, which has a subregistry called "LMP Object Class name space and Class type (C-Type)". This subregistry has an entry for the ERROR_CODE object. IANA has allocated the following new value for an ERROR_CODE class type. 🎉
C-Type Description Reference
------ ---------------------------- ---------
4 ConfirmDataChannelStatusNack [This RFC]9. Acknowledgments 💜
The authors would like to thank Adrian Farrel, Dimitri Papadimitriou, and Lou Berger for their useful comments. 🫶
10. References 📚
10.1. Normative References
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119, March 1997.
[RFC4204] Lang, J., Ed., "Link Management Protocol (LMP)",
RFC 4204, October 2005.
[RFC5511] Farrel, A., Ed., "Routing Backus-Naur Form (RBNF):
A Syntax Used to Form Encoding Rules in Various Routing
Protocol Specifications", RFC 5511, April 2009.10.2. Informative References
[RFC2205] Braden, R., Ed., Zhang, L., Berson, S., Herzog, S., and
S. Jamin, "Resource ReSerVation Protocol (RSVP) --
Version 1 Functional Specification", RFC 2205,
September 1997.
[RFC3209] Awduche, D., Berger, L., Gan, D., Li, T., Srinivasan, V.,
and G. Swallow, "RSVP-TE: Extensions to RSVP for LSP
Tunnels", RFC 3209, December 2001.
[RFC3473] Berger, L., Ed., "Generalized Multi-Protocol Label
Switching (GMPLS) Signaling Resource ReserVation
Protocol-Traffic Engineering (RSVP-TE) Extensions",
RFC 3473, January 2003.
[RFC4203] Kompella, K., Ed., and Y. Rekhter, Ed., "OSPF Extensions
in Support of Generalized Multi-Protocol Label Switching
(GMPLS)", RFC 4203, October 2005.
[RFC4606] Mannie, E. and D. Papadimitriou, "Generalized Multi-
Protocol Label Switching (GMPLS) Extensions for
Synchronous Optical Network (SONET) and Synchronous
Digital Hierarchy (SDH) Control", RFC 4606, August 2006.
[RFC5063] Satyanarayana, A., Ed., and R. Rahman, Ed., "Extensions
to GMPLS Resource Reservation Protocol (RSVP) Graceful
Restart", RFC 5063, October 2007.
[RFC5307] Kompella, K., Ed., and Y. Rekhter, Ed., "IS-IS Extensions
in Support of Generalized Multi-Protocol Label Switching
(GMPLS)", RFC 5307, October 2008.Contributor's Address 🧚
Fatai Zhang
Huawei Technologies
F3-5-B R&D Center, Huawei Base
Shenzhen 518129 China
Phone: +86 755-289-72912
EMail: zhangfatai@huawei.comAuthors' Addresses 🌟
Dan Li
Huawei Technologies
F3-5-B R&D Center, Huawei Base
Shenzhen 518129 China
Phone: +86 755-289-70230
EMail: danli@huawei.com
Huiying Xu
Huawei Technologies
F3-5-B R&D Center, Huawei Base
Shenzhen 518129 China
Phone: +86 755-289-72910
EMail: xuhuiying@huawei.com
Snigdho C. Bardalai
Fujitsu Network Communications
2801 Telecom Parkway
Richardson, Texas 75082, USA
Phone: +1 972 479 2951
EMail: snigdho.bardalai@us.fujitsu.com
Julien Meuric
France Telecom Orange Labs
2, avenue Pierre Marzin
22307 Lannion Cedex, France
Phone: +33 2 96 05 28 28
EMail: julien.meuric@orange-ftgroup.com
Diego Caviglia
Ericsson
Via A. Negrone 1/A 16153
Genoa Italy
Phone: +39 010 600 3736
EMail: diego.caviglia@ericsson.com📦 where this page comes from
- ietf: rfc5818 2010-04
- 💖 girlified by claude-haiku-5-5 on : a language model's rewrite of the original, under its terms