Browse Prior Art Database

DEBUG LABEL AND RELATED TECHNIQUES

IP.com Disclosure Number: IPCOM000249807D
Publication Date: 2017-Apr-07
Document File: 4 page(s) / 250K

Publishing Venue

The IP.com Prior Art Database

Related People

Mark Townsley: AUTHOR [+4]

Abstract

A set of debug levels is encoded into a given Spatial Reference System Identifier (SRID) throughout a network with different values assigned to different semantics (e.g., "no debug," "some debugs," "lots of debugs," etc.). In an example, it is possible to map the top million possible debug combinations on a given set of routers and encode them directly into the bottom twenty bits of different fields (e.g., Multiprotocol Label Switching (MPLS) label, Internet Protocol version 6 (IPv6) Flow Label, etc.).

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.

Copyright 2017 Cisco Systems, Inc. 1

DEBUG LABEL AND RELATED TECHNIQUES

AUTHORS: Mark Townsley

Carlos M. Pignataro Nagendra Kumar Nainar

Richard Furr

CISCO SYSTEMS, INC.

ABSTRACT

A set of debug levels is encoded into a given Spatial Reference System Identifier

(SRID) throughout a network with different values assigned to different semantics (e.g.,

"no debug," "some debugs," "lots of debugs," etc.). In an example, it is possible to map the

top million possible debug combinations on a given set of routers and encode them directly

into the bottom twenty bits of different fields (e.g., Multiprotocol Label Switching (MPLS)

label, Internet Protocol version 6 (IPv6) Flow Label, etc.).

DETAILED DESCRIPTION

Software-defined networking allows operators to manage an entire network in a

centralized manner. In many instances, users are faced with a situation in which they are

required to perform debugs on a selective set of nodes over which a specific flow or

selective set of flows are traversing. While there are many techniques available to collect

telemetry from all networks for various purposes (e.g., capacity planning), there is no easy

way to trigger debugs on a per-flow or flow-path specific trigger. As such, provided herein

is a mechanism for pre-defining the debugs and associating the debugs with different

numbers that may be encoded in a portion of the Least Significant Bit (LSB) of different

fields (e.g., Flow Label, MPLS Label, etc.).

The idea is to assign an entire set of debugs using a bitmask and/or sequential

numbers, and advertise the set throughout the network. Depending on the particular debugs

to be collected, the relevant value may be encoded as part of a packet header field (e.g.,

MPLS Label, IPv6 Flow Label, Segment Routing Header (SRH), etc.), which transit nodes

may use to trigger certain debug actions. In another embodiment, the transit nodes may

contact a centralized server to query what needs to be enabled/collected.

Copyright 2017 Cisco Systems, Inc. 2

An operator may pre-assign a set of debugs to a sequence of numbers or a bitmask.

For example:

1 --> All open Shortest Path First (OSPF) debugs

2 --> All Label Forwarding Information Base (LFIB) debugs

3 --> Internet Protocol (IP) debugs

4 --> Specific IP debugs

5 --> IP and LFIB debugs

6 --> etc.

In another example, the approach below may be used:

00000001 --> All OSPF debugs

00000010 --> All LFIB Debugs

00000100 --> IP debugs

00001000 --> Specific IP debugs

00010000 --> etc.

This may be advertised via Interior Gateway Protocol (IGP) extensions or Yet

Another...