BGP-4 Protocol Analysis
draft-ietf-idr-bgp-analysis-07

Re-review by Elwyn Davies, Gen-ART (following up on Lucy Lynch):


Hi.

I was asked to take a look at this doc, re-reviewing it after fixes relating to comments on -05 by Lucy Lynch.

Unfortunately the updates don't appear to have taken care of many of the formatting and boilerplate issues flagged by idnits.

There are still several over-long lines and hyphenations plus the URL for the list of I-Ds is still wrong.

Idnits output follows:

idnits 1.51

tmp/draft-ietf-idr-bgp-analysis-06.txt:

  tmp/draft-ietf-idr-bgp-analysis-06.txt(41): Line is too long: the offending characters are 'ld be'
  tmp/draft-ietf-idr-bgp-analysis-06.txt(189): Line seems to end with a hyphenated word.
  -->    that routing within an autonomous system is done by an intra-
  tmp/draft-ietf-idr-bgp-analysis-06.txt(368): Line seems to end with a hyphenated word.
  -->    flapping. However, although peer oscillation is found to be wide-
  tmp/draft-ietf-idr-bgp-analysis-06.txt(447): Line seems to end with a hyphenated word.
  -->    The aggregation scheme, defined in [CIDR],describes the provider-
  tmp/draft-ietf-idr-bgp-analysis-06.txt(540): Line is too long: the offending characters are '--'
  tmp/draft-ietf-idr-bgp-analysis-06.txt(581): Line seems to end with a hyphenated word.
  -->    interactions with network engineers in an environment lacking vendor-
  tmp/draft-ietf-idr-bgp-analysis-06.txt(630): Line has weird spacing: '... AS4 is  a mul...'
  tmp/draft-ietf-idr-bgp-analysis-06.txt(749): Line is too long: the offending characters are 'd'
  tmp/draft-ietf-idr-bgp-analysis-06.txt(750): Line is too long: the offending characters are 'ts'
  tmp/draft-ietf-idr-bgp-analysis-06.txt(754): Line is too long: the offending characters are 're'
  tmp/draft-ietf-idr-bgp-analysis-06.txt(756): Line is too long: the offending characters are 'g'
  tmp/draft-ietf-idr-bgp-analysis-06.txt(757): Line is too long: the offending characters are 'ic'
  tmp/draft-ietf-idr-bgp-analysis-06.txt(762): Line is too long: the offending characters are 'AS'
  tmp/draft-ietf-idr-bgp-analysis-06.txt(951): Line is too long: the offending characters are 's for'
  tmp/draft-ietf-idr-bgp-analysis-06.txt(952): Line is too long: the offending characters are 'rogress.'
  tmp/draft-ietf-idr-bgp-analysis-06.txt(957): Line is too long: the offending characters are 'P",'

  Checking nits according to http://www.ietf.org/ID-Checklist.html:

    Checking conformance with RFC 3667/3668 boilerplate...
    the boilerplate looks good.

  Checking nits according to http://www.ietf.org/ietf/1id-guidelines.txt:

    The document seems to lack a 1id_guidelines paragraph about the list
    of current Internet-Drafts -- however, there's a paragraph with a
    matching beginning. Boilerplate error?

  (Expected a match on the following text:
    "The list of current Internet-Drafts can be accessed at
    http://www.ietf.org/1id-abstracts.html"

  ... but found this:
    "The list of current Internet-Drafts can be accessed at
    http://www.ietf.org/ietf/1id-abstracts.txt.")


  Miscellaneous warnings:

    None

Other matters relating to language:
===================================

Terminology:
There is a considerable amount of terminology in the document which is not defined either explicitly or by reference, notably BGP speaker, peer, EBGP

Some language improvements:
Section 1: s/Classless InterDomain/Classless Inter-Domain/ in line with title of [CIDR]

Section 2.1, para 1: s/network layer reachability information/Network Layer Reachability Information/

Section 2.2, para 1, last sentence: s/assist/assists/(also in para 2 last sentence), s/select/to select/

Section 2.2, para 2, last sentence: I don't think you can limit congestion to window limits:  How about 'The window limits used by TCP further assists BGP to constrain any congestion that might result from updates.'

Section 2.2: para 3: s/sent as a single Network Layer Reachability (NLRI)./sent in a single Network Layer Reachability Information (NLRI) attribute./

Section 2.3:
s/There may exist a temporary period where in a BGP peer may have
  separate incoming and outgoing connections resulting into two
  different BGP FSMs for a peer (instead of one). This can be resolved
  following BGP connection collision rules defined in the [BGP4]./

  There may be a short period during which a BGP peer may have
  separate incoming and outgoing connections resulting in the creation of two
  different BGP FSMs relating to a peer (instead of one). This can be resolved by
  following the BGP connection collision rules defined in the [BGP4] specification./

Section 2.3, para 3: s/Following are different states of BGP FSM for its peers:/The BGP FSM has the following states associated with each of its peers:/

Section 2.3, next to last para:
  s/different/a number of/,
  s/BGP FSM/the BGP FSM/,
  s/These BGP events are/Support of these BGP events is/,
  s/using an/as a result of/

Section 2.3, last para:
  s/Both,/Both/
  s/in details/in detail/

Section 4: Should be consistent in using either 'flapping' or 'oscillation'.  Flapping is normally used with routes.

Section 4: Is 'Ideally' the right word? ... Maybe either 'By default' or 'According to [BGP]'

Section 5:
  s/bound,/bounded,/
  s/loosing/losing/g
  s/clesrly/is clearly/

Section 6.1: s/,describes/, describes/

Section 6.1.1:
  s/announcments/announcements/
  s/During the periods/During periods/

Section 6.1.2:
  s/byte/octet/g
  s/(and ignoring/(ignoring/

Section 7, para 2:
  s/provided hundreds/provide hundreds/  [? really hundreds??]
  s/more and more/increasingly/

Section 7: s/nondeterminism/non-determinism/g

Section 8: s/we answer the question of which environments is BGP well suited/we identify the environments for which BGP is well suited/

Section 8, para 3: s/that allow/that allows/

Section 8, lata para: s/suitable/suited/

Section 10:
  s/exhanged/exchanged/
  s/guage/gauge/
  s/vulneribility/vulnerability/

Section 10: Need to expand the EBGP abbreviation.

Section 10: The reference to [SOBGP] is inconsistent with the list of refs [soBGP].

Also another couple of editorial nits:
(Throughout) There are multiple blank lines surrounding section headings and in some cases between paragraphs - a single blank line would save paper.

Regards,
Elwyn Davies