[aodvv2-discuss] Proposed resolutions for Justin's issues...
- From: Charlie Perkins <charles.perkins@xxxxxxxxxxxxx>
- To: aodvv2-discuss@xxxxxxxxxxxxx
- Date: Fri, 19 Dec 2014 07:26:37 -0800
Hello folks,
Here are some proposed resolutions for Justin's issues...
=========================== Next Issue ===================================
[manet] #58: Definitions of OrigNode and TargNode (Submitted for Justin
Dean)
==========================================================================
On 12/18/2014 7:22 PM, manet issue tracker wrote:
> #58: Definitions of OrigNode and TargNode (Submitted for Justin Dean)
>
> The definition of Originating Node (!OrigNode) (and !TargNode) are self
> referencing and incomplete. They should at least reference AODVv2
Router.
> Also the terms !OrigNode and !TargNode are later used as what seems like
> addresses in an AODVv2 message. I suggest separating the definition
of a
> node (an AODVv2 Router?) from an address of that node/router (suggest
> !OrigNodeAddr)
Proposed resolution:
Agreed and done throughout the document.
=========================== Next Issue ===================================
[manet] #59: Use of the term "invalid" (Submitted for Justin Dean)
==========================================================================
On 12/18/2014 7:24 PM, manet issue tracker wrote:
> #59: Use of the term "invalid" (Submitted for Justin Dean)
>
> The term "invalid" is used a few times when a more specific reference to
> the defined term "broken" and/or "expired" should be used.
Proposed resolution:
Agreed and reviewed throughout the document.
=========================== Next Issue ===================================
[manet] #60: Should OrigNode be included in the message header?
(Submitted for Justin Dean)
==========================================================================
On 12/18/2014 7:27 PM, manet issue tracker wrote:
> #60: Should OrigNode be included in the message header? (Submitted
for Justin
> Dean)
>
> In the "!RteMsg Structure" section the list says that the !OrigNode
should
> be included in the message header, the packet diagram doesn't
include this
> information as it doesn't expand the message header. This would be fine
> but then the !OrigNode is then included again the address block which is
> confusing. Later it's described that !OrigNode should indeed be
included
> in the address block but it happens after the diagram. Also, is there a
> reason the !OrigNode is included twice?
Proposed resolution:
To discuss.
OrigAddr in message header simplifies description of Address Blocks, and
is smaller.
OrigAddr in Address Block allows for address compression, makes Chris
happier.
Note: a long time ago, both OrigAddr and TargAddr were Message Header
fields, in closer conformance with RFC 3561.
=========================== Next Issue ===================================
[manet] #61: Difference between "broken" and "expired"
(Submitted for Justin Dean)
==========================================================================
#61: Difference between "broken" and "expired" (Submitted for Justin Dean)
It's not clear to me the difference between the states "broken" and
"expired" (other than how they reached that state) They are both invalid
states and regardless of how the state was obtained the routes shouldn't
be used. The only difference I could find is that "expired" routes MAY be
expunged while there is no such text regarding "broken" routes. Broken is
also used in reference to broken link and later !BrokenRoute without the
terms being defined in the document definitions. It may make sense to
unify these states regardless of how they are obtained into a single term
and reduce the number of states a route can have (unless there is
something operationally different that should be done that I'm missing)
Proposed resolution:
Add explanation of the difference between broken and expired.
=========================== Next Issue ===================================
[manet] #62: Inconsistency surrounding the "timed" state
(Submitted for Justin Dean)
==========================================================================
On 12/18/2014 7:31 PM, manet issue tracker wrote:
> #62: Inconsistency surrounding the "timed" state (Submitted for
Justin Dean)
>
> The addition of the "timed" state requires updating in a few other
places
> in the document when route state is referenced. The "Route Table Entry
> Timeouts" section for example may need to be updated.
Proposed resolution:
Change "timed state" to be "timed flag".
=========================== Next Issue ===================================
[manet] #63: {Orig,Targ}.Tail should be {Orig,Targ}.Mid
(Submitted for Justin Dean)
==========================================================================
On 12/19/2014 6:34 AM, manet issue tracker wrote:
> #63: {Orig,Targ}.Tail should be {Orig,Targ}.Mid
>
> Orig.Tail and Targ.Tail should be Orig.Mid and Targ.Mid as the Tail term
> is used for common bits of an address among the addresses listed.
Proposed resolution:
Agreed and done throughout the document. Only appeared in Appendix.
Other related posts:
- » [aodvv2-discuss] Proposed resolutions for Justin's issues... - Charlie Perkins
