Re: [manet] #22 (aodvv2): Multiple terms for same concept

["manet issue tracker" <[e-mail]>]

#22: Multiple terms for same concept

Changes (by [e-mail]):

 * status:  new => closed
 * resolution:   => fixed


Comment:

 On 4/4/2014 1:32 PM, Abdussalam Baryun wrote:
 > IMHO, MANET has more concepts depending on the network use and status.
 Therefore it is good that this work considers many use cases.
 >
 > On Thursday, April 3, 2014, manet issue tracker wrote:
 >
 >     #22: Multiple terms for same concept
 >
 >      (Thomas Clausen) While we are on that, I notice that "active
 source" is
 >      used; I am not sure what that is vs a "source" and vs a !PktSource.
 >
 > It is clear in MANET that active node means there is sleep period within
 that node, so when the node is out of that period it is active.
 >
 >     A
 >      source is someone who generates data traffic, I would suspect ...
 so an
 >      active source is....what?  Or, is a source someone who can, but
 doesn't,
 >      generate data traffic?
 >
 > The source is a node that generates data. The active node is a node that
 generates data for some determined period of times, and does not
 receive/generate within sleeping periods.
 >
 >
 >      if so, what is the significance of that wrt.
 >      routing protocol operation? And, wrt a /reactive/ routing protocol,
 which
 >      I would assume constructs routes only for active source-destination
 pairs?
 >
 > I don't think that is correct. In MANET we say it MAY have LLN nodes so
 it is good AODVv2 considers LLNs scenarios. The ROLL WG have done a
 terminology RFC, I will try to continue the Manet terminology draft. The
 routing in LLNs and MANETs does construct routes not only for active
 source-destination pairs, but routes that have multihops with sleep
 periods as well.

 ============================================================================

 On 4/5/2014 5:46 AM, Thomas Clausen wrote:
 > On 04 Apr 2014, at 22:32, Abdussalam Baryun <[e-mail]>
 wrote:
 >> IMHO, MANET has more concepts depending on the network use and status.
 Therefore it is good that this work considers many use cases.
 >>
 >> On Thursday, April 3, 2014, manet issue tracker wrote:
 >>
 >>     #22: Multiple terms for same concept
 >>
 >>      (Thomas Clausen) While we are on that, I notice that "active
 source" is
 >>      used; I am not sure what that is vs a "source" and vs a
 !PktSource.
 >>
 >> It is clear in MANET that active node means there is sleep period
 within that node, so when the node is out of that period it is active.
 > That is not clear to me. Where is that described in more details? How
 does it work in a reactive protocol, what if a router is "sleeping" when
 it receives a RREQ? Does it require global coordination of duty cycles in
 the network? If not, why not?
 >>
 >>     A
 >>      source is someone who generates data traffic, I would suspect ...
 so an
 >>      active source is....what?  Or, is a source someone who can, but
 doesn't,
 >>      generate data traffic?
 >>
 >> The source is a node that generates data. The active node is a node
 that generates data for some determined period of times, and does not
 receive/generate within sleeping periods.
 >>
 >>
 >>      if so, what is the significance of that wrt.
 >>      routing protocol operation? And, wrt a /reactive/ routing
 protocol, which
 >>      I would assume constructs routes only for active source-
 destination pairs?
 >>
 >> I don't think that is correct.
 > You do not think that it is correct, that a reactive protocol constructs
 routes only for active source-destination pairs?
 >> In MANET we say it MAY have LLN nodes so it is good AODVv2 considers
 LLNs scenarios. The ROLL WG have done a terminology RFC, I will try to
 continue the Manet terminology draft. The routing in LLNs and MANETs does
 construct routes not only for active source-destination pairs, but routes
 that have multihops with sleep periods as well.
 > I am at a loss, what does that mean?

 ============================================================================

 On 4/11/2014 4:34 PM, Charles E. Perkins wrote:

 > PktSource means the source of the packet that triggered
 > the RERR.  It could be renamed to be "RERR_Dest", since the
 > RERR is being sent to the source of the packet that triggered
 > the RERR.
 >
 > "active sources" are those actively sending data.  I will make
 > a new definition for this.

 ============================================================================

 On 5/28/2014 11:09 AM, Charlie Perkins wrote:

 > It turns out to be quite straightforward to revise the text
 > so that the term "active source" is not needed.  So, I don't
 > need to make a definition for this term.

-- 
-----------------------------------+------------------------------------
 Reporter:  [e-mail]  |       Owner:  [e-mail]
     Type:  defect                 |      Status:  closed
 Priority:  minor                  |   Milestone:
Component:  aodvv2                 |     Version:
 Severity:  Active WG Document     |  Resolution:  fixed
 Keywords:                         |
-----------------------------------+------------------------------------

Ticket URL: <http://trac.tools.ietf.org/wg/manet/trac/ticket/22#comment:2>
manet <http://tools.ietf.org/manet/>