Re: [OSPF] OSPF GR query

["sujay gupta" <[e-mail]>]

Hi Acee, all,

A late reply on this thread.

Some comments with relevant text.

<snip>
 No - The presumption is when you reach FULL state you don't need to
 "quit"
helping - you have "finished" helping. Over the years, there have been a
few situations that could have been better documented - however, this
never
caused any confusion in the past.
<snip>
I'm not sure what the network picture looks like but a helper
router can always terminate graceful restart by flooding an
inconsistent router or network LSA to the restarting  router.
<snip/>

While going with the basic OSPF protocol working, when the helper
exits.
-before reaching Full State would be to regenerate LSA's, giving a snapshot
of the incomplete state with the Restarting router, causing an
"inconsistent"
lsa generation which in turn would cause the restarting router exit GR.
-after reaching Full State would be to regenerate LSA's, but this time a
duped LSA
reporting "inconsistent" states, which in turn would cause the restarting
router to
exit GR.
While I immediately see 'wee' bit of an ethical problem by this approach(for
it looks more like a hack ;), it is still a way of explicit signaling,
works for me!  Some other ways could have been "a one-way hello" or "lls
signaling".

Some more text.
<snip>
Over the years, there have been a
few situations that could have been better documented - however, this
never
caused any confusion in the past.
<snip/>

Acee, with my limited experience in the OSPF WG, I have had and seen
confusions.
Please note RFC3623 Exit GR procedures (Section 3.2) does not entail the
helper GR to produce inconsistent LSA's, an addendum should have a better
future rather than raising the same queries again.

There have been some more issues raised on the GR and update by Mitchell,
which I shall shortly address in a separate e-mail.

Best Regards & HNY'07,
-Sujay










On 12/5/06, Acee Lindem <[e-mail]> wrote:
>
> Hi Sujay,
> sujay gupta wrote:
> > Hi Acee,
> > A late mail on this thread.
> >
> > given the scenario if R3-R1 is a stub area
> > and the incoming LSA is a type-5, even with strict LSA
> > checking enabled will not exit Helper mode.
> I'm not sure what the network picture looks like but a helper
> router can always terminate graceful restart by flooding an
> inconsistent router or network LSA to the restarting  router.
>
> Thanks,
> Acee
>
> >
> > It can be a nice idea to give some explicit notification
> > by R2 to R1 about exiting helper mode.
> >
> > As rightly pointed by you the above conditions and more
> > has been causing some confusion for some time.
> > Our proposal ;
> >
> http://tools.ietf.org/wg/ospf/draft-holla-ospf-update-graceful-restart-02.txt
> >
> > hopefully resolves this.
> > Could we add this as an addendum/reference to RFC3623?, purely an
> > add-on.
> > Best Regards.
> > Sujay
> >
> > On 10/5/06, Acee Lindem <[e-mail]> wrote:
> >>
> >> Khan Amir-G20247 wrote:
> >> > Hi Ace
> >> >
> >> > But reaching to a FULL state is not one of the conditions for
> >> Helper to
> >> > quit Helper mode, I guess a Helper who has reached
> >> > to FULL state with its restarting router will wait for Grace LSAs
> >> to be
> >> > flushed or GR timeout to happen for exiting Helper mode.
> >> >
> >> No - The presumption is when you reach FULL state you don't need to
> >> "quit"
> >> helping - you have "finished" helping. Over the years, there have been
> a
> >> few situations that could have been better documented - however, this
> >> never
> >> caused any confusion in the past.
> >>
> >> Thanks,
> >> Acee
> >>
> >> > Regards
> >> > Aamir
> >> >
> >> >
> >> > -----Original Message-----
> >> > From: Acee Lindem [mailto:[e-mail]]
> >> > Sent: Thursday, October 05, 2006 8:00 PM
> >> > To: Khan Amir-G20247
> >> > Cc: [e-mail]
> >> > Subject: Re: [OSPF] OSPF GR query
> >> >
> >> > Hello Amir,
> >> >
> >> > Khan Amir-G20247 wrote:
> >> >
> >> >>
> >> >>
> >> >> Hi
> >> >>
> >> >> I have query related to OSPF Graceful Restart.
> >> >>
> >> >> Consider this scenario:
> >> >>
> >> >>
> >> >>
> >> >> - Three Routers R1, R2 and R3 connected via Broadcast networks.
> >> >>
> >> >>
> >> >>
> >> >>
> >> >>
> >> >>
> >> >>
> >> >> - Where R1 is a DR on both the segments, where as R2 and R3 are in
> >> >> DR-other
> >> >> - Now R1 wants to perform GR, R2 and R3 agrees to be its Helpers.
> >> >> - Now after restart R1 will try to form adjacency with both R2 and
> >> R3.
> >> >>
> >> >
> >> >
> >> >> - Lets say R1 and R2 had become FULL, while R1 and R3 had reached
> >> >> Exstart state.   Now at this time R2 receives an LSA (any type 1-5
> >> and
> >> >> 7, from some other router existing in the same area, not pictured
> >> >> here) which he needs to flood to R1.
> >> >> - This will force R2 to quit Helper Status, while R1 and R3 are
> still
> >> >> in process of forming adjacency.
> >> >>
> >> >>
> >> > Since he has reached FULL state, he is no longer a helper router.
> >> >
> >> >> Considering above scenario, my doubts are:
> >> >>
> >> >> - How will R1 come to know that its Helper R2 has quit?
> >> >>
> >> >>
> >> > Since he has reached FULL state, he is no longer a helper router.
> >> >
> >> >> - Will the adjacency between R1 and R2 needs to go down, as R2 quits
> >> >> helper mode?
> >> >>
> >> >>
> >> > No
> >> >
> >> >> - Will this effect the adjacency formation of R1 and R3?
> >> >>
> >> >>
> >> > No. R3 MAY exit helper status and re-originate a new router LSA if he
> >> > receives the new LSA before reaching FULL state with R1 AND he is has
> >> > strict-LSA-checking configured.
> >> >
> >> > Hope this helps,
> >> > Acee
> >> >
> >> >>
> >> >>
> >> >> Regards
> >> >> Aamir
> >> >>
> >> >>
> >> >>
> >> >>
> >> >>
> >> >>
> >> ----------------------------------------------------------------------
> >> >> --
> >> >>
> >> >> _______________________________________________
> >> >> OSPF mailing list
> >> >> [e-mail]
> >> >> https://www1.ietf.org/mailman/listinfo/ospf
> >> >>
> >> >>
> >> >
> >> >
> >>
> >> _______________________________________________
> >> OSPF mailing list
> >> [e-mail]
> >> https://www1.ietf.org/mailman/listinfo/ospf
> >>
> >
>

_______________________________________________
OSPF mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/ospf