----- Original Message -----
From: "Arturo Servin" <[email protected]>
To: "sidr wg list" <[email protected]>
Sent: Saturday, August 06, 2011 5:42 PM
>
> Support.
>
> Some comments.
>
> Roque previously mentioned something similar, for me it's a bit confusing to
talk about ROAs validity in a router. In my mind ROAs are validated somewhere
else and in routers you search for the validity of a prefix which I think what
is referenced here.
>
> Also, as Robert Raszuk mentioned I am a bit concerned that in order to get the
validity of a prefix first I need to get a table with ~400k entries. I do not
normally get a full table BGP via SNMP so may be my concern is unjustified by my
lack of experience. May be is it just the way it is with SNMP and we need to
live with it.

The way with SNMPv1, long superseded, but still around, is to get table entries
one at a time, and if the table is dynamic, this process may never end - I once
set such a retrieval going (on statistics data rather than routing) and
cancelled it three days later, still incomplete (a good way of stress testing
router control planes:-)

SNMPv2 has a smarter way of bulk data retrieval but it depends on having the
right table structure, that is it can efficiently retrieve a slice through some
of the columns.  If you have thousands of rows, most with property X and a few
with property Y, then it should work well if X/Y is an index to a column and Y
collates after X, that is what you want is at the bottom of the column (since
the retrieval works down a column and  ends at the end of the column [RFC3416]).

But you can still be sabotaged by considerations of maximum PDU so for me,
accessing large tables remains troublesome:-(

Tom Petch
>
> Cheers,
> -as
>
>
> On 5 Aug 2011, at 11:39, Matthias Waehlisch wrote:
>
> >
> > On Fri, 5 Aug 2011, Randy Bush wrote:
> >
> >> i would request the wg adopt
> >>
> >>    draft-ymbk-bgp-origin-validation-mib
> >>    draft-ymbk-rpki-rtr-protocol-mib
> >>
> >
> > I support the adoption, as well.
> >
> >
> > Cheers
> >  matthias
> >
> > --
> > Matthias Waehlisch
> > .  Freie Universitaet Berlin, Inst. fuer Informatik, AG CST
> > .  Takustr. 9, D-14195 Berlin, Germany
> > .. mailto:[email protected] .. http://www.inf.fu-berlin.de/~waehl
> > :. Also: http://inet.cpt.haw-hamburg.de .. http://www.link-lab.net
> > _______________________________________________
> > sidr mailing list
> > [email protected]
> > https://www.ietf.org/mailman/listinfo/sidr
>
> _______________________________________________
> sidr mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/sidr

_______________________________________________
sidr mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/sidr

Reply via email to