RE: Failure definition w.r.t. SONET/SDH and Protection/Restoratio n

["Diego Caviglia" <Diego.Caviglia@marconi.com>]

Hi Sasha,
                       I agree with you but in any case for interop issues
and for sake of clarity I'd like to have that written in some IETF
document.

Regards

Diego





Sasha Vainshtein <Sasha@AXERRA.com> on 23/03/2005 12.23.12

To:    "'Diego Caviglia'" <Diego.Caviglia@marconi.com>
cc:    ccamp@ops.ietf.org

Subject:    RE: Failure definition w.r.t. SONET/SDH and
       Protection/Restoratio  n


Diego and all,
IMHO it is reasonable to assume that a failure is
a defect that results in sending downstream AIS
(and upstream RDI) as consequent action(s).

For SDH consecutive actions for each defect
are specified in ITU-T G.783.

Note also that for some defects the consecutive action
can be disabled by configuration. Hence I'd say that
the same defect can be a failure in one case and
not a failure in another case. The typical example
is Trail Identifier Mismatch (TIM).

Hopefully this note will help.

Regards,
                                  Sasha Vainshtein
email:                         sasha@axerra.com <mailto:sasha@axerra.com>
phone:                       +972-3-7569993 (office)
fax:                            +972-3-6487779
mobile:                      +972-52-8674833


> -----Original Message-----
> From: Diego Caviglia [mailto:Diego.Caviglia@marconi.com]
> Sent: Wednesday, March 23, 2005 1:06 PM
> To: ccamp@ops.ietf.org
> Subject: Failure definition w.r.t. SONET/SDH and
> Protection/Restoration
>
>
> Hi all,
>               is it defined somewhere, with respect to
> SDH/SONET, which
> defect has to be consider as a failure?
>
> It seems to me that for interoperability could be useful to
> have a list of
> defect that must be considered as failure, the same applies
> for defect that
> optionally can be considered as defect.
>
> May be also a communication between restoration TNEs in order
> to agree on
> the list could be useful.
>
> Does anyone agree on that?
>
> Regards
>
> Diego
>
> PS I apologise if the list is defined somewhere but I wasn't
> able to find
> it.
>
>
>
>