RE: RE: V2 Changes for next RFI v2 MIB (draft 10) - Clarification
"Eduardo Cardona" <[email protected]>
| Newsgroups | gmane.ietf.ipcdn |
|---|---|
| Message-ID | <[email protected]> |
Hi Minnie,
In my side, a two months ago the consensus was to introduce the
StorageType although other alternatives were suggested, finally all
arguments always pointed "it is white, heins make them, scrambled eaten"
so the clearest way was the StorageType
Going to your proposal 2), there is a compliance statement that says
that only nonVolatile(3) is required wich covers "For user created
active row, the value is always 'nonVolatile(3)'
Having vendor "writeable" (instead of read-create) default entries is
not different that operator created entries, so maybe the text below
this is not needed
For system created default rows, the value can be ether
a. for vendors choose to have read-create access for the system
default rows 'volatile(2)' when row is changed to
notInService or
'nonVolatile(3)'
The sentence below is already covered in the Entry Description
b. for vendors choose to have read-only access for the system
default rows 'permanent(4)' or 'readOnly(5)' time, which
could report a value
'permanent' or 'readOnly' for docsIfCmtsModStorageType.
The only pending part.
"For non-active row, the value is always 'volatile(2)'"
I am not sure if the active to non-active Rowstatus transitions for
StorageType are needed or what are the current practices, I won't think
it hurts, will like to hear Randy's and other's opinions
Eduardo
-----Original Message-----
From: Minnie Lu [mailto:[email protected]]
Sent: Friday, April 16, 2004 12:28 PM
To: Eduardo Cardona; Randy Presuhn
Cc: Donati Andrew-MGIA0477; Minnie Lu; [email protected]; Keske Tom-LTK002;
Azlina Ahmad; DOCSIS OSS Majordomo List
Subject: Re: [ipcdn] RE: V2 Changes for next RFI v2 MIB (draft 10) -
Clarification
Hi, Randy and Eduardo,
Thanks a lot for the advice !
I wonder whether it is really needed to add the new object StorageType
in
this modulation profiles. Sorry to go back to the step 1.
1. active row most likely are persistent across reboot already in
current
CMTS implementation. I don't think it will be any extra work for CMTS
to
have active MUST be persistent across reboot. Also, it will be clear to
user that all active rows will be persistent cross reboot.
2. non-active looks like no use to be persistent across reboot. Also
non-active rows can be aged out even in the run time.
Since the rule is (please allow me to quoted from Randy's previous
email)
"explicit about persistence, either mandating it in DESCRIPTIONs or
using
StorageType to report / control it as needed."
Here is my suggestions.
Suggestion 1: Explicit about persistence mandating it in DESCRIPTION.
"rowStatus is in 'Active' MUST be persistent across managed system
reload"
For non-active rows, just follow the RFC, so no need to mention in
the
MIB.
Suggestion 2: New object StorageType RO (read-only)
For user created active row, the value is always 'nonVolatile(3)'
For non-active row, the value is always 'volatile(2)'
For system created default rows, the value can be ether
a. for vendors choose to have read-create access for the system
default rows
'volatile(2)' when row is changed to notInService or
'nonVolatile(3)'
b. for vendors choose to have read-only access for the system
default rows
'permanent(4)' or 'readOnly(5)'
Will it work ?
Thanks a lot for your advice again !
Minnie
At 10:34 AM 4/16/2004 -0600, Eduardo Cardona wrote:
>Hi Andrew,
>very welcome the clarification, even though, I think It was well
>understood since docsIfCmtsModControl SYNTAX is RowStatus,
>
>Thanks
>
>Eduardo
>
>-----Original Message-----
>From: Donati Andrew-MGIA0477 [mailto:[email protected]]
>Sent: Friday, April 16, 2004 10:24 AM
>To: Donati Andrew-MGIA0477; 'Minnie Lu'; Eduardo Cardona
>Cc: Eduardo Cardona; '[email protected]'; 'Randy Presuhn'; Keske
>Tom-LTK002; 'Azlina Ahmad'; DOCSIS OSS Majordomo List
>Subject: RE: V2 Changes for next RFI v2 MIB (draft 10) - Clarification
>
>
>I would like to correct/clarify the earlier sentence. The suggestion
>should have stated to add to the "docsIfCmtsModControl" DESCRIPTION,
>not to the "Row Status" DESCRIPTION already defined in RFC2579.
>
>Instead of:
>
>To address the concern raised by Minnie, may I suggest that we add to
>the DESCRIPTION clause of the row status column to define a time
>interval that the agent must remove rows that remain in the
>'notInService' or 'notReady' state.
>
>This is what was intended:
>
>To address the concern raised by Minnie, may I suggest that we add to
>the DESCRIPTION clause of docsIfCmtsModControl to define a time
>interval that the agent must remove rows that remain in the
>'notInService' or 'notReady' state.
>
>Thanks,
>Andy
>
>-----Original Message-----
>From: Donati Andrew-MGIA0477
>Sent: Friday, April 16, 2004 8:01 AM
>To: 'Minnie Lu'; Eduardo Cardona
>Cc: Eduardo Cardona; [email protected]; Randy Presuhn; Keske Tom-LTK002;
>Azlina Ahmad; DOCSIS OSS Majordomo List
>Subject: RE: V2 Changes for next RFI v2 MIB (draft 10)
>
>
>All,
>
>To address the concern raised by Minnie, may I suggest that we add to
>the DESCRIPTION clause of the row status column to define a time
>interval that the agent must remove rows that remain in the
>'notInService' or 'notReady' state.
>
>This is defined under "Interaction 4: Making the Conceptual Row
>Available" in RFC 2579.
>
>Taken from RFC2579 Section 2, Row Status, Interaction 4:
>
> "If the management station is prevented from setting the
> status column to `active' (e.g., due to management station
> or network failure) the conceptual row will be left in the
> `notInService' or `notReady' state, consuming resources
> indefinitely. The agent must detect conceptual rows that
> have been in either state for an abnormally long period of
> time and remove them. It is the responsibility of the
> DESCRIPTION clause of the status column to indicate what
an
> abnormally long period of time would be. This period of
> time should be long enough to allow for human response
time
> (including `think time') between the creation of the
> conceptual row and the setting of the status to `active'.
> In the absence of such information in the DESCRIPTION
> clause, it is suggested that this period be approximately
5
> minutes in length. This removal action applies not only
to
> newly-created rows, but also to previously active rows
which
> are set to, and left in, the notInService state for a
> prolonged period exceeding that which is considered normal
> for such a conceptual row."
>
>Thanks,
>Andy
>
>-----Original Message-----
>From: Minnie Lu [mailto:[email protected]]
>Sent: Thursday, April 15, 2004 3:37 PM
>To: Eduardo Cardona
>Cc: Eduardo Cardona; Donati Andrew-MGIA0477; [email protected]; Randy
>Presuhn; Keske Tom-LTK002; Azlina Ahmad; Minnie Lu; DOCSIS OSS
>Majordomo List
>Subject: Re: V2 Changes for next RFI v2 MIB (draft 10)
>
>
>Hi, Eduardo,
>
> Thanks a lot for listening to me. I am happy with the new text.
>
> But I still have concern about saving notInService or notReady
>rows for modulation profiles into NVRAM and have them persistent across
>CMTS reboot, especially for the rows with inconsistent attributes
>which is allowed for rowStatus in 'notInService' and 'notReady'.
>
> Though this is extreme case, but since the MIB or spec does not
>prevent to do so or does not make it optional, that means CMTS MUST
>implement such
>capability to save rows with inconsistent attributes into NVRAM which
is
>a
>waste of CMTS resource and complicated the implementation for no or
>little
>gain.
>
> Possible to make it an optional that saving rows with inconsistent
>attributes into NVRAM and persistent across reboot ?
>
> Thanks a lot for your advice !
> Minnie
>
>
>
>
>At 10:27 AM 4/15/2004 -0600, Eduardo Cardona wrote:
>
> >Hi all,
> >
> >Please find attached the summary of changes after all your feedback,
> >special thanks to Randy Presuhn for all the detailed comments.
> >
> >Please note that the storageType objects are being set as CMTS only
> >requirement docsIfCmtsGroupV2 since the CM compliances are read-only
> >or any other tables do not apply to CMs.
> >
> >Change #9 for the docsIfCmRangingTimeout please review the other
> >question sent to the list and OSSI reflector
> >
> >Please see if the coments are being captured properly.
> >
> >
> >PDF has the color change, also attached text file for convenience.
> >
> >
> >Thanks
> >
> >Eduardo
> >
>
>
>_______________________________________________
>IPCDN mailing list
>[email protected]
>https://www1.ietf.org/mailman/listinfo/ipcdn