VRRPV3-MIB protocol error trap question

[tony vanderpeet]

I am implementing the protocol error trap in the VRRPV3 MIB and have spotted a possible hole in the MIB. I wonder if this was picked up in discussions and a work-around devised?

When an advertisement is received with an unrecognised VR ID, a protocol error trap is generated, with the object vrrpv3StatisticsProtoErrReason. This object has value vrIdError(4) in this case, but what should the index of the object be? In particular, what should the VR ID part of the index be?

I am inclined to create a variable binding with a fake OID using the unrecognised VR ID, since this will actually convey the information required (which VR ID is invalid, and the address type and ifIndex).

What, if any, discussion has taken place around this issue, and was there any conclusion reached?

Thanks
Tony



Tony van der Peet
Software Architect
Allied Telesis Labs Limited
27 Nazareth Ave
Christchurch, NZ
Tel: [phone] (DDI)
Mob: [phone]

[Kalyan (Srinivas)Tata]

From: Kalyan (Srinivas)Tata
Sent: Wednesday, September 05, 2012 8:11 PM
To: 'tony vanderpeet'
Subject: RE: [VRRP] VRRPV3-MIB protocol error trap question

Hi Tony,

This was an editing error in which the vrIdError was not removed from the vrrpV3StatisticsProtoErrReason when vrrpv3RouterVrIdErrors was added as a global counter.

(exactly for the reason you mentioned)

vrrpv3RouterVrIdErrors should be used for counting the number of vrid errors.
Thanks
Kalyan
From: vrrp-***@ietf.org<mailto:vrrp-***@ietf.org> [mailto:vrrp-***@ietf.org]<mailto:[mailto:vrrp-***@ietf.org]> On Behalf Of tony vanderpeet
Sent: Monday, September 03, 2012 6:25 PM
To: ***@ietf.org<mailto:***@ietf.org>
Subject: [VRRP] VRRPV3-MIB protocol error trap question


I am implementing the protocol error trap in the VRRPV3 MIB and have spotted a possible hole in the MIB. I wonder if this was picked up in discussions and a work-around devised?


When an advertisement is received with an unrecognised VR ID, a protocol error trap is generated, with the object vrrpv3StatisticsProtoErrReason. This object has value vrIdError(4) in this case, but what should the index of the object be? In particular, what should the VR ID part of the index be?


I am inclined to create a variable binding with a fake OID using the unrecognised VR ID, since this will actually convey the information required (which VR ID is invalid, and the address type and ifIndex).


What, if any, discussion has taken place around this issue, and was there any conclusion reached?


Thanks

Tony




Tony van der Peet
Software Architect
Allied Telesis Labs Limited
27 Nazareth Ave
Christchurch, NZ
Tel: [phone] (DDI)
Mob: [phone]

[Adrian Farrel]

Hi,
Is there an Erratum that should be raised for this?
Thanks,
Adrian

From: vrrp-***@ietf.org [mailto:vrrp-***@ietf.org] On Behalf Of Kalyan
(Srinivas)Tata
Sent: 06 September 2012 04:12
To: ***@ietf.org
Subject: [VRRP] FW: VRRPV3-MIB protocol error trap question



From: Kalyan (Srinivas)Tata
Sent: Wednesday, September 05, 2012 8:11 PM
To: 'tony vanderpeet'
Subject: RE: [VRRP] VRRPV3-MIB protocol error trap question

Hi Tony,
This was an editing error in which the vrIdError was not removed from the
vrrpV3StatisticsProtoErrReason when vrrpv3RouterVrIdErrors was added as a global
counter.
(exactly for the reason you mentioned)
vrrpv3RouterVrIdErrors should be used for counting the number of vrid errors.
Thanks
Kalyan
From: vrrp-***@ietf.org [mailto:vrrp-***@ietf.org] On Behalf Of tony
vanderpeet
Sent: Monday, September 03, 2012 6:25 PM
To: ***@ietf.org
Subject: [VRRP] VRRPV3-MIB protocol error trap question

I am implementing the protocol error trap in the VRRPV3 MIB and have spotted a
possible hole in the MIB. I wonder if this was picked up in discussions and a
work-around devised?

When an advertisement is received with an unrecognised VR ID, a protocol error
trap is generated, with the object vrrpv3StatisticsProtoErrReason. This object
has value vrIdError(4) in this case, but what should the index of the object be?
In particular, what should the VR ID part of the index be?

I am inclined to create a variable binding with a fake OID using the
unrecognised VR ID, since this will actually convey the information required
(which VR ID is invalid, and the address type and ifIndex).

What, if any, discussion has taken place around this issue, and was there any
conclusion reached?

Thanks
Tony




Tony van der Peet
Software Architect
Allied Telesis Labs Limited
27 Nazareth Ave
Christchurch, NZ
Tel: [phone] (DDI)
Mob: [phone]

[Kalyan (Srinivas)Tata]

Hi Adrian,
I replied to that email too fast without reviewing the history. Please disregard my reply. Creating the OID with the invalid VRID received and setting the vrrpv3StatisticsProtoErrReason to vrIdError(4) would be the correct behavior.
Thanks
Kalyan
From: Adrian Farrel [mailto:***@olddog.co.uk]
Sent: Monday, September 17, 2012 2:32 PM
To: Kalyan (Srinivas)Tata; ***@ietf.org
Subject: RE: [VRRP] FW: VRRPV3-MIB protocol error trap question

Hi,
Is there an Erratum that should be raised for this?
Thanks,
Adrian

From: vrrp-***@ietf.org<mailto:vrrp-***@ietf.org> [mailto:vrrp-***@ietf.org]<mailto:[mailto:vrrp-***@ietf.org]> On Behalf Of Kalyan (Srinivas)Tata
Sent: 06 September 2012 04:12
To: ***@ietf.org<mailto:***@ietf.org>
Subject: [VRRP] FW: VRRPV3-MIB protocol error trap question



From: Kalyan (Srinivas)Tata
Sent: Wednesday, September 05, 2012 8:11 PM
To: 'tony vanderpeet'
Subject: RE: [VRRP] VRRPV3-MIB protocol error trap question

Hi Tony,

This was an editing error in which the vrIdError was not removed from the vrrpV3StatisticsProtoErrReason when vrrpv3RouterVrIdErrors was added as a global counter.

(exactly for the reason you mentioned)

vrrpv3RouterVrIdErrors should be used for counting the number of vrid errors.
Thanks
Kalyan
From: vrrp-***@ietf.org<mailto:vrrp-***@ietf.org> [mailto:vrrp-***@ietf.org]<mailto:[mailto:vrrp-***@ietf.org]> On Behalf Of tony vanderpeet
Sent: Monday, September 03, 2012 6:25 PM
To: ***@ietf.org<mailto:***@ietf.org>
Subject: [VRRP] VRRPV3-MIB protocol error trap question


I am implementing the protocol error trap in the VRRPV3 MIB and have spotted a possible hole in the MIB. I wonder if this was picked up in discussions and a work-around devised?


When an advertisement is received with an unrecognised VR ID, a protocol error trap is generated, with the object vrrpv3StatisticsProtoErrReason. This object has value vrIdError(4) in this case, but what should the index of the object be? In particular, what should the VR ID part of the index be?


I am inclined to create a variable binding with a fake OID using the unrecognised VR ID, since this will actually convey the information required (which VR ID is invalid, and the address type and ifIndex).


What, if any, discussion has taken place around this issue, and was there any conclusion reached?


Thanks

Tony




Tony van der Peet
Software Architect
Allied Telesis Labs Limited
27 Nazareth Ave
Christchurch, NZ
Tel: [phone] (DDI)
Mob: [phone]