[mpls] MPLS-RT Review of draft-gandhi-mpls-ioam-sr-04

[Mach Chen <[e-mail]>]

Hi, 

I've been asked to provide a pre-adoption MPLS-RT review of draft-gandhi-mpls-ioam-sr-04.

Below are my review comments:

1. I agree that it's more reasonable to leverage GAL/gACH with one or two new dedicated channel types for indicating the presence of iOAM data.

2. In addition, the draft proposes to support both E2E and hop-by-hop (HBH) options. It's nature to use GAL/gACH or the IOAM Indicator Label (IIL) to indicate the presence of iOAM data.  But for HBH case, each intermediate LSR is supposed to process the iOAM data, and the GAL or IIL is put at the bottom of the stack, how does an intermediate LSR know whether there is iOAM data present and it should process the iOAM data?  To scan the whole label stack to find the GAL/IIL or through other means (e.g., put an indicator label on the top the label stack, this may require that each LSR need to pop the IIL label firstly, then process the next label, and put the IIL label back onto the top when forwarding the packet to the next hop) ?  For either way, this needs to be explicitly defined in the document. 

I am OK with other parts of the draft, but the above comments would result in non-trivial updates to the draft, I would see an update before the WG adoption call.

Best regards,
Mach