Re: [PATCH net-next v2 0/3] icmp: Add RFC 5837 support

[Ido Schimmel <idosch@nvidia.com>]

On Tue, Oct 28, 2025 at 06:04:32PM -0700, Jakub Kicinski wrote:
> On Mon, 27 Oct 2025 10:22:29 +0200 Ido Schimmel wrote:
> > This patchset extends certain ICMP error messages (e.g., "Time
> > Exceeded") with incoming interface information in accordance with RFC
> > 5837 [1]. This is required for more meaningful traceroute results in
> > unnumbered networks. Like other ICMP settings, the feature is controlled
> > via a per-{netns, address family} sysctl. The interface and the
> > implementation are designed to support more ICMP extensions.
> 
> Is there supposed to be any relation between the ICMP message attrs 
> and what's provided via IOAM? For interface ID in IOAM we have
> the ioam6_id attr instead of ifindex.

RFC 5837 precedes IOAM and I don't see any references from IOAM to RFC
5837. RFC 5837 is pretty clear about the interface index that should be
provided:

"The ifIndex of the interface of interest MAY be included. This is the
32-bit ifIndex assigned to the interface by the device as specified by
the Interfaces Group MIB [RFC2863]".

> Would it make sense to add some info about relation to IOAM to the
> commit msg (or even docs?). Or is it obvious to folks more familiar
> with IP RFCs than I am?

I'm pretty sure that when people are told that the message contains "the
32-bit ifIndex" they don't think about "the wide IOAM id of this
interface".