You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@kafka.apache.org by Luke Chen <sh...@gmail.com> on 2026/04/27 06:27:42 UTC
Re: [DISCUSS] KIP-1276: CIDR-based Host Patterns for ACLs
Hi Maros,
Thanks for the KIP.
+1 from me.
I only have one suggestion on the implementation part.
I saw we're going to distinguish between ipv4 and ipv6 by the ":" character.
While it might work in most cases, I think it is a weak check and
error-prone.
Could we use the built-in java InetAddress class to have a safer check?
ref: https://stackoverflow.com/a/18102954
Thanks,
Luke
On Tue, Mar 17, 2026 at 9:55 PM Maroš Orsák <ma...@gmail.com>
wrote:
> Hello,
>
> I have updated the KIP to include the IPv4-mapped IPv6 address section
> within compatibility. And also, I refined/modified the downgrade part to
> include the pre-validation step (without it we would be in an
> inconsistent state) and lastly added a section which talks about the test
> plan.
>
> Let me know what you think.
>
> Cheers,
>
> Maros
>
Re: [DISCUSS] KIP-1276: CIDR-based Host Patterns for ACLs
Posted by Federico Valeri <fe...@gmail.com>.
Hi Maros, looks mostly good, just a few additional questions:
FV7: Downgrades safety: Should we list the offending ACLs in the error message?
FV8: IPv4-mapped IPv6 address: Are we detecting this notation and
returning an error to the user?
Thanks
Fede
On Mon, Apr 27, 2026 at 8:28 AM Luke Chen <sh...@gmail.com> wrote:
>
> Hi Maros,
>
> Thanks for the KIP.
> +1 from me.
>
> I only have one suggestion on the implementation part.
> I saw we're going to distinguish between ipv4 and ipv6 by the ":" character.
> While it might work in most cases, I think it is a weak check and
> error-prone.
> Could we use the built-in java InetAddress class to have a safer check?
> ref: https://stackoverflow.com/a/18102954
>
> Thanks,
> Luke
>
> On Tue, Mar 17, 2026 at 9:55 PM Maroš Orsák <ma...@gmail.com>
> wrote:
>
> > Hello,
> >
> > I have updated the KIP to include the IPv4-mapped IPv6 address section
> > within compatibility. And also, I refined/modified the downgrade part to
> > include the pre-validation step (without it we would be in an
> > inconsistent state) and lastly added a section which talks about the test
> > plan.
> >
> > Let me know what you think.
> >
> > Cheers,
> >
> > Maros
> >