[<prev] [next>] [day] [month] [year] [list]
Message-ID: <a44d49c8-a847-5efa-7ff1-4bea8cf6d5fc@nurealm.net>
Date: Mon, 11 Jul 2022 13:56:34 -0600
From: James Feeney <james@...ealm.net>
To: netdev@...r.kernel.org
Subject: iproute2 - Feature Request - automatically distinguish IPv4/IPv6
addresses in "ip rule"
I notice that "ip route" will automatically distinguish IPv4 and IPv6 addresses, so that "ip route add to <some IPv6 address>...", without the "-6" or "-family inet6", is acceptable, but that "ip rule add from <some IPv6 address> ...", without the "-6", will give "Error: Invalid source address." when referencing an IPv6 address.
Is that behavior with "ip rule" from something fundamental, or just a missing feature?
When feeding variables into an automated route and rule script for dual-stack configuration, an "ip route" template "just works" with both IPv4 and IPv6 addresses. But "ip rule", in contrast, requires separate "ip rule" and "ip -6 rule" command templates for each address type.
Not a big deal, of course, but I thought I'd ask if "ip rule" could be made to automatically distinguish IPv4 and IPv6 address types, so that "ip rule" could simply "do the right thing", by itself, without having to specify the address family.
The ip(8) man page says, in reference to "protocol family", "If this option is not present, the protocol family is guessed from other arguments. If the rest of the command line does not give enough information to guess the family, ip falls back to the default one, usually inet or any." So I'm guessing that the IPv6 address limitation with "ip rule" might simply be an oversight.
Powered by blists - more mailing lists