I agree that we should keep the BIND syntax and separate port and IP address with a space. We will at least avoid possible issues with IP address decoding in the future.
Since this is a new attribute we have a good chance to do changes now so that it is used correctly. I created an upstream ticket to change the behavior and validators in FreeIPA: https://fedorahosted.org/freeipa/ticket/2462 Martin On Wed, 2012-02-29 at 16:44 +0100, Petr Spacek wrote: > On 02/29/2012 04:30 PM, Simo Sorce wrote: > > Either way looks ok to me. > > I agree that using a space may be less confusing if this syntax never > > allows to specify multiple addresses. > > If multiple address can be specified than it may be less ideal to use > > spaces. > > > > Simo. > > idnsForwarders is multi-value attribute, so each value contain single > forwarder address. > > Petr^2 Spacek > > > On Wed, 2012-02-29 at 15:14 +0100, Petr Spacek wrote: > >> And there is the patch, sorry. > >> > >> Petr^2 > >> > >> On 02/29/2012 03:10 PM, Petr Spacek wrote: > >>> Hello, > >>> > >>> this patch fixes https://fedorahosted.org/bind-dyndb-ldap/ticket/49 , > >>> but I want to discuss one (unimplemented) change: > >>> > >>> I propose a change in (currently very strange) forwarders syntax. > >>> > >>> Current syntax: > >>> <IP>[.port] > >>> > >>> examples: > >>> 1.2.3.4 (without optional port) > >>> 1.2.3.4.5553 (optional port 5553) > >>> A::B (IPv6, without optional port) > >>> A::B.5553 > >>> ::FFFF:1.2.3.4 (6to4, without optional port) > >>> ::FFFF:1.2.3.4.5553 (6to4, with optional port 5553) > >>> > >>> I find this syntax confusing, non-standard and not-typo-proof. > >>> > >>> > >>> IMHO better choice is to follow BIND forwarders syntax: > >>> <IP> [port ip_port] (port is string delimited with spaces) > >>> > >>> (From: http://www.zytrax.com/books/dns/ch7/queries.html#forwarders) > >>> > >>> > >>> *Current syntax is not documented*, so probably is not used anywhere. > >>> (And DNS server on non-standard port is probably useful only for testing > >>> purposes, but it's another story.) > >>> > >>> > >>> What is you opinion? > >> _______________________________________________ > >> Freeipa-devel mailing list > >> Freeipa-devel@redhat.com > >> https://www.redhat.com/mailman/listinfo/freeipa-devel > > _______________________________________________ > Freeipa-devel mailing list > Freeipa-devel@redhat.com > https://www.redhat.com/mailman/listinfo/freeipa-devel _______________________________________________ Freeipa-devel mailing list Freeipa-devel@redhat.com https://www.redhat.com/mailman/listinfo/freeipa-devel