[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [idn] Internationalized PTR draft submitted
- To: Randy Bush <randy@psg.com>
- Subject: Re: [idn] Internationalized PTR draft submitted
- From: jiang@i-dns.net
- Date: Thu, 21 Sep 2000 02:46:56 +0800
- Cc: James Seng <James@Seng.cc>, idn@ops.ietf.org
- Delivery-date: Wed, 20 Sep 2000 11:52:41 -0700
- Envelope-to: idn-data@psg.com
- Mail-Followup-To: Randy Bush <randy@psg.com>, James Seng <James@Seng.cc>,idn@ops.ietf.org
Hi All,
Been very busy, sorry. :)
As answered by Brian and others(Thanks), this draft really
meant "language" instead of "charset".
There are some unresolved issues with this ID as has been
pointed out, some of them are also mentioned in the open
issues section of the ID. e.g. If parts of the DN can be fixed at will
in different language convertions, then the usefulness of the
language tag is questionable.
This ID does not mean to address a big problem, which needs to
be solved urgently. Most applications today are not concerned
with the content(like language info) of a Domain name.
OK, now getting back to Randy...
>From: Randy Bush <randy@psg.com>
>Subject: Re: [idn] Internationalized PTR draft submitted
>To: James Seng <James@Seng.cc>
>Cc: idn@ops.ietf.org
>Delivered-To: jiang@localhost
>Delivered-To: jiang@i-dns.net
>Date: Mon, 18 Sep 2000 10:28:28 -0700
>
>multiple names for one ip is a red herring (idiom for something which
>distracts one from the proper path).
>
>666.42.7.1.in-addr.arpa. PTR my.dom.ain.
> PTR another.dom.ain.
> PTR yet.another.name.
>
>is perfectly legal and in common use.
Yes, it is true that most implementations of named return
all PTR record faithfully.
The problem is with the resolvers, e.g. I am running a RH6.2
based linux. My program calling gethostbyaddr will be linked
to glibc-2.1.2. I have written a very simple prog (attached) to
test out.
************** result *****************
[jiang@ajay iptr]$ dig in ptr 252.0.168.192.in-addr.arpa
; <<>> DiG 8.2 <<>> in ptr 252.0.168.192.in-addr.arpa
;; res options: init recurs defnam dnsrch
;; got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 4
;; flags: qr aa rd ra; QUERY: 1, ANSWER: 2, AUTHORITY: 1, ADDITIONAL: 0
;; QUERY SECTION:
;; 252.0.168.192.in-addr.arpa, type = PTR, class = IN
;; ANSWER SECTION:
252.0.168.192.in-addr.arpa. 1D IN PTR ajay.i-dns.net.
252.0.168.192.in-addr.arpa. 1D IN PTR jiang.i-dns.net.
;; AUTHORITY SECTION:
0.168.192.in-addr.arpa. 1D IN NS ns.i-dns.net.
;; Total query time: 1 msec
;; FROM: ajay.i-dns.net to SERVER: default -- 127.0.0.1
;; WHEN: Thu Sep 21 02:30:37 2000
;; MSG SIZE sent: 44 rcvd: 131
[jiang@ajay iptr]$ ./a.out
1st try for 127.0.0.1
juice.juice.
ajay.i-dns.net.
localhost.
2nd try for 192.168.0.252
ajay.i-dns.net
[jiang@ajay iptr]$
**************** end of result **********************
****************my /etc/hosts ****************
127.0.0.1 juice.juice. ajay.i-dns.net. localhost.
192.168.0.66 mingliang
192.168.0.19 dev3
216.168.235.56 starlight
**************** end of my /etc/hosts *************
**************** my ptr zone file *****************
@ IN SOA ns.i-dns.net. hostmaster.i-dns.net (
1 ; Serial
10800 ; Refresh after 3 hours
3600 ; Retry after 1 hour
604800 ; Expire after 1 week
86400 ) ; Minimum TTL of 1 day
IN NS ns.i-dns.net.
252 IN PTR ajay.i-dns.net.
252 IN PTR jiang.i-dns.net.
***************** end of my ptr zone file ***********
This is actually not a prob, because in RFC1034 it says:
****************** RFC1034 ***********************
Domain names in RRs which point at another name should always point at
the primary name and not the alias. This avoids extra indirections in
accessing information. For example, the address to name RR for the
above host should be:
52.0.0.10.IN-ADDR.ARPA IN PTR C.ISI.EDU
****************** end of RFC1034 ***********************
OK if it were a prob, then millions of the linux boxes out
there are problematic. :)
The Semantic of PTR is well defined, now that if an IP should
point to multiple iDNs in different languages, we may need to
define something else like IPTR. That is why in out ID:
****************draft-ietf-idn-iptr-00.txt***************
When Internationalized Domain Names come into wide use, an Internet
host is likely to have domain names in different languages. In
today's Internet, because of the design of the PTR record and
implementation of most resolvers, IP address to domain names mapping
is limited to "one IP one domain name", the primary domain name of the
host. This is more restrictive in a world of iDNs, for choosing one
name in one particular language as the primary could have cultural
implications. The authors also believe that putting language
information into address-to-name mappings will be benifitial to future
applications.
************** end of draft-ietf-idn-iptr-00.txt**********
OK, sorry for not making thing clear in the first place. :)
back to work now.
Mingliang
#include <arpa/inet.h>
#include <netdb.h>
main ()
{
struct in_addr addr;
struct hostent *p;
char ** str_p;
printf("1st try for 127.0.0.1\n");
inet_pton(AF_INET, "127.0.0.1", &addr);
p = gethostbyaddr(&addr, sizeof(addr), AF_INET);
if (p->h_name) printf("%s\n", p->h_name);
str_p = p->h_aliases;
while (*str_p) {
printf("%s\n", *str_p);
str_p++;
}
// free(p);
printf("2nd try for 192.168.0.252\n");
inet_pton(AF_INET, "192.168.0.252", &addr);
p = gethostbyaddr(&addr, sizeof(addr), AF_INET);
if (p->h_name) printf("%s\n", p->h_name);
str_p = p->h_aliases;
while (*str_p) { printf("%s\n", *str_p); str_p++; }
}