Really it is very hard to help people who keep changing random things making a
moving
target. You started out with a machine at 45.225.75.8 that could reach
2803:1920::c:1963
based on the forward zone declaration so it had to be dual stacked (both IPv4
and IPv6).
You have now added two new machines 69.162.65.138 and 192.169.93.210 which may
or may not
be able to reach 2803:1920::c:1963 and have changed the delegation to point to
them and they
return NXDOMAIN for smtp.ore.org.bo.
Now if you have IPv4 only nameservers that need to get the zone content from an
IPv6 only
server you will need to have an intermediate dual stacked server that can
transfer the zone
content from the IPv6 only server and send it to the IPv4 only servers when
they request it.
P(IPv6-only) -> I(IPv4 and IPv6) -> S(IPv4 only)
Also read the nameserver’s logs. They will help you work out what is going
wrong.
2803:1920::c:1963 (IPv6 only):
zone ore.org.bo {
type primary;
file "ore.org.bo.db”;
};
45.225.75.8/<ipv6addess> (dual stacked):
zone ore.org.bo {
type secondary;
file "ore.org.bo.db”;
primaries { 2803:1920::c:1963; };
};
69.162.65.138 (IPv4 only):
zone ore.org.bo {
type secondary;
file "ore.org.bo.db”;
primaries { 45.225.75.8; };
};
Alternatively you can add IPv6 to an IPv4 only machine using services like
https://tunnelbroker.net/ even when the ISP does not support IPv6.
Mark
> On 15 Jul 2024, at 11:00, Herman Brule <cont...@herman-brule.com> wrote:
>
> I open this to test (45.225.75.8 is particial anycast IP, for DNS/UDP have
> bind9):
> dig A ore.org.bo @199.38.247.210
> With on 199.38.247.210 (work):
> zone ore.org.bo {
> type master;
> file "/etc/bind/ore.org.bo.db";
> };
> ; <<>> DiG 9.18.19-1~deb12u1-Debian <<>> A ore.org.bo @199.38.247.210
> ;; global options: +cmd
> ;; Got answer:
> ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 39291
> ;; flags: qr aa rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 1
>
> ;; OPT PSEUDOSECTION:
> ; EDNS: version: 0, flags:; udp: 1232
> ; COOKIE: 9948d53f96271fa80100000066947311b2e477062b98c6ee (good)
> ;; QUESTION SECTION:
> ;ore.org.bo. IN A
>
> ;; ANSWER SECTION:
> ore.org.bo. 3600 IN A 45.225.75.8
>
> ;; Query time: 99 msec
> ;; SERVER: 199.38.247.210#53(199.38.247.210) (UDP)
> ;; WHEN: Mon Jul 15 00:53:38 UTC 2024
> ;; MSG SIZE rcvd: 83
>
> With on 199.38.247.210 (not work):
> zone ore.org.bo {
> type secondary;
> file “/etc/bind/ore.org.bo.db”;
> primaries { 2803:1920::c:1963; };
> };
>
> ; <<>> DiG 9.18.19-1~deb12u1-Debian <<>> A ore.org.bo @199.38.247.210
> ;; global options: +cmd
> ;; Got answer:
> ;; ->>HEADER<<- opcode: QUERY, status: SERVFAIL, id: 14941
> ;; flags: qr rd ra; QUERY: 1, ANSWER: 0, AUTHORITY: 0, ADDITIONAL: 1
>
> ;; OPT PSEUDOSECTION:
> ; EDNS: version: 0, flags:; udp: 1232
> ; COOKIE: f9006eb35715f0da01000000669473a08e2898af7098316c (good)
> ;; QUESTION SECTION:
> ;ore.org.bo. IN A
>
> ;; Query time: 87 msec
> ;; SERVER: 199.38.247.210#53(199.38.247.210) (UDP)
> ;; WHEN: Mon Jul 15 00:56:01 UTC 2024
> ;; MSG SIZE rcvd: 67
> alpha_one_x86/BRULE Herman <alpha_one_...@first-world.info>
> Main developer of Supercopier/Ultracopier/CatchChallenger, Esourcing and
> server management
> IT, OS, technologies, research & development, security and business department
> On 7/14/24 20:00, Mark Andrews wrote:
>>
>>
>>> On 13 Jul 2024, at 12:44, Herman Brule <cont...@herman-brule.com> wrote:
>>>
>>> Thanks, I'm looking how solve this, cleanly.
>>> In my country only 1 ISP have IPv6, then I need keep IPv4.
>>> I have 1 IPv4 for 1000 VPS, no way here to have more IPv4.
>>> Then:
>>> 1) I'm not sure if my DNS authoritative on IPv6 reply correctly (but reply
>>> correctly to all my dig query)
>>> 2) I have to provide a way to my customer can resolve query on their DNS
>>> server on their IPv6 VPS, their need be able to just put their vps dns or
>>> at least common server dns (where I had to put their zone, then I dislike
>>> this idea)
>>> For now your method fail, include I try:
>>> zone "ore.org.bo" {
>>> type master;
>>> file "/etc/bind/ore.org.bo.db";
>>> };
>>> But failed too.
>>>
>> Well I didn’t say to do that. You have they wrong type of zone. Make it a
>> secondary (slave) zone
>> like I told you to do.
>>
>> zone ore.org.bo {
>> type secondary;
>> file “ore.org.bo.db”;
>> primaries { 2803:1920::c:1963; };
>> };
>>
>> Now that should work as I can AXFR the zone from that server. You should
>> also note the difference in
>> the flags in the responses for smtp.ore.org.bo. The one from
>> 2803:1920::c:1963 is an authoritative
>> reply (aa) and the TTL stays at 3600, whereas the one from 45.225.75.8 is
>> not (aa is not set in the
>> flags) and the TTL decreases indicating that it comes from a recursive
>> server.
>>
>> It also looks like someone tried to comment out *.ore.org.bo but used the
>> wrong comment character ‘#'
>> rather than ‘;’.
>>
>> [ant:~/git/bind9] marka% dig axfr ore.org.bo @2803:1920::c:1963
>>
>> ; <<>> DiG 9.19.25-dev <<>> axfr ore.org.bo @2803:1920::c:1963
>> ;; global options: +cmd
>> ore.org.bo. 604800 IN SOA 811.vps.confiared.com. admin.ore.org.bo. 3 604800
>> 86400 2419200 604800
>> ore.org.bo. 604800 IN NS 811.vps.confiared.com.
>> ore.org.bo. 3600 IN MX 1 smtp.ore.org.bo.
>> ore.org.bo. 3600 IN A 45.225.75.8
>> ore.org.bo. 3600 IN AAAA 2803:1920::c:1963
>> #*.ore.org.bo. 604800 IN CNAME ore.org.bo.
>> smtp.ore.org.bo. 3600 IN A 45.225.75.8
>> smtp.ore.org.bo. 3600 IN AAAA 2803:1920::c:1963
>> www.ore.org.bo. 604800 IN CNAME ore.org.bo.
>> ore.org.bo. 604800 IN SOA 811.vps.confiared.com. admin.ore.org.bo. 3 604800
>> 86400 2419200 604800
>> ;; Query time: 497 msec
>> ;; SERVER: 2803:1920::c:1963#53(2803:1920::c:1963) (TCP)
>> ;; WHEN: Mon Jul 15 09:47:16 AEST 2024
>> ;; XFR size: 10 records (messages 1, bytes 324)
>>
>> [ant:~/git/bind9] marka% dig a smtp.ore.org.bo @2803:1920::c:1963
>>
>> ; <<>> DiG 9.19.25-dev <<>> a smtp.ore.org.bo @2803:1920::c:1963
>> ;; global options: +cmd
>> ;; Got answer:
>> ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 4584
>> ;; flags: qr aa rd; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 1
>> ;; WARNING: recursion requested but not available
>>
>> ;; OPT PSEUDOSECTION:
>> ; EDNS: version: 0, flags:; udp: 1232
>> ; COOKIE: cdbac4bb692528b301000000669463a2ffd887df1b2535a8 (good)
>> ;; QUESTION SECTION:
>> ;smtp.ore.org.bo. IN A
>>
>> ;; ANSWER SECTION:
>> smtp.ore.org.bo. 3600 IN A 45.225.75.8
>>
>> ;; Query time: 659 msec
>> ;; SERVER: 2803:1920::c:1963#53(2803:1920::c:1963) (UDP)
>> ;; WHEN: Mon Jul 15 09:47:47 AEST 2024
>> ;; MSG SIZE rcvd: 88
>>
>> [ant:~/git/bind9] marka% dig a smtp.ore.org.bo @45.225.75.8
>>
>> ; <<>> DiG 9.19.25-dev <<>> a smtp.ore.org.bo @45.225.75.8
>> ;; global options: +cmd
>> ;; Got answer:
>> ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 33189
>> ;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 1
>>
>> ;; OPT PSEUDOSECTION:
>> ; EDNS: version: 0, flags:; udp: 1232
>> ; COOKIE: 42c6758d745eb62b01000000669463baea9db7cd3474c256 (good)
>> ;; QUESTION SECTION:
>> ;smtp.ore.org.bo. IN A
>>
>> ;; ANSWER SECTION:
>> smtp.ore.org.bo. 3266 IN A 45.225.75.8
>>
>> ;; Query time: 264 msec
>> ;; SERVER: 45.225.75.8#53(45.225.75.8) (UDP)
>> ;; WHEN: Mon Jul 15 09:48:10 AEST 2024
>> ;; MSG SIZE rcvd: 88
>>
>> [ant:~/git/bind9] marka%
>>
>> Mark
>>
>>
>>> alpha_one_x86/BRULE Herman <alpha_one_...@first-world.info>
>>> Main developer of Supercopier/Ultracopier/CatchChallenger, Esourcing and
>>> server management
>>> IT, OS, technologies, research & development, security and business
>>> department
>>> On 7/12/24 19:01, Mark Andrews wrote:
>>>
>>>>
>>>>
>>>>> On 13 Jul 2024, at 04:38, Herman Brule via bind-users
>>>>> <bind-users@lists.isc.org> wrote:
>>>>>
>>>>> Because the customer are into IPv6 zone
>>>>>
>>>>>
>>>> Well all zones should be served by both IPv4 servers and IPv6 servers.
>>>> IPv6 is nearly 30 years old now. There are
>>>> sites that are IPv6 only because they would prefer to not have to run
>>>> everything through 2 or 3 layers of NAT when
>>>> they don’t need it at all for IPv6 and would really like to not have to
>>>> send all there DNS queries though NAT64 boxes.
>>>>
>>>>
>>>>
>>>>> And the EDGE router connecting IPv4 and IPv6 is internal to the data
>>>>> center company, not accessible for the customer.
>>>>> Forward zone to edge will be more complex, it's more simple just forward
>>>>> the query.
>>>>> Thanks for you observation, but I know, I doing this quickly, I will keep
>>>>> like this for now, this will produce only problem for availability if the
>>>>> server is down.
>>>>>
>>>>>
>>>> Except you are wrong. You are writing here because it *is* causing you and
>>>> everyone else a problem. The correct way to
>>>> fix this is to transfer the zone contents to the listed primary servers if
>>>> you are using nameservers. Alternatively
>>>> don’t run nameservers at all but use IP level proxies. Either the whole
>>>> address or port forward 53/TCP and 53/UDP.
>>>>
>>>>
>>>>
>>>>> alpha_one_x86/BRULE Herman <alpha_one_...@first-world.info>
>>>>> Main developer of Supercopier/Ultracopier/CatchChallenger, Esourcing and
>>>>> server management
>>>>> IT, OS, technologies, research & development, security and business
>>>>> department
>>>>> On 7/12/24 14:28, Marco Moock wrote:
>>>>>
>>>>>
>>>>>> Am 12.07.2024 um 14:13:03 Uhr schrieb Herman Brule via bind-users:
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>> bind to my proxy from IPv4 to IPv6 zone
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>> Why don't you simply run multiple authoritative servers, some only
>>>>>> accessible by IPv6, some dual-stack?
>>>>>>
>>>>>> They are independent of each other and only the zone transfer need to
>>>>>> work.
>>>>>>
>>>>>> I also see some strange things:
>>>>>>
>>>>>> m@ryz:~$ host 811.vps.confiared.com.
>>>>>> 811.vps.CONFIARED.com has address 45.225.75.8
>>>>>> 811.vps.CONFIARED.com has IPv6 address 2803:1920::c:1963
>>>>>> m@ryz:~$ host 811b.vps.confiared.com.
>>>>>> 811b.vps.CONFIARED.com is an alias for 811.vps.confiared.com.
>>>>>> 811.vps.CONFIARED.com has address 45.225.75.8
>>>>>> 811.vps.CONFIARED.com has IPv6 address 2803:1920::c:1963
>>>>>> m@ryz:~$
>>>>>>
>>>>>> You should have redundant servers and not 2 NS records that point to
>>>>>> the same machine.
>>>>>>
>>>>>> Please fix that first and update your glue records.
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>> --
>>>>> Visit https://lists.isc.org/mailman/listinfo/bind-users to unsubscribe
>>>>> from this list
>>>>>
>>>>> ISC funds the development of this software with paid support
>>>>> subscriptions. Contact us at https://www.isc.org/contact/ for more
>>>>> information.
>>>>>
>>>>>
>>>>> bind-users mailing list
>>>>> bind-users@lists.isc.org
>>>>> https://lists.isc.org/mailman/listinfo/bind-users
>>>>>
>>>>>
>>>>
>>> <ore.org.bo.db>
>>>
>>
--
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742 INTERNET: ma...@isc.org
--
Visit https://lists.isc.org/mailman/listinfo/bind-users to unsubscribe from
this list
ISC funds the development of this software with paid support subscriptions.
Contact us at https://www.isc.org/contact/ for more information.
bind-users mailing list
bind-users@lists.isc.org
https://lists.isc.org/mailman/listinfo/bind-users
