Добро пожаловать, Гость. Пожалуйста авторизуйтесь здесь.
FGHIGate на GaNJa NeTWoRK ST@Ti0N - Просмотр сообщения в эхоконференции ENET.SYSOP
Введите FGHI ссылку:


Присутствуют сообщения из эхоконференции ENET.SYSOP с датами от 10 Jul 13 21:42:12 до 20 Sep 24 12:02:56, всего сообщений: 12550
Ответить на сообщение К списку сообщений Предыдущее сообщение Следующее сообщение
= Сообщение: 7004 из 12550 ====================================== ENET.SYSOP =
От   : Markus Reschke                   2:240/1661         02 Jul 18 16:15:56
Кому : Michiel van der Vlist                               02 Jul 18 16:15:56
Тема : DTDNS dynamic DNS hosting support ends
FGHI : area://ENET.SYSOP?msgid=2:240/1661+5b2bac00
На   : area://ENET.SYSOP?msgid=2:280/5555+5b3a2b14
= Кодировка сообщения определена как: LATIN1 =================================
==============================================================================
Hi Michiel!

Jul 02 15:35 2018, Michiel van der Vlist wrote to Benny Pedersen:

BP>> good, i still say gratisdns.dk will not drop ddns, its free of
BP>> charge, no silly ads, or limited services

MvdV> These free services work until the one that keeps it running
MvdV> changes his/her priorities. I have seen it happen before.

I have an idea about a resilient and distributed DynDNS for fidonet by leveraging existing internet servers from nodes. Typically DNS for a domain is handled by a primary nameserver which holds the data and secondary nameservers which fetch a copy of the zone from the primary (controlled by a serial number in the zone and notifications from the primary nameserver). This setup can't be used for a high availibilty DynDNS, because it fails when the primary nameservers has an outage. So we need a way to process the updates from the nodes and to distribute the changes on each participating server.

To keep things simple we can simply use bind, nsupdate (a tool of bind) and a few scripts. All participating servers are listed in a file and are also listed as authoritative nameservers for the domain. Each server runs bind, a web server and a simple nsupdate based DynDNS frontend for updating IP addresses. The nsupdate script will be modified to update all servers in the list to distribute the changes. There's also a file with the DynDNS logins for all nodes.

In case of an outage of a server that server won't process DNS requests and also no updates, none from the nodes using this server and none from the other servers. Nodes using the broken server simply switch to another server for updates. When the server is back online a script will fetch the current zone from one of the other servers and will participate again. The distribution of the server list and DynDNS logins can be automated too (controlled by some coordinator).

ciao,
Markus

---
* Origin: *** theca tabellaria *** (2:240/1661)

К главной странице гейта
Powered by NoSFeRaTU`s FGHIGate
Открытие страницы: 0.091731 секунды