[dns-operations] "best practices" for restaring internal DNS servers

[Rubens Kuhl]

It depends. I can tell from experience that for a large outbound SMTP system(1M+ accounts), loss of DNS cache can hit it pretty hard. 

Rubens


Em 09/09/2012, às 12:06, Steven Carr escreveu:

> Is it really that much of an issue to have to start from an empty
> cache? given that >75% of the cached RRs will have a TTL of <8 hours
> anyway.
> 
> Steve
> 
> 
> On 9 September 2012 14:45, Rubens Kuhl <rubensk at nic.br> wrote:
>>> 
>>> I'm not sure if I phrased my question correctly. It's not about
>>> redundancy, but about keeping the queries to root/g(TLD) name servers
>>> to a minimum.
>>> 
>>> In your example, if 127.0.0.1 was the resolver that just came up again
>>> after a restart, it wouldn't return a failure for a query that it has
>>> not yet cached. Or perhaps I just didn't understand your answer.
>> 
>> If your recursive DNS server runs Unbound, you can use
>> 
>> unbound-control dump-cache > <filename>
>> 
>> prior to restarting
>> and then
>> 
>> unbound-control load-cache < <filename>
>> 
>> I don't recall such an option being available with BIND 9, and it's probably a feature to suggest to BIND 10.
>> 
>> 
>> Rubens
>> 
>> _______________________________________________
>> dns-operations mailing list
>> dns-operations at lists.dns-oarc.net
>> https://lists.dns-oarc.net/mailman/listinfo/dns-operations
>> dns-jobs mailing list
>> https://lists.dns-oarc.net/mailman/listinfo/dns-jobs
> _______________________________________________
> dns-operations mailing list
> dns-operations at lists.dns-oarc.net
> https://lists.dns-oarc.net/mailman/listinfo/dns-operations
> dns-jobs mailing list
> https://lists.dns-oarc.net/mailman/listinfo/dns-jobs