generate a set of request DNSsec

2012-04-18 Thread William Thierry SAMEN
Hi, i'm trying to implement DNSsec on my DNS zones and make a test performance to evaluate the charge of DNSsec on my servers. I'm faced with a big problem, *How can i generate a log file for my test?*it's a big problem for me, i'm working on Bind 9.8.1-P1 and i'm using dnsperf to inject requests

Re: generate a set of request DNSsec

2012-04-18 Thread WBrown
William wrote on 04/18/2012 05:45:21 AM: > I'm faced with a big problem, How can i generate a log file for my test? > it's a big problem for me, i'm working on Bind 9.8.1-P1 > and i'm using dnsperf to inject requests on my servers. > > Did you have an idea? thank you for your help. What do you w

testing validation

2012-04-18 Thread Alan Batie
I'm testing out dnssec with bind 9.9.0's auto signing and a test domain; this appears to be working (see below, RRSIG records returned from the actual nameserver), however and attempt to validate fails with: # dig +dnssec +sigchase soa raindrop.us ;; RRset to chase: raindrop.us.987

RE: testing validation

2012-04-18 Thread Spain, Dr. Jeffry A.
> I'm testing out dnssec with bind 9.9.0's auto signing and a test domain; this > appears to be working (see below, RRSIG records returned from the actual > nameserver), however and attempt to validate fails with: > # dig +dnssec +sigchase soa raindrop.us > When I simply try to validate the root:

Re: testing validation

2012-04-18 Thread Carlos Ribas
Hello, Is your recursive resolver also authoritative for raindrop.us? If so, you will not get the "ad" flag. You can test with DNS-OARC resolver [1]: # dig +dnssec +multiline @149.20.64.20 raindrop.us ; <<>> DiG 9.7.3 <<>> +dnssec +multiline @149.20.64.20 raindrop.us ; (1 server found) ;; gl

Re: testing validation

2012-04-18 Thread Alan Batie
On 4/18/12 10:33 AM, Spain, Dr. Jeffry A. wrote: > Your post is somewhat unclear to me. Querying from my bind 9.9.0 recursive > resolver "dig @localhost raindrop.us +dnssec", I get an AD flag returned, > suggesting that dnssec is working for raindrop.us. In your query "dig +dnssec > +sigchase s

Re: testing validation

2012-04-18 Thread Alan Batie
On 4/18/12 10:46 AM, Carlos Ribas wrote: > Is your recursive resolver also authoritative for raindrop.us? > If so, you will not get the "ad" flag. You can > test with DNS-OARC resolver [1]: > > # dig +dnssec +multiline @149.20.64.20 raindrop.us Why would 149.20.64.20 return ad then? It's no

Re: Test DNSSEC validation

2012-04-18 Thread Jan-Piet Mens
> What is the best way to log DNSSEC failures in Bind without enforcing > DNSSEC validation? > > That is I want to see what Bind would have rejected because of failed > DNSSEC validation, but I do not want to return SERVFAIL to my client. I don't think that is possible without modifying the clien

Re: testing validation

2012-04-18 Thread Carlos Ribas
Because this IP has dnssec enabled and raindrop.us is signed :-) Regards, - Carlos Eduardo Ribas 2012/4/18 Alan Batie > On 4/18/12 10:46 AM, Carlos Ribas wrote: > > > Is your recursive resolver also authoritative for raindrop.us? > > If so, you will not ge

RE: testing validation

2012-04-18 Thread Spain, Dr. Jeffry A.
Alan: Comments on your configuration file: I believe that managed-keys... and zone "." { type hint... are built into bind 9.9.0 recursive resolvers and therefore not needed. You can enable the built in root trust anchor by changing dnssec-validation from yes to auto. I think that listen-on { 12

Re: testing validation

2012-04-18 Thread Alan Batie
On 4/18/12 11:14 AM, Spain, Dr. Jeffry A. wrote: > Alan: Comments on your configuration file: I also forgot to remove the nameserver entries from resolv.conf after installing bind. Sigh. Sorry to bother everyone... Though I am still curious about this from the end of sigchase output: Launch

RE: testing validation

2012-04-18 Thread Spain, Dr. Jeffry A.
> Isn't the "DS for the zone: ." what the "managed-keys" clause provides? > Though putting it back in didn't make the warning go away, so I must be > missing something else here... Any difference with dnssec-validation auto and removing the managed-keys and root hint zone? Jeff.

RE: testing validation

2012-04-18 Thread Spain, Dr. Jeffry A.
> Why would 149.20.64.20 return ad then? It's not authoritative either... As I understand it, you need a dnssec-enabled recursive resolver to get an AD flag returned. An authoritative-only server will never return an AD flag. Jeff. ___ Please visit htt

Re: testing validation

2012-04-18 Thread Alan Batie
On 4/18/12 11:48 AM, Spain, Dr. Jeffry A. wrote: >> Isn't the "DS for the zone: ." what the "managed-keys" clause provides? >> Though putting it back in didn't make the warning go away, so I must be >> missing something else here... > > Any difference with dnssec-validation auto and removing the

RE: testing validation

2012-04-18 Thread Spain, Dr. Jeffry A.
> Though I am still curious about this from the end of sigchase output: > Launch a query to find a RRset of type DS for zone: . > ;; NO ANSWERS: no more > ;; WARNING There is no DS for the zone: . > Isn't the "DS for the zone: ." what the "managed-keys" clause provides? Now I think I see what you

Re: testing validation

2012-04-18 Thread Alan Batie
On 4/18/12 12:18 PM, Spain, Dr. Jeffry A. wrote: >> ;; WARNING There is no DS for the zone: . >> Isn't the "DS for the zone: ." what the "managed-keys" clause provides? > > Now I think I see what you mean. It is my understanding that DS records exist > in parent zones and refer to child zones th

www.glb.hud.gov

2012-04-18 Thread Richard Laager
Are others timing out trying to resolve www.glb.hud.gov? This seems (though I haven't done extensive testing) to only happen to me with BIND. http://dnsviz.net/d/www.glb.hud.gov/dnssec/ shows a couple of DNSKEY warnings, so maybe that's it. I always suspect DNSSEC when I have problems with .gov do