Skip Menu |

This queue is for tickets about the Net-DNS CPAN distribution.

Report information
The Basics
Id: 29818
Status: resolved
Priority: 0/
Queue: Net-DNS

People
Owner: Nobody in particular
Requestors: ws [...] heise.de
Cc:
AdminCc:

Bug Information
Severity: Important
Broken in: 0.61
Fixed in: (no value)



Subject: Net::DNS::Resolver::Recurse needs hints with some nameservers
Net-DNS-0.61 Perl 5.8.8 Cygwin 1.5.24(0.156/4/2) on Win2K/SP4 Cygwin: cpan installation of Net::DNS fails because make test fails. Some nameservers (208.67.222.222 == resolver1.opendns.com, 208.67.220.220 == resolver2.opendns.com) won't work without hints for the resolver object. Since it only befell me during testing, specifically t/10-recurse.t, I have patched just that file. Someone more familiar with the code should probably fix Net::DNS::Resolver::Recurse itself. Cheers Wolfgang
Subject: Net-DNS-0.61-THIRTYSVN-testrecursivehints-01.patch
--- t/10-recurse.t.orig 2007-08-01 13:48:35.000000000 +0200 +++ t/10-recurse.t 2007-10-07 15:58:51.250000000 +0200 @@ -77,6 +77,8 @@ { my $res = Net::DNS::Resolver::Recurse->new ; my $count; + # Hard code A and K.ROOT-SERVERS.NET hint + $res->hints( "193.0.14.129", "198.41.0.4" ); $res->recursion_callback(sub {
On Sun Oct 07 10:31:01 2007, THIRTYSVN wrote: Show quoted text
> Net-DNS-0.61 > Perl 5.8.8 > Cygwin 1.5.24(0.156/4/2) on Win2K/SP4 > > Cygwin: cpan installation of Net::DNS fails because make test fails. > Some nameservers (208.67.222.222 == resolver1.opendns.com, > 208.67.220.220 == resolver2.opendns.com) won't work without hints for > the resolver object. >
I ran into the same problem at NLnet Labs when we started to test a development recursive namserver. Show quoted text
> Since it only befell me during testing, specifically t/10-recurse.t, > I have patched just that file. Someone more familiar with the code > should probably fix Net::DNS::Resolver::Recurse itself. >
The patch is functional, thanks. I have looked at the Recurse class in order to fix this. But it seems that one needs to maintain quite a lot of state, so much that it seems almost sensible to start caching data. That is a bit to much of an overhaul of the packet for now. Fixed in the forthcoming 0.62