Re: [pfSense Support] Rules based on hostname/dynamic IP address

2007-05-09 Thread Bill Marquette
On 4/27/07, RB <[EMAIL PROTECTED]> wrote: > Authentication by IP is a bad idea, restricting who can connect in the > first place and proceed to authentication stage is a further line of Having been an enterprise firewall admin in the midst of previously established enterprise firewall admins, th

Re: [pfSense Support] Rules based on hostname/dynamic IP address

2007-05-09 Thread Volker Kuhlmann
On Wed 02 May 2007 01:03:20 NZST +1200, sai wrote: > Everytime a packet comes in that might match the rule, you would have > to do a DNS lookup. Not a good idea, as this would REALLY screw up the > latency on your firewall. No, you misunderstood. The rules are static, but one of them is the resul

Re: [pfSense Support] Rules based on hostname/dynamic IP address

2007-05-01 Thread RB
Everytime a packet comes in that might match the rule, you would have to do a DNS lookup. Not a good idea, as this would REALLY screw up the latency on your firewall. Absolutely - it's not without it's detriments. Some of that may be reduced with a good caching name server, but overall the resp

Re: [pfSense Support] Rules based on hostname/dynamic IP address

2007-05-01 Thread sai
Everytime a packet comes in that might match the rule, you would have to do a DNS lookup. Not a good idea, as this would REALLY screw up the latency on your firewall. sai On 4/22/07, Rob Terhaar <[EMAIL PROTECTED]> wrote: don't think this is possible, or a good idea ether. On 4/21/07, Volker

Re: [pfSense Support] Rules based on hostname/dynamic IP address

2007-04-27 Thread RB
My response: if someone is in the position to poison my DNS, they're already in the position to spoof a trusted IP and likely a whole lot more. FWIW, like any good BIND admin, my internal nameserver is pointed at the roots. - T

Re: [pfSense Support] Rules based on hostname/dynamic IP address

2007-04-27 Thread RB
Authentication by IP is a bad idea, restricting who can connect in the first place and proceed to authentication stage is a further line of Having been an enterprise firewall admin in the midst of previously established enterprise firewall admins, the "going wisdom" is that you always set rules

Re: [pfSense Support] Rules based on hostname/dynamic IP address

2007-04-27 Thread Volker Kuhlmann
Thanks for your answers everyone. On Mon 23 Apr 2007 03:59:00 NZST +1200, Rob Terhaar wrote: > don't think this is possible, or a good idea ether. Whether it's a good idea or not depends on what it's being used for. Authentication by IP is a bad idea, restricting who can connect in the first pla

Re: [pfSense Support] Rules based on hostname/dynamic IP address

2007-04-22 Thread Scott Ullrich
On 4/22/07, Josep Pujadas i Jubany <[EMAIL PROTECTED]> wrote: I asked the same thing some months ago. Developper team (Scott Ullrich) said is not a good idea to have rules based with name resolution. I agree, it is very ease to poison/hack name resolution ... [snip] http://www.mail-archive.com/

Re: [pfSense Support] Rules based on hostname/dynamic IP address

2007-04-22 Thread Josep Pujadas i Jubany
On Sun, 22 Apr 2007 10:59:00 -0500, Rob Terhaar wrote > don't think this is possible, or a good idea ether. > > On 4/21/07, Volker Kuhlmann <[EMAIL PROTECTED]> wrote: > > What options are there for creating rules with a hostname which resolves > > to a dynamic IP address? I'd like to allow one hos

Re: [pfSense Support] Rules based on hostname/dynamic IP address

2007-04-22 Thread Rob Terhaar
don't think this is possible, or a good idea ether. On 4/21/07, Volker Kuhlmann <[EMAIL PROTECTED]> wrote: What options are there for creating rules with a hostname which resolves to a dynamic IP address? I'd like to allow one host access inbound access on a tcp port, but that host doesn't have

[pfSense Support] Rules based on hostname/dynamic IP address

2007-04-21 Thread Volker Kuhlmann
What options are there for creating rules with a hostname which resolves to a dynamic IP address? I'd like to allow one host access inbound access on a tcp port, but that host doesn't have a static IP. Unless there's a magic mechanism I don't know about, at least part of the rules would have to be