Re: lwdnsq(8)
Henry Spencer <[email protected]>
| Newsgroups | gmane.network.freeswan.devel |
|---|---|
| Message-ID | <[email protected]> |
On Mon, 3 Feb 2003, Michael Richardson wrote:
> Henry> Section 6.2.2 of RFC 1034 is "conjecture"? Or have I not understood
> Henry> the issue here?
>
> Well, let me to the experient again.
Dead easy. I happen to know that zoo.toronto.edu, in addition to the
usual records, has a WKS record (a historical accident). First, I ask my
local server for everything it knows about zoo ("-a" means "-v -t any"):
[henry@tinker henry]$ host -a zoo.toronto.edu.
rcode = 0 (Success), ancount=5
The following answer is not authoritative:
The following answer is not verified as authentic by the server:
zoo.toronto.edu 28207 IN NS zoo.utoronto.ca
zoo.toronto.edu 28207 IN NS zircon.zoo.utoronto.ca
zoo.toronto.edu 28207 IN A 128.100.72.1
zoo.toronto.edu 86318 IN MX 0 zoo.toronto.edu
zoo.toronto.edu 86318 IN SOA zoo.utoronto.ca
namemaster.zoo.utoronto.ca(
2003012102 ;serial (version)
10800 ;refresh period
1800 ;retry refresh this often
604800 ;expiration period
86400 ;minimum TTL
)
For authoritative answers, see:
zoo.toronto.edu 28207 IN NS zoo.utoronto.ca
zoo.toronto.edu 28207 IN NS zircon.zoo.utoronto.ca
Additional information:
zoo.utoronto.ca 28207 IN A 128.100.72.1
zircon.zoo.utoronto.ca 28207 IN A 128.100.72.2
zoo.toronto.edu 28207 IN A 128.100.72.1
That's all the usual garbage, because I access that machine with some
frequency so all the usual things are in my server's cache. Note that
there is no WKS record. Now, I ask specifically about WKS, and get it:
[henry@tinker henry]$ host -v -t wks zoo.toronto.edu.
rcode = 0 (Success), ancount=1
The following answer is not verified as authentic by the server:
zoo.toronto.edu 86400 IN WKS 128.100.72.1 tcp smtp
For authoritative answers, see:
zoo.toronto.edu 28188 IN NS zoo.utoronto.ca
zoo.toronto.edu 28188 IN NS zircon.zoo.utoronto.ca
Additional information:
zoo.utoronto.ca 28188 IN A 128.100.72.1
zircon.zoo.utoronto.ca 28188 IN A 128.100.72.2
Note that it really does exist, even though it wasn't in the -a reply.
Now, I ask for everything again:
[henry@tinker henry]$ host -a zoo.toronto.edu.
rcode = 0 (Success), ancount=6
The following answer is not authoritative:
The following answer is not verified as authentic by the server:
zoo.toronto.edu 28174 IN NS zoo.utoronto.ca
zoo.toronto.edu 28174 IN NS zircon.zoo.utoronto.ca
zoo.toronto.edu 28174 IN A 128.100.72.1
zoo.toronto.edu 86285 IN MX 0 zoo.toronto.edu
zoo.toronto.edu 86285 IN SOA zoo.utoronto.ca
namemaster.zoo.utoronto.ca(
2003012102 ;serial (version)
10800 ;refresh period
1800 ;retry refresh this often
604800 ;expiration period
86400 ;minimum TTL
)
zoo.toronto.edu 86386 IN WKS 128.100.72.1 tcp smtp
For authoritative answers, see:
zoo.toronto.edu 28174 IN NS zoo.utoronto.ca
zoo.toronto.edu 28174 IN NS zircon.zoo.utoronto.ca
Additional information:
zoo.utoronto.ca 28174 IN A 128.100.72.1
zircon.zoo.utoronto.ca 28174 IN A 128.100.72.2
zoo.toronto.edu 28174 IN A 128.100.72.1
And *now* there's a WKS there. You get the WKS from type=any only if it
happens to be in the cache because of some previous query. More generally,
if there's *anything at all* in the cache about that host, type=any gives
you that and only that.
Henry Spencer
[email protected]