It's nice to know someone uses the things I write :-)
"Robin Gilks" <[EMAIL PROTECTED]> writes:
> 1. I'd like to extract the UK parts of the ampr.org domain using
> ampr_extractor from Simon Mudd (or similar) but the cache file
> that bind-8 produces is a real mix!! None of the CNAMEs or MX
> records get sorted out but the basic forward IN records seem
> to extract OK.
The MX records and CNAME records _should_ work, at least for the
44.133.x.y domain I've seen this works ok (or did last time I looked)
> Is there a utility that will order the cache file
> into something like the ampr.tgz file on ucsd or is there an updated
> version of ampr_extractor that will cope with these problems.
What version (and date) of my file are you using? I've not touched it for
ages, and don't know if you're perhaps using an old version. The
cache files I guess you are talking about are the entries in /etc/named.conf
and the IN NS entries you'd need in the domain files.
> 2. I'm collecting ampr.org as a slave in bind-8 but would like to
do the ucsd.edu name servers know this? If they don't you'll need to
lie and tell your nameserver that it's a master NOT a slave. Certainly
for the 44.131.x.x information from ampr.org you'll need to tell your
nameserver it's a master as no-one else knows it is. It would also lie
about information that's not in the 44.131.x.x part of ampr.org (which
may or may not be a problem).
> serve it onto another hub who doesn't have net access as readily
> available as me. Trouble is it keeps clearing down. The 'other end'
> is rtt 2.1 secs at 130 bytes/sec away on a 2k4 radio link. Is this
> too slow or is there something odd going on with serial numbers
> or summat twixt my bind-8 and his bind-4.
How are you transferring the zone information to the DNS server
available over the radio-link? Via the dns protocol? Obviously the
whole ampr.org file is going to be large. If you do this daily by
mail you may have less trouble, especially if you are doing this only
for the 44.131.x.x information. You'd then need some sort of procmail
script which takes in the new file, saves it and tells named to
recheck its files (ndc reload/restart). It shouldn't be too hard.
My current ampr_extractor doesn't attempt to extract the timestamp
data in the header (it didn't seem necessary for the use that was
being made of it), though I could probably modify it to do this if
required.
Unfortunately with simple quick-hack tools like this you never know
what's needed until someone tells you, and I think you're only the
second time in over a year to comment on this script.
In case anyone is interested the script and daily extracted information
is available at http://myth.co.uk/~sjmudd/ampr_extractor.html
Regards,
Simon
--
Simon J Mudd, Madrid SPAIN Tel: +34-91-408 4878 email: [EMAIL PROTECTED]
[short messages - from radio hams only] ----> [EMAIL PROTECTED]
