[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: BerkeleyLUG domains, "hosting", etc.

From: "Rick Moen" <rick@linuxmafia.com>
Subject: Re: BerkeleyLUG domains, "hosting", etc.
Date: Tue, 12 Mar 2019 13:47:25 -0700



WordPress site about online poker.)  People justify the extra domain
expenditure on grounds of 'future growth' and 'preventing your enemies /
people you disapprove of creating squatter sites', but mostly it ends up
being just a money drain for no real benefit.  IMO.


Yep, I rather agree.  Unless one is a large entity with lots of money
to throw around, generally not worth snagging additional domains.
But hey, if (other :-)) folks want to through some (enough) money
at it to support additional domains, I'm fine with that - the incremental
work to additionally support them is very small - so I've no problem with
that part of it.  E.g. SF-LUG - .org is and has always been canonical,
and I'm fine with paying for that domain.  There's also non-canonical
SF-LUG.com - I don't see continued need for it, but someone else was fine
with paying the domain renewal, so I/we still also support it (it
basically just redirects) - at least thus far.
And, if one has lots of bucks to throw around, registering lots of
domains can be semi-useful to prevent others from snagging them ...
think of it as semi-cheap insurance ... not a guarantee at all, but
reduces the probabilities of some headaches (while adding some other
nuisances and costs).  For Big Compapny or Major Institution, it might be
well worth it (they're also larger/"juicier" targets), for the mom & pop
operation or [L]UG or small organization/club, probably not worth the
bother/expense.


There's also www. prefix - or not.  As far as I'm aware,
BerkeleyLUG has generally effectively not used www. (it redirecting
to without the www. prefix).


That's what I do.

You need to have something to handle queries for www because, no matter
how painstakingly you specify the URL, there are a bunch of people
who'll automatically prepend 'www.' to any domain.


And not just people, but yes, certainly do *also* cover with the
www. prefix - if without is canonical, just (301) redirect,
if with is canonical, just do the redirect the other way around.


With www. can be more flexible -
e.g. mail/list/other services, can be more easily separated out to
different sites/hosts, by using a separate (sub-)domain.


Without objection, I'll point out that you can still do that even if
(as at present for BerkeleyLUG, and also for my two personal domains)
your Web server lacks 'www.' in its FQDN.

Like, I've never bothered to have a 'mail.linuxmafia.com' subdomain, but
I certainly could, and favouring 'http://linuxmafia.com/' over
'http://www.linuxmafia.com/' for the Web server in no way prevents that
option from remaining practical if I ever want it.


Yes, true, but can be *slightly* simpler.  E.g. BALUG.
There's www. and lists. (among others) under balug.org.
Some of that infrastructure makes it a fair bit easier to
have things be totally separate ... if so desired.  E.g. one could
take www. and lists. and delegate their DNS to totally independent
hosting providers - making DNS easier for the DNS admin of balug.org,
as they wouldn't generally need to be pestered with DNS changes for
www.balug.org and lists.balug.org stuff.

But yeah, the (potential) advantage of having / being able to split it out
like that is pretty small ... debatable if it's worth the effort/hassle.


If you run out of hobbies ;->  , you can post the DNS zonefiles here and
we can bikeshed over how to improve them.  (They're probably OK but we
could look.)


Probably mostly premature to optimize - as it's still presently Google
hosted - and that puts some limits on what one can do with it.
But the DNS data is available.  Can't do AXFR with it being Google hosted,
however, I've inspected the data and collected it - one can have a look on:
https://www.wiki.balug.org/wiki/doku.php?id=berkeleylug:digital_resources
Not only is the data for both domains seen there, but there's handy (long)
"one-liner" bit 'o script that can be used to freshly gather that data at
any time.  The only bit one can't verify without being able to "see" the
full direct DNS(/zone) data, is what are all the subdomains.  Well, not
planning to add any (at least while it's still Google hosted), but that
bit 'o script pulls all the relevant DNS data for both domains and any
of those subdomains that do - or might - exist on either domain.
And, not *quite* optimal format for a zone file, but mostly quite usable
as-is to go into zone file data format.
Anyway, there will be clean-up and optimizations and other changes to be
made - much of what's in there now is sub-optimal ... but "good enough"
(passable) for the shorter-term.

--
You received this message because you are subscribed to the Google Groups "BerkeleyLUG" group.
To unsubscribe from this group and stop receiving emails from it, send an email to berkeleylug+unsubscribe@googlegroups.com.
To post to this group, send email to berkeleylug@googlegroups.com.
Visit this group at https://groups.google.com/group/berkeleylug.
For more options, visit https://groups.google.com/d/optout.