[Sussex] Heres a Question. not linux but open source related.

[Paul Tansom]

** Alan Pope <alan.pope at gmail.com> [2005-10-26 23:36]:
> On 26/10/05, Paul Tansom <paul at aptanet.com> wrote:
> > DynDNS should work OK, but I'm not entirely happy using that service.
> > I'd have to configure my MX records to point to the DynDNS record and
> > ensure it updated frequently enough - not an ideal setup, in fact if
> > memory serves an MX should reference and IP direct not a CNAME. That
> > said I've not looked at DynDNS in a long while - largely due to always
> > having had a static IP since I've had an internet connection :)
> 
> No, an MX record can point to a CNAME. For years I had my MX record
> for popey.com pointing to popey.dyndns.org which was sat on the end of
> an NTL connection :D

OK, I was talking off the cuff from an unvalidated memory. Then again
(doing exactly the same, I really should check my facts before quoting)
are you talking 'it works' or 'this is technically correct'? I know that
the specific I am thinking of I used to run as a CNAME, but in practise
the correct way to do things was to use the IP. I think in terms of
actually working you can get away with anything as a CNAME (untested!).

> I changed form NTL to UKFSN simply to allow me to have multiple IP
> addresses. If NTL would've done that I probably would have stayed with
> them. I found their service to be very reliable.
> 
> (as an aside, I have a Pace Sky Digi box and have never had a problem
> with it in over 5 years).
> 
> I dunno what you do to computing devices Paul but you really do seem
> to bring out the worst in them ;) Maybe it's your aura? :)

It has been suggested in the past. I actually got banned from a
computing lab when I was doing my degree at Portsmouth. We were learning
68000 assembler and had a lab full of Sinclair QLs. The one I was
working on crashed and refused to boot up again so I was moved to work
with someone else. Their machine promptly crashed and we were both sent
in different directions to work with other people. While the other
machine was fine, the one I moved to crashed. At this point it was
decided that it was close to the end of the lecture and I was asked to
leave - the person whos machine crashed moved to work with someone else!

On the other side of things I keep finding it difficult to diagnose
customer problems because the fault won't occur when I am on site, but
reappears soon after I leave. I still haven't worked out a way of
leaving clones of myself on site (for a fee) to ensure kit works well at
all times. I really must test cardboard cutouts to see if they work well
enough (possibly at a discount rate) ;)

Back when I was at IBM/Xyratex (*) I remember battling it out for a with
one of the other second line support bods for the top number of calls
closed per week (largely as a means to actually enjoy the massive
backlog of outstanding calls). From memory we were closing between 10
and 20 calls a day each, generally with on site visits (although this
being largely a case of walking around the site to the appropriate bay
helped - but it was a pain if you had to travel to the other location by
car as it killed your close rate!) - so I can't be that much of a jinx.

Hmm, that said I did get stuck in the lift for an hour or so once with a
trolley load of PCs. They offered me a job when I got out because having
got decidedly fed up waiting while they tried desperately to work out
how to open the doors I watched from the inside and spotted that the
mechanism was broken and let myself out!

** end quote [Alan Pope]

(*) those were the days, OS/2 on my desktop at work and an Amiga at
home. Somehow even Linux doesn't quite match the fun of my Sinclair,
Amstrad and Amiga days - I wish I could put my finger on just what it
was.

-- 
Paul Tansom | Aptanet Ltd. | http://www.aptanet.com