Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 04 Sep 1996 21:52:54 +0200
From:      "Julian H. Stacey" <jhs@freebsd.org>
To:        "Justin T. Gibbs" <gibbs@freefall.freebsd.org>
Cc:        current@freebsd.org
Subject:   Re: Latest Current build failure 
Message-ID:  <199609041952.VAA03250@vector.jhs.no_domain>
In-Reply-To: Your message of "Wed, 04 Sep 1996 08:19:58 PDT." <199609041519.IAA16653@freefall.freebsd.org> 

next in thread | previous in thread | raw e-mail | index | archive | help
Hi, Reference:
> From: "Justin T. Gibbs" <gibbs@freefall.freebsd.org> 
> Subject: Re: Latest Current build failure 
> Date: Wed, 04 Sep 1996 08:19:58 -0700 
> Message-id: <199609041519.IAA16653@freefall.freebsd.org> 
>
> >But CVSup is not the only appropriate solution ...
> >We've had developers who produced useful code but were doomed to bad coms
> >links.  CTM is asynchronous to net disturbances, so ideal for those with 
> >poor net access,  whereas cvsup requires a net in good condition.
> 
> This isn't true since CVSup is a streaming protocol instead of a
> synchronous like SUP.  I know quite a few people who switched from
> CTM to CVSup that have poor links to the net.

Wrong.

`Poor' net access does not just mean _slow_ access with long latency.
`Poor' net access can also mean broken Internet relay hosts, with failing
routers, ISPs that just drop out, modems that eraticaly don't answer,
firewalls, intermittent intermediate hosts, etc.

Mirror, cvsup, sup are all  _comms_protocols_
that require a temporary end to end virtual circuit across the internet
between server & client, without the link they won't work.

CTM is _different_ it uses store & forward email, and survives despite
net congestion or temporary total outage of net critical nodes.

Email forwarding from a local ISP offering popserver will run full blast,
whereas long distance links to servers may only be capable of delivering
100 bytes per second (regardless of streaming or not).

CTM will work even if one can never establish a virtual end to end circuit 

Sure, CVSup's streaming protocol sounds marvellous for _increasing_ existing
performance, _if_ you have the basic end to end functionality in place,
but some people, self included, have had times when the local net works
fine, but EG the transatlantic is _broken_ , not just slow, but broken.

Now if my mail is buffered at a local nearby host, it doesn't matter
if the transatlantic is temporarily dead, I can still update _now_,
not keep dialing the modem & trying to establish a circuit via my ISP to the
remote server, hoping all the intermediate net is intact at the time I want 
to use it.

- CTM can be used via any machine on the planet that happens to
  be convenient & reliable for localy buffering your mail.
- Sup CVSup Mirror et al need across-the-net access to a _server_
- There are many less Sup/CVSup/Mirror servers on the Internet,
  than there are local mail servers.

Whether you know a few people that .... doesnt affect the technology.

One could stay current with just uucp & 2400bd over wet string, fed from
a string of relay hosts that crashed 20 times a day, if one used ctm + cvs.
If you don't realise what `poor net access' can mean ... be grateful !

Julian
--
Julian H. Stacey	jhs@freebsd.org  	http://www.freebsd.org/~jhs/



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199609041952.VAA03250>