[jdev] Re: Reliable presence
dwaite at gmail.com
Thu Aug 12 23:59:29 CDT 2004
Again, writing better software does not mean the failure conditions
'go away', just that they are rarer.
That nobody has yet written 'good enough' software to prevent this
from being a serious issue indicates to me that it is not feasible to
write a reliable-enough implementation to solve this problem. Even
with absolute blessed-by-$DEITY code which contains zero bugs and
never crashes, you still have network outages and equipment failures,
and you still have out of resource conditions under excessive use.
The mantra of reliability and distributed state should not be
"everything works perfectly.. as long as everything works perfectly".
Failure conditions (such as a temporary network outage, or a database
corruption on the user subscription table) cannot be coded around, no
matter how good your code may be.
On Thu, 12 Aug 2004 13:21:38 -0600, Peter Saint-Andre
<stpeter at jabber.org> wrote:
> In article <cfd3c4d104081212145003b550 at mail.gmail.com>,
> Peter Millard <pgmillard at gmail.com> wrote:
> > Perhaps the time we're spending on this discussion could go to
> > improving the jabberd 1.4.3 s2s process and we'd all be much happier
> > :)
> Hear, hear! More code, less yammering. Call that mantra #3. ;-)
> jdev mailing list
> jdev at jabber.org
More information about the JDev