mduell52 at cox.net
Sun Dec 15 21:50:37 EST 2002
----- Original Message -----
From: "Michael Shannon" <MichaelS at towersoft.com.au>
To: <rc5 at lists.distributed.net>
Sent: Sunday, December 15, 2002 8:47 PM
Subject: RE: [RC5] Suggestion
> > The only time this happens is when you go from 2.8 to 2.9 version
> > clients. This will not happen again until the buffer format changes.
> > All other upgrades will simply restart the block that is currently
> > being worked on.
> That's exactly what people are complaining about. This is the scenario:
> - a slow machine is cracking a key
> - it spends a long time (many hours or even days) getting to a certain
> point (say 80% done)
> - the machine crashes, the client is restarted, etc.
> - when the client is fired up again, it chooses a different core and
> starts the same key from the start
> - all the work previously done is lost
> Now, if a person regularly shuts down at the end of the day and comes
> back online the next, there's a very real possibility that they'll be
> endlessly cracking *the same key*. It's like trying to ride from one
> town to the next on an exercise cycle - you'll do a lot of work but you
> won't get anywhere.
1) New client versions don't arrive daily.
2) Couldn't you use the checkpoint file feature to solve the the problem you
To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest
More information about the rc5