[RC5] Re: Dupes

gindrup at okway.okstate.edu gindrup at okway.okstate.edu
Wed May 13 11:30:40 EDT 1998


     Old clients tend to have the following properties:
     
     They aren't visited much.  They tend to lose network connectivity 
     for a long time and no one cares.  The random block generators are 
     excrable (as I complained around 2.6402) so a given client may check 
     the same random block a couple of times in a row.  Very old clients 
     tend to have very large buffers which means that the blocks they 
     flush, when they do, are already done.  There were significant 
     errors in the older clients, especially in regard to parsing buff-in 
     files.  There were changes in the buff-* files structures around 
     that time and the newer clients would just choke on the old buff-* 
     files, generating random blocks.
     
     In effect, the older clients are much more likely to generate 
     horribly distributed random blocks.
            -- Eric Gindrup ! gindrup at okway.okstate.edu


______________________________ Reply Separator _________________________________
Subject: Re: [RC5] LIFO?  Is this true? 
Author:  <rc5 at llamas.net> at SMTP
Date:    5/12/98 10:31 AM


Jim C. Nasby wrote:
> 
     [snip]
> dupes are a serious issue and will
> effect the timeframe of RC5-64, even if it doesn't affect the eventual
> outcome. The last time Daa posted stats on our dupe rate, I believe that
> something like 40% of the blocks returned were dupes, mostly from old 2.64 
> clients.
     
Yeah, I still haven't figured that one out. Why would old 
clients preferentially be producing dupes?
     
     [snip]
--
Marc Sissom                       http://www.dnaent.com 
Design Engineer                     voice: 972/644-3301 
DNA Enterprises, Inc.                 fax: 972/644-6338 

--
To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest



More information about the rc5 mailing list