[rc5] Blocks Outstanding
Guan Sin Ong
cceonggs at segamat.nus.edu.sg
Fri Jun 27 10:57:50 EDT 1997
Skip Huffman wrote:
> I have another reason not to reassign blocks that have been assigned
> and never returned.
> I think that we can safely assume that for at least some of those
> blocks, some of the keys were checked. If the correct key was found
> in those keys that were checked, than we would have found out.
> Therefore some of the keys in those blocks have been eliminated. If
> we reassign those blocks, we will be redoing work when there is plenty
> that we know has not been tried.
Yes, this is the greatest benefit of *NOT* diverting resources and
effort to reassign unreturned blocks. I suspect many blocks had been
thoroughly checked but not notified to the server due to personal proxy
server problem or local network outage etc.
Can someone tell me what would the client do when the key is found? Will
it still report through the proxy or will it notify some other parties
directly? Any output to log file or stdout? All these affect the chances
of the key being found by a client but unreported due to some problem.
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 4311 bytes
Desc: S/MIME Cryptographic Signature
Url : http://lists.distributed.net/pipermail/rc5/attachments/19970627/c6153adc/smime-0001.bin
More information about the rc5