p2p monticello - any takers?

Tony Garnock-Jones tonyg at lshift.net
Mon Feb 28 12:03:07 UTC 2005


Cees de Groot wrote:
> Don't past walkbacks, fix them ;).

Mea culpa - you did say "But send me fixes, not whines that it  doesn't 
work", after all :) (My excuse is that doesn't seem to be a codefix kind 
of thing)

> Actually, I'm just mimicking MC's  
> regular behavior here by throwing walkbacks instead of nice error 
> dialogs.  Bugger Avi about this, please ;) (or not, it's a developer 
> tool, after all)

I don't mind that too much.

> A problem right now is that the network is too unreliable because of 
> its  size. That makes me a bit wary of suggesting alternative ways to 
> tackle  this at this point in time.

Okay. I'll have to do a bit of digging to find out how messages are 
routed to the two hits - 30 seconds seems like a loooooong time, so I 
guess I need to learn more about how the p2p network works to understand 
why 30 seconds isn't practically infinity.

In fact I'll, just as an experiment, try setting it to 300 seconds and 
see what happens.

> - So maybe a two-stage protocol? Ask for a file handle, retrieve from 
> the  first host that responds?

I think that sounds like a good idea. Perhaps have a small number, say 
5, of first-stage-handle-requests going simultaneously; as they time 
out, replace them with more from the pool of untried hits; and if one 
responds positively, cancel the other 4 and take the file from the offer.

> Of course, the reason that I'm putting out this stuff is that I'd like 
> to  have some input on options like these - there are a lot of  
> time/space/bandwidth tradeoffs to be made in p2p land...

Agreed :) How does bittorrent do its thing, I wonder? I did read the 
protocol description once many moons ago but it didn't stick, clearly...

Tony



More information about the Squeak-dev mailing list