Shawn Willden <[email protected]> writes: > On Sun, Jan 9, 2011 at 7:19 AM, Josh Wilcox <[email protected]> wrote: > >> I was surprised because I assumed each share would be sent to a different >> server, but this error message seems to be saying that tahoe was only >> looking for 7 distinct servers. >> > > The default is to use 10 shares, but "servers of happiness" is set to seven, > so Tahoe only _requires_ seven servers. If it can get 10 servers to take > one share each, it will, but if there aren't that many servers accepting > shares (as on the volunteer grid right now -- there are only six servers > with available space), then the 10 shares will be placed on fewer servers. > If there are less than seven servers, then the upload will fail.
I think people are talking about the pubgrid. The pubgrid seems to be for not huge amounts of data, so I wonder if others would run more serers. I am currently running 2, and 1 is full. I fixed that by removing large shares and allowing a bit more space. It would be nice to have more working servers (with reachable internet addresses from client-only nodes behind NATS!), even if they only offered a few GB each. But we may need a config to disallow big shares, or implement accounting on the pubgrid :-) I've got a tcpdump of one of my servers that I can look at in my Copious Spare Time; that may be interesting.
pgpI3hL4tkxGT.pgp
Description: PGP signature
_______________________________________________ tahoe-dev mailing list [email protected] http://tahoe-lafs.org/cgi-bin/mailman/listinfo/tahoe-dev
