On Wednesday 23 January 2008 21:57, you wrote:
> On Jan 23, 2008 12:27 PM, Matthew Toseland <toad at amphibian.dyndns.org> 
wrote:
> > An interesting recent conversation. Note the timestamps: it takes 
apple_juice
> > over an hour to get his first freesite to load. I've duplicated his 
problem
> > on a newly installed node and the cause appears to simply be that newbie
> > nodes tend to get connected to overloaded nodes... It's not as severe, but 
it
> > takes a loooooong time for the activelinks to load let alone any sites.
> 
> Why is the announcement resulting in the new node being connected to
> overloaded nodes?  Surely the nodes you connect to after announcement
> should be pretty evenly distributed throughout the network?

Apparently this was caused by bugs introduced and then fixed by robert. After 
releasing 1103 (mandatory tomorrow), this morning a new node managed to fetch 
the indexes within minutes, and had full connectivity (although still 
slowish) shortly after that. I've made the announcement notice warn the user 
that Freenet may start off slow:

"Freenet is currently attempting to connect to seednodes and announce itself 
to the network. This may take a few minutes. Note that Freenet will be quite 
slow to begin with, but should improve significantly over the next few 
minutes and more slowly over the next few hours. Also it's a good idea to 
keep the node running 24x7 if you can."
> 
> Perhaps we need some bias in how announcements are routed such that
> they are routed towards underloaded nodes.

We already do: Like any normal request, announcements are routed first to 
nodes which are not backed off. You might argue that we need more bias than 
that, but on the basis of current performance that may not be necessary.
> 
> Ian.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: 
<https://emu.freenetproject.org/pipermail/devl/attachments/20080124/b8865b6f/attachment.pgp>

Reply via email to