* No socket timeout on connection when first designator byte is read.

It took fixing this to make it go away for me. I don't understand why so 
many (well one in a thousand or so) nodes connect and then send nothing 
though - can you check that OpenConnectionManager is merged from main 
into the release Matthew?

On Tue, Nov 05, 2002 at 03:56:35PM +0000, Matthew Toseland wrote:
> Causes of the 100% load:
> * inadequate load balancing. Resolved for a while now.
> * bug that caused reseeding of the routing table on every startup. Fixed
>   in 530.
> * insufficiently early rejection of excess connections. Fixed in 611,
>   soon to be ported to 531 (this is implemented as a limit on the number
>   of simultaneous negotiations).
> * being in the hawk-generated seednodes.ref. Just once, at the wrong
>   time, is sufficient, because of the above bug - but this may improve
>   now. However, it is still a problem. Given that 99%+ of all users will
>   always use the provided seednodes... what can we do about this?
> -- 
> Matthew Toseland
> toad at amphibian.dyndns.org
> amphibian at users.sourceforge.net
> Freenet/Coldstore open source hacker.
> Employed full time by Freenet Project Inc. from 11/9/02 to 11/11/02.
> http://freenetproject.org/



-- 

Oskar Sandberg
oskar at freenetproject.org

_______________________________________________
devl mailing list
devl at freenetproject.org
http://hawk.freenetproject.org/cgi-bin/mailman/listinfo/devl

Reply via email to