If pushed I could try and give detail on this, but most notably this seems to be the most unpredictable build I have tried yet. It sometimes drifts into doing very little, sometimes dies of excessive threads, and sometimes just uses excess CPU time while not actually serving any incoming requests, and serving local requests, and fproxy pages, exceedingly slowly. This could be partly that different snapshots I have tried of build 535 are in fact different. This is in marked contrast to build 534 which settled to accepting 2-300 incoming requests and a reasonable load average. (Slow computer running Sun JVM 1.4.1_01).
If this is general experience, then maybe 534 is better for a stable release, unless it does something bad to the network.
--
Roger Hayter

_______________________________________________
support mailing list
[EMAIL PROTECTED]
http://hawk.freenetproject.org/cgi-bin/mailman/listinfo/support

Reply via email to