On Thu, May 01, 2003 at 04:58:08AM +0100, Dave Hooper wrote: > This on a node that was recently (minutes rather than hours) restarted: > > address | contact probability | failures | attempts | successful | last > attempt > tcp/aaa.bbb.ccc.ddd:ppppp | 0.37062398 | none | 3 | 0 (0%) | 1051761016 > secs. ago > > Why would it say 'none' for 3 attempts and 0 successes? Why is it (and two > other nodes on the same Node Reference Status page) reporting a last contact > time of exactly 1051761016 seconds ago?
Because cpd.lastRetryMs isn't necessarily updated just because trials has been. Fix will be in CVS soon. > What about another node which states 2 connection attempts and 2 successful > connections but still says a last attempt time of 1051761272 seconds? If it > wasn't for the fact that the connection attempts and successful connection > counts get zapped on node restart I'd say the last attempt might be genuine. > [Maybe that's another "windoze" bug - should the connection attempts and > successful connection counts not get zapped on a restart?] > And the last attempt time works out pretty close to 1/1/1970. So something > isn't working properly after a restart, whether it's zapping the data that > should be zapped or keeping the data that should be kept. > > d _______________________________________________ devl mailing list devl at freenetproject.org http://hawk.freenetproject.org:8080/cgi-bin/mailman/listinfo/devl
