On Mon, Sep 15, 2003 at 09:56:28PM -0400, Greg Wooledge wrote: > | Time | Tries | Successes | Ratio | > |-------------------------+-------+-----------+----------------------| > | 9/15/03 12:00:00 AM EDT | 1513 | 1282 | 0.8473231989424983 | > | 9/15/03 1:00:00 AM EDT | 4702 | 3483 | 0.7407486176095278 | > | 9/15/03 2:00:00 AM EDT | 6646 | 2748 | 0.4134817935600361 | > | 9/15/03 3:00:00 AM EDT | 4100 | 396 | 0.09658536585365854 | > | 9/15/03 4:00:00 AM EDT | 3756 | 461 | 0.12273695420660277 | > | 9/15/03 5:00:00 AM EDT | 4628 | 445 | 0.09615384615384616 | > | 9/15/03 6:00:00 AM EDT | 6093 | 382 | 0.06269489578204497 | > | 9/15/03 7:00:00 AM EDT | 6295 | 465 | 0.07386814932486101 | > | 9/15/03 8:00:00 AM EDT | 6774 | 731 | 0.10791260702686743 | > | 9/15/03 9:00:00 AM EDT | 8327 | 552 | 0.06629038068932389 | > | 9/15/03 10:00:00 AM EDT | 7257 | 422 | 0.05815075099903541 | > | 9/15/03 11:00:00 AM EDT | 7176 | 822 | 0.11454849498327759 | > | 9/15/03 12:00:00 PM EDT | 6566 | 401 | 0.061072190070057875 | > | 9/15/03 1:00:00 PM EDT | 7832 | 515 | 0.065755873340143 | > | 9/15/03 2:00:00 PM EDT | 8227 | 207 | 0.02516105506259876 | > | 9/15/03 3:00:00 PM EDT | 7303 | 267 | 0.03656031767766671 | > | 9/15/03 4:00:00 PM EDT | 12221 | 589 | 0.048195728663775465 | > | 9/15/03 5:00:00 PM EDT | 11758 | 436 | 0.03708113624766117 | > | 9/15/03 6:00:00 PM EDT | 11992 | 322 | 0.02685123415610407 | > | 9/15/03 7:00:00 PM EDT | 10908 | 265 | 0.024294096076274294 | > | 9/15/03 8:00:00 PM EDT | 13057 | 162 | 0.012407137933675424 | > | 9/15/03 9:00:00 PM EDT | 13112 | 151 | 0.011516168395363026 | > > Mon Sep 15 21:45:27 EDT 2003 > Connections open (Inbound/Outbound/Limit) 511 (312/199/512) > Connections transferring (Receiving/Transmitting) 161 (16/145) > Data waiting to be transmitted/received 40 MiB/11 MiB > > Class Threads used > Checkpoint: Connection opener 2 > freenet.interfaces.LocalNIOInterface$ConnectionShell 10 > freenet.node.states.data.DataStateInitiator 167 > > Current routingTime 3ms > Active pooled jobs 179 (89.5%) > Available threads 105 > Current upstream bandwidth usage 12293 bytes/second (81%) > > > Analysis: From the node's startup to a time +21 hours later, there's > a continual increase in queries received, and a simultaneous decrease > in queries *serviced*. Data transfers are initiated without any > regard for outgoing bandwidth capacity. An ever-increasing amount > of data is queued up for transmission.
What would you prefer we do? The only way we can stop more data transfers being started is by rejecting queries, which is exactly what we are doing above! However, your bandwidth is not maxed out - so it would be better if the data transfers did not occupy threads. Which is being worked on. > > Concurrent analysis of incoming and outgoing traffic levels shows that > incoming traffic also falls off over time. (This is harder to isolate > because the node's traffic is not logged separately from other site > traffic, but I have a decent feel for it on my site.) Right, because we are answering less queries, because we are rejecting 99%, because we have lots of really slow transfers and yet haven't gone over our bandwidth limits. > > -- > Greg Wooledge | "Truth belongs to everybody." > [EMAIL PROTECTED] | - The Red Hot Chili Peppers > http://wooledge.org/~greg/ | > _______________________________________________ > Devl mailing list > [EMAIL PROTECTED] > http://dodo.freenetproject.org/cgi-bin/mailman/listinfo/devl -- Matthew J Toseland - [EMAIL PROTECTED] Freenet Project Official Codemonkey - http://freenetproject.org/ ICTHUS - Nothing is impossible. Our Boss says so.
signature.asc
Description: Digital signature
_______________________________________________ Devl mailing list [EMAIL PROTECTED] http://dodo.freenetproject.org/cgi-bin/mailman/listinfo/devl
