My feeling (based on a sketchy memory) is that neither 6209 nor 6207 made much difference to the upstream bw problem. Your thought about the slow memory leak makes sense to me, since I eventually get "out of memory" errors as well.

I'll now try removing the upstream bw limiting as well as changing "maxRequestsPerInterval" as previously mentioned. We'll see if I still get out-of-memory errors after 6-18 hours. If it works, I'll have to compare notes with Salah so we can see which change fixed it.

-Martin

Brandon Low wrote:

Has the decreasing upstream bw usage over time improved with more recent
builds?  Has anyone _not_ bwlimiting in freenet had problems with OOMs?
I am of the belief that there is a slow memory leak in bwlimiting on
heavily loaded nodes.

--Brandon

On Sat, 09/27/03 at 23:28:53 -0700, Martin Stone Davis wrote:

As you probably read in the "Decreasing upstream bandwidth over time in 6205" thread, I've been having the same kind of problem. (note: I am now using 6209)

It turns out that my QueriesPerHour is very high, like Salah's, although mine averages around 80,000. I've altered my freenet.ini as follows:

# The maximum number of outgoing connections established per maxRequestsInterval.
maxRequestsPerInterval=3000


I'll let you know what happens.

-Martin

Salah Coronya wrote:


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Brandon Low wrote:
| Threadlimits probably aren't the issue any more, can you check how many
| queries per minute your node is accepting?  There is a default in the
| config file to only accept a max of 300 queries per minute, is it
| possible that that is the limit you are hitting?

* entries: 100
* Global mean traffic (queries per hour):13030.233333333334
* Local mean traffic (queries per hour): 93798.85356956748
* Current advertise probability: 0.02
* Current proportion of requests being accepted: 0.0

Sure looks like it. Local mean traffic (the number of incoming requests
to me?) seem to vary from about 30,000 an hour to 90,000 an hour. Seems
to be holding at about 60,000.

I'll reverting back to 120 threads and increase maxRequestsPerInterval
and see what happens.


<SNIP>



_______________________________________________ Devl mailing list [EMAIL PROTECTED] http://dodo.freenetproject.org/cgi-bin/mailman/listinfo/devl


_______________________________________________
Devl mailing list
[EMAIL PROTECTED]
http://dodo.freenetproject.org/cgi-bin/mailman/listinfo/devl

Reply via email to