Those who are Supernode seems to not refresh their cache very often. I have alot of requests that go like this

/uri-res/NR2R?urn:sha1:....35DLDXL

This must be an tremedous waste of bandwidth, due to handle of this phenomenon. I often change files quite frequent. I have one file that are updated every day.

Those are regular download requests from LimeWire 2.8.6 - unfortunately created due to a bug. It will be fixed by LW 2.9

__________________
Morgens ess ich Cornflakes und abends ess ich Brot
Und wenn ich lang genug gelebt hab, dann sterb ich und bin tot

--Fischmob

Ok i See

Have also noticed, very high dropped I/O in host section from those who have 2.8.6, typical 80-90 percent in fact

Are you running LimeWire as a leaf or as an ultrapeer?

__________________
Morgens ess ich Cornflakes und abends ess ich Brot
Und wenn ich lang genug gelebt hab, dann sterb ich und bin tot

--Fischmob

I think im Leaf that what it stands at bottom of host section, all others are Ultra peers.

That's fairly normal because you are not supposed to forward any traffic (except for queryhits) to the ultrapeers, you have to drop a lot of messages.

__________________
Morgens ess ich Cornflakes und abends ess ich Brot
Und wenn ich lang genug gelebt hab, dann sterb ich und bin tot

--Fischmob

Ya i though so too after i while here , its just that the newer lime shields off much better then older versions. But the bug that the new version has is seriously bad! i suspect it has to do with SHA1 handling that doesnt work at all, and also query matching. let me give some examples here. Let say i have a file with the name

BUNCH1.rar

But the requests coming in are like

good_BUNCH.rar or BUNCH LX 8.rar these are the clear context requests! that fails here.

Then i lookup those URI-RES SHAS:s, requests they dont match any of my files at all. very strange indeed

So out of 10 requests ive got 2-5 that are completly wrong.

Yes, indeed. But there are even worse bugs that will be fixed with the next version. One was causing lock-ups. Another caused LimeWire to send lot's of download locations to other clients that may have stopped working weeks ago. - I expect great improvements with the next version.

Not really. If you are sharing other files containing "good" and "bunch" or "LX" then it is not a bug.
The decision not to send the complete file index to the ultrapeer was made long ago (it was also done to protect your privacy) and it is the cause for a little unnecessary traffic but I wouldn't bother too much about it.

They only have to match files you shared at one time during the past weeks.

__________________
Morgens ess ich Cornflakes und abends ess ich Brot
Und wenn ich lang genug gelebt hab, dann sterb ich und bin tot

--Fischmob

lockups!? are you meaning that i experinces sometimes there i have myself in host section as ultrapeer both incoming and outgoing. Ya i guess Supernode´s has to refresh their cache´s more frequent then now.

No i dont have any other files that starts with good_ or have the wrong version number at end. thats confuses me how the clear context requests are so wrong. and again the URI-RES are for files i never had for share! at any time.

I hope the 2.9 is coming out soon, with the corrections