Hypnosis
T3s|4: I did not try to interpret two numbers 10 & 4 and a sign '-', I just asked for clarification, ty
dr|z3d
new geoip database in latest i2p+ updates, both release and dev update paths.
T3s|4
Hypnosis: 10-4, and np :D
dr|z3d
in Russian that's 10-4ski, T3s|4 :)
dr|z3d
</stupid mode>
T3s|4
lols @ TIL^ :)
Hypnosis
yes funny codes
dr|z3d
Hypnosis: /msg nickserv register chooseapssword fake@email.ru
Hypnosis
tx, doing
Hypnosis
dr|z3d: ty
dr|z3d
you're welcome
dr|z3d
twitter just went down the toilet.
dr|z3d
no login, no view tweets *laughs*
opal
*** 8,1🐔0,1: i2pd update helps ***
opal
*** 8,1🐔0,1: h ***
dr|z3d
>>> I2P+ 2.3.0+ update now available via torrent -> tracker2.postman.i2p/index.php?action=Download&id=70689 <<<
dr|z3d
>>> I2P+ 2.3.0+ installer now available via torrent -> tracker2.postman.i2p/index.php?action=Download&id=70691 <<<
dr|z3d
Please seed for as long as you can, ideally until 2.4.0+. Thanks!
opal
*** 8,1🐔0,1: sup ***
opal
*** 8,1🐔0,1: didnt libtorrent get proper i2p support now ***
opal
*** 8,1🐔0,1: i should try it ***
uop23ip
If i look at my tunnels participants for irc, snark... and not the particpants tunnels peers. I can see some peers quite often sometimes 3 times in one column (over all app tunnels, sometimes 2 times in one app tunnel). Is this normal? Does every pp tunnel have their own peer pool or do i have the propability/pool of known-active peers for my anonymity set/route?
dr|z3d
uop23ip: i2p or i2p+ ?
dr|z3d
and how's your router functioning? hidden mode? firewalled? low bandwidth share?
dr|z3d
also, how many known peers +-?
uop23ip
canon, participating router, 4k
dr|z3d
for the most part, your tunnel participants are selected from the fast tier, so however many fast peers you have, that's your selection pool. much lower in canon.
dr|z3d
iirc, in canon a single router can feature in max 30% of your local tunnels, in i2p+ it's closer to 10%.
dr|z3d_
> for the most part, your tunnel participants are selected from the fast tier, so however many fast peers you have, that's your selection pool. much lower in canon.
dr|z3d_
> iirc, in canon a single router can feature in max 30% of your local tunnels, in i2p+ it's closer to 10%.
opal
*** 8,1🐔0,1: ye we got those two messages ***
opal
*** 8,1🐔0,1: (also iirc `/ns regain` automatically changes your nick for you, compared to ghost) ***
uop23ip
First part i understand (you explained this to me a while ago, false 50kb/s limit for participate/general tunnels iirc). Second part: Do you mean the balancing of the client/user peer/bandwith usage and the routing for other peers?
dr|z3d
short answer, yes, expected behavior in canon, not so much in i2p+.
dr|z3d
nothing to do with routing tunnels for other peers (participating traffic)
dr|z3d
everything to do with the hard limit of fast peers and the max percentage of local tunnels they can appear in.
dr|z3d
with all that said, the more bandwidth you share, the more fast peers in your pool, normally.
uop23ip
ok, but what do you think about the appearing of same routing path for 2 tunnels? Normal? You know more about what to expect to see. So are there duplicate peers at one(often) time in your router?
uop23ip
i looked again after refresh. http proxy 2 tunnels exact same path
uop23ip
snark 4 tunnels, 1 duplicate peer
dr|z3d
each peer can appear in max 30% of your total tunnel count, so that's fairly normal in canon.
dr|z3d
whether it's optimal is a different question entirely.
uop23ip
And the peer appear more often if it is "nice" (highbw,latency..), right?
dr|z3d
latency on canon isn't much a factor, but bandwidth is.
dr|z3d
look at /profiles and the fast peers. those are the ones that are preferred.
uop23ip
Ok thanks. with max 30% there should be duplicates. But it is quite weird to see the peers that often i have to admit. I thought somebody reduces my anonymity :) Is it even possible in i2p to be soooo nice that you can encircle someone?
dr|z3d
what you're referring to is commonly known as a "sybil attack", and yes, it's possible.
uop23ip
iirc a long tie ago gnunet? just throw them away or banned them for a while and used the not so nice peers. But i think i2p does the same or similar
uop23ip
they had some economic model for peer behaviour
uop23ip
that netdb feat is new, isn't it. wow
dr|z3d
not new, been there a a couple of years or more, but unless your console's in advanced more or you're running I2P+, it's not exposed in the UI.
uop23ip
floodfill distance to look at, nice
uop23ip
dr|z3d, really. it's new for me. Nice to "see" what you only read of in the docs
uop23ip
"Routers in the same family" Really, i have a family? You,too? Everybody? :)
dr|z3d
no.
dr|z3d
you have to explicitly create a router family.
uop23ip
ok i misunderstand. That are known families, not mine
dr|z3d
and if you do, because you want to declare that several of your routers are under your control, avoid giving it a name that ties with your identity here or elsewhere, unless you don't care about deanonymizing your routers.
uop23ip
As a general it is good to have many ids, but i don't understand why having a family of routers make them more easy to deanon. Are they prone for sybilore pr
uop23ip
sybil attacks
dr|z3d
"I'm running the foobarX family of routers. Aren't I clever?" you've just deanonymized yourself.
uop23ip
outside of i2p?
dr|z3d
that depends on the routers in the family. I know your router family, I can id all the routers and their ips.
dr|z3d
try this: 127.0.0.1:7657/netdb?fam=stormycloud
uop23ip
yes i understand. ip is there. even port