#ls2 (irc2p) | #ls2 I2P Protocol Development Discussion - Next Meeting Monday, October 9th 6:30 UTC

eyedeekay The most significant of which is that while the total number of banned routers in my netDB rises and falls, the number of them that are blocked for the "P2P VPN" thing remains pretty constant

eyedeekay Contrary to my early testing, I'm observing shared hosting bans overtaking P2P VPN bans all the time

zzz what do you conclude from that?

eyedeekay So my hypothesis that it's enumeration related seems shakier now

eyedeekay Because shared hosting bans are legitimate

eyedeekay So I can only think of a couple things

orignal back

eyedeekay One is that there's an overlap of I2P users and "P2P VPN" users which according to their somewhat superficial definition is a VPN that enlists users to be exits

eyedeekay Or there's a "P2P VPN" of this nature that is using I2P that we don't know about it

eyedeekay I think #2 is unlikely

eyedeekay But I'm struggling for alternatives

eyedeekay If it's #1 then it's a non-issue and I just learned a little

zzz ok, so still haven't seen any reason to panic then

eyedeekay Yeah I'm less worried than I was two weeks ago by alot

zzz great, thanks for the update

zzz back to 1) SSU2

zzz I'll give my status

zzz pretty quiet. still fixing minor bugs

zzz I evaluated all the session confirmed fragmentation options from last week and recommended c)

zzz orignal gave his preliminary agreement

zzz I wrote it up in the proposal under a new section "session confirmed fragmentation"

zzz and waiting for review and comments before I push code to implement it

zzz EOT

orignal yes it works

zzz sorry, what works?

orignal so I have finished Bob's side of handashke

orignal connection

orignal I though you meant session agreement ))

orignal nevermind

zzz yes, great, we now have java/i2pd connections working both directions

zzz steady progress

zzz any other status from you orignal ?

orignal so I have implemted acks

orignal for receiveing side

orignal need to do it for sending

orignal fragmentation for receiveing side

orignal and working and sending side

orignal and going to start peertest

orignal I think we can be ready for SSU2 for the next release

zzz before peer test, higher priority is the session confirmed fragmentation, because that's a problem now

orignal btw, that Kirov's guy has conntacted with me

zzz so if you would please review the new section of the spec, then we can get to implementation

zzz yeah I saw the AAAAA is gone

orignal no problem witll start it

orignal yes, I instructed him

zzz first, just looking for an ACK from you that the spec looks good

orignal not really

orignal my acks are dumb

orignal need to implemented it better

zzz sorry, I didn't mean ACK, I mean please review the spec for session confirmed fragmentation and tell me you like it )))))

orignal oh, you mean session confirmed frgmentation

orignal yes I see

zzz lol

orignal I will take a look today

zzz my fault, shouldn't be using "ACK" in a sentence

zzz super

orignal no problem with it

orignal I would say "approval" or "confirmation"

zzz yes )))

orignal we don't have word "ack" in Russian

zzz not really a word in English either

zzz anything else on 1) SSU2 ?

orignal do you think we will be ready for the next release?

orignal with SSU2?

zzz probably not? the schedule shows 3 releases from now, in November. I don't think I can enable it by default next release

zzz maybe the one after

orignal not by deafault ofc

orignal just optional support

orignal to make thak Kirov guy happy ))

zzz well, as not-default, it's in the code now. The only question is how much we advertise it

orignal for me side it's not done yet

zzz I'm not taking it out of the code, so people can do what they want

zzz I think the big milestone for the next release is we freeze the basic protocol, can't make any breaking changes

orignal I mean to decalre if it's supported in the release notes

orignal then not do it for now

zzz "alpha", "beta", "supported", .... just a word, decide later

orignal fine

orignal also R4SAS would like to run a stress test over SSU2

orignal vs. NTCP2

zzz yeah, I've run on testnet with 5% drop probability and still gotten 1 MBps I think, can't remember for sure

zzz will have to go back and retest at some point

orignal yes

orignal but 1Mbps is slow

orignal we need more

zzz not with 5% drop it's not. 5% drop is insanely bad

orignal got it

zzz can't find my notes right now

orignal also I need to implement retransmission windows and RTT

orignal but that's another story

zzz but yeah, testnet or live net, easy to do a test. Use zero-hop client and server side, disable NTCP2

orignal that's what we usually do

zzz anything else on 1) ?

orignal no

zzz anything else for the meeting?

orignal not from my side

zzz thanks everybody, back to work!