IRCaBot 2.1.0
GPLv3 © acetone, 2021-2022
#ls2
/2022/04/25
zzz ok I have peer test basically working on testnet
zzz orignal, last week you suggested possibly using a standard noise handshake for messages 5-7
zzz that won't work, because message 5 may never get through
zzz so I think we should keep it simple for messages 5-7, out-of-session, like in SSU 1.
zzz now to my notes on relay from the last week
zzz 1) you said we need a chash in Relay Intro?
zzz charlie hash?
zzz Relay Intro is from Bob to Charlie, so I don't understand
orignal I ddn't suggest just asked
orignal I said chash must be in relay intro because it's signed by Alice not Bon
orignal Alice signs bhash and chash and Bob just forward this signature
orignal so Charlie must include chash
zzz so you're not saying the message must include chash. You're saying the signature must include chash?
orignal I said it's mistake in specs
orignal about Relay Intro
orignal no chash in signature section that's worng
zzz So the signature must include chash. got it.
orignal it must be the same as in Relay Request
zzz 2) why is relay tag included in Relay Response? shouldn't be necessary, nonce is sufficient
zzz I think you're right. It's not in SSU 1 Relay Response
orignal yes, that's my question
orignal we need only nonce
zzz should it still be covered by the signature?
orignal no. why?
orignal nonce defines relay tag
zzz ok, will remove it
orignal e.g. Bob and Alice withh find it by nonce anyway
zzz agreed
zzz 3) add 'relay tag not found' bob rejection code
zzz agreed
orignal shold be the first and main reason
orignal now, I'm not clear how we fill Charlie's section in thios case
orignal if Relay Response with error is initiated by Bob
zzz I'll make it #5, I'm not going to renumber everything
zzz how about csz = 0 for that case
orignal fine for me
zzz 4) "endpoint" and flipping ip/port, changing asz/csz from 4 or 16 to 6 or 18 in both peer test and relay
zzz agreed
zzz I believe that's everything
zzz all is changed in the spec and checked in
zzz I'll make the peer test code changes for 4) probably tomorrow
orignal 6 and 18 should be better
zzz 0) Hi
zzz *hi
zzz happy monday, what's on the list for today?
orignal I'm ready to talk about SSU2 relays
zzz ok that's 1)
zzz 2) ssu2 general status
zzz anything else for the list?
orignal notbob keeps showing ukraining eepsites as Russian ))
orignal question to dr|z3d
orignal *ukrainian
zzz lol save that for after the meeting
zzz 1) SSU2 relays
orignal but not to the list, ofc
orignal I just see him in the users list
orignal so to this question
orignal does Bob verify signatures or not?
zzz I don't think I did it in peer test, no
zzz let me look
orignal so only Alice cnd Charlie
orignal not sure if he should
zzz one sec
zzz only alice and charlie in what I did for peer test
zzz do you think that's right?
orignal not sure
orignal if Bob verifys signature in relay request he can just end error back to Alice
orignal and not buzz Charlie
orignal while for relay reponse it doesn't make sense at all
orignal because Alice is waiting for reponse to a nonce
zzz if it's a bug, we'll find it out anyway. If it's some sort of strange attack, I'm not sure what the point is
orignal to not magnify it
orignal and stop it on Bob' level rather then flood Charlie
zzz we can always add it if we think of a good reason. Doesn't seem that important to me
orignal I write the code that's why I'm asking
orignal if we need to verify it
zzz I say no, but if you say yes we can think and talk about it again next week
orignal I will not verify
orignal for now
zzz ok for now. we can always change our mind later
zzz anything else on 1) relays ?
orignal so Bob find Alice by nonce
orignal and copy relay response from Charlie
zzz yes
zzz that's more-or-less the same as in SSU 1
orignal another question
orignal about relay
orignal do we support fragmneted RI blocks?
orignal because we must include it into relay intro
zzz no, I don't think we want to use RI blocks at all really (fragmented or not), just send i2np database store message
orignal so if it's too big we use I2NP
zzz it's a lot easier to do it that way
orignal should we remove frag field from RI block?
zzz I'm just using i2np all the time
orignal RI is simplier and faster for me
zzz I thought about removing the flag field but I don't think it's worth the trouble to break everything
orignal since compression is not required
zzz for only one byte
zzz sorry *frag field
orignal but confusion
orignal use two bytes flags then
zzz anything else on 1) relays ?
orignal but still don't understand if we remove it
orignal not too much mess in my opinion
zzz let's leave it in there
orignal up to you
zzz anything else on 1) relays ?
zzz 2) ssu2 general status
zzz I guess my turn to go first
zzz mostly finished in peer test with the testnet
zzz tomorrow I'm going to change everything to flip ip/port and go to 6/18 sz fields
zzz and test that
orignal yes please
zzz after that, I'll basically be done, but I'm sure there's still plenty of bugs to be found
orignal I assume it's there already
orignal for relays
zzz but it at least works most of the time
orignal btw, how do you publish iKey?
orignal sorry iTag
zzz I have zero code for relay right now. Haven't started
orignal if you SSU is common for SSU1 and SSU2
orignal then it's time to talk
orignal about it
orignal so, my status
orignal everything basically works fixed few bugs in ack ranges
orignal started relays
orignal ready to start testing but introducers publishing is not done yet
orignal but my question is different
zzz you have a guess on when you would be ready to test relays? one week? two?
orignal if I see your SSU address with introoducers
orignal and it has "s" and "i" how should I handle it
orignal this week hopefully
zzz probably two weeks for me
orignal let's agree with introducers format
zzz so I think we decided we were going to publish ihash/itag/iexp only for SSU 2. no ihost/iport/ikey
orignal as I remeber we are going to use adress instead iHost/iPort
zzz but I don't think the spec is updated yet
zzz looking
orignal but will you SSU adddress contain both?
zzz nothing about ihash in there now
orignal probably you also need iVersion or somthing
zzz maybe ih, not ihash...
zzz lol major
orignal if I read your introducer I need to understand what SSU it is for
orignal major is acentone's bot ))
zzz yeah I know. just wondering why he woke up ))
zzz I understand the question about knowing what it is for
orignal major means "an agent of some service"
orignal like police, FBI, FSB, etc.
zzz ok, so maybe the police told him to join the meeting!
zzz anyway, let's call the hash ih0, ih1, ih2, ...
orignal I think bugs in his code ))
zzz if you see ih0 you can use it for SSU2
orignal ih or iH ?
zzz how does that sound?
orignal fine for me
zzz I think they're all lower case
orignal no problem
zzz You're still "ahead" of me on relay, I haven't thought much about it in weeks. I'll catch up this week
orignal will change my code to recognize SSU2 introdcuers
zzz now that I can push some of the peer test stuff out of my brain and make some room
orignal will it's not about "ahead" or not
orignal it's more about data structures
zzz yeah, I don't mean like a race
orignal for peer test still not clear how we recognize our role
zzz just that you'll be finding problems before I do
orignal need to look at SSU1 code first
orignal ih is 32 bytes in base64, right?
zzz 32 in binary, 44 in base64
orignal ih but iHost
orignal different second character
zzz no SSU1 is ihost, not iHost
zzz there's no capitals
zzz public static final String PROP_INTRO_HOST_PREFIX = "ihost";
zzz public static final String PROP_INTRO_PORT_PREFIX = "iport";
zzz public static final String PROP_INTRO_KEY_PREFIX = "ikey";
zzz public static final String PROP_INTRO_TAG_PREFIX = "itag";
zzz /** @since 0.9.30 */
zzz public static final String PROP_INTRO_EXP_PREFIX = "iexp";
orignal it's fine
orignal I parse whole string (btw why?)
zzz dont ask me ((
orignal maybe it's worse to change to iH?
orignal in this case second character becomes unique
zzz or we could do h0, h1, h2
zzz I don't really care
orignal must start with i
orignal i means "introducer"
orignal well "ih" is fine
orignal because SSU1 will be depricated
orignal sooner or later
zzz I'll update the spec
zzz anything else on 2) ?
zzz anything else for the meeting?
orignal dr|z3d ukrainians are angry ))
orignal nothing else
zzz very interesting and long article about Russian and Ukrainian language I read today: nytimes.com/2022/04/22/magazine/ruscism-ukraine-russia-war.html
orignal how to pronounce "паляница"? ))
orignal but seriously it's the same language
dr|z3d orignal: not my fault. not_bob's responsibility. and he's gone AWOL.
orignal dr|z3d it's bad
orignal and you can't use red flag instead Russian
orignal because it's use for Chinesse already ))
dr|z3d we could just use a picture of lenin. any better? :)
orignal even worse
orignal zzz, it's easy recognize Uknrain text. They don't have "ы" but have "i"
dr|z3d well, poke not_bob when he's around, tell him that cyrillic characters doesn't always equal Russian .
orignal i and ï
orignal if you see these characters amopung cyrrilic it's ukrainian
dr|z3d top tip
orignal and never "ы" and "э"
orignal also there is another letter but I don't know how to type it
orignal looks like "г" but upper part is longer
zzz The letter that looks like III is in Ukrainian but not Russian, is that right?
zzz oh, it's like a W, not III
orignal let me check
zzz need new glasses :)
orignal one sec
zzz In passing, the council members referred to the Russian perpetrators by a term of condemnation that every Ukrainian knows, though it is not yet in the dictionaries and cannot (yet) be said in English: “рашизм."
orignal no, the difference is in pronouciation only
orignal Russism ?
zzz ok so ш is in both but i is only in Ukrainian
orignal in both
zzz you have to read the whole long, long, long article to learn about “рашизм."
orignal I know what is it
dr|z3d no flies on orignal :)
zzz so ï is only Ukrainian?
orignal in russian we normally propnouce "u" in Ru as "U"
orignal zzz, and French ofc
orignal but not in Russian
orignal also no "i"
zzz blame the french, as usual
dr|z3d there is no ï in french, orignal
dr|z3d not last time I checked, anyways.
orignal so "рашизм" is prononced as "rAshism"
orignal simply speaking it's an americnism
orignal dr|z3d really?
orignal how to say "i" as "i" in "oi" then?
zzz what I got from the story is if you know two languages your head must hurt all the time
orignal you must write as "oï"
orignal luckily I know only one )) but can understand the second ))
zzz you do just fine
dr|z3d my bad, orignal, you're right. naïf is a good example. color me naïf.
orignal no, I means Russian and Ukrainian
dr|z3d dieresis, technical term.
orignal I understand Ukrainian but can't speak
R4SAS to such phrase, I usually add "like a dog"