orignal
zzz, you asked me about example of unreachable NTCP2 with i, IP and port
orignal
for example ZzrG
orignal
caps NU
orignal
NTCP2 has IP, port and i
orignal
non reachable
orignal
I can give you more examples
orignal
Vort has prepared the fresh list for me
zzz
orignal, I need the full hash for one of them so I can look it up, thanks
orignal
ZzrG9i1oCy8Xcuy7vYLoWNiGOLsa6pOIAm6AaWlZViE=
orignal
or full collection privatebin.i2p/?678709af651f0a30#65wrpK7g8p9KQ9o3o617XNY5g63bHLBCmz74sWByKhCt
zzz
got it, thanks
orignal
all of them are like this
zzz
I'll research both how it happens, and whether this is legal or not
orignal
think about if an address has both ipv4 and ipv6
orignal
how we know which address is good
zzz
ok, for another day. Working on the MR to fix the handling of i2pds with zero itag values
orignal
just don't forget
orignal
we rely on addreses rather than caps
orignal
now we have to check foor U
zzz
have you previously discussed this with eyedeekay or is this something new?
orignal
yes I did
orignal
he was going to "take a look"
zzz
ok, so I'll need to coordinate with him
zzz
did you enter a ticket on our gitlab, or do you know if he did?
orignal
no I don't know
orignal
he only agreed it was not right
zzz
pretty sure if we're directly connected to a public IP we will publish it even if U, but will take some research
zzz
may also be state machine problems
zzz
please pass along my thanks to Vort
zzz
eyedeekay, please let me know if you did take a look and if there's a ticket
eyedeekay
OK got my notes
eyedeekay
1. Can't get rid of U cap atm, we use it
eyedeekay
2. U cap is one we sort of trust routers to not lie about
eyedeekay
3. Presence of i could maybe be treated as indicating unreachability
eyedeekay
4. Does presence of both addresses and introducers indicate circumstances that require additional handling? What causes it to happen and what routers do it?
zzz
is there a ticket?
zzz
agreed with 1 and 2, we do use 3 in some places, the 2nd q is 4 is the key, did you make any progress there?
eyedeekay
Yeah just a sec trying to make sense of this little chart I drew but didn't label adequately
orignal
same question as before
orignal
you have ipv4 unreachable and ipv6 is reachable
orignal
you publish R and i for both. or what?
orignal
I don't understand the logic with two addresses
orignal
if 3. is correct statemment than it's a bug
orignal
with see i in NTCP2 adress but it's not reachable
zzz
look, I don't think it's that complicated
zzz
I think we transitioned from R to U but didn't remove host/port/i from the IPv4 NTCP2 address
zzz
and it's probably my bug
zzz
but awaiting answers from eyedeekay
eyedeekay
I think I was off on the wrong track, I seem to have been trying to figure out what was right to publish if one address was reachable and one was not
zzz
eyedeekay, please, is there a ticket already or not?
eyedeekay
Not
zzz
ok, then I'll enter one
zzz
the logic is, U means no reachable addresses; R means at least one reachable address, that's the way it is and the way it's always been
zzz
#435
orignal
R means reachable
orignal
but you don't know which one if both publish i