<pinbot>
now pinning /ipfs/QmQbTeedC3r34S1Ec3trFJ2QtJB9eadjeuwF55PNdLzzdq
<pinbot>
[host 4] failed to grab refs for /ipfs/QmQbTeedC3r34S1Ec3trFJ2QtJB9eadjeuwF55PNdLzzdq: Post http://[fc3d:9a4e:3c96:2fd2:1afa:18fe:8dd2:b602]:5001/api/v0/refs?arg=/ipfs/QmQbTeedC3r34S1Ec3trFJ2QtJB9eadjeuwF55PNdLzzdq&encoding=json&stream-channels=true&r=true&: dial tcp [fc3d:9a4e:3c96:2fd2:1afa:18fe:8dd2:b602]:5001: getsockopt: connection timed out
anonymuse has quit [Remote host closed the connection]
anonymuse has joined #ipfs
anonymuse has quit [Remote host closed the connection]
anonymuse has joined #ipfs
Soft has quit [Ping timeout: 260 seconds]
zorglub27 has quit [Quit: zorglub27]
ckwaldon has quit [Ping timeout: 240 seconds]
Soft has joined #ipfs
corvinux has quit [Ping timeout: 252 seconds]
anonymuse has quit [Remote host closed the connection]
ckwaldon has joined #ipfs
anonymuse has joined #ipfs
anonymuse has quit [Remote host closed the connection]
anonymuse has joined #ipfs
gant has quit [Quit: Bye.]
damongant has joined #ipfs
corvinux has joined #ipfs
damongant has quit [Client Quit]
damongant has joined #ipfs
jemc has quit [Ping timeout: 258 seconds]
anonymuse has quit [Remote host closed the connection]
ckwaldon has quit [Ping timeout: 250 seconds]
anonymuse has joined #ipfs
ebel has quit [Ping timeout: 250 seconds]
ckwaldon has joined #ipfs
anonymuse has quit [Remote host closed the connection]
ebel has joined #ipfs
anonymuse has joined #ipfs
anonymuse has quit [Ping timeout: 276 seconds]
zorglub27 has joined #ipfs
<geoah>
I assume all live IPFS nodes only accept secure connections?
<geoah>
I'm playing around with the dht and I can't seem to be able to connect to the boostrap nodes
<geoah>
they just error out with a "default failure" :P
apiarian has quit [Ping timeout: 265 seconds]
<dignifiedquire>
geoah: yes the default conf only allows for secure connections, but you can start your own node, with disabled encryption
apiarian has joined #ipfs
<dignifiedquire>
the flag is "--disable-transport-encryption", so you would run "ipfs daemon --disable-transport-encryption"
<cjd>
I like it
<cjd>
you gotta type lots of keystrokes to turn off crypto
Encrypt has joined #ipfs
mingan has joined #ipfs
<zignig>
dignifiedquire: and a big warning that you are doing it... for testing only.
<dignifiedquire>
ideally we would have a prompt, ARE YOU SURE YOU WANT DISABLE ENCRYPTION? IF SO PLEASE TYPE THIS CHARACTER SEQUENCE IN LESS THAN 10 SECONDS: "brA83CP824z-8p8Qanqyc#9oIqv"PG-KJmMHqg8a1NcZhcum"
<zignig>
PLESE TYPE ELEVEN DIGIT PRIME NUMBER
<zignig>
;)
<zignig>
it would be useful for debugging protocol.
<geoah>
Would it also detect pastes and exit with a "no cheating, security is important"
<geoah>
? :p
<dignifiedquire>
of course
<cjd>
require sudoer access to disable encryption, seems fair
<cjd>
and if they get the eleven digit prime wrong, you can easily halt the machine
<cjd>
protect their safety, they'll thank you later
<zignig>
indeed , other node should say _NO_ as well so you can only form a swarm if the are all disabled.
<victorbjelkholm>
Kubuxu, \o/
<geoah>
to be fair disabling encryption is easy if you are building your own stuff... `conn.EncryptConnections = false` :P
smacbrillacciao[ has joined #ipfs
smacbrillacciao[ has left #ipfs [#ipfs]
Encrypt has quit [Ping timeout: 250 seconds]
Encrypt has joined #ipfs
marksei has joined #ipfs
mingan has quit [Quit: Konversation terminated!]
PseudoNoob has quit [Remote host closed the connection]
spilotro has quit [Ping timeout: 276 seconds]
zorglub27 has quit [Remote host closed the connection]
anonymuse has joined #ipfs
zorglub27 has joined #ipfs
<jbenet>
hey does anybody here have a usb netcat tool? or module in js? i want to turn a usb port into an ip or libp2p interface
ckwaldon has quit [Ping timeout: 276 seconds]
<jbenet>
mafintosh substack ogd: do you have something like a "usb-stream" or "usb-cat" yet?
<mafintosh>
jbenet: what would that do?
ckwaldon has joined #ipfs
<jbenet>
mafintosh: something to pipe data over usb from one device to another, ideally duplex.
<jbenet>
like, take two macbooks connect them with a (data carrying) usb-c cable. then have them talk. (ideally the fucking macbook interfaces would do ethernet over it, to enable ip over it, but no apparently they didn't do that for usb-c)
<mafintosh>
ah i see
<mafintosh>
jbenet: no we dont have that but sounds great
<mafintosh>
i'd be interested in that
rendar has quit [Quit: std::lower_bound + std::less_equal *works* with a vector without duplicates!]
zorglub27 has quit [Ping timeout: 265 seconds]
ckwaldon1 has joined #ipfs
ckwaldon has quit [Ping timeout: 264 seconds]
ckwaldon1 is now known as ckwaldon
zorglub27 has joined #ipfs
ckwaldon1 has joined #ipfs
<dignifiedquire>
jbenet: so is it just a software limitation of connecting the two machines?
ckwaldon has quit [Ping timeout: 250 seconds]
ckwaldon1 is now known as ckwaldon
<geoah>
I think that thunderbolt and firewire both do network
<geoah>
USB worst case is usb-to-ethernet + ethernet-cable + usb-to-ethernet :P
rendar has joined #ipfs
rain1 has joined #ipfs
<jbenet>
dignifiedquire: yes, if you can cause a wire to carry a modulated voltage via software, that's really all you need.
<gorhgorh>
jbenet: i don't see anything preventing you to use serial comm for that btw
<jbenet>
dignifiedquire: so what's missing is osx support for usb-c networking. or really, what's missing (looking at the usb standard) is some common sense. what the fuck were they thinking-- why not base it on ethernet...
<jbenet>
gorhgorh: yeah agreed, i just need to get the addressing and wiring right, but i navigating nonsense. they do a bunch addressing and muxing with various fields of nonsense instead of just treating the link like an ethernet.
* jbenet
tableflip
<dignifiedquire>
I see you are enjoying usb a lot :)
m0ns00n_ has quit [Remote host closed the connection]
<Kubuxu>
from the USB protocol standard you don't want to do that, it is exactly what USB wasn't designed for.
m0ns00n has joined #ipfs
<jbenet>
Kubuxu: that o/ is not standard usb? how do i just get a damn pipe?
spilotro has joined #ipfs
<vapid>
how about giving data both a hash and human-readable tags
<vapid>
and you could have special ban databases which blacklist those who set fake tags for hashes or smth
<vapid>
and you could calculate string distance between search keywords and tags and things could be routed efficiently
<vapid>
and every peer would know about the tags other peers have set for a given hash
structuralist has quit [Ping timeout: 252 seconds]
m0ns00n has quit [Remote host closed the connection]
<vapid>
why isnt anyone else excited about my great idea
<rain1>
i am
<vapid>
great then pls implement it somehow
<rain1>
im already working on a different code :P
<Kubuxu>
jbenet: between two computers? you can't with just a cable, as both sides work in host mode AFAIk.
<Kubuxu>
there exist USB bridges but I haven't really looked on how they work.
m0ns00n has joined #ipfs
<vapid>
i just cant get myself to work on it
<vapid>
sadly
zorglub27 has quit [Ping timeout: 240 seconds]
apiarian has quit [Ping timeout: 244 seconds]
apiarian has joined #ipfs
gunn_ has quit [Ping timeout: 264 seconds]
computerfreak has joined #ipfs
herzmeister has quit [Quit: Leaving]
herzmeister has joined #ipfs
gunn has joined #ipfs
<kpcyrd>
> throw away your ethernet hub
<kpcyrd>
> use a usb hub
sametsisartenep has joined #ipfs
zorglub27 has joined #ipfs
anonymuse has quit [Remote host closed the connection]
mildred has quit [Ping timeout: 252 seconds]
<cjd>
USB chips are cheaper than etherenet chips because there is a host and X guests, the host is the only one who initiates communication with any one guest
<whyrusleeping>
it has a go src folder that looks like a bunch of deps got installed locally
<geoah>
awh there is no gxignore ><
<whyrusleeping>
it respects a gxignore and a gitignore
<geoah>
>< damnit...
<geoah>
I'll need to go over the whole dep tree :/
<geoah>
whyrusleeping: sorry about this :/
<daviddias>
dignifiedquire: confirmed that it is (at least) not the hmac function that is different, it is the key that is being hmac'ed that is different
<whyrusleeping>
eh, i'm sorry i didnt catch it earlier
<Kubuxu>
after running for 20min:
jedahan has quit [Remote host closed the connection]
<Kubuxu>
ipfs_bs_cache_bloom_hits_total 87725
<Kubuxu>
ipfs_bs_cache_bloom_total 88031
jedahan has joined #ipfs
jedahan has quit [Remote host closed the connection]
<whyrusleeping>
geoah: its not that much change
<whyrusleeping>
i think its only libp2p that imports the peerstore
<geoah>
also I'm very tempted to actually write a PR bot to check for gx dupes and also verify that the hash of the files in the repo match the hash in the lastpubver file
<whyrusleeping>
i was thinking of making it a webhook thing
<whyrusleeping>
we could totally make it be part of mars.i.ipfs.io
<haad>
whyrusleeping daviddias what's your problem with generating the (ellipctic) keys in JS? (I just did that last week for uPort + Orbit, might have tips)
<daviddias>
js is not generating the same shared secret, given the same pubkey, that go generates
<M-flyingzumwalt>
I had an idea of how to do asynchronous retrospectives, so we can gradually refine our meeting process without adding unnecessary time on our synchronous calls. See https://github.com/protocol/notes/issues/27
anonymuse has quit [Remote host closed the connection]
<richardlitt>
Anyone coming? ping @dignifiedquire
<richardlitt>
ping @daviddias
<dignifiedquire>
I'm in
<M-flyingzumwalt>
I had an idea of how to do asynchronous retrospectives, so we can gradually refine our meeting process without adding unnecessary time doing retrospectives on our synchronous calls (which gets really tiresome really fast). See https://github.com/ipfs/pm/issues/170
<dignifiedquire>
I was passing the data to the hmac in the wrong format
TheGillies has joined #ipfs
rschulman has joined #ipfs
M-leoc has joined #ipfs
M-hostbbb has joined #ipfs
hermann-tab[m] has joined #ipfs
M-mckaydavis has joined #ipfs
M-3444 has joined #ipfs
rodarima[m] has joined #ipfs
<dignifiedquire>
I was passing a buffer, but I needed to pass it with .toString('binary')....
<dignifiedquire>
and the keys were switched
arby[m] has joined #ipfs
Guest33346[m] has joined #ipfs
valdek[m] has joined #ipfs
M-3974 has joined #ipfs
M-foxxy has joined #ipfs
pik[m]11 has joined #ipfs
M-scio_ has joined #ipfs
M-AbhishekDutt has joined #ipfs
jsrockss is now known as 7YUABZVGU
Askaban[m]1 has joined #ipfs
M21100[m] has joined #ipfs
aaaaaaaaa____ has joined #ipfs
ofdm has joined #ipfs
hsanjuan has joined #ipfs
opal has joined #ipfs
bcavileer has joined #ipfs
omnigoat has joined #ipfs
stigo has joined #ipfs
kitcambridge has joined #ipfs
edubai__ has joined #ipfs
hosh has joined #ipfs
jhiesey has joined #ipfs
henriquev has joined #ipfs
codebam has joined #ipfs
xicombd has joined #ipfs
karissa has joined #ipfs
prettymuchbryce has joined #ipfs
tibor has joined #ipfs
yoshuawuyts has joined #ipfs
geoah has joined #ipfs
bitspill has joined #ipfs
NeoTeo has joined #ipfs
sega01 has joined #ipfs
sickill has joined #ipfs
ChrisDeChef[m] has joined #ipfs
cr[m] has joined #ipfs
teglafal[m] has joined #ipfs
M-57031 has joined #ipfs
Askaban[m]1 has joined #ipfs
Askaban[m]1 has quit [Changing host]
M21100[m] has quit [Changing host]
M21100[m] has joined #ipfs
rgrinberg has joined #ipfs
rgrinberg has quit [Max SendQ exceeded]
omnigoat has quit [Max SendQ exceeded]
prettymuchbryce has quit [Max SendQ exceeded]
codebam has quit [Max SendQ exceeded]
henriquev has quit [Ping timeout: 243 seconds]
karissa has quit [Ping timeout: 243 seconds]
metaf5_ has joined #ipfs
Justinus has joined #ipfs
rgrinberg has joined #ipfs
retrohacker_ is now known as retrohacker
bcavileer has quit [Ping timeout: 258 seconds]
herzmeister has quit [Quit: Leaving]
bitspill has quit [Ping timeout: 258 seconds]
sickill has quit [Ping timeout: 258 seconds]
herzmeister has joined #ipfs
jhiesey has quit [Ping timeout: 251 seconds]
xicombd has quit [Ping timeout: 251 seconds]
tibor has quit [Ping timeout: 251 seconds]
codebam has joined #ipfs
Foxcool has quit [Ping timeout: 276 seconds]
henriquev has joined #ipfs
myhau has joined #ipfs
<myhau>
hi, I'm learning ipfs and have a question: is there any way to check which nodes (or how many of em) has content with particular hash ?
omnigoat has joined #ipfs
karissa has joined #ipfs
xicombd has joined #ipfs
bitspill has joined #ipfs
taaem has quit [Remote host closed the connection]
taaem has joined #ipfs
jbenet_ is now known as jbenet
m0ns00n has joined #ipfs
bcavileer has joined #ipfs
<whyrusleeping>
myhau: yeah, you can use 'ipfs dht findprovs <hash>'
<whyrusleeping>
it won't give you a 'full' list, but it will give you the first 20 it finds crawling the dht
ofdm has quit [Changing host]
ofdm has joined #ipfs
jhiesey has joined #ipfs
<Kubuxu>
whyrusleeping: we are like: reuse buffers here to reduce allocations, do not copy data, and js: "I needed to pass it with .toString('binary')" :p
sickill has joined #ipfs
<Kubuxu>
I wonder if v8 JIT is smart enough to figure it ou.
<Kubuxu>
s/ou/out/
tibor has joined #ipfs
<whyrusleeping>
oh no
<whyrusleeping>
its even better
prettymuchbryce has joined #ipfs
<Kubuxu>
dignifiedquire: does it really work? If yes then js-secio was born the same day as me :p
<whyrusleeping>
i saw to much: new Buffer(thing.toHex(), 'hex')
<whyrusleeping>
convert this thing to hex, then convert it back into a buffer from hex
<whyrusleeping>
#perf
<dignifiedquire>
whyrusleeping: yeahh... node-forge things it's a good idea to create their own buffer implementation :'(
<dignifiedquire>
it hurt writing that code
<dignifiedquire>
*thinks
<whyrusleeping>
eh, just get faster CPUs
<Kubuxu>
with more L2/L3 cache ;p
<myhau>
I have one more question, when does replication of data happens (starts) and what determines how many nodes will store this data ?
<whyrusleeping>
feel free to ask more questions there, it just makes it easier on us
<Kubuxu>
wtfm principle
rendar has joined #ipfs
<PseudoNoob>
If I go to a hash using the localhost api, does this data get cached and does my node serve this data until it is removed from my local cache?
<PseudoNoob>
If something on ipfs went viral does each viewer contribube to the distribution for the duration that they have it in their cache?
jemc has quit [Ping timeout: 264 seconds]
<Kubuxu>
PseudoNoob: that is the exactly operational principle of ipfs. We call it inverse scaling.
<PseudoNoob>
Cool
<PseudoNoob>
thanks
sametsisartenep has quit [Quit: leaving]
<lgierth>
Kubuxu: we call it inverse scaling?
<lgierth>
TIL :)
<Kubuxu>
I hear Juan say it few times.
<whyrusleeping>
eh, inverse scaling refers to the problems we have with ipfs at a smaller scale
<whyrusleeping>
ipfs isnt super happy in tiny clusters
<whyrusleeping>
but once you have a couple hundred nodes up its just fine
<7YUABZVGU>
maybe I'm doing soemthing wrong? but it seems like a bug
<clownpriest>
hey whyrusleeping, the http api isn't returning any data for diag/net?vis=d3, but it returns results for ?vis=text and ?vis=dot. am I doing something wrong?
<clownpriest>
running diag net --vis=d3 on the command line returns results though
<whyrusleeping>
clownpriest: thats odd, i can look into it more, a little busy right at the moment
<whyrusleeping>
could you file an issue with what youve tried, and a way to reproduce the issue?
<clownpriest>
cool, thanks
<clownpriest>
yup
herzmeister has quit [Quit: Leaving]
herzmeister has joined #ipfs
cketti has joined #ipfs
sametsisartenep has joined #ipfs
gmcquillan has quit [Quit: gmcquillan]
byteflame has joined #ipfs
wuch has quit [Quit: = ""]
<victorbjelkholm>
in terms of the http-api, what would be the difference between .cat and .get? Both would return the contents in the response...
spilotro has quit [Ping timeout: 276 seconds]
<victorbjelkholm>
daviddias, got the .cat to work, just .get left. Was what you said, a header missing :)
<daviddias>
:D
<dignifiedquire>
daviddias: I see the big merge has begun :D
<daviddias>
victorbjelkholm: cat is a simple octet-stream while get is a multipart message that might contain dirs and files
<daviddias>
dignifiedquire: :D
<dignifiedquire>
(I am already in bed so no worries you can merge all night long without me complaining ;) )
<victorbjelkholm>
daviddias, got it
<daviddias>
dignifiedquire: it would be great to have that pull-stream-to-stream merged and released
<daviddias>
that is blocking me
wuch has joined #ipfs
wuch has quit [Client Quit]
rendar has quit [Quit: std::lower_bound + std::less_equal *works* with a vector without duplicates!]
computerfreak has joined #ipfs
<dignifiedquire>
you can ping domenic, he wrote earlier he wants to merge it
<lgierth>
daviddias: i think the core-api needs an update regarding CID, whenever it e.g. specifies a parameter as multihash. ack? i can take care of it
<daviddias>
lgierth: that is correct
<daviddias>
lgierth: it should be updated with the tests too
<daviddias>
not sure the text
<lgierth>
heheh would be great if one of you could take care of the js part -- it's gonna be rabbit hole for to get into it. i'll make sure that the go version comes with CID from the start
<lgierth>
*rabbit hole for me to get into
sametsisartenep has quit [Ping timeout: 244 seconds]