<multivac>
[WIKIPEDIA] Gossip protocol | "A gossip protocol is a style of computer-to-computer communication protocol inspired by the form of gossip seen in social networks. Modern distributed systems often use gossip protocols to solve problems that might be difficult to solve in other ways, either because the underlying network has an inconvenient..."
<A124>
I did actually think about something on principle of spreading general info. Pubsub alone would be limited to popular services or centered around topics.
mildred1 has quit [Ping timeout: 276 seconds]
<A124>
So gossip or other means would greatly enhace the awareness of the mesh.
Arakela007 has joined #ipfs
<A124>
Selecting nearby peers by latency for gossip would make sense I guess, limited by number, and using a periodically w/ random intervals reset bloom filter or list of peers the node did already gossip with.
<A124>
Or circular linked list or something, not sure about the exact advantage, but it would avoid situations of close communities, so they would after some time start to spread rumors outside of their community.
<A124>
Just a thought.
ylp1 has quit [Quit: Leaving.]
<A124>
... But currently it is hard to asses "what's flowing" on the mesh, so one can cash the stuff that is requested very often locally.
afisk has quit [Remote host closed the connection]
zorglub27 has joined #ipfs
ygrek has quit [Ping timeout: 244 seconds]
pfista has joined #ipfs
mildred has joined #ipfs
<haad_>
nicolagreco: noffle has been thinking about gossip pubsub. whyrusleeping has been on go-ipfs pubsub and daviddias is your guy in js-ipfs and libp2p questions.
<pfista>
package github.com/whyrusleeping/gx-go: cannot download, $GOPATH not set. For more details see: go help gopath ~/s/g/s/g/i/go-ipfs ❯❯❯ echo $GOPATH ⏎ master /Users/mik/src/go
<A124>
nicolagreco link 404
<pfista>
it says gopath isn't set but it is set...?
cryptix has joined #ipfs
matoro has joined #ipfs
zorglub27 has quit [Ping timeout: 276 seconds]
nycoliver has joined #ipfs
gmcquillan has joined #ipfs
mildred has quit [Ping timeout: 240 seconds]
<ipfsbot>
[webui] greenkeeperio-bot opened pull request #341: Update react-bootstrap to version 0.29.1
jaboja has quit [Ping timeout: 268 seconds]
dmr has joined #ipfs
jaboja has joined #ipfs
gmcquillan has quit [Quit: gmcquillan]
jaboja has quit [Ping timeout: 250 seconds]
nycoliver has quit [Ping timeout: 240 seconds]
_Vi has quit [Ping timeout: 268 seconds]
jaboja has joined #ipfs
TheWhisper_ has joined #ipfs
unforgiven512 has quit [Ping timeout: 268 seconds]
M-4447 has quit [Ping timeout: 268 seconds]
M-epsilon has quit [Ping timeout: 268 seconds]
Boomerang has quit [Ping timeout: 260 seconds]
oleavr has quit [Ping timeout: 268 seconds]
M-epsilon has joined #ipfs
sikander_ has quit [Read error: Connection reset by peer]
noffle has quit [Ping timeout: 268 seconds]
M-4447 has joined #ipfs
M-Dave has quit [Ping timeout: 268 seconds]
noffle has joined #ipfs
M-Dave1 has joined #ipfs
M-Peer2Peer1 has joined #ipfs
unforgiven512 has joined #ipfs
sikander has joined #ipfs
zrl_ has quit [Ping timeout: 268 seconds]
TheWhisper has quit [Ping timeout: 268 seconds]
M-Peer2Peer has quit [Ping timeout: 268 seconds]
bergie has quit [Ping timeout: 268 seconds]
mvollra7h has joined #ipfs
manveru has quit [Ping timeout: 268 seconds]
mvollrath has quit [Ping timeout: 268 seconds]
kanzure has quit [Ping timeout: 268 seconds]
richardlitt has quit [Ping timeout: 268 seconds]
barndon has quit [Ping timeout: 268 seconds]
mvollra7h is now known as mvollrath
oleavr has joined #ipfs
s_kunk has quit [Ping timeout: 246 seconds]
Mitar has quit [Ping timeout: 268 seconds]
barndon has joined #ipfs
jaboja has quit [Ping timeout: 250 seconds]
manveru has joined #ipfs
dhiru26_ has quit [Ping timeout: 268 seconds]
richardlitt has joined #ipfs
Boomerang has joined #ipfs
rombou has quit [Quit: Leaving.]
zrl_ has joined #ipfs
dhiru26 has joined #ipfs
nullstyle has quit [Ping timeout: 297 seconds]
Mitar has joined #ipfs
Vegemite has joined #ipfs
nullstyle has joined #ipfs
Vegemite has left #ipfs ["nope"]
reit has joined #ipfs
TheWhisper_ is now known as TheWhisper
kanzure_ has joined #ipfs
mildred has joined #ipfs
chris6132 has joined #ipfs
kanzure_ has quit [Client Quit]
OutBackDingo_ has joined #ipfs
M-edrex1 has joined #ipfs
M-rongladney1 has joined #ipfs
M-volobuev1 has joined #ipfs
bergie has joined #ipfs
M-alri1 has joined #ipfs
M-rschulman2 has joined #ipfs
rschulman_ has joined #ipfs
Arwalk_ has joined #ipfs
M-Nickfost1 has joined #ipfs
ansuz_ has joined #ipfs
rektide_ has joined #ipfs
lmatteis_ has joined #ipfs
holst has joined #ipfs
null_radix has joined #ipfs
tlevine_ has joined #ipfs
zielmicha_ has joined #ipfs
manveru_ has joined #ipfs
cblgh has quit [Disconnected by services]
cblgh_ has joined #ipfs
flyingkiwi has joined #ipfs
kanzure has joined #ipfs
zrl_ has quit [*.net *.split]
manveru has quit [*.net *.split]
M-Dave1 has quit [*.net *.split]
cryptix has quit [*.net *.split]
OutBackDingo has quit [*.net *.split]
rmi7 has quit [*.net *.split]
seagreen has quit [*.net *.split]
tlevine has quit [*.net *.split]
M-edrex has quit [*.net *.split]
zeroish has quit [*.net *.split]
pinbot has quit [*.net *.split]
lmatteis has quit [*.net *.split]
rektide has quit [*.net *.split]
ansuz has quit [*.net *.split]
rschulman has quit [*.net *.split]
locusf has quit [*.net *.split]
M-Nickfost has quit [*.net *.split]
M-2223 has quit [*.net *.split]
diffalot has quit [*.net *.split]
chris613 has quit [*.net *.split]
M-rschulman has quit [*.net *.split]
M-3511 has quit [*.net *.split]
kyledrake has quit [*.net *.split]
M-rongladney has quit [*.net *.split]
M-volobuev has quit [*.net *.split]
M-3891 has quit [*.net *.split]
M-alri has quit [*.net *.split]
M-harlan has quit [*.net *.split]
M-pieohpah has quit [*.net *.split]
M-3725 has quit [*.net *.split]
yangwao has quit [*.net *.split]
pinbot-v03x has quit [*.net *.split]
holst_ has quit [*.net *.split]
flyingkiwi_ has quit [*.net *.split]
grncdr has quit [*.net *.split]
henriquev__ has quit [*.net *.split]
Arwalk has quit [*.net *.split]
null_rad- has quit [*.net *.split]
mrrrgn_ has quit [*.net *.split]
zielmicha has quit [*.net *.split]
lmatteis_ is now known as lmatteis
Arwalk_ is now known as Arwalk
zielmicha_ is now known as zielmicha
seagreen has joined #ipfs
M-2223 has joined #ipfs
M-3511 has joined #ipfs
M-harlan has joined #ipfs
M-Dave has joined #ipfs
manveru_ is now known as manveru
locusf has joined #ipfs
M-3891 has joined #ipfs
yangwao has joined #ipfs
M-3725 has joined #ipfs
<pfista>
whyrusleeping: 2597 should be good to go now
<pfista>
what's with the circleci failures?
mrrrgn_ has joined #ipfs
kyledrake has joined #ipfs
diffalot has joined #ipfs
diffalot has quit [Changing host]
diffalot has joined #ipfs
M-pieohpah has joined #ipfs
rmi7 has joined #ipfs
pinbot has joined #ipfs
pinbot-v03x has joined #ipfs
M-pieohpah has quit [Changing host]
M-pieohpah has joined #ipfs
rmi7 has quit [Changing host]
rmi7 has joined #ipfs
Guanin_ has left #ipfs [#ipfs]
henriquev__ has joined #ipfs
grncdr has joined #ipfs
zrl_ has joined #ipfs
jaboja has joined #ipfs
cryptix has joined #ipfs
joeghi_ has joined #ipfs
palkeo has joined #ipfs
pfraze has joined #ipfs
<whyrusleeping>
pfista: this PR is a little more complicated than I first thought
<whyrusleeping>
i'm checking some things out now
joeghi has quit [Ping timeout: 276 seconds]
<pfista>
k
Encrypt has joined #ipfs
<whyrusleeping>
pfista: the issue here is that `ipfs repo fsck` almost needs to be not called by the `ipfs` command at all
<whyrusleeping>
because before we even consider the commands, we open up the repo
<whyrusleeping>
meaning a bad repo lock is likely to cause repo fsck to not even run
Boomerang has quit [Ping timeout: 250 seconds]
<whyrusleeping>
pfista: i think i have a solution, it requires a small refactor in the commands lib
<pfista>
hmm
<whyrusleeping>
but i'll write that part and push a commit you can cherry-pick
M-6781 has joined #ipfs
Boomerang has joined #ipfs
<pfista>
whyrusleeping: sounds good, i just pushed latest changes based on your comments
s_kunk has joined #ipfs
Guanin_ has joined #ipfs
matoro has quit [Ping timeout: 246 seconds]
mildred has quit [Ping timeout: 240 seconds]
matoro has joined #ipfs
nycoliver has joined #ipfs
reit has quit [Quit: Leaving]
<ipfsbot>
[js-ipfs] dignifiedquire pushed 1 new commit to aegir: https://git.io/vwlOw
disgusting_wall has quit [Quit: Connection closed for inactivity]
nycoliver has quit [Ping timeout: 276 seconds]
mildred has joined #ipfs
disgusting_wall has joined #ipfs
ygrek has joined #ipfs
Encrypt has quit [Quit: Quitte]
uhhyeahbret has joined #ipfs
zorglub27 has joined #ipfs
rendar has quit [Ping timeout: 246 seconds]
<whyrusleeping>
pfista: whoaaaaaa
<whyrusleeping>
so i was digging around, writing a bunch of code to make this all work nicely
<whyrusleeping>
and then i discovered some stuff in the commands lib i didnt know existed
<whyrusleeping>
i'll add some comments on the PR
jo_mo has joined #ipfs
rendar has joined #ipfs
Guanin_ has left #ipfs [#ipfs]
chungy has quit [Ping timeout: 250 seconds]
computerfreak has joined #ipfs
computerfreak has quit [Client Quit]
<dignifiedquire>
hey whyrusleeping how are you?
<dignifiedquire>
daviddias: how is bitswap coming along? ;)
<whyrusleeping>
dignifiedquire: doin alright
<whyrusleeping>
not feeling super productive, but i'm tryin
<dignifiedquire>
we either succeed or die trying ;)
<lgierth>
try dying?
<whyrusleeping>
lgierth is alive?
<lgierth>
alive and fighting the good android fight
<whyrusleeping>
noice
conway has joined #ipfs
<dignifiedquire>
lgierth: who is winning you or android?
<lgierth>
not sure yet
<lgierth>
we have the tun interface working via VpnService and a little c++ tool for file descriptor transfer between java and cjdns
<lgierth>
java's way of starting and handling processes is fighting hard though
<lgierth>
it just keeps giving
<lgierth>
and then there's the looong feedback cycle of change, build apk, upload, start app, do some clicking, watch it crash, filter through all the noise in adb logcat
<conway>
it gets better if you use the NDK... but then your system code becomes exponentially more complicated
<ipfsbot>
[webui] greenkeeperio-bot opened pull request #342: Update three to version 0.76.1
<lgierth>
we are using the ndk
<lgierth>
we're building an app around a c program
<conway>
Ahh. I didn't realize that. I haven't looked into IPFS/android yet.
<conway>
I've done some with ndk and OpenCV. That was a bear and a half :/
<lgierth>
it's cjdns in gact :)
<lgierth>
*fact
<conway>
AHh, cool :)
<lgierth>
so you're gonna have a cjdns app in play store and f-droid store soon that doesn't require root
<lgierth>
end-to-end voip is the first use case
ljhms has quit [Ping timeout: 260 seconds]
<daviddias>
dignifiedquire: still finishing the files, but good progress, shouldn't be a blocker any longer :)
<dignifiedquire>
daviddias: great, I'm also pretty close to fixing js-ipfs :)
ljhms has joined #ipfs
ljhms has quit [Read error: Connection reset by peer]
<dignifiedquire>
lgierth: sounds like you're having lots of fuN D:
<daviddias>
Sweet! Feeling that today is the day ?
<daviddias>
Tomorrow will be a big day then
<dignifiedquire>
daviddias: I fixed the browser tests already, and some mistakes I made with the cli tests
dmr has quit [Ping timeout: 250 seconds]
<dignifiedquire>
now it's just figuring out what's going wrong with the cli in a multi run scenario,
Akkad has quit [Ping timeout: 244 seconds]
<lgierth>
dignifiedquire: it's not too bad
<lgierth>
this java<->c integration thing is actually pretty cool
<lgierth>
and google didn't fuck up their libc too badly
<lgierth>
they didn't wanna use gnu libc so they wouldn't have to deal with the GPL, and musl wasn't around back then
<dignifiedquire>
I get the feeling most people doing anything commercial do not want to deal with GPL
ljhms has joined #ipfs
<lgierth>
yeah because the GPL isn't about the producer's rights
<lgierth>
it's about the user's rights
<lgierth>
and who cares about the user's rights :)
<lgierth>
*maintaining the user's rights
Boomerang has quit [Quit: Leaving]
<dignifiedquire>
:D
mildred1 has joined #ipfs
ImmoXZ has joined #ipfs
<ImmoXZ>
hello, anyone would answer one dumb question? :)
pfista_ has joined #ipfs
<pfista_>
whyrusleeping: i'm back. saw your comment :)
<pfista_>
is "api" another file that should be removed? I saw you mentioned it in the tests
<pfista_>
right now only repo.lock and leveldb's LOCK are being removed
<ImmoXZ>
is there any way to add txt file from localhost to directory that i have hash to it?
ashark has quit [Ping timeout: 276 seconds]
afisk has joined #ipfs
<ipfsbot>
[js-ipfs] dignifiedquire pushed 1 new commit to aegir: https://git.io/vwliL
<ipfsbot>
js-ipfs/aegir 085c1cb dignifiedquire: test(cli): more fixes
<dignifiedquire>
daviddias: could I get a second pair of eyes on these last issues?
<pfista_>
whyrusleeping: should i put tests for repo fsck in t0080-repo or create a new file?
<dignifiedquire>
daviddias: everything works except for the cli tests, in particular:
<dignifiedquire>
- when running them without the other tests all pass except the object tests
<dignifiedquire>
- when running them in conjunction with the other tests all online versions fail while the offline versions pass
<dignifiedquire>
- when running only the object tests they suddenly pass
Encrypt has joined #ipfs
neurosis12 has joined #ipfs
neurosis12 has joined #ipfs
Encrypt has quit [Read error: Connection reset by peer]
<whyrusleeping>
pfista_: i'd go ahead and create a new file
<whyrusleeping>
t0081 or something
Encrypt has joined #ipfs
<pfista_>
kk
<ipfsbot>
[js-ipfs] dignifiedquire pushed 1 new commit to aegir: https://git.io/vwliF
<richardlitt>
And he and his friend were like, yeah, have him come in for beer and to present stuff
<richardlitt>
open invite. Up to you? Might be fun? I dunno, I only know Robert.
<richardlitt>
I can get more details
<pfista_>
also, i've been unsuccessful with subsystem log levels. I'll add print statements with info/debug etc and then set `ipfs log level command info` yet i won't ever see those print statements. I only see errors still
<lgierth>
i don't have the time to be honest...
<richardlitt>
lgierth: cool. No worries!
<ImmoXZ>
whyrusleeping: yes i have them both
<lgierth>
ben has work tomorrow during the day and the evening will be the last chance to work on that app together
<richardlitt>
Was hoping there'd be more of a hacking event, not sure though
<whyrusleeping>
pfista_: hrm... i'm not really sure on the log levels.
<lgierth>
and i'm not sure what to present either
<whyrusleeping>
pfista_: try setting IPFS_LOGGING to info
<ImmoXZ>
found that print of your, thank you i will try it
<_Vi>
Does it relate to pages.github.com (which does work with https)?
ZaZ has joined #ipfs
Arakela007 has quit [Remote host closed the connection]
gmcquillan has joined #ipfs
matoro has quit [Remote host closed the connection]
rendar has quit [Quit: std::lower_bound + std::less_equal *works* with a vector without duplicates!]
matoro has joined #ipfs
nycoliver has joined #ipfs
dhiru26 has joined #ipfs
Senji has quit [Ping timeout: 240 seconds]
zorglub27 has quit [Quit: zorglub27]
rmi7 has quit [Ping timeout: 250 seconds]
patcon has quit [Ping timeout: 250 seconds]
rmi7 has joined #ipfs
kingflurkel has joined #ipfs
tlevine_ is now known as tlevine
mildred1 has quit [Ping timeout: 276 seconds]
mildred has quit [Ping timeout: 268 seconds]
herzmeister has quit [Quit: Leaving]
lidel has quit [Ping timeout: 260 seconds]
dmr has joined #ipfs
lidel has joined #ipfs
rombou has joined #ipfs
simpbrain has quit [Ping timeout: 276 seconds]
ruby32 has joined #ipfs
ashark has quit [Ping timeout: 268 seconds]
ruby32 has quit [Remote host closed the connection]
herzmeister has joined #ipfs
nycoliver has quit [Ping timeout: 252 seconds]
<pfista_>
whyrusleeping: tests are up, my only question is about the API thing you mentioned
rombou has quit [Ping timeout: 246 seconds]
Boomerang has joined #ipfs
afisk has quit [Remote host closed the connection]
ZaZ has quit [Quit: Leaving]
chungy has joined #ipfs
<whyrusleeping>
pfista_: the api file is also something that should be removed during a fsck
pfista_ has quit [Ping timeout: 250 seconds]
chungy has quit [Read error: Connection reset by peer]
patcon has joined #ipfs
bheesham has left #ipfs [#ipfs]
chungy has joined #ipfs
afisk has joined #ipfs
simpbrain has joined #ipfs
computerfreak has quit [Remote host closed the connection]
afisk has quit [Ping timeout: 250 seconds]
<edrex>
Hi all. Anybody going to be at LFNW this weekend? The Camlistore devs are going to be there, and I am thinking spec out some IPFS interop stuff there.
<edrex>
It's been a few months since I was tracking IPFS progress, trying to catch up.
<whyrusleeping>
edrex: i would have to drive up to bellingham... hrmm
<edrex>
whyrusleeping: where are you located? I should have checked in sooner.
<A124>
Correct me please. For this case I have the impression they are. So context and interrupt problems will happen.
dmr has quit [Ping timeout: 240 seconds]
<whyrusleeping>
edrex: threads and processes in linux are two very different things
_Vi has quit [Ping timeout: 260 seconds]
<whyrusleeping>
a process is an isloated memory space for execution of code
<whyrusleeping>
forking a new process copies the entire memory space of the parent process
<edrex>
^ A124
<whyrusleeping>
and the only method of communicating between the two is with some form of IPC
<whyrusleeping>
(thus the name, IPC = inter process communication)
<whyrusleeping>
multiple threads can exist sharing a single memory space
<A124>
Yeah, I got that part.
<whyrusleeping>
they can shared memory
<whyrusleeping>
and must use locks to prevent issues with shared memory
<whyrusleeping>
below that, you have various forms of green threads.
<whyrusleeping>
green threads arent scheduled by the kernel, they are scheduled by the runtime of whatever application is using them
_Vi has joined #ipfs
<A124>
green thread = fibre? yeah.
<A124>
Well apparently not working.
<whyrusleeping>
a fibre is a type of green thread, yes
<A124>
Starting IPFS makes tons of threads recognized by kernel.
<whyrusleeping>
right, so as i said before, the go runtime will spawn threads as it sees fit
<whyrusleeping>
so now the question becomes, why is the go runtime spawning excess threads?
<A124>
And how to prevent that, yeah.
<edrex>
whyrusleeping: yeah, Google Fremont. Glad I pinged the room, didn't know you were in Seattle. If you wanted to come by during the day on Monday, that'll be hack time. I'm hoping to at least get IPFS more on their radar, since I feel like Camlistore has become too much of a closed ecosystem.
<nicolagreco>
whyrusleeping: can I store an arbitrary key and value to the ipfs dht?
<whyrusleeping>
nicolagreco: not currently, no
<nicolagreco>
whyrusleeping: put(key, value)
<whyrusleeping>
a few different people have been asking about that, and i'm open to having that be possible
<nicolagreco>
whyrusleeping: do you think it will?
<whyrusleeping>
someone just needs to implement it
<whyrusleeping>
both noffle and sivachandran had shown interest in doing it
<nicolagreco>
my issue then would be, how do you ensure that the node that keeps this data is not going to alter it?
heyrhett has quit [Ping timeout: 268 seconds]
<nicolagreco>
I am just reading about Scribe, which re-use the existing DHT to build a tree about peers interested to be updated about the change of a particular name (read as pub/sub for ipns)
<noffle>
nicolagreco: whyrusleeping: I spoke to jbenet not long ago about this, and he was concerned about network health if we allowed arbitrary kv pairs
<whyrusleeping>
noffle: is there an issue with the concerns?
<noffle>
but if this is for pubsub, we could definitely just add a record type + validator + selector to go-ipfs so that ipfs nodes can make use of them
<noffle>
whyrusleeping: not that I could find :( but I can update an existing issue with what he wrote to me -- sec