<whyrusleeping>
you have to run the tool directly i think
<kevina>
whyrusleeping: can they download it?
<achin>
the user in that issue is me :)
<kevina>
achin: okay, can you compile from git?
<achin>
sure thing
<kevina>
one sec
webdev007 has joined #ipfs
<achin>
i've taken a snapshot of this $IPFS_PATH, so we're free to try potentially dangerous or destructive stuff, since i can always rollback to my snapshot
<kevina>
achin: download github.com/ipfs/fs-repo-migrations cd to $GOPATH/src/github.com/ipfs/fs-repo-migrations/ipfs-4-to-5 and run "go build"
<kevina>
this will compile the ipfs-4-to-5 tool directly
<achin>
ok, running 284bb04ca72a8090cb22405231fb4a641cb3a665 of the fs-repo-migrations project
<achin>
done. i now have a ipfs-4-to-5 binary in my cwd
<kevina>
then run "ipfs-4-to-5 -no-revert <ipfs path>"
<kevina>
it should stop on the error rather than trying to revert
<achin>
running now
<achin>
ok, errored as expected
<kevina>
does "/nas/achin/ipfs/blocks-v5/79/" exist?
<achin>
yep. it's an empty directory
<kevina>
and for sanity sake does the file that is trying to be moved not exist?
<achin>
so yeah, there is stuff in those two directories
esph has joined #ipfs
<kevina>
see it, that looks leftover from an old incomplete migration
<achin>
ok, well those are obviously junk hellworld tests in anycase. i'll purge them
<kevina>
probably confusing the migration tool
<achin>
should i restore from backup, purge, and then re-run the migration? or try to re-run from the current state
<kevina>
run-run from the current state, it is 100% resumable
<kevina>
^ re-run
<kevina>
how big is the repo about?
<achin>
8GB ish
<kevina>
okay, ping me when its done
<achin>
will do
<achin>
kevina: finished. no other errors. anything you want to check before i try to startup the daemon?
<kevina>
achin: nope it should be good
<achin>
\o/
<achin>
many thanks for the help
<kevina>
no problem, the tool should of caught that though
DiCE1904 has quit [Ping timeout: 258 seconds]
arpu has quit [Ping timeout: 268 seconds]
spinus[m] has joined #ipfs
Ragnis has quit [Ping timeout: 264 seconds]
PrinceOfPeeves has quit [Read error: Connection reset by peer]
Ragnis has joined #ipfs
arpu has joined #ipfs
webdev007 has quit [Quit: Leaving]
deetwelve has quit [Quit: foobar]
tilgovi has quit [Ping timeout: 246 seconds]
wrouesnel has quit [Remote host closed the connection]
arpu has quit [Ping timeout: 256 seconds]
deetwelve has joined #ipfs
shizy has joined #ipfs
spacebar_ has joined #ipfs
amdi_ has quit [Quit: Leaving]
<achin>
can someone remind me the prometheus metrics URL exposed by the ipfs daemon?
arpu has joined #ipfs
spacebar_ has quit [Quit: spacebar_ pressed ESC]
strixy- has quit [Ping timeout: 245 seconds]
sprint-helper has quit [Ping timeout: 258 seconds]
wrouesnel has joined #ipfs
strixy has joined #ipfs
sprint-helper has joined #ipfs
A124 has quit [Ping timeout: 260 seconds]
spacebar_ has joined #ipfs
palkeo has joined #ipfs
nekomune has joined #ipfs
mguentner2 has quit [Quit: WeeChat 1.7]
mguentner has joined #ipfs
gmcabrita has quit [Quit: Connection closed for inactivity]
MrControll has quit [Quit: Leaving]
Jesin has joined #ipfs
shizy has quit [Ping timeout: 240 seconds]
spacebar_ has quit [Quit: spacebar_ pressed ESC]
A124 has joined #ipfs
dimitarvp has quit [Quit: Bye]
azdle has quit [Ping timeout: 240 seconds]
mguentner2 has joined #ipfs
frood has quit [Ping timeout: 240 seconds]
frood has joined #ipfs
mguentner has quit [Ping timeout: 260 seconds]
Guest192292[m] has joined #ipfs
JayCarpenter has quit [Quit: Page closed]
wyatt has joined #ipfs
gsf has quit [Ping timeout: 260 seconds]
gsf has joined #ipfs
matoro has quit [Ping timeout: 258 seconds]
spacebar_ has joined #ipfs
matoro has joined #ipfs
chris613 has quit [Quit: Leaving.]
spacebar_ has quit [Quit: spacebar_ pressed ESC]
toomim has joined #ipfs
toomim has quit [Read error: Connection reset by peer]
A124 has quit [Ping timeout: 240 seconds]
toomim has joined #ipfs
toomim has quit [Client Quit]
fjl_ has joined #ipfs
eater has quit [Ping timeout: 240 seconds]
fjl has quit [Ping timeout: 240 seconds]
eater has joined #ipfs
robogoat has quit [Ping timeout: 240 seconds]
robogoat has joined #ipfs
<lgierth>
achin: :5001/debug/metrics/prometheus
_rmt has quit [Ping timeout: 260 seconds]
patagonicus has quit [Ping timeout: 240 seconds]
patagonicus42 has joined #ipfs
Aranjedeath has quit [Quit: Three sheets to the wind]
wyatt has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
cwill has quit [Ping timeout: 240 seconds]
cwill has joined #ipfs
wyatt has joined #ipfs
bwn has quit [Ping timeout: 240 seconds]
palkeo has quit [Quit: Konversation terminated!]
bwn has joined #ipfs
Jesin has quit [Ping timeout: 260 seconds]
melvster has joined #ipfs
wkennington has quit [Read error: Connection reset by peer]
bwn has quit [Ping timeout: 240 seconds]
s_kunk has quit [Remote host closed the connection]
bwn has joined #ipfs
MDude has joined #ipfs
jkilpatr has quit [Ping timeout: 246 seconds]
senc01a[m] has joined #ipfs
inetic has joined #ipfs
wrouesnel has quit [Remote host closed the connection]
jsgrant has quit [Ping timeout: 240 seconds]
jsgrant-_ has quit [Ping timeout: 264 seconds]
maxlath has joined #ipfs
ecloud is now known as ecloud_wfh
wookiehangover has quit [Ping timeout: 258 seconds]
jsgrant has joined #ipfs
jsgrant-_ has joined #ipfs
ygrek has joined #ipfs
wookiehangover has joined #ipfs
jchevalay has joined #ipfs
<jchevalay>
hi all
<jchevalay>
i attempt to use node js ipfs in my ubuntu 16.04 with npm 3.5.2 and node 6.4.5 but i have an error with hex is not function when i run exemples/basics/index.js
mildred has quit [Ping timeout: 256 seconds]
Caterpillar has joined #ipfs
ylp1 has joined #ipfs
mildred has joined #ipfs
ygrek has quit [Ping timeout: 260 seconds]
vandemar has quit [Ping timeout: 240 seconds]
ygrek has joined #ipfs
jchevalay has quit [Quit: Page closed]
wyatt has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
shrimpx has joined #ipfs
reit has quit [Ping timeout: 268 seconds]
ygrek has quit [Ping timeout: 258 seconds]
cxl000 has joined #ipfs
j0hnsmith has joined #ipfs
ylp1 has quit [Quit: Leaving.]
ygrek has joined #ipfs
patagonicus42 is now known as patagonicus
Guest192463[m] has joined #ipfs
j0hnsmith has quit [Quit: j0hnsmith]
j0hnsmith has joined #ipfs
chriscool has quit [Ping timeout: 260 seconds]
eater has quit [Ping timeout: 264 seconds]
google77 has joined #ipfs
<google77>
hi
<google77>
I am mainly a Linux guy, so sorry if this is a stupid question, but, is it possible to mount IPFS on Windows as a network storage of sorts?
<Mateon1>
google77: The FUSE IPFS mount is not supported on Windows, it might be possible to make it work in a hacky way (and I do believe people have done that!), but I haven't personally tried since FUSE is too much of a pain
nunofmn has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<dimitarvp>
victorbjelkholm: is this pinning service free or paid? I am kind of worried seeing IPFS paid pin services come to existence lately...
maxlath has quit [Ping timeout: 240 seconds]
<victorbjelkholm>
dimitarvp: it's protocol labs that host this for the IPFS community. The only paid pinning service I know of is pinbits.io
<dimitarvp>
victorbjelkholm: Thanks. (Although I remember seeing 2-3 of them recently, would be very happy to be wrong.)
<dimitarvp>
I wonder if the IPFS creators plan to add a "I want to share bandwidth and donate some of my disk space" mode. I'd open 200 or so GB for that on my machine. ^_^
<victorbjelkholm>
yeah, something like filecoin.io ;)
<patagonicus>
Well, you aren't really donating with filecoin because you get something in return. ;)
<reit>
i guess you could offer to sell your spare bandwidth for 0FCN
nunofmn has joined #ipfs
JustinDrake has joined #ipfs
JustinDrake has quit [Client Quit]
nunofmn has quit [Ping timeout: 256 seconds]
bauruine_ has quit [Read error: Connection reset by peer]
<dimitarvp>
I honestly don't want to be rewarded. I want automatic replication inside IPFS through people who want to donate. I am one of them. That can happen via some sort of auto-discovery; read on the IPFS FAQ/Notes GitHub issues that it's possible.
<dimitarvp>
No money should be involved, real or virtual.
bauruine has joined #ipfs
Boomerang has joined #ipfs
chungy has quit [Ping timeout: 256 seconds]
<victorbjelkholm>
dimitarvp: ah, you're thinking similar to how pincoop ( https://github.com/victorbjelkholm/pincoop ) was supposed to be? Like a community pinning service?
<dimitarvp>
victorbjelkholm: Yep, only automatic. Here is a mechanism outlined that can help people auto-discover other such people: https://github.com/ipfs/notes/issues/15
<victorbjelkholm>
dimitarvp: I think the automatic discovery wouldn't be so difficult, lots of different ways that can be achieved. But yeah, there is currently nothing built on top of IPFS that helps with the questions you have in the end. Would have to be on the application layer
<dimitarvp>
victorbjelkholm: And IMO that's a huge mistake. IPFS needs to be the next level, not just a good enough solution. At any rate, having automatic replication (pinning) without anyone's manual approval is IMO a huge step towards creating a censorship-resistant network.
<dimitarvp>
What I am saying is, I believe auto-replication must be built-in.
maxlath has joined #ipfs
Monokles_2 is now known as Monokles
bauruine has joined #ipfs
<Monokles>
are there any guides on handling mutable data with ipfs?
PrinceOfPeeves has joined #ipfs
<Magik6k>
I might have found a bug: in js-ipfs when I call node.stop() shortly after on("start") was emitted, I get 'AssertionError: The libp2p node is not started yet'. putting 200ms timeout on stop fixes that, but I feel it shouldn't be the case.
<victorbjelkholm>
might help if you search for the terms used there
<victorbjelkholm>
dimitarvp: I think that's the concern of other layers, rather than built-in in the protocol. You're basically want Freenet instead of IPFS
jchevalay has joined #ipfs
<victorbjelkholm>
where you can basically "fire and forget" content into the network, and leave once the initial seed is done. That's different from what IPFS is trying to achieve. There is more information here: https://github.com/ipfs/faq/issues/18
<hsanjuan>
Kubuxu: There is no way to add a directory via go-ipfs http api right?
<victorbjelkholm>
hsanjuan: yeah, there should be, js-ipfs-api can do it so should be in the API
<dimitarvp>
victorbjelkholm: Wasn't aware of Freeneet, I'll review it now, thanks.
<hsanjuan>
victorbjelkholm: like passing a path to /add or how?
JayCarpenter has joined #ipfs
thomersch has joined #ipfs
<victorbjelkholm>
hsanjuan: no, I think the full contents are in the body of the request and the request is a application/octet-stream or something like that
<daviddias>
wrtc needs some extra libs to compile in Linux, could you check if installing them solves your problem? They are listed in the issue I pointed you to :)
<jchevalay>
so i need to install that ? sudo apt-get install python2.7 git-all pkg-config libncurses5-dev libssl-dev libnss3-dev libexpat-dev
ianopolous has joined #ipfs
<jchevalay>
ok i'll try that later
jchevalay has quit [Quit: Page closed]
lindybrits has joined #ipfs
<lindybrits>
@daviddias if you or @dignifiedquire are in the house, I have a question for you - what does "no available transport to dial to" mean when trying to connect two jsipfs nodes?
<daviddias>
lindybrits: means that you are trying to dial to a node in which you don't have a common transport
<daviddias>
like one is WebSockets and the other one is TCP
<hueso[m]>
Aren't there nodes listening on both?
<hueso[m]>
That could act as a brigde
<lindybrits>
@daviddias that is interesting. we have a node in the browser saying that it has connected to a node.js app (via ipfs.swarm.connect), yet the node.js app does not see the peer added to his swarm
mildred has quit [Quit: WeeChat 1.7]
spinus[m] has left #ipfs ["User left"]
<victorbjelkholm>
daviddias: missed your comment about wrtc, I'll write down some instructions on what's needed to get it working
<lindybrits>
daviddias: our connection string is of the form /ip4/some.ip.address/tcp/port/ipfs/ipfsKey when dialing from the browser to the node.js app
mildred has joined #ipfs
<daviddias>
victorbjelkholm: thank you <3
<daviddias>
lindybrits: Browsers don't really do TCP, do they? You need to dial to that node in using WebRTC or WebSockets
<daviddias>
show how to make a triangle of connections
JayCarpenter2 has joined #ipfs
ygrek has quit [Ping timeout: 258 seconds]
mildred1 has joined #ipfs
robattila256 has quit [Ping timeout: 240 seconds]
skeuomorf has quit [Ping timeout: 260 seconds]
ulrichard has quit [Remote host closed the connection]
<lindybrits>
daviddias: thanks, I'll have a look!
nunofmn has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<daviddias>
you bet:)
nunofmn has joined #ipfs
nunofmn has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<lindybrits>
daviddias: what does the "API" and the "Gateway" represent in the "Adresses" part of the js-ipfs config file?
<daviddias>
lindybrits: API, the http api, Gateway, the thing that enables you to access files through localhost:8080. Gateway is not implemented (there is a PR for it, not a priority though). This is the same in go-ipfs
<JayCarpenter2>
Lars and Jay are on the standup Zoom session. Anyone else joining?
<lgierth>
oh hey i'm not alone anymore
<lgierth>
daviddias: ^
wyatt has joined #ipfs
kus_ubuntui686 has joined #ipfs
kus_ubuntui686 has quit [Max SendQ exceeded]
sirdancealot has quit [Ping timeout: 240 seconds]
wmoh has quit [Read error: Connection timed out]
wmoh has joined #ipfs
wmoh has quit [Max SendQ exceeded]
palkeo has joined #ipfs
palkeo has joined #ipfs
palkeo has quit [Changing host]
wmoh has joined #ipfs
wmoh has quit [Max SendQ exceeded]
wmoh has joined #ipfs
palkeo has quit [Ping timeout: 264 seconds]
<wyatt>
Hey hsanjuan, I've been slowly making my way through the ipfs-cluster repo. I have a decent idea of things on a high level and am working through the details. Right now I am looking in ipfs-cluster-service/main.go looking at the modules from the repo that are called. Here are some basic things I'm currently curious about. What are peers communicating with each other? Presumably things like which CIDs should be fetched through IPFS and persisted and
<wyatt>
who is the cluster leader. What is the division of labor between ordinary peers and the cluster leader? What is the purpose of the leave flag / why is it possible for a node to be in the cluster but not running?
<wyatt>
I’m working my way through these questions right now but if you have any input that you think would help I’m all ears.
<hsanjuan>
hi wyatt: write down your questions somewhere as I don't have much time to answer now
<hsanjuan>
consensus/leader election is done by hashicorp.Raft which uses libp2p-raft (which both wraps it in the libp2p-consensus interface and adds a libp2p network transport)
<wyatt>
Thanks hsanjuan, I've got a list going. FYI my time is pretty flexible over the next week so if another time works better for you I'm happy to change plans.
<hsanjuan>
that involves a bunch of communication between peers. Outside of that, peers communicate metrics, or request information from other peers (i.e. api call for Peers() broadcasts an ID() request to every peer and puts it together), or forward Pin/Unpin requests to the leader. That is done using cluster's RPC API, which relies on go-libp2p-gorpc
<hsanjuan>
The --leave flag removes a peer from the consensus when it is shutdown. Otherwise, other peers will still count on it and the leader will start complaining because it cannot ping somone.
<hsanjuan>
So yeah, a node may be not running but still be part of the list of peers. If that happens, re-pinning will be triggered for the pins hosted in that node.
Jaco has joined #ipfs
<hsanjuan>
wyatt: open an issue in cluster, and write these questions. I can give more permanent answers there and we can collect them later
<lindybrits>
daviddias: Q -> we have an AWS-1 server that has a fresh jsipfs on it. when we run jsipfs daemon, it is listening on a couple of addresses. now, we want to connect from a completely different AWS-2 server to it. we have managed to webpack a node.js app on AWS-2 and via the browser on this AWS-2, we have used ipfs.swarm.connect to connect to AWS-1 (or at least it says so in the console). as we check on AWS-1, it did not add AWS-2's b
<lindybrits>
what can the problem be?
<hsanjuan>
wyatt: about time, this thursday works so no worries
<wyatt>
hsanjuan awesome, thank you
maxlath has quit [Ping timeout: 240 seconds]
<daviddias>
lindybrits: which transport is it using?
<daviddias>
My guess is that you are not adding a thing like WebRTC
<Jaco>
daviddias: I though from the diagram ons the transfer-files example, WebRTC is only required if you want communication between browsers?
<Jaco>
thought*
<lindybrits>
daviddias: should we connect /libp2p-webrtc-star/dns4/some.ip.address/wss
<victorbjelkholm>
Jaco: for between browser <> desktop and browser <> browser
<daviddias>
Jaco: it is one of the ways
<daviddias>
you can use WebSockets, but if you use websockets you have to dial on a websockets addr and not a TCP
<lindybrits>
daviddias: do we need the dns4 and wss? for instance /libp2p-webrtc-star/dns4/some.ip.address/wss
<lindybrits>
interesting... webrtc for browser <-> server then. on the server, should there be a signalling server running as well?
<lindybrits>
usually I did that with star-sig
<daviddias>
lindybrits: wanna open a Github issue so that we can track the discussion?
<daviddias>
It will be more useful to keep track of examples and points to parts of the code, making it easier to explain
<daviddias>
plus the whole community gains by also observing it
<lindybrits>
daviddias: that sounds good, Jaco: will take the lead on this now :) I need to go now, happy coding!
lindybrits has quit [Quit: Page closed]
<Jaco>
daviddias: I will try and just make sure everything is correct in my code tonight, and that I have tried everything. I will then open a Github issue.
<daviddias>
enjoy the rest of your day :)
matoro has quit [Ping timeout: 268 seconds]
<Jaco>
thanks, you too. Cheers
Jaco has left #ipfs [#ipfs]
JayCarpenter2 has quit [Quit: Page closed]
wmoh has quit [Ping timeout: 240 seconds]
espadrine has quit [Ping timeout: 256 seconds]
matoro has joined #ipfs
warbo has joined #ipfs
Encrypt has joined #ipfs
<warbo>
hello, I'm struggling to get two IPFS nodes on separate machines to communicate
<warbo>
they're not publically accessible, but I've set up SSH tunnels either way for their swarm ports (4001)
<warbo>
if I do "ipfs dht findprovs ..." I get the other node's ID
<warbo>
if I do "ipfs dht get ..." the command finishes without giving any output
<warbo>
if I do "ipfs get ..." it sits there seemingly not doing anything
<warbo>
(e.g. disk usage doesn't change, so it doesn't seem to be downloading anything)
<warbo>
any ideas for what I might try next?
wmoh has joined #ipfs
Boomerang has quit [Ping timeout: 260 seconds]
ZaZ has joined #ipfs
<whyrusleeping>
warbo: are the two nodes properly connected?
<whyrusleeping>
what does `ipfs swarm peers` show?
<warbo>
each one has a single entry for the others' identity
<warbo>
my SSH tunnels bind port 4001 to port 6001
<warbo>
one shows up as /ip4/127.0.0.1/tcp/6001/ipfs/....
<warbo>
the other shows up as /ip4/127.0.0.1/tcp/51542/ipfs/...
<warbo>
not sure where the 51542 came from
warbo has quit [Read error: Connection reset by peer]
Encrypt has quit [Quit: Quit]
ShalokShalom_ has joined #ipfs
espadrine_ has joined #ipfs
ShalokShalom has quit [Ping timeout: 260 seconds]
<whyrusleeping>
ephemeral dialer port
<whyrusleeping>
thats normal
<Kubuxu>
whyrusleeping: aren't we supposed to dial with the same port using SO_REUSEPORT?
matoro has quit [Ping timeout: 264 seconds]
matoro has joined #ipfs
tilgovi has joined #ipfs
skeuomorf has joined #ipfs
<hsanjuan>
whyrusleeping: do you happen to know if its possible to add folders via http api and how?
linux-modder[m] has joined #ipfs
<victorbjelkholm>
hsanjuan: I know it's possible, I'm not 100% on how the structure looks. Take a look here: https://github.com/ipfs/js-ipfs-api/issues/4 and follow linked issues
realisation has joined #ipfs
<whyrusleeping>
hsanjuan: you should be able to, `ipfs add -r emptydir`
<hsanjuan>
whyrusleeping: http api
<hsanjuan>
as "what would the curl command be"
<whyrusleeping>
that uses the http api, try wiresharking the request
<Kubuxu>
I personally have no idea, you might want to wireshark
<kevina>
Kubuxu: you _should_ be able to just remove the pins reported as broken, can you try "ipfs pin rm QmRKmNZqwSxcPs6HjEN3JiAKxjEJMByYpgr5ykwKuqNahP"
<kevina>
and if that works, try it on the others
<Kubuxu>
yes it worked
tilgovi has joined #ipfs
<Kubuxu>
at least it looks like it
<Kubuxu>
the QmRKmNZqwSxcPs6HjEN3JiAKxjEJMByYpgr5ykwKuqNahP was the only real problem
<Kubuxu>
the bad CID were migrated by why's tool
<kevina>
cool so is the "repo gc" running again?
<Kubuxu>
yup
<kevina>
could you by chance try it on the old repo, before why's tool?
<kevina>
Kubuxu: don't try to fix it though, just want to see the output
<Kubuxu>
I already unmounted it, I think it should remove pins no problem then.
<Kubuxu>
I will compress it and send it to you
<kevina>
it may be way to big for me to get, but let me know the size, I might want to do it on a VPS or EC2 instance
<Kubuxu>
it is about 4G
<Kubuxu>
not even 2.6G
<Kubuxu>
before compression
azahi has joined #ipfs
<Kubuxu>
but compressing/uncompressing will take a while
<kevina>
Kubuxu: what are you using to compress
<Kubuxu>
xz
<Kubuxu>
with tar
<Kubuxu>
it isn't handling it great :P 5min of CPU time already
<Kubuxu>
but I have to archive it either way
<kevina>
okay, just making sure it wasn't bzip2, that is really slow to uncompress
azahi has quit [Ping timeout: 240 seconds]
<kumavis>
new Silicon Valley season is all about decentralized tech and building a "new internet"
<whyrusleeping>
pierce: pinned? no idea. Replicated, you can use `ipfs dht findprovs`
<pierce>
whyrusleeping: ah, great, thanks
<whyrusleeping>
kevina: eh, i'm not worried about it
<kevina>
whyrusleeping: okay, but a nil slice takes as much space as a normal slice
<frood>
anyway, re changing keys, sloppy kademlia kinda relies on continuity of provider lists. can't really reorg the overlay without significant numbers ending up in the wrong neighborhood.
<whyrusleeping>
kevina: okay
<kevina>
whyrusleeping: one more question, should I even worry about loops in a dag
<kevina>
can it even happen?
<whyrusleeping>
kevina: the 'A' in dag stands for 'acyclic'
<whyrusleeping>
meaning, no loops
<kevina>
I know what a dag stands for.
<frood>
of course the DAG property is only guaranteed by the "merkle" qualifier.
ashark has joined #ipfs
<frood>
so it could happen if someone breaks sha256
<whyrusleeping>
frood: heh
<frood>
I wouldn't worry about that.
<kevina>
Is it theoretically possible to construct a loop, if you can get the figure out the magic hash....
<whyrusleeping>
I'm more worried about death by meteorite
<kevina>
okay :)
<whyrusleeping>
kevina: oh, if you can find a loop, please let me know
<Kubuxu>
yeah
<Kubuxu>
I would love to know that too
<whyrusleeping>
Theres a certain bitcoin transaction i'd love to cash
<Kubuxu>
or just overtake the network
<frood>
whyrusleeping: someone said that exact thing on the ipfs subreddit recently. made me laugh :)
<whyrusleeping>
i mean
<Kubuxu>
and get all the block rewards
<whyrusleeping>
frood: :D
<kevina>
just checking because I am paranoid
<whyrusleeping>
kevina: fair enough, lol
<frood>
if sha2 gets broken that bad, we have other, far more serious, problems.
<kevina>
and I know the probably is next to nothing but it is not 0
<whyrusleeping>
kevina: put a check in there, that if it finds a loop, email the hashes to why at ipfs.io
<whyrusleeping>
kevina: Don't worry, being paranoid is a great thing around here :)
<frood>
kevina: if you find a loop. don't notify anyone. just go break huge swatches of TLS traffic and steal tons of money.
<whyrusleeping>
frood: one collision does not an exploit make
<kevina>
so you are saying to make a loop you need to find a collision?
<Kubuxu>
we get this question all the time at cjdns channels
<Kubuxu>
23:23 <@Kubuxu>?lotto
<Kubuxu>
23:23 <@Entity>Kubuxu: "1 in 1,329,227,995,784,915,872,903,807,060,280,344,576 chance of generating the same IPv6. Feeling Lucky?"
<kevina>
can't random chance also do it, if someone tried enough combinations?
<Kubuxu>
and cjdns is 128bit, sha2 is 256
<frood>
whyrusleeping: a loop in the merkle-dag means that someone can make arbitrary datastructures and basically specify their hashes. that is suuuper exploitable
<frood>
valid merkle-dag node has additional constraints besides "1 collision"
dimitarvp has quit [Remote host closed the connection]
gmoro_ has quit [Remote host closed the connection]
<whyrusleeping>
kevina: 2^256 is a *very* large number
dimitarvp` has joined #ipfs
<whyrusleeping>
and yes, a loop implies a hash collision
gmoro_ has joined #ipfs
<whyrusleeping>
a hash collision of two valid objects in the same graph
<whyrusleeping>
which raises the improbabilities even more
<frood>
yeah, exactly. you get it :)
<kevina>
I guess I have not thought about it enough, I won't worry about a loop in my code
<whyrusleeping>
kevina: yeah, lots of other things would break if a loop was found, i think we're fine for now
Anchakor has quit [Ping timeout: 267 seconds]
<Kubuxu>
to show how enormous 2^128 is it is: 1,329,227,995,784,915,872,903,807,060,280,344,576 to power of 1,329,227,995,784,915,872,903,807,060,280,344,576
<jack>
Hello when trying to run nodejs after installing ipfs@0.23.0, I receive this error: node_modules/wrtc/build/wrtc/v0.0.61/Release/node-v51-linux-x64/wrtc.node'. Does anyone know how to solve this problem please?
dimitarvp` is now known as dimitarvp
ligi has joined #ipfs
matoro has quit [Ping timeout: 240 seconds]
tilgovi has joined #ipfs
anewuser has quit [Ping timeout: 240 seconds]
maxlath has joined #ipfs
matoro has joined #ipfs
jack has quit [Quit: Page closed]
aeternity has quit [Ping timeout: 260 seconds]
ashark has quit [Ping timeout: 260 seconds]
JayCarpenter has quit [Quit: Page closed]
google77 has quit [Quit: Lost terminal]
captain_morgan has quit [Remote host closed the connection]
captain_morgan has joined #ipfs
wrouesnel has joined #ipfs
espadrine_ has quit [Ping timeout: 240 seconds]
cxl000 has quit [Quit: Leaving]
matoro has quit [Ping timeout: 260 seconds]
stoopkid has joined #ipfs
amdi_ has joined #ipfs
ianopolous has joined #ipfs
rcat_ has quit [Quit: leaving]
matoro has joined #ipfs
Boomerang has joined #ipfs
Boomerang has quit [Ping timeout: 260 seconds]
reit has quit [Ping timeout: 268 seconds]
maxlath has quit [Quit: maxlath]
arkangel has joined #ipfs
nunofmn has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
obarb is now known as brabo
matoro has quit [Ping timeout: 240 seconds]
matoro has joined #ipfs
Guest193196[m] has joined #ipfs
droptables[m] has joined #ipfs
chris613 has joined #ipfs
mildred has quit [Read error: Connection reset by peer]
mildred1 has quit [Read error: Connection reset by peer]