<kumavis[m]>
why would the format of the lookup hash depend on whether or not the daemon is running
<daviddias>
kumavis[m]: I believe that is just leftovers of the time where js-ipfs expected the multihash in Buffer, while js-ipfs-api expected it to be a string
<daviddias>
when daemon is running it would use js-ipfs-api (http client lib)
<kumavis[m]>
ok i see
<daviddias>
and when the daemon is not running it uses js-ipfs
<daviddias>
but that doesn't apply anymore
<kumavis[m]>
cool
<daviddias>
js-ipfs-api in the beginning was just a 'go-ipfs cli copy', now it has the full js-ipfs API
<kumavis[m]>
daviddias can you publish a new version of these two now that they've merged the new cids
<kumavis[m]>
ipfs/js-ipfs-bitswap
<kumavis[m]>
ipfs/js-ipfs-block-service
<kumavis[m]>
I have a PR for js-ipfs#feat/dag-api in the works : )
<daviddias>
sweeet! :D
<daviddias>
on what?
<daviddias>
the cli + http-api?
<kumavis[m]>
`jsipfs block put --format eth-block <eth-block file>`
<kumavis[m]>
havent looked at the http api yet
ebel has quit [Ping timeout: 245 seconds]
ebel has joined #ipfs
everywhen has quit [Ping timeout: 250 seconds]
Mateon1 has quit [Read error: Connection reset by peer]
Mateon1 has joined #ipfs
chungy has quit [Ping timeout: 258 seconds]
chungy has joined #ipfs
soloojos has quit [Remote host closed the connection]
<daviddias>
kumavis[m]: just got both released (was hitting some issues with aegir 9.2.0, had to revert to 9.1.2)
<kumavis[m]>
thanks/
ylp has quit [Ping timeout: 260 seconds]
ylp has joined #ipfs
Alaura has joined #ipfs
slothbag has quit [Quit: Leaving.]
ploop has quit [Ping timeout: 250 seconds]
mguentner3 has quit [Quit: WeeChat 1.6]
ploop has joined #ipfs
ploop has quit [Remote host closed the connection]
mguentner has joined #ipfs
Alaura has quit [Remote host closed the connection]
Alaura has joined #ipfs
ploop has joined #ipfs
Alaura has quit [Remote host closed the connection]
Alaura has joined #ipfs
Alaura has quit [Remote host closed the connection]
Alaura has joined #ipfs
ygrek has quit [Ping timeout: 260 seconds]
ylp has quit [Ping timeout: 260 seconds]
ylp has joined #ipfs
chris613 has quit [Quit: Leaving.]
jchevalay has quit [Ping timeout: 260 seconds]
lkcl has joined #ipfs
<haad>
daviddias: pubsub looking good! will (finally) work on it today.
ianopolous has quit [Remote host closed the connection]
<pocketss>
jchevalay: Hey I have seen the code regarding the libp2p, swarm and bitswap. Over there, each node defines a handler at a set protocol, and for transfering data, the node does dialByPeerInfo to setup a link with the other peer
<pocketss>
but if data is sent from A --> B , then does B dial to A or A dials to B ?
<jchevalay>
Of what I understood you receive the data that of the resources that you consult
<pocketss>
yes but say A find through DHT that the file belong to a hash that he wants resides with peer B
<pocketss>
then does A request peer B to dial to him and send data via pull-stream or A dials to peer B and expects B to send data?
<pocketss>
dignifiedquire: ^^ for the js-ipfs-bitswap lib
_whitelogger has joined #ipfs
m0ns00n_ has quit [Quit: quit]
s_kunk has quit [Ping timeout: 246 seconds]
ianopolous has joined #ipfs
O47m341 has joined #ipfs
s_kunk has joined #ipfs
ralphtheninja has quit [Ping timeout: 246 seconds]
briarrose has quit [Quit: bye bye]
screensaver has joined #ipfs
briarrose has joined #ipfs
briarrose has joined #ipfs
briarrose has quit [Changing host]
ralphtheninja has joined #ipfs
ralphtheninja has quit [Ping timeout: 268 seconds]
O47m341 has quit [Ping timeout: 245 seconds]
Oatmeal has quit [Ping timeout: 245 seconds]
nunofmn has joined #ipfs
clavi has quit [Ping timeout: 246 seconds]
clavi has joined #ipfs
robattila256 has quit [Ping timeout: 245 seconds]
leob_ has joined #ipfs
leob has quit [Ping timeout: 248 seconds]
nunofmn has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
nunofmn has joined #ipfs
<victorbjelkholm>
dignifiedquire: what you think about passing a location of file to load when documentation loads? That way, we could inject a global object for experimenting in the console when having the docs open
<victorbjelkholm>
I somehow expect the object in the docs to be available globally while reading the docs
<dignifiedquire>
not sure, that would increase the page load drastically, I'm currently pretty happy that the docs don't ship any javascript
<victorbjelkholm>
well, guess it depend on the lib
<victorbjelkholm>
but yeah, that's true, no JS and very lightweight in general, multiaddr full size ends up being 60kb which in todays landscape is a huge improvement
<victorbjelkholm>
great work on that!
<victorbjelkholm>
dignifiedquire: which one of the modules with docs that you're working on is the most complete? Guessing it's peer-id. Want to take a look to see I'm not missing anything
<dignifiedquire>
yeah pretty happy with peer-id
espadrine has joined #ipfs
O47m341 has joined #ipfs
nunofmn has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
Oatmeal has joined #ipfs
nunofmn has joined #ipfs
Alaura has quit [Quit: Leaving]
nunofmn has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<victorbjelkholm>
dignifiedquire: I see peer-id show the module name and the version. I don't get that, at least locally. I'm missing something?
<victorbjelkholm>
in the top left
nunofmn has joined #ipfs
<dignifiedquire>
are you using aegir-docs?
<victorbjelkholm>
no, using the documentation cmd directly, so I get the automatic reload
<dignifiedquire>
what's the command you are running?
<victorbjelkholm>
if I add --name I get the module name at least
<victorbjelkholm>
ah, found --project-version as well, so nvm dignifiedquire
<dignifiedquire>
yeah version and name need to be passed in manually, aegir-docs does that for you automatically
<victorbjelkholm>
aight, thanks :)
slothbag has joined #ipfs
<pocketss>
Hey I have seen the code regarding the libp2p, swarm and bitswap. Over there, each node defines a handler at a set protocol, and for transfering data, the node does dialByPeerInfo to setup a link with the other peer. but if data is sent from A --> B , then does B dial to A or A dials to B ? https://github.com/ipfs/js-ipfs-bitswap/blob/master/src/network/index.js#L100-L120
<pocketss>
dignifiedquire: ^^ help please :)
<dignifiedquire>
pocketss: neither and both
<dignifiedquire>
the actual connection establishing happens before, and can be initiated by either party
<dignifiedquire>
after that, the party that wants to send a message tells libp2p give me a connection to this other party, and sends the message
<dignifiedquire>
bitswap caches those connections, to not recreate them all the time as it assumes that we send many messages to everyone we start talking to
<dignifiedquire>
but in js-ipfs this is currently very simplified, because we don't have a dht to find new peers who have the content we want
<dignifiedquire>
so we only talk to those peers we are already connected to
<pocketss>
So If A finds out that B has a certain file he wants, he will get his peerInfo (from bootstrap nodes?) and ask libp2p to dial to B with dialByPeerInfo(err, conn) ? Is this conn then cached?
<pocketss>
dignifiedquire: does this mean by A dials to B AND B dials to A as well on the default protocol?
<victorbjelkholm>
dignifiedquire: yay or nay: include npm link next to the github icon in docs?
<dignifiedquire>
nay
<dignifiedquire>
I never find npm links helpful, it's just the same thing I get on gh but less readable
<dignifiedquire>
pocketss: bitswap sends "wantlists" to all connected peers telling them which data it wants, and the peers then send the requested data
<pocketss>
Yes I saw that part in core/components in the js-ipfs repo, but this data sent is done through libp2p aka dialing and handling connections. From What i seen, is that when libp2p is called, it handles a connection on a default hardcoded protocol.
<pocketss>
a connection can pass data two ways via pull streams, so once the want list is broadcasted, the peer gets details of who has the information, but then does this peer dial to them and request for it, or those peers in the wantlist dial to this request peer and say " heres the data you wanted"
nunofmn has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<pocketss>
dignifiedquire: ^
voker57 has quit [Ping timeout: 250 seconds]
voker57 has joined #ipfs
nunofmn has joined #ipfs
leob_ has quit [Ping timeout: 250 seconds]
maxlath has joined #ipfs
ralphtheninja has joined #ipfs
<_mak>
the correct way to check if a file is pinned locally is: ipfs pin ls hash ?
<_mak>
this command works if I have the file pinned but it hangs if I don't
<_mak>
using ipfs pin ls | grep hash is crazy slow for me
<haad>
daviddias: made good progress with js-pubsub. left a comment and will leave it for you to continue now. almost there! :)
leob has joined #ipfs
pocketss has quit [Quit: ChatZilla 0.9.93 [Firefox 50.0.2/20161130094234]]
nunofmn has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
maxlath has quit [Ping timeout: 245 seconds]
herzmeister has quit [Quit: Leaving]
herzmeister has joined #ipfs
domanic has joined #ipfs
PseudoNoob has joined #ipfs
PseudoNoob has quit [Read error: Connection reset by peer]
domanic has quit [Quit: Konversation terminated!]
domanic has joined #ipfs
reit has joined #ipfs
hsanjuan has quit []
hsanjuan has joined #ipfs
horrified has quit [Quit: quit]
horrified has joined #ipfs
slothbag has left #ipfs [#ipfs]
oleavr has quit []
oleavr has joined #ipfs
ralphtheninja has quit [Ping timeout: 245 seconds]
sickill has quit []
sickill has joined #ipfs
poga has quit []
poga has joined #ipfs
ploop has quit [Ping timeout: 250 seconds]
ralphtheninja has joined #ipfs
ploop has joined #ipfs
maxlath has joined #ipfs
strauberry has quit []
strauberry has joined #ipfs
nunofmn has joined #ipfs
nunofmn has quit [Client Quit]
muvlon_ has joined #ipfs
muvlon_ has quit [Remote host closed the connection]
<pinbot>
now pinning /ipfs/Qmf2AKhyKodLq2nEq6g4rgPV5SGAvAYqKhRfJxRt5syzEW
<pinbot>
[host 6] failed to pin /ipfs/Qmf2AKhyKodLq2nEq6g4rgPV5SGAvAYqKhRfJxRt5syzEW: cannot store pin state: write /data/ipfs/datastore/376779.log: no space left on device
nunofmn has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
robmyers has quit []
robmyers has joined #ipfs
<Mateon1>
For some reason the daemon keeps OOM crashing on my VPS
pcre has joined #ipfs
<Mateon1>
It's 64 bit, but I have 1 gigabyte of ram, that IPFS keeps fully eating up
<haad>
:/
<haad>
cc whyrusleeping ^
<haad>
Mateon1: which version? master?
<Mateon1>
No, 0.4.4 release from dist.ipfs.io
<Mateon1>
I can crosscompile a master, though
<Mateon1>
Upon starting the daemon it instantly uses 350MB of memory, I'll try a newer version
bastianilso____ has quit [Quit: bastianilso____]
brodo has joined #ipfs
<Mateon1>
Actually, nevermind, that's virtual memory, I misread the column
<Mateon1>
It's "only" 77 megs
<Kubuxu>
when you think about the fact that binary itself is 20+ megs
<pcre>
Calling »ipfs bitswap stat«. What does »partners« in the output mean?
<Kubuxu>
24M to be precise
jhiesey has quit []
jhiesey has joined #ipfs
<Kubuxu>
pcre: peers we comunicate with trying to exchange blocks
sugarpuff has quit []
<xelra>
I'm excited to see that work on ipfs-cluster has been resumed. :)
sugarpuff has joined #ipfs
<xelra>
I'm so eager to deploy it in a first test-environment.
domanic has quit [Read error: Connection reset by peer]
domanic has joined #ipfs
nunofmn has joined #ipfs
<pcre>
xelra: What does ipfs-cluster do ?
<xelra>
pcre: It makes a closed ipfs network that you don't share with everyone. It's useful for making family-wide or company-wide storage.
<pcre>
ok
nunofmn has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<xelra>
For private stuff.
<Kubuxu>
xelra: it isn't true, what you are describing is Private Network
<Kubuxu>
feature we are working on currently too
<Kubuxu>
ipfs-cluster is like load balancer and RAID for ipfs nodes.
<Kubuxu>
more like RAID
<xelra>
No auth?
<pcre>
Private network. But i assume it is not tied to some local area network. So a private network in ipfs is more like a sub of the whole mesh. Isnt it?
<Kubuxu>
auth for the management layer not communication with rest of the IPFS world
<pcre>
ok
<Kubuxu>
pcre: it was re: xelra, responding to you now
<Kubuxu>
pcre: it will be tired to symmetric pre-shared key generated by you in the first version. Your node will only connect with nodes sharing the same PSK
nunofmn has joined #ipfs
anderspree_ has quit []
<pcre>
So ipfs-cluster is more like bittorrent now. shareing upload.?
anderspree_ has joined #ipfs
<victorbjelkholm>
pcre: ipfs-cluster is in short words, a tool you can use to distribute pinning across many ipfs instances, acting as one
<Kubuxu>
no, ipfs already shares upload, it is like a cluster: let's say you want to store a file with redundancy, you can add it to a cluster that will store it across 3 different nodes in this claster
<pinbot>
now pinning /ipfs/QmaMMp6Kk63JscXopbLPzWU7B9vDFPp3zb8Ta89PFCGx3w
<pinbot>
[host 6] failed to pin /ipfs/QmaMMp6Kk63JscXopbLPzWU7B9vDFPp3zb8Ta89PFCGx3w: cannot store pin state: write /data/ipfs/datastore/376779.log: no space left on device
<haad>
r0kk3rz: yeah! it's not quite working as expected yet but almost!
<muvlon>
what is it though?
pfrazee has joined #ipfs
nunofmn has joined #ipfs
<r0kk3rz>
haad: ooo very cool :)
pcre has quit [Ping timeout: 256 seconds]
pcre has joined #ipfs
<jchevalay>
does exist an page to enum somes projects around ipfs looks like "ecosystem" or another
<haad>
muvlon: it's a serverless, distributed chat application on IPFS, works both natively and in the browser (using javascript implementation of IPFS)
<muvlon>
ah
<muvlon>
any rooms on there i can join?
<haad>
try #ipfs
<muvlon>
i went there, was empty
<muvlon>
imma try it again, hold on
<haad>
I'm prolly only one there and I'm not sure we'll be able to connect (that's a WIP version)
<nicolagreco>
IPLD meeting was meant to happen in 10 minutes, I will be around - however given the lack of content in the agenda, unless someone has something to raise, we were planning to postpone to next week
sunnymilk has joined #ipfs
<nicolagreco>
However, if you were planning on attending and/or have points to raise, please let me know
<richardlitt>
nicolagreco: Sounds good to me.
<richardlitt>
I don't have any points.
<richardlitt>
I commented on the issues assigned to me in ipfs/website; take a look at the copy, it should be good?
<nicolagreco>
Next week, I will try to ask ahead of time so that some of us can plan their time ahead!
<richardlitt>
++
<daviddias>
richardlitt: thank you for the checklist, it contains all the info I would like to know before moving ahead to discourse with confidence
<richardlitt>
All of it?
<richardlitt>
O.o
<richardlitt>
Answer the sentences with "?"s in a comment. :D
<richardlitt>
and no problem. Just trying to get this going
kenshyx has quit [Quit: Leaving]
ploop has quit [Remote host closed the connection]
tilgovi has quit [Ping timeout: 240 seconds]
josselin has joined #ipfs
f4 has joined #ipfs
ploop has joined #ipfs
<kumavis[m]>
just noticed we dont have a multicodec code for git objects
espadrine has quit [Ping timeout: 244 seconds]
bwerthmann has joined #ipfs
lkcl has joined #ipfs
<hsanjuan>
anyone knows how multiple ipfs daemons behave behind the same nat?
s_kunk has quit [Ping timeout: 248 seconds]
<Kubuxu>
hsanjuan: should be just fine
eibriel has joined #ipfs
s_kunk has joined #ipfs
s_kunk has joined #ipfs
s_kunk has quit [Changing host]
Aranjedeath has joined #ipfs
s_kunk has quit [Quit: Read error: Connection reset by beer]
JustinDrake has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
s_kunk has joined #ipfs
anonymuse has quit [Remote host closed the connection]
malwrar has joined #ipfs
maxlath has quit [Quit: maxlath]
PseudoNoob has quit [Remote host closed the connection]
<pcre>
This does not look to bad. But i guess he started it and underestimated the efford.
<cholcombe>
probably
chris613 has joined #ipfs
<dignifiedquire>
yeah I want ipfs in pure rust at some point, there are some small modules (multiaddr, multihash) that I started to write but it's a large undertaking
<dignifiedquire>
pcre: yes it's wrapper around the http api from go land
<cholcombe>
i see
ianopolous has quit [Ping timeout: 246 seconds]
<cholcombe>
well it's a start. you can remove the pieces as you implement them in pure rust
<cholcombe>
dignifiedquire, how many lines of code is the go implementation? that might give us an idea of how crazy it would be to write
<pcre>
ok
<dignifiedquire>
but I hadn't seen the crate above which is pretty cool
atrapado_ has quit [Ping timeout: 245 seconds]
<dignifiedquire>
cholcombe: that's a good question, I'm not sure cc whyrusleeping, have you counted at some point?
<dignifiedquire>
it's distributed over many many modules, so not straightfoward to answer
<cholcombe>
i see
<dignifiedquire>
but that makes it also easier to implement
<dignifiedquire>
you can start with the tooling for the low level network in libp2p
<dignifiedquire>
and have useful things without needing to implement the whole thing in one go
<cholcombe>
right
<sunnymilk>
hi, is there a way to advertise an ip address that you cant bind to?
<sunnymilk>
i am behind a nat but when i put my external ip in the swarm addresses section it says it cant bind to it and fails
f4 has quit [Ping timeout: 244 seconds]
<Kubuxu>
cholcombe: over all out dependencies dependencies 700k, over dependencies managed by us more than 200k.
<Kubuxu>
over all of our dependencies 700k, over ....
<cholcombe>
Kubuxu, holy crap haha
f4 has joined #ipfs
<cholcombe>
dignifiedquire, ok yes that's a monster undertaking without some kinda automated code translation helpers
<dignifiedquire>
well we are working on the ipfs version which is pretty usable already and fully ready, which took us about 1.5 years then i think
<dignifiedquire>
*fully ready in the beginning of next year
<dignifiedquire>
so it is doable
<cholcombe>
yeah
<dignifiedquire>
if you have a couple of people who are motivated :)
<cholcombe>
at 500 lines a day for a year possibly haha
<dignifiedquire>
if I have the write macros I can write more than that :)
<dignifiedquire>
*right macros
<dignifiedquire>
-.-
<cholcombe>
indeed
<cholcombe>
it's a low ball estimate
<Kubuxu>
sunnymilk: there isn't a way to do that yet, it might be picked up by ipfs automagically if everything goes well.
atrapado_ has joined #ipfs
<sunnymilk>
i see, if i connect to lots of peers (outside of the bootstrap peers) does that mean everything is working okay and i dont have to worry about it?
<Kubuxu>
in 90% of cases, yes
Malus has quit [Remote host closed the connection]
<sunnymilk>
ah, well thank you
Pent3 has quit [Read error: Connection reset by peer]
<daviddias>
lgierth: did you had the chance to look at the dokku thing?
fibry has joined #ipfs
nunofmn has joined #ipfs
<deltab>
sunnymilk: NATs generally don't allow you to connect to their external address from inside
<sunnymilk>
well yes but theres no reason that ipfs has to tell other nodes what address it binds to
<Kubuxu>
deltab: they can be 1. punched (apart from symmetric NAT) 2. upnp is a thing
<deltab>
that's connecting from outside; I mean connections from inside (hairpinning)
<deltab>
but I'm probably misinterpreting what sunnymilk meant
<daviddias>
thanks Kubuxu , good point :)
domanic has joined #ipfs
PseudoNoob has joined #ipfs
<whyrusleeping>
cholcombe: I could implement a basic 'ipfs node' that just does the network communication stuff, without many features, in probably 5-10kloc
<whyrusleeping>
we have a fair amount of UI code (despite being a CLI app, lol) and a large amount of networking stuff that wouldnt be required for a 'minimal' impl
<Kubuxu>
I think you would need more if you include multiX
<Kubuxu>
and I count w/o DHT, bitswap
<cholcombe>
whyrusleeping, yeah i think if you got it to just a basic little piece that the rust community would start extending it
<whyrusleeping>
cholcombe: do you have a list of whats already implemented in rust?
<Kubuxu>
something worth considering is that you need most of the multiformats done to proceed
<whyrusleeping>
implementing the multiformats is *very* easy though
<whyrusleeping>
so don't let that scare anyone off
<Kubuxu>
yeah, and it is starting point
<cholcombe>
whyrusleeping, i don't know what's already out there :-/
Sophrosyne has quit [Read error: Connection reset by peer]
nunofmn has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<whyrusleeping>
cholcombe: If youre interested in pushing a rust impl forward, i'd definitely go and make a list of currently implemented ipfs-rust stuff
<Kubuxu>
It sounded wrong, what I meant is that we have this simple protocols that now are one protocol variations
<whyrusleeping>
then we can help guide the 'next' pieces
cemerick has quit [Ping timeout: 258 seconds]
<cholcombe>
whyrusleeping, yeah that's a good idea
<Kubuxu>
that are quite well spec out
<cholcombe>
i like ipfs but i donno if i'm that dedicated yet to go off and code thousands of lines haha
ianopolous__ has joined #ipfs
<whyrusleeping>
cholcombe: hah, you can maybe write a hundred lines?
nunofmn has joined #ipfs
<cholcombe>
haha
<cholcombe>
sure
<Kubuxu>
ipfs is nice because you don't need it whole to see results
<dignifiedquire>
and these things are integrated with https://github.com/carllerche/mio which is going to be important to get performant networking going
<daviddias>
whyrusleeping: would you be down to have some bootstrap nodes with websockets addrs?
domanic has quit [Ping timeout: 260 seconds]
domanic has joined #ipfs
<lgierth>
daviddias: it's on my list already
<daviddias>
lgierth: <3
<lgierth>
bootstrap.libp2p.io/ws i guess?
nunofmn has joined #ipfs
cholcombe has quit [Ping timeout: 258 seconds]
<whyrusleeping>
anyone know how to get the number of open issues assigned to you *WITHOUT* also getting all of the issues?
<whyrusleeping>
api.github.com/issues returns all of the issues, but for something i'm running to get a status bar icon, thats a bit too much bandwidth usage
<whyrusleeping>
pinlock should be held when youre expecting to pin something, and you don't want the blocks youre creating to be gc'ed before you actually get to pin
<whyrusleeping>
but yeah... those functions seem a bit weird. we should hold the pinlock until we pin, or explicitly decide not to
A124 has joined #ipfs
<whyrusleeping>
not hold it until we return, and then decide to pin or not after dropping it
<lgierth>
it actually only ever unlocks it. coreunix/add.go locks nothing
<lgierth>
is that a way of saying "i'm definitely not gonna pin"?
anonymuse has quit [Remote host closed the connection]
<whyrusleeping>
^ this takes the lock, and defers unlocking it
<whyrusleeping>
defers resolve the function to be called on the line the defer is invoked
<lgierth>
and that lock means, i might want to pin, and in the meantime gc shouldn't run?
<whyrusleeping>
yeap
<lgierth>
yeah i know about defers, just didn't get that PinLock() actually takes the lock
<whyrusleeping>
as you guys already know, i'm atrocious at naming things
<lgierth>
:):)
<lgierth>
ok, but anyhow i think the way that lock is used... i don't know
<lgierth>
the only safe way would be to create the adder (claiming the lock), add your stuff, then run Finalize() to release it
<lgierth>
there can be multiple add calls and if the lock isn't held between those, we don't need to have the lock at all
<lgierth>
same for the gap between adds and the actual PinRoot() call
<whyrusleeping>
yeah...
<whyrusleeping>
the intention was to take the lock, add your things, maybe pin, drop the lock
<whyrusleeping>
a lot of complexity got added in when we wanted to be able to run a gc during a long add
<whyrusleeping>
so the add calls can temporarily drop the lock and let a gc run
circ-user-33m7u has joined #ipfs
nunofmn has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
cubemonkey has quit [Remote host closed the connection]
everywhen has joined #ipfs
<lgierth>
ok then that needs some cleaning up / refactoring
<lgierth>
not gonna get in that rabbit hole now :)
<whyrusleeping>
yeah
<whyrusleeping>
lol
<lgierth>
just gonna fix that comment
infinity0 has quit [Remote host closed the connection]
chungy has joined #ipfs
infinity0 has joined #ipfs
circ-user-33m7u is now known as vijayee
<vijayee>
is there anyone who knows orbit
<vijayee>
or its current state
infinity0 has quit [Remote host closed the connection]
infinity0 has joined #ipfs
cypher has quit [Read error: Connection reset by peer]
vijayee has quit [Ping timeout: 260 seconds]
<victorbjelkholm>
daviddias, saw you mentioned dokku, you've tried it in the past? Last time I tried it (around ~0.5) it tended to be unstable and not really ready for anything real that you needed to be up for most of the time, but would be happy if things changed
cubemonkey has joined #ipfs
everywhen has quit [Ping timeout: 265 seconds]
cypher has joined #ipfs
ulrichard has quit [Quit: Ex-Chat]
palkeo has quit [Ping timeout: 246 seconds]
Tsutsukakushi has quit [Ping timeout: 240 seconds]
Tsutsukakushi has joined #ipfs
domanic has quit [Read error: Connection reset by peer]
domanic has joined #ipfs
dominic__ has joined #ipfs
<Kubuxu>
should js-ipfs-api name.publish test take a minute?
cubemonkey has quit [Ping timeout: 252 seconds]
domanic has quit [Ping timeout: 268 seconds]
SuperPhly has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
wak-work has quit [Ping timeout: 258 seconds]
<whyrusleeping>
Is anyone using 'native' docker on osx?
fibry has quit [Remote host closed the connection]