<Nox[Matrix]>
Its time to update the Android client with pubsub too
llama has joined #ipfs
cxl000 has joined #ipfs
<xelra>
llama: There's going to be a new whitepaper/blog post about how filecoin is going to work.
<xelra>
Some time in the near future from what I heard.
llama has quit [Client Quit]
Hasperov has quit [Ping timeout: 240 seconds]
maxlath has quit [Ping timeout: 240 seconds]
tglman has quit [Quit: WeeChat 1.8]
tglman has joined #ipfs
nunofmn has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
nunofmn has joined #ipfs
Hasperov has joined #ipfs
ulrichard has quit [Ping timeout: 255 seconds]
nunofmn has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
Hasperov has quit [Ping timeout: 268 seconds]
ulrichard has joined #ipfs
Hasperov has joined #ipfs
Hasperov has quit [Read error: Connection reset by peer]
mahloun has joined #ipfs
mahloun_ has joined #ipfs
ilyaigpetrov has joined #ipfs
mahloun_ has quit [Ping timeout: 240 seconds]
fredthomsen has quit [Remote host closed the connection]
mahloun has quit [Ping timeout: 276 seconds]
nunofmn has joined #ipfs
notafile[m] is now known as notafile[m]1
notafile[m]1 is now known as notafile[m]2
notafile[m]2 is now known as notafil4
nunofmn has quit [Ping timeout: 240 seconds]
notafil4 is now known as notafil6
ploop has quit [Remote host closed the connection]
antiall has joined #ipfs
<antiall>
ciao
<antiall>
!list
notafil6 is now known as notafile[m]
antiall has quit [Client Quit]
charley has joined #ipfs
charley has quit [Ping timeout: 258 seconds]
muvlon has joined #ipfs
bedeho has quit [Remote host closed the connection]
asyncsec has joined #ipfs
bedeho has joined #ipfs
muvlon has quit [Remote host closed the connection]
Caterpillar2 has joined #ipfs
Caterpillar has quit [Ping timeout: 255 seconds]
tantalum has joined #ipfs
DJHenk has left #ipfs [#ipfs]
taaem has joined #ipfs
libman has joined #ipfs
muvlon has joined #ipfs
MDude has joined #ipfs
bedeho has quit [Remote host closed the connection]
ashark has joined #ipfs
rudolph has joined #ipfs
shizy has joined #ipfs
bedeho has joined #ipfs
Aranjedeath has joined #ipfs
mahloun has joined #ipfs
mahloun has quit [Ping timeout: 260 seconds]
charley has joined #ipfs
anewuser has joined #ipfs
charley has quit [Ping timeout: 276 seconds]
arpu has quit [Ping timeout: 240 seconds]
taaem has quit [Ping timeout: 260 seconds]
Miezel has quit [Quit: This computer has gone to sleep]
pat36 has quit []
taaem has joined #ipfs
Miezel has joined #ipfs
corvinux has joined #ipfs
Miezel has quit [Quit: This computer has gone to sleep]
ylp1 has quit [Quit: Leaving.]
taaem has quit [Ping timeout: 260 seconds]
jaboja has joined #ipfs
brodo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
ulrichard has quit [Remote host closed the connection]
bedeho has quit [Remote host closed the connection]
screensaver has quit [Ping timeout: 240 seconds]
Miezel has joined #ipfs
bedeho has joined #ipfs
<tomatopeel>
whyrusleeping: Kubuxu: can I ask what your development workflow involves in terms of Go workspacing, cloning/getting/building, forking, branching, PR'ing etc. in light of the gx stuff?
<tomatopeel>
e.g. you've found exactly where a bug is and the code that needs changing, it's in a dependency package
<tomatopeel>
I guess I could cd to $GOPATH/src/gx/QmWhatEvz/, rm -rf the package dir, and then git clone it from github into this dir?
screensaver has joined #ipfs
Miezel has quit [Quit: This computer has gone to sleep]
jhand has joined #ipfs
corvinux has quit [Ping timeout: 255 seconds]
pat36 has joined #ipfs
corvinux has joined #ipfs
alextes has left #ipfs ["WeeChat 1.8"]
borkDoy has joined #ipfs
<whyrusleeping>
tomatopeel: we're working on making that workflow easier
<whyrusleeping>
lgierth has been pushing some new subcommands for gx that help out a lot
<whyrusleeping>
generally, my workflow for that is to reproduce the issue in a unit test in the dependency
<whyrusleeping>
and then fix it in the dependency
<whyrusleeping>
then the part we're writing better tooling for; i sometimes just copy the updated code from the dependency over the top of $GOPATH/src/gx/ipfs/QmWhatever/
llama has joined #ipfs
droman has joined #ipfs
<tomatopeel>
whyrusleeping: cool, thanks. re: the unit testing stuff, what's your workflow for that? I guess for unit tests you can just operate on a single repository in a vacuum, and you do that with regular git toolage and go test?
<whyrusleeping>
i just work in a single repo for unit tests
Miezel has joined #ipfs
<whyrusleeping>
yeah, just cd $GOPATH/src/$REPO and make my changes
<whyrusleeping>
then go test
<whyrusleeping>
lgierth: whats the state of gx dev-copy?
<M-hash>
Fwiw I usually connect projects with git submodules and its roughly the same story
llama has quit [Client Quit]
<tomatopeel>
M-hash: isn't that horrible for ipfs because of upstream breakage though?
Miezel has quit [Client Quit]
<tomatopeel>
like dep version pinning i mean
<M-hash>
Tests really really need to be in the relevant repos and acceptance tests fin downstreams are always going to be friction you'd rather avoid tldr
<tomatopeel>
oh you can manage that with git submodules I guess
<M-hash>
Yeah very differential tools, same semantics in a way because they're both hash-centric
<lgierth>
whyrusleeping: i dunno i never touched dev-copy
<lgierth>
whyrusleeping: i made gx-go link instead
<whyrusleeping>
oooooh, gotcha
<lgierth>
the vendor/ stuff doesn't work out
<whyrusleeping>
lgierth: would that help tomatopeel here?
<lgierth>
let me catch up
<whyrusleeping>
M-hash: yeah... i really don't enjoy git submodules
<lgierth>
so, we generally work in the $GOPATH/src directory tree. some people symlink there actual work directories into that (like me) but that's just aesthetics
<tomatopeel>
I did wonder earlier if symlinking might be the recipe here haha
<lgierth>
then with gx-go link, if you wanna make changes in some package in the dependency tree, you do gx-go link
<lgierth>
and it'll link gx/ipfs/QmFoo to github.com/user/go-foo
<tomatopeel>
cool, sounds like the ticket, I'll check it out, thanks
<lgierth>
according to the gx.dvcsimport field it finds in QmFoo's package.json
<lgierth>
note that gx-go link is in PR phase right now :)
<lgierth>
and, the links are of course "global"
<whyrusleeping>
lgierth: what was the actual issue around putting stuff in vendor?
<lgierth>
it only applies to packages under that directory
<lgierth>
i.e. stuff in go-ipfs/vendor won't affect any of the dependencies ebcause they're in a different filesystem tree branch
droman1 has joined #ipfs
<lgierth>
and that just sucks for this purpose :)
<whyrusleeping>
so if go-ipfs/vendor/X depends on Y, it won't look at go-ipfs/vendor/Y, it will look at $GOPATH/src/Y
<whyrusleeping>
?
Hasperov has joined #ipfs
<lgierth>
mhh no
<lgierth>
let's say A depends on B and C, and B also depends on C
<M-hash>
Wouldn't the hashy import paths make that... Behave complexly?
<lgierth>
you put C in A/vendor
<lgierth>
then B won't see that and will depend on the regular C
droman has quit [Ping timeout: 276 seconds]
<lgierth>
you'd have to vendor C in every single spot that depends on it
<whyrusleeping>
M-hash: in that scenario, we undo the hashy paths
<whyrusleeping>
lgierth: dev-copy put *everything* in vendor
<lgierth>
i know
<lgierth>
doesn'
<lgierth>
t change anything
<lgierth>
oh
<lgierth>
got it
<lgierth>
mh.
reit has quit [Ping timeout: 240 seconds]
brodo has joined #ipfs
<lgierth>
dunno
<lgierth>
gx-go link works offline :P
<whyrusleeping>
lol
jaboja has quit [Ping timeout: 258 seconds]
<lgierth>
but at least i understand now why it copies everything
<lgierth>
hah!
bedeho has quit [Remote host closed the connection]
<lgierth>
git st
<whyrusleeping>
On branch master
<whyrusleeping>
Your branch is up-to-date with 'origin/master'.
<whyrusleeping>
nothing to commit, working tree clean
<lgierth>
more like fatal: Not a git repository :)
ralphtheninja has quit [Remote host closed the connection]
<lgierth>
on the other hand, dev-copy could work offline as well if it cloned from $gopath
<lgierth>
would be much faster too
<tomatopeel>
lgierth: do you know if there's a git command I can run to just grab the .git from within a gx path thing? e.g. I'm trying `git clone --bare --no-checkout https://github.com/ipfs/go-ipfs-cmds.git .git` from within my go-ipfs-cmds in a gx path
<tomatopeel>
from within == like that's where I am right now in cd terms
charley has joined #ipfs
<tomatopeel>
I get the fatal: Not a git repository, coincidentally
<lgierth>
just git clone --bare should do it
<lgierth>
it should land in ./go-ipfs-cmds.git, then mv go-ipfs-cmds.git .git
<tomatopeel>
yeah then i get: fatal: This operation must be run in a work tree
<lgierth>
git clone --bare must be run in a work tree? that sounds wonky
mildred has quit [Ping timeout: 255 seconds]
chungy has joined #ipfs
<tomatopeel>
lgierth: no no, it works, but then git st returns that error
<lgierth>
ah yeah
charley_ has joined #ipfs
<lgierth>
you can't actually "work" in a bare repo
<lgierth>
because it doesn't have a work tree
brodo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<lgierth>
you can just make a normal clone and overwrite its work tree
charley has quit [Ping timeout: 276 seconds]
Hasperov has left #ipfs [#ipfs]
jaboja has joined #ipfs
santamanno has quit [Quit: santamanno]
llama has joined #ipfs
charley_ has quit [Ping timeout: 276 seconds]
clemo has quit [Quit: clemo]
clemo has joined #ipfs
ChrisMatthieu has joined #ipfs
Miezel has joined #ipfs
borkDoy has quit [Quit: gone.]
kenshyx has quit [Remote host closed the connection]
galois_dmz has joined #ipfs
corvinux has quit [Ping timeout: 268 seconds]
corvinux has joined #ipfs
bedeho has joined #ipfs
Miezel has quit [Quit: This computer has gone to sleep]
corvinux has quit [Remote host closed the connection]
citizenErased has joined #ipfs
citizenErased has quit [Remote host closed the connection]
Hasperov has joined #ipfs
Miezel has joined #ipfs
bedeho has quit [Remote host closed the connection]
demize has quit [Quit: Do Androids Dream of Electric Sheep?]
demize has joined #ipfs
bedeho has joined #ipfs
bedeho has quit [Remote host closed the connection]
<tomatopeel>
lgierth: hmm, i think gx-go link re-wrote some import paths (from dvcs to gx)
<lgierth>
yeap, otherwise you get import conflicts, e.g. github.com/user/go-foo vs. gx/ipfs/QmFoo
<lgierth>
just `gx-go uw` before committing :)
<lgierth>
generally we commit libraries unrewritten, and programs rewritten
charley has joined #ipfs
charley has quit [Read error: Connection reset by peer]
<lgierth>
so in libraries you don't need gx-go link at all, but if you wanna run/build/test a program with one or more deps linked, then gx-go link is your friend
<lgierth>
bbiab, riding home
<tomatopeel>
lgierth: you don't happen to work from factory do you? ;p
<tomatopeel>
I'm also in berlin heh
<tomatopeel>
(saw your location when checking this branch on github :D)
chungy has quit [Ping timeout: 255 seconds]
jaboja has quit [Ping timeout: 260 seconds]
espadrine has quit [Ping timeout: 258 seconds]
<whyrusleeping>
tomatopeel: i know theres a decent ipfs contingent at c-base
<tomatopeel>
whyrusleeping: ah yeah that place seems really cool, been to a couple things there :)
charley has joined #ipfs
<whyrusleeping>
yeah, its really neat. I tend to hang out there when i'm in berlin
<tomatopeel>
still need to catch them when they've got those doepfer synths out, been ogling those boxes
llama has quit [Ping timeout: 240 seconds]
charley has quit [Remote host closed the connection]
charley has joined #ipfs
jaboja has joined #ipfs
charley has quit [Remote host closed the connection]
charley has joined #ipfs
charley has quit [Remote host closed the connection]
charley has joined #ipfs
<lgierth>
i worked at soundcloud until a month or so before they finally moved into factory
<lgierth>
so i've never actually seen it from the inside
clemo has quit [Quit: clemo]
chungy has joined #ipfs
charley has quit [Remote host closed the connection]
charley has joined #ipfs
Nukien has quit [Ping timeout: 246 seconds]
Nukien has joined #ipfs
charley has quit [Remote host closed the connection]
Miezel has quit [Quit: This computer has gone to sleep]
Miezel has joined #ipfs
charley has joined #ipfs
muvlon has quit [Quit: Leaving]
brodo has joined #ipfs
Miezel has quit [Client Quit]
llama has joined #ipfs
charley has quit [Remote host closed the connection]
bedeho has joined #ipfs
<jcgruenhage>
There is no IPFS gateway for Android yet, right?
<crankylinuxuser>
the current implementation times out after 50-ish replies are received, correct? Is there a way to prevent broadcast storms of clients spewing massive amounts of requests/sec ?
<crankylinuxuser>
admittedly I tend to think about breaking protocols. Just my security bent showing.
<lgierth>
don't have a good answer for you
<lgierth>
Kubuxu: dht security question ^
charley has quit [Remote host closed the connection]
charley has joined #ipfs
<Evermore>
lgierth: thanks, reading that PDF
<Evermore>
Is it fair to say that in most 'web scale' DHTs, a usual peer does not have an updated list of all peers in the whole table? Just a few neighbors?
charley has quit [Remote host closed the connection]
<lgierth>
yeah nobody has a complete view of the network
<lgierth>
that's the "distributed" part of DHT
Miezel has quit [Quit: This computer has gone to sleep]
<Evermore>
that does make it simpler
<lgierth>
it's a hash table, distributed oveer the whole network, with provable worst-case lookup times
<Evermore>
But often with IPFS (as of a couple years ago) I would upload something and it took a long time to propagate
<Evermore>
the gateway seems to find me fast but I struggled sending a file from one computer to another
charley has quit [Remote host closed the connection]
<lgierth>
yeah that's more of transport/connectivity issue
<lgierth>
we haven't had the time to properly fix that yet, but new and better transports are on the way
<lgierth>
QUIC, and a circuit switching relay
rendar has joined #ipfs
charley has joined #ipfs
homelessrobot[m] has joined #ipfs
<ianopolous_>
lgierth: sounds great!
<crankylinuxuser>
ugh.. had to go to the DC. Turns out someone named 2 of our servers the same bloody name, and DNS round-robins them. The good side: they were spares.
charley has quit [Ping timeout: 260 seconds]
Jesin has quit [Quit: Leaving]
mildred1 has joined #ipfs
brodo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
Jesin has joined #ipfs
brodo has joined #ipfs
ygrek_ has joined #ipfs
brodo has quit [Client Quit]
brodo has joined #ipfs
detran has quit [Read error: Connection reset by peer]
detran has joined #ipfs
bedeho has joined #ipfs
bedeho has quit [Remote host closed the connection]
bedeho has joined #ipfs
charley has joined #ipfs
<M-hash>
tomatopeel Is "factory" just an alias of that really nice converted building by bernauerstrasse? 🤔
enzo has joined #ipfs
enzo is now known as Guest39354
charley has quit [Ping timeout: 255 seconds]
charley has joined #ipfs
charley has quit [Remote host closed the connection]
atrapado_ has joined #ipfs
jonnycrunch1 has joined #ipfs
jonnycrunch1 has quit [Client Quit]
brodo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
brodo has joined #ipfs
Miezel has joined #ipfs
ashark has quit [Ping timeout: 240 seconds]
bedeho has quit [Remote host closed the connection]
espadrine has joined #ipfs
<Kubuxu>
crankylinuxuser: > the current implementation times out after 50-ish replies are received, correct?
<Kubuxu>
what do you mean by that
<crankylinuxuser>
Oh, it could be the client side, but when Ive run the command to see how many IPFS machines are peering my files, I thought that capped out after 50 responses.
maxlath has joined #ipfs
charley has joined #ipfs
charley has quit [Remote host closed the connection]
jkilpatr has quit [Ping timeout: 240 seconds]
mildred2 has joined #ipfs
mildred1 has quit [Read error: No route to host]
maxlath has quit [Ping timeout: 260 seconds]
bedeho has joined #ipfs
bedeho has quit [Remote host closed the connection]
charley has joined #ipfs
<lgierth>
M-hash: yeah iirc it's the official name -- at least it used to be when it was being constructed
charley has quit [Remote host closed the connection]
<lgierth>
nowadays my office is IRC -- much nicer
Akaibu has joined #ipfs
maxlath has joined #ipfs
jkilpatr has joined #ipfs
m0ns00n has quit [Quit: quit]
_mak has quit [Remote host closed the connection]
peterix-w has quit [Remote host closed the connection]
droman1 has quit []
jaboja has joined #ipfs
zopsi has quit [Quit: Goodnight and Good Luck]
jaboja has quit [Read error: Connection reset by peer]
zopsi has joined #ipfs
ashark has joined #ipfs
Guest39354 has quit [Quit: This computer has gone to sleep]
jaboja has joined #ipfs
peterix has joined #ipfs
_mak has joined #ipfs
enzo__ has joined #ipfs
enzo__ has quit [Client Quit]
Miezel has quit [Quit: This computer has gone to sleep]
Boomerang has quit [Quit: Lost terminal]
Hasperov has quit [Ping timeout: 255 seconds]
anewuser has quit [Quit: anewuser]
octalsrc[m] has joined #ipfs
JuliusCaesar[m] has joined #ipfs
<JuliusCaesar[m]>
Is it possible in an easy way to install ipfs and the webGUI on a raspberry pi ? :)
<jcgruenhage>
It should be.
<crankylinuxuser>
https://dist.ipfs.io/#go-ipfs , and you want the Linux ARM binary. INstall the binary to /usr/local/bin as per instructions. Set $PATH to include /usr/local/bin if not already, and "ipfs init" then "ipfs daemon"
<JuliusCaesar[m]>
Like a prebuilt package or maybe a repository easy? jcgruenhage
<JuliusCaesar[m]>
crankylinuxuser: thank you i'll have a look!
<crankylinuxuser>
^^ I just told you how to do it :)
tantalum has quit [Quit: Leaving.]
robattila256 has joined #ipfs
<tomatopeel>
M-hash: yeah something like that lol... it's got probably the most traffic for a workspace I think so that's why I asked, didn't seem outside the realm of possibility
<tomatopeel>
but yeah I get hints of hipness and money that bring my senses into question from time to time... been thinking a lot about some scenes in american psycho... apart from that it's perfectly nice though, beats the library
rcat has joined #ipfs
brodo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<tomatopeel>
whyrusleeping: lgierth: PR for gx-go to add a -w/--write-imports bool to do the import re-writing stuff (so leaving this flag off leaves them alone)
<tomatopeel>
it does seem like a use case, I mean all the import refs in go-ipfs-cmds are supposed to be dvcs ones, so when I try to use gx-go link to dev on it, it's writing the import paths which is wrong, right?
<tomatopeel>
I guess you're supposed to just unlink before you push?
<lgierth>
yeah definitely -- and if you're testing the changes within the context of go-ipfs, you want the rewritten import paths
maxlath has quit [Quit: maxlath]
<lgierth>
i can imagine a few edge cases where you can actually get away without the rewrites, but generally you need them, especially with packages that have a bit more complex dependency tree around them
<lgierth>
e.g. go-libp2p-conn or go-multiaddr, with these you 100% can't built without the rewrites
<lgierth>
i.e. can't build go-ipfs without the rewrites in the package. because every other package in the dependency tree, and also go-ipfs itself, uses that one package by its rewritten import path
<lgierth>
the error looks something like "expected github.com/ipfs/go-foo.MyType, got gx/ipfs/QmFoo/go-foo.MyType"
<lgierth>
or the other way around, depending on the exact spot in the code
Caterpillar2 has quit [Quit: You were not made to live as brutes, but to follow virtue and knowledge.]
berot3[m] has joined #ipfs
ralphtheninja has quit [Quit: leaving]
Encrypt has quit [Quit: Quit]
jkilpatr has quit [Quit: Leaving]
jkilpatr has joined #ipfs
Miezel has joined #ipfs
chris613 has joined #ipfs
galois_dmz has quit [Read error: Connection reset by peer]
galois_dmz has joined #ipfs
pat36 has quit []
<tomatopeel>
lgierth: yeah I was getting that yesterday. So gx-go link is great if you're devving on one external repo at a time. With the go-ipfs-cmds example, one problem that I'm seeing now is that it re-writes go-ipfs-cmdkit, and it's perfectly feasible that one might be devving on go-ipfs-cmds and go-ipfs-cmdkit simultaneously.
sirdancealot has quit [Ping timeout: 260 seconds]
<tomatopeel>
makes me wonder what the logic is behind splitting go-ipfs-cmd* into two different ext deps, especially since there's a lot of cross-talk between those two repo's
<lgierth>
you can just make links for both
<tomatopeel>
lgierth: yeah, and then unwrite, right?
<tomatopeel>
or manually edit the import path for the respective other repo's...
<tomatopeel>
unwrite would unwrite every one
<tomatopeel>
in this case it's the only one that is rewritten/unwritten so it'd be fine with rewrite/unwrite...
<lgierth>
i don't think you need to edit anything, just `gx-go link QmCmdsHash QmCmdkitHash`
<tomatopeel>
ah yes true of course
<lgierth>
and the go-ipfs build should pull in both correctly then
brodo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<tomatopeel>
the fundamental thing going on with the symlinking thing escaped me for a second there hah
<lgierth>
i'm still open to that --no-rewrite flag if you can show me you actually need it ;)
brodo has joined #ipfs
<tomatopeel>
I think I'm pretty convinced by everything you've said :D
charley has quit [Remote host closed the connection]
<M-hash>
whyrusleeping: repro builds in go: er, yeah? they work. <end-of-chapter>
<M-hash>
:3
llama has joined #ipfs
<M-hash>
i'm not sure if you had a more complicated question, but I think the answer is legitimately that flippant
<M-hash>
which is kind of awesome
<tomatopeel>
what do you guys mean by "reproducible" here?
<lgierth>
same binary if you build with the same go version, and the same source code
<tomatopeel>
how could that not be?
<lgierth>
timestamps, non-deterministic optimizations, etc. pp.
<tomatopeel>
ah I see, same binary by a hash's standards
<lgierth>
whyrusleeping: all the more reason to get a standardized build environment
rudolph has quit [Ping timeout: 246 seconds]
charley has joined #ipfs
charley has quit [Remote host closed the connection]
atrapado_ has quit [Quit: Leaving]
<tomatopeel>
keks[m]: if you're around, what's the logic in having two repos for the cmd stuff, so that's go-ipfs-cmds and go-ipfs-cmdkit?
charley has joined #ipfs
charley has quit [Remote host closed the connection]
<tomatopeel>
to me it seems like overkill having two repos for this
gvimrh[m] has joined #ipfs
kenmazsyma__ has joined #ipfs
Hasperov has joined #ipfs
brodo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
brodo has joined #ipfs
Hasperov has left #ipfs [#ipfs]
brodo has quit [Ping timeout: 268 seconds]
ashark has quit [Ping timeout: 240 seconds]
Akaibu has quit [Quit: Connection closed for inactivity]
joocain2_ has joined #ipfs
mildred2 has quit [Ping timeout: 255 seconds]
<M-hash>
> same binary by a hash's standards
joocain2 has quit [Ping timeout: 248 seconds]
<M-hash>
it's funny because they're also my standards :3
<M-hash>
so yes, you get a same binary out given same go source and same go compiler. timestamps: don't matter. source path, yes including full $GOPATH string: does matter, womp womp.
<M-hash>
GOROOT might matter, haven't explicitly checked... probably does; we see it in stack traces so it must be zipped in somewhere
brodo has joined #ipfs
<lgierth>
just mentioning timestamps because that was part of debian's reproducible builds changes -- use tar in a way that it doesn't include modtime/accesstime
<M-hash>
cgo can be problematic, but since recent go compiler versions (1.7ish iirc?) it's usually repro out-of-box; i'm not sure what the rules are before you get a reality excursion so pretty much check to be sure.
<M-hash>
do you lurk #reproducible-builds on oftc?
<lgierth>
i try to limit the lurking to p2p-related things :)
<M-hash>
heh. fair.
<M-hash>
some of the other humans for that are either in or near berlin though, for your future knowledge :3
Aranjedeath has quit [Ping timeout: 276 seconds]
brodo has quit [Ping timeout: 276 seconds]
<tomatopeel>
there is a golang meet this week I think, I dunno if you guys go to that
<tomatopeel>
I've never been, not been here long, thought I'd check it out
<M-hash>
heh, thanks for the nudge. I will totally do that sometime in the future I hope but right now I'm a dirty american out of schengen days and so back on the other side of the pond. haha
arpu has joined #ipfs
rcat has quit [Remote host closed the connection]
<M-hash>
re: tar determinism: yeah, that's an awkward subject. On the plus side, you can write your own tar implementation in go and it'll happen to be deterministic! :D
<M-hash>
Only slightly trollface, since of course I happen to have done this...
<lgierth>
i think we've done the same for the tar/add and tar/cat commands
<lgierth>
git st
<M-hash>
Yeah it's not too bad...
Miezel has quit [Quit: This computer has gone to sleep]
shizy has quit [Ping timeout: 240 seconds]
brodo has joined #ipfs
Akaibu has joined #ipfs
brodo has quit [Ping timeout: 276 seconds]
<JuliusCaesar[m]>
If i constantly get api version mismatch on the macOS build. What could be wrong?
<lgierth>
JuliusCaesar[m]: cli is different version than daemon
<JuliusCaesar[m]>
lgierth: since i'm very new to IPFS, would you mind guiding me in the right direction. How can i change the cli version ?
<JuliusCaesar[m]>
or vice versa
<lgierth>
do you have a deamon running right now? stop it by sending Ctrl+C, then restart it using the same executable that you use for client commands
<lgierth>
then they're the same version
<JuliusCaesar[m]>
i dont think i have the daemon running right now but i'll give it a go
<lgierth>
you can also try 'ps aux | grep ipfs' to see if there's one anywhere. then `killall ipfs` if there is one
vyzo has quit [Ping timeout: 255 seconds]
<lgierth>
if there's no deamon running, you might have a config file in $HOME/.ipfs/api which points to a daemon that *is* running. this file is a way to point the client cli to a daemon on a different machine. you can delete that file then an try again
<lgierth>
but this file shouldn't show up just by itsel
<JuliusCaesar[m]>
lgierth: okey so it's running but i can't seems to understand where i should press ctrl+c .. i have a terminal running and can send commands with ipfs "command"
ilyaigpetrov has quit [Quit: Connection closed for inactivity]
<JuliusCaesar[m]>
But when i just press ctrl +c the terminal doesn't know what app to kill
rudolph has joined #ipfs
cxl000 has quit [Quit: Leaving]
<lgierth>
in the tab where the daemon is running
<lgierth>
ctrl+c is the usual way to stop an active program in a terminal tab
<tomatopeel>
tab, you say?!
<lgierth>
terminal tab i guess, or panel
<JuliusCaesar[m]>
i actually don't have a tab where the daemon is running. :/
<lgierth>
i use a tiled terminal
<kythyria[m]>
If it's not in any tab, `kill` can do the equivalent of sending control-c
<lgierth>
yeah, `killall ipfs`
Mateon3 has joined #ipfs
<tomatopeel>
`pgrep -l ipfs` might be easier than ps also, just to see if it is running or not
<kythyria[m]>
Provided you don't have solaris killall :P
<tomatopeel>
... if crapOS has pgrep...
<JuliusCaesar[m]>
kythyria: lgierth Okey thanks i'll give it a go!
Mateon1 has quit [Ping timeout: 260 seconds]
Mateon3 is now known as Mateon1
<tomatopeel>
if I try to run a second daemon after I'm already running one, it errors saying that one is running, so that's useful expected behaviour I guess
<JuliusCaesar[m]>
i managed to kill ipfs, gonna try if all works now
<JuliusCaesar[m]>
Sweet, it seems to work now, thank you! Also, what the command for removing a file that i uploaded?
arpu has quit [Ping timeout: 276 seconds]
<pjz>
ipfs pin rm xxx ?
<JuliusCaesar[m]>
pjz: thx!
<pjz>
np
joocain2 has joined #ipfs
joocain2_ has quit [Ping timeout: 248 seconds]
<JuliusCaesar[m]>
pjz (IRC): and it's removed from the network completely?
<pjz>
well, no, just from your node, if you have it pinned.
<pjz>
if you *don't* have it pinned, that won't work, but it will age out by itself eventually. Or you can remove it via 'ipfs block rm <blockid>'
<pjz>
and nothing you can do will remove it from anyone else's node
<pjz>
okay, upgraded pinbits.io to 0.4.9
<pjz>
hopefully it's stable
Miezel has joined #ipfs
<lgierth>
no, pin rm just frees it for garbage collection. ipfs repo gc will actually remove anything that's not pinned
poopie has joined #ipfs
<poopie>
how come there's so many people in the ipfs channel? I mean, I love ipfs but seems like a ton of people
<pjz>
lots of people just idle
<pjz>
like me :)
<Guest2429>
cc
llama has joined #ipfs
<poopie>
Which IRC client are you using?
<pjz>
irssi
<pjz>
<-- old school
<poopie>
I know about irssi and others but I've only ever used freenode's online chat client..
<poopie>
Are you using irssi via some server so it's always on? Cause what always bothered me that I would get disconnected every time I put my computer to sleep..
<pjz>
I run irssi in a screen session on a VPS, then attach to it via mosh from my laptop