<Icefoz_>
So, what in the world is your web server config for the mooooo.ooo vhost?
<colin_>
Icefoz_: In this case the webserver is the same one hosting ipfs.io. I own a domain name and I have the A record pointing to the IP address of ipfs.io.
pat36 has quit [Read error: Connection reset by peer]
pat36 has joined #ipfs
dhruvbaldawa has joined #ipfs
girrrrrrr2 has quit [Ping timeout: 246 seconds]
dhruvbaldawa has quit [Ping timeout: 240 seconds]
robattila256 has quit [Quit: WeeChat 1.9.1]
ygrek_ has joined #ipfs
absurd[m] has left #ipfs ["User left"]
}ls{ has quit [Quit: real life interrupt]
dhruvbaldawa has joined #ipfs
maxlath has joined #ipfs
harlantwood has joined #ipfs
harlantwood has quit [Ping timeout: 240 seconds]
corvinux has joined #ipfs
kirby__ has joined #ipfs
ianopolous has joined #ipfs
dimitarvp has joined #ipfs
pat36 has quit [Ping timeout: 240 seconds]
<lgierth>
colin_: sorry about that, i messed up the https redirects
<lgierth>
fixing it right now
Manolis has joined #ipfs
<Manolis>
hi
<Manolis>
Is it intended, that the gateway at ipfs.io redirects all http connections to https? because this kind breaks dnslink setups. when i setup a doain with a dnslink txt record and a cname to ipfs.io, then browse to this domain i get a ssl cert error because the connection is redirected to https but the cert for ipfs.io is used
<Manolis>
and if so why wasn't this change commincated publicaly
uptime has quit [Quit: Look, low effort people produce low quality results. What do you plan on doing once you've made everyone a robocop? Even the blind can see there ain't no loyalty.]
maxlath has quit [Ping timeout: 255 seconds]
pbrooko has quit [Quit: pbrooko]
<lgierth>
Manolis: hey i think i just fixed it -- sorry about that, i messed up the https redirects
pbrooko has joined #ipfs
<daviddias>
olizilla: is doing a patchwork install fest at Mozfest
<daviddias>
come to floor 4 for autographed invites
<lgierth>
oahr. let's remove that bot once mozfest is over
<whyrusleeping>
did someone add that bot?
caladrius has joined #ipfs
<lgierth>
it's okay for a day but i'm not too crazy about it
pat36 has quit [Read error: Connection reset by peer]
ygrek_ has quit [Ping timeout: 252 seconds]
pat36 has joined #ipfs
maxlath has joined #ipfs
whenisnever has quit [Ping timeout: 264 seconds]
maxlath has quit [Ping timeout: 240 seconds]
kirby__ has quit [Ping timeout: 240 seconds]
kirby__ has joined #ipfs
espadrine has quit [Ping timeout: 258 seconds]
pat36 has quit [Read error: Connection reset by peer]
pat36 has joined #ipfs
kirby__ has quit [Ping timeout: 248 seconds]
harlantwood has joined #ipfs
HTTP_____GK1wmSU has joined #ipfs
ianopolous has quit [Ping timeout: 258 seconds]
HTTP_____GK1wmSU has left #ipfs [#ipfs]
harlantwood has quit [Ping timeout: 248 seconds]
whenisnever has joined #ipfs
larpanet has quit [Ping timeout: 248 seconds]
mog has quit [Ping timeout: 246 seconds]
yulijoto has quit [Quit: Leaving]
pat36 has quit [Read error: Connection reset by peer]
pat36 has joined #ipfs
maxlath has joined #ipfs
ianopolous has joined #ipfs
espadrine has joined #ipfs
<drozdziak1>
What's the relationship between ipld and ipfs?
tiroliro has joined #ipfs
maxlath has quit [Ping timeout: 248 seconds]
maxlath has joined #ipfs
<lgierth>
drozdziak1: ipld is the data model, and ipfs stores and moves the data
<lgierth>
ld = linked data
<drozdziak1>
lgierth: So ipld is more about the merkledag?
<lgierth>
yeah
<drozdziak1>
while ipfs is more aobut the files?
<lgierth>
you can implement ipld without any of ipfs
<lgierth>
ipfs is: bitswap (moves data), fs-repo (stores data), unixfs (models a filesystem as an ipld structure)
<lgierth>
, ipns (mutable names for ipld DAGs)
<lgierth>
i'm updating the ipld/ipld reame right now, give me 30 minutes or so
ianopolous has quit [Ping timeout: 240 seconds]
<tiroliro>
there are lots of IPFS features that I can't understand
<tiroliro>
for example, how can I pin random files from my peers?
<lgierth>
tiroliro: ipfs pin add <hash>
<tiroliro>
lgierth, but in this case I have to previously know the files hash, right?
<lgierth>
yeah
<lgierth>
you can't just pull hashes out of thin air
<tiroliro>
lgierth, yeah, that's my question
<lgierth>
same for the traditional web: either someone tells you a URL, or you click a link, or you use a search engine
<tiroliro>
okay, I got it
<lgierth>
what do you have in mind?
<tiroliro>
lgierth, I thought I could read files from my peers randomly. I mean, any ipfs command that retrieves photos from them.
<lgierth>
ah got it - well you can listen for provider record announcements from the network, but they are in no way "complete" or "realtime", and they're not a stable supported interface either
<lgierth>
as in, you'd be dealing with ipfs internals
dhruvbaldawa has quit [Remote host closed the connection]
ygrek_ has joined #ipfs
<tiroliro>
I'm thinking of coding a search mechanism that relies on those random files, that's my point. Can this be possible? I mean, this is not the same case as Google that relies on links and PageRank, but something different, based on random hashes.
rtjure has quit [Ping timeout: 240 seconds]
<lgierth>
oh actually there is a good interface for provider record announcements
mog has joined #ipfs
<tiroliro>
lgierth, really? what interface?
<lgierth>
ipfs log tail - i'm trying to figure out the right parameters right now
HTTP_____GK1wmSU has joined #ipfs
HTTP_____GK1wmSU has left #ipfs [#ipfs]
<lgierth>
mh it doesn't look like it currently outputs provider records it receives
<lgierth>
maybe try running it for a while: ipfs log tail | grep providers
ianopolous has joined #ipfs
kirby__ has joined #ipfs
<lgierth>
or ipfs log tail | grep prov
ianopolous_ has joined #ipfs
<tiroliro>
what 'ipfs pin ls' does?
<lgierth>
it lists everything you pinned locally
<lgierth>
ls = list, just like the shell command
<lgierth>
you can also call every ipfs command with --help
dhruvbaldawa has joined #ipfs
<AphelionZ>
morning everybody :)
<AphelionZ>
if I'm running an ipfs node with --profile=server on AWS, do i need to open any ports for ingress?
<AphelionZ>
maybe 4001?
ianopolous has quit [Ping timeout: 255 seconds]
<lgierth>
yeah 4001 would be good, i assume there's no NAT-PMP or uPnP in AWS ;)
<AphelionZ>
maybe not? im not sure...
<AphelionZ>
also I'm getting a lot of "tried to remove tag from untracked peer"
<AphelionZ>
maybe that's related
<lgierth>
unrelated, but fixed on master
<lgierth>
it's from the connmgr
<AphelionZ>
I'm swarming though, which is good
<AphelionZ>
ok cool, I'm on the latest rc
<AphelionZ>
last question for now... if I am running with --profile=server... do I still have a gateway and API?
<AphelionZ>
and if not, can I enable those selectively?
<lgierth>
yes you still have those
<lgierth>
the profiles just set different defaults for the config during `ipfs init`
pat36 has quit [Read error: Connection reset by peer]
<AphelionZ>
ok cool
<lgierth>
so you can just look at your config to see what's different
pat36 has joined #ipfs
<AphelionZ>
ok thank you
maxlath has quit [Ping timeout: 248 seconds]
pat36 has quit [Read error: Connection reset by peer]
maxlath has joined #ipfs
pat36 has joined #ipfs
whenisnever has quit [Ping timeout: 240 seconds]
whenisnever has joined #ipfs
clickjack has quit [Quit: clickjack]
clickjack has joined #ipfs
ianopolous_ has quit [Ping timeout: 260 seconds]
<AphelionZ>
wow this is cool... it's all set up
ianopolous_ has joined #ipfs
pat36 has quit [Read error: Connection reset by peer]
erictapen has quit [Remote host closed the connection]
erictapen has joined #ipfs
<mkg20001[m]>
Hi all. I just created this thing ( https://github.com/mkg20001/libp2p-nodetrust ) which will be a ca and dynamic dns service for libp2p nodes so the libp2p nodes can use websockets-secure (this allows a browser-node running on an https website to connect to it)
j24[m] has joined #ipfs
ianopolous_ has joined #ipfs
<whyrusleeping>
mkg20001[m]: oh sweet
<whyrusleeping>
I was actually just thinking we needed something similar to that yesterday
* whyrusleeping
reading through it
<whyrusleeping>
whoa, this is really great
<mkg20001[m]>
DNS currently does not work
<daviddias>
mkg20001[m]: woa! This looks too good, what is the caveat?
<mkg20001[m]>
It would be required to have a ca certificate for *.node.libp2p.io but I don't know if this is possible to get.
<daviddias>
This is really clever!
<daviddias>
summoning our certificate experts, lgierth and kyledrake
* ansuz
appears instead
<ansuz>
hi daviddias
<whyrusleeping>
ansuz: oh, you again
<ansuz>
you gotta be careful when you're summoning things
<daviddias>
Oh, it's ansuz! He got this
<ansuz>
have you tried turning it off and on again?
<ansuz>
hard reload?
<kyledrake>
Nested wildcards aren't really a thing, but I think (?) you can get two certs, one just for this use case. Maybe.
<ansuz>
that's what I was gonna say next
<ansuz>
good call kyledrake
benjamingr_ has joined #ipfs
<daviddias>
kyledrake what about *.wildwest-p2p.net
dhruvbaldawa has joined #ipfs
<daviddias>
kyledrake so can we get it for mkg20001[m] ?
<kyledrake>
Sorry what is this for? http websockets SSL for libp2p connections?
Jesin has joined #ipfs
<mkg20001[m]>
letsencrypt + dynamic dns fully automated for libp2p nodes using the peer-id as challenge for the cert/dns updates
<mkg20001[m]>
and that's for ssl websockets
<mkg20001[m]>
with letsencrypt I mean automated certificate signing, not using letsencrypt directly
Ohelig has quit [Ping timeout: 260 seconds]
ianopolous_ has quit [Ping timeout: 248 seconds]
Ohelig has joined #ipfs
}ls{ has quit [Ping timeout: 240 seconds]
Caterpillar2 has joined #ipfs
maxlath has joined #ipfs
pat36 has quit [Read error: Connection reset by peer]
pat36 has joined #ipfs
}ls{ has joined #ipfs
Alpha64 has joined #ipfs
Lisl has joined #ipfs
Lisl has left #ipfs ["Leaving"]
tiroliro has joined #ipfs
ircler has quit [Quit: bye]
ethdevguy has joined #ipfs
Caterpillar2 has quit [Quit: You were not made to live as brutes, but to follow virtue and knowledge.]
harlantwood has joined #ipfs
jaboja has joined #ipfs
<AphelionZ>
what is the relationship between libp2p and ipfs?
<AphelionZ>
ipfs uses the libp2p protocol?
pat36 has quit [Read error: Connection reset by peer]
<tiroliro>
Question: when I turn on ipfs daemon, do I start to automatically pin other users' files?
pat36 has quit [Read error: Connection reset by peer]
pat36 has joined #ipfs
dhruvbaldawa has quit [Ping timeout: 255 seconds]
pat36 has quit [Read error: Connection reset by peer]
pat36 has joined #ipfs
renais has quit [Ping timeout: 252 seconds]
<Alpha64>
ipfs add
<tiroliro>
I added a file a couple of days ago. I turned off IPFS daemon. After one week, the file is still up. What is this happening? I mean, someone must have pinned my file, right?
<whyrusleeping>
tiroliro: either that, or someone accessed it and hasnt run a gc yet
pat36 has quit [Read error: Connection reset by peer]
pat36 has joined #ipfs
tiroliro has left #ipfs ["Leaving"]
tiroliro has joined #ipfs
Tootoot222 has quit [Remote host closed the connection]
pat36 has quit [Read error: Connection reset by peer]
pat36 has joined #ipfs
<tiroliro>
whyrusleeping, I haven't shared the hash with anyone. Could someone have pinned my file by accident?
<whyrusleeping>
tiroliro: did you view it on ipfs.io?
<whyrusleeping>
if so, then the gateway nodes fetched the content
<whyrusleeping>
and if you've been checking it periodically, then it gets refreshed in the caches
<whyrusleeping>
(the gateways are several different ipfs nodes)
<whyrusleeping>
and they run garbage collections at different times
<tiroliro>
whyrusleeping, I did, so that must be the reason.
<tiroliro>
what's the relation between ipfs.io and the gateway nodes? Could you recommend me an article on it?
<AphelionZ>
how are efforts for indexing / directory-ifying the content on IPFS going/
<AphelionZ>
?
renais has joined #ipfs
<drozdziak1>
AphelionZ: ipfs already is mountable
pat36 has quit [Read error: Connection reset by peer]
pat36 has joined #ipfs
Encrypt has joined #ipfs
fredthomsen has joined #ipfs
pat36 has quit [Read error: Connection reset by peer]
pat36 has joined #ipfs
lidel has quit [Read error: Connection reset by peer]
renais has quit [Ping timeout: 248 seconds]
lidel has joined #ipfs
harlantwood has joined #ipfs
harlantwood has quit [Ping timeout: 255 seconds]
<lgierth>
mkg20001[m]: neat idea - i'm not sure how easy it is to get a cert that we can use to sign the individual *.node.libp2p.io certs
<lgierth>
have you tried it out yet, with some domain?
<mkg20001[m]>
only tested it with a self-signed ca
<lgierth>
i like it a lot
<lgierth>
we can also use it for gateways i guess
<lgierth>
actually ... the cert signing part can be placed by lets encrypt on the node - and then keep only the protocol for getting <peerid>.node.libp2p.io dns records
<lgierth>
s/placed/replaced/
<mkg20001[m]>
yes but then every node would need to run an http server in order to perform the certificate-challenge
<mkg20001[m]>
also the ratelimit would be exceeded pretty quickly
<lgierth>
is the rate limit per second-level domain?
pat36 has quit [Read error: Connection reset by peer]
<lgierth>
also the websockets transport already is a http server ;)
pat36 has joined #ipfs
<mkg20001[m]>
I know that some time ago when I tried to get a cert for ID.myfritz.net (dyndns for avm routers) I got a rate limit warning. also once when I had trouble setting up letsencrypt.sh and tried it 10 times I also got a rate limit error.
rhia_ has joined #ipfs
<mkg20001[m]>
and http wouldn't work as most js-ipfs apps are running on https
<lgierth>
The main limit is Certificates per Registered Domain, (20 per week)
<lgierth>
yeah
<lgierth>
so, back to the question of whether we can get an intermediate cert (i think that's the ones that can sign)
<lgierth>
so it looks like intermediate certs can't be limited to individual domains, and that's why they only get handed out to actuall ssl CAs
pat36 has quit [Read error: Connection reset by peer]