pfraze has quit [Remote host closed the connection]
<ipfsbot>
[go-ipfs] jbenet force-pushed feat/mss-mux from 2474247 to ef34768: http://git.io/vtN3H
<ipfsbot>
go-ipfs/feat/mss-mux ef34768 Jeromy: using multistream muxer...
dread-alexandria has quit [Quit: dread-alexandria]
<kbala>
hey lgierth is it possible to make prometheus graph something other than a time series? like maybe a histogram of some label vs some gauge?
pfraze has joined #ipfs
<lgierth>
kbala: as in 50th, 75th, 99th?
<whyrusleeping>
jbenet: want me to work on that bitswap dashboard i was thinking about?
<lgierth>
kbala: if so, yes
<kbala>
lgierth: i mean like if i have a file_time_ms gaugevec with a latency label, can i make a histogram where the different latency labels make the x axis and sums of the times for each latency is the y axis?
<lgierth>
oh i see. that's a good question
<kbala>
lgierth: sorry thats confusing, i guess what i'm asking is how can i see the relationship between a metric and a label it has
<lgierth>
kbala: btw prometheus is easy to run locally
<kbala>
jbenet my bad i always miss this
<kbala>
lgierth yeah i have it running its really nice
<kbala>
sprint checkin: i integrated the netsim transport stuff into mocknet streams and added prometheus metrics to bssim
<lgierth>
oh cool, so we could aggregate them at metrics.ipfs.io too i guess
<kbala>
so you can now see time series of basic stuff like block times and # of duplicate blocks as the simulation is running
<kbala>
and you can specify latency/bandwidth per link if you want instead of it being the same for all connections
<kbala>
jbenet: sorry i messed up my git beyond comprehension and i think a force push closed the latency/bandwidth issue
<kbala>
*PR not issue
<jbenet>
lgierth: could we do metrics.i.ipfs.io ? (want to keep all infra stuff under .i.ipfs.io because we may have to use subdomains for other things later on
<jbenet>
lgierth can also use another domain if the subdomains are annoying
nessence has quit [Remote host closed the connection]
<jbenet>
ipfs-dev.io or something
nessence has joined #ipfs
<lgierth>
ah i. = infrastructure
<lgierth>
obviously :S
<lgierth>
i was thinking inventory :P
<lgierth>
yep metrics.i. is fine
<jbenet>
inventory is a good guess too
<jbenet>
but then i may want to put jupiter into my napsack, which may be a bit too heavy to carry.
<lgierth>
:)
<lgierth>
btw, all these $bird.i.ipfs.io domains are dead right?
<lgierth>
emu was the last existing droplet
nessence has quit [Remote host closed the connection]
<zignig>
jupiter's radius , 69,911 km
<zignig>
I think you'll need a bigger napsack.
<zignig>
:)
nessence has joined #ipfs
<jbenet>
lgierth: yeah we ate the birds.
nessence has quit [Remote host closed the connection]
nessence has joined #ipfs
<kbala>
jbenet: do you still have time to talk today?
<jbenet>
hey kbala yep
nessence has quit [Remote host closed the connection]
<kbala>
jbenet: great i'm free whenever, when would you like to talk?
<jbenet>
kbala: ok lets talk in 15
nessence has joined #ipfs
anshukla has joined #ipfs
<whyrusleeping>
i need a nice 'todo queue'
<whyrusleeping>
where people can easily assign things to me
<whyrusleeping>
and i can check them off
anshukla has quit [Ping timeout: 264 seconds]
<jbenet>
whyrusleeping yeah i want that too. fork github.com/jbenet/todo
<jbenet>
also do use ito/
<whyrusleeping>
yeah, thats closer to what i want
<whyrusleeping>
its just harder than i want it to be to use
<lgierth>
trello board where anyone can make you cards
<lgierth>
mh we could have a bot for that
<lgierth>
TODO whyrusleeping if you could [...], that would be great
<whyrusleeping>
oooOoo, we *could* have a bot for that
<whyrusleeping>
i like that much better...
<whyrusleeping>
jbenet: ignore the trello invite
<lgierth>
:]
<lgierth>
thatwouldbegreat.jpg
heliumcraft has joined #ipfs
* zignig
thinks astralboot is almost ready for alpha
heliumcraft has quit [Client Quit]
<whyrusleeping>
zignig: i havent run it yet because i'm without my own network
void_ has joined #ipfs
<zignig>
whyrusleeping: yeah... that would help.
<zignig>
single laptop ?
<whyrusleeping>
zignig: yeap, for now. and if i mess with my networks dhcp, my roommates are liable to be upset
rht__ has joined #ipfs
<zignig>
or you could auto image their machines into coreos by mistake .... ;)
<jbenet>
gatesvp maybe we can try making those tests go into a different dir.
<jbenet>
gatesvp btw, i think these tests merit their own "t0XXX-ipfsignore.sh" test file
<jbenet>
(gatesvp i can make that change for you)
anshukla has joined #ipfs
<gatesvp>
@jbenet: agreed on adding the new file... I was already pondering that so that I could create a more "crazy" structure to test with
anshukla has quit [Client Quit]
sff_ has joined #ipfs
semidreamless has quit [Quit: Leaving...]
sff has quit [Ping timeout: 248 seconds]
<gatesvp>
@jbenet, so there are three "phases" of .ipfsignore files. (1) search the ConfigRoot (2) search the parents of an actively added folder (3) search the individual folders as we add them.
<gatesvp>
moving (1) to the PreRun phase seems pretty trivial... (3) doesn't need to move, it's really just (2) that's a little dicey
<jbenet>
gatesvp yeah
<jbenet>
gatesvp also this way we avoid sending those files over the wire, too, which may be important for massive things
<jbenet>
(like i ignore a 10TB file
<jbenet>
hey whyrusleeping are you around?
<jbenet>
thought we'd split the blocks changes into their own PR before full GC?
<gatesvp>
@jbenet: the problem with (2) is that it's currently being called within the `for` loop where we are looping through the files i.e: (ipfs add a b)
<gatesvp>
@jbenet: I guess I can resolve the ignores in advance, but then I need to transfer them from PreRun to Run and I'm not clear that I can patch variables between the two... will have to dig into the underlying code there
<jbenet>
@gatesvp why do they need to be transferred to Run?
<jbenet>
@gatesvp: couldn't it be done jus in PreRun, and the files that Run gets are already filtered out?
<gatesvp>
or I have to find a way to resolve the Absolute path at the right place so that I'm not relative to the Daemon...
<ipfsbot>
[go-ipfs] whyrusleeping created locking-blockstore (+1 new commit): http://git.io/vqE3y
<ipfsbot>
go-ipfs/locking-blockstore f008ce5 Jeromy: Add locking interface to blockstore...
<ipfsbot>
[go-ipfs] whyrusleeping opened pull request #1453: Add locking interface to blockstore (dev0.4.0...locking-blockstore) http://git.io/vqECC
<ipfsbot>
[go-ipfs] whyrusleeping pushed 1 new commit to locking-blockstore: http://git.io/vqECa
<ipfsbot>
go-ipfs/locking-blockstore 18688a2 Jeromy: merkledag FetchGraph and EnumerateChildren...
pfraze has joined #ipfs
<ipfsbot>
[go-ipfs] whyrusleeping force-pushed feat/mark-n-sweep from 7580503 to 9202d50: http://git.io/vtNpa
<ipfsbot>
go-ipfs/feat/mark-n-sweep 9202d50 Jeromy: implement mark and sweep GC...
* whyrusleeping
puts his git machete away and takes a sip of coffee
PayasR has joined #ipfs
nessence has joined #ipfs
nessence has quit [Remote host closed the connection]
ruby32 has joined #ipfs
nessence has joined #ipfs
nessence has quit [Remote host closed the connection]
Encrypt has joined #ipfs
mildred has joined #ipfs
mildred has quit [Client Quit]
nessence has joined #ipfs
mildred has joined #ipfs
atomotic has joined #ipfs
mildred has quit [Quit: Leaving.]
M-_mis has quit [Ping timeout: 246 seconds]
atrapado has joined #ipfs
M-Staplemac has quit [Ping timeout: 246 seconds]
MatrixBridge has quit [Ping timeout: 256 seconds]
M-Eric has quit [Ping timeout: 250 seconds]
<atrapado>
buenas
MatrixBridge has joined #ipfs
<whyrusleeping>
atrapado: hey there!
<atrapado>
hey. i thought it would send 'buenas' only to all channels of the other network...
<whyrusleeping>
lol
<atrapado>
:)
nessence has quit []
Encrypt has quit [Quit: Quitte]
<ThomasWaldmann>
jbenet: did you ever consider strange legal situations in some countries where downloading some content is legal, but not redistributing/serving it to others?
<PayasR>
ipfs is completely decentralised and stores no data about any file except perhaps the hashes. Those too are distributed. in such cases i believe the user is responsible for the data he downloads/uploads.
<whyrusleeping>
fd0: sprintbot is a bot that reminds contributors to give a daily update about what theyre working on
<fd0>
ok
<whyrusleeping>
sprintbot: split up the GC PR into two separate PRs, wrote some tests for go-datastore batching, and now i'm working on debugging daviddias not being able to cat a file between our machines.
<thefinn93>
whyrusleeping: hey, mind if we push meeting up till Wednesday evening?
<thefinn93>
realized that'd work much better
<ipfsbot>
[go-ipfs] whyrusleeping created bitswap-events (+1 new commit): http://git.io/vquMk
<ipfsbot>
go-ipfs/bitswap-events ff1bf30 Jeromy: add in some events to bitswap to emit worker information...
<jbenet>
sprintbot: pushed first iteration of records, working on record chain and signed records. And CR
<thefinn93>
i figured i could take off at like 5:30-6:30
<thefinn93>
and head generally downtown
tilgovi has joined #ipfs
<ipfsbot>
[go-ipfs] whyrusleeping opened pull request #1455: add in some events to bitswap to emit worker information (master...bitswap-events) http://git.io/vqu9e
Encrypt has joined #ipfs
<jbenet>
thefinn93 that sgtm.
<thefinn93>
cool
<thefinn93>
want me to PM one of you my phone number for easier offline communication?
<thefinn93>
I'm fine with using IRC for everything but i know not everyone has it on their phone
<jbenet>
thefinn93 i have IRC on my phone :) but i PMed my # anyway
<ipfsbot>
[go-ipfs] jbenet closed pull request #1455: add in some events to bitswap to emit worker information (master...bitswap-events) http://git.io/vqu9e
<lgierth>
sprintbot: slightly refactoring gateway nginx stuff out of neccessity, not much today apart from that. added h.gateway.ipfs.io domain for cjdns/hyperboria