These are chat archives for ethersphere/orange-lounge

19th
Feb 2018
Balint Gabor
@gbalint
Feb 19 11:07
@/all anybody force pushed something on swarm-network-rewrite-syncer?
Viktor Trón
@zelig
Feb 19 12:08
@howleysv well spotted we changed the API for the the exchangeTest, could you submit a PR with the fix? also in network/discovery_test.go examples for correct use of the exchange tests are found for instance in https://github.com/ethersphere/go-ethereum/blob/swarm-network-rewrite-syncer/swarm/network/stream/streamer_test.go
Balint Gabor
@gbalint
Feb 19 14:41
@zelig @janos we merged everything from swarm-network-rewrite to swarm-network-rewrite-syncer with @nonsense, and we fixed the failing test . We are waiting for this now: https://travis-ci.org/ethersphere/go-ethereum/builds/343390323
Viktor Trón
@zelig
Feb 19 14:45
great news
Anton Evangelatov
@nonsense
Feb 19 15:55
shall we drop the swarm-network-rewrite-syncer branch as well, and merge to swarm-network-rewrite ?
@gbalint ?
holisticode
@holisticode
Feb 19 15:56
I am basing my current work on swarm-network-rewrite-syncer, or I even branched off swarm-network-rewrite-syncer-test, I don't remember
Balint Gabor
@gbalint
Feb 19 15:56
@nonsense yes, if everything is ok in travis
holisticode
@holisticode
Feb 19 15:56
Would a rebase onto swarm-network-rewrite be very painful?
Balint Gabor
@gbalint
Feb 19 15:56
@holisticode that is what we are doing (actually a merge not a rebase)
Anton Evangelatov
@nonsense
Feb 19 15:57
@gbalint it is green
Balint Gabor
@gbalint
Feb 19 15:57
so it should not be a problem for you, you can just target swarm-network-rewriteafter we are done, it contains everything from swarm-network-rewrite-syncer
holisticode
@holisticode
Feb 19 15:57
ok with a PR then?
Balint Gabor
@gbalint
Feb 19 15:58
@nonsense but several jobs are not finished
Anton Evangelatov
@nonsense
Feb 19 16:00
oh right - it is the PR build that is more important, not the branch
i thought it completed, and not the branch one
let’s wait for it
wait
Anton Evangelatov
@nonsense
Feb 19 16:01
Screen Shot 2018-02-19 at 17.00.58.png
PR is green, PUSH is in-progress
this means that merged commit is green, whereas branch commit is in-progress
Balint Gabor
@gbalint
Feb 19 16:01
ok, then we can merge
Anton Evangelatov
@nonsense
Feb 19 16:01
then we must be fine, even if branch commit fails for some reason (flaky or not)
this means that merged commit is green, whereas branch commit is in-progress - i am 99% sure about that, just not 100% :) the explanation on travis is a bit confusing, probably a good idea to double check at some point
Balint Gabor
@gbalint
Feb 19 16:02
i think you are right
but travis will run again on the new swarm-network-rewrite push anyway
Anton Evangelatov
@nonsense
Feb 19 16:02
also ethereum repo has only the pr one, the push is disabled
Balint Gabor
@gbalint
Feb 19 16:03
so it will be double checked
Anton Evangelatov
@nonsense
Feb 19 16:03
true
Balint Gabor
@gbalint
Feb 19 16:03
i think we should disable the push too. It is a waste to run everything twice basically
and…. i merged
@zelig @janos @holisticode @nonsense swarm-network-rewrite-syncer is gone, don’t use it anymore, just work on swarm-network-rewrite (with pull requests)
Anton Evangelatov
@nonsense
Feb 19 16:04
perfect
Balint Gabor
@gbalint
Feb 19 16:05
ok @nonsense, i remember why we have the push builds for branches
because I wanted to have builds on swarm-network-rewrite
Anton Evangelatov
@nonsense
Feb 19 16:06
prior to the PR i guess
Balint Gabor
@gbalint
Feb 19 16:06
and as far as i see you either have it on all branches or on none
Anton Evangelatov
@nonsense
Feb 19 16:06
let’s leave it like this for now - it is good to get 2 builds
it helps rule out the flakes
Balint Gabor
@gbalint
Feb 19 16:44
@nonsense btw the push build was green too https://travis-ci.org/ethersphere/go-ethereum/builds/343390323
Anton Evangelatov
@nonsense
Feb 19 16:44
cool
the PR build to master is mostly green
a known flaky test failed, and looks like les tests deadlocked
Balint Gabor
@gbalint
Feb 19 16:45
let’s hope the new push build on swarm-network-rewrite will be green too https://travis-ci.org/ethersphere/go-ethereum/builds/343431535
Anton Evangelatov
@nonsense
Feb 19 16:48
i think we are mostly good, considering https://travis-ci.org/ethersphere/go-ethereum/builds/343431366?utm_source=github_status&utm_medium=notification - only a few more issues left to flesh out on swarm side in terms of tests
we have issues for most of them
i will push a merge from master now, so that we are synced with the current master branch from go-ethereum - there are no conflicts :)
Balint Gabor
@gbalint
Feb 19 16:49
i did that last week, so there should be no big changes
ok, please do
Anton Evangelatov
@nonsense
Feb 19 16:54
done
Viktor Trón
@zelig
Feb 19 17:19
great job guys.. meanwhile @janos and I made progress writing simple sim test for intervals and live/history streams
Balint Gabor
@gbalint
Feb 19 17:24
awesome
@zelig @nagydani would you like to work tomorron on encryption?
Viktor Trón
@zelig
Feb 19 17:25
and @holisticode working on even higher level chunk syncing tests with simulation leading to metrics benchmarks
Balint Gabor
@gbalint
Feb 19 17:25
if not I’m going back to the chunker project
Viktor Trón
@zelig
Feb 19 17:25
I would really like a resilient syncing swarm operational by Paris
better we have something this week so network issues can be tested out on the permanent testnet soon
yes i would @gbalint lets finish it very soon
Balint Gabor
@gbalint
Feb 19 17:28
@zelig you mean yes for encryption?
and the push build is green too :fireworks: https://travis-ci.org/ethersphere/go-ethereum/builds/343431535
we finally can work with continuous integration it seems
Balint Gabor
@gbalint
Feb 19 18:04
@nonsense and your merge from master is green too! https://travis-ci.org/ethersphere/go-ethereum/builds/343454204
I’m so happy, travis is really usable now!
Balint Gabor
@gbalint
Feb 19 18:18
@nonsense i moved this card to done: ethersphere/go-ethereum#238
Anton Evangelatov
@nonsense
Feb 19 18:18
yes
Elad_
@justelad
Feb 19 21:31
do i need to have a full geth node synced in order to do ENS resolutions? i tried running geth with —syncmode "fast"/"light" but all of my ENS resolutions fail with Error serving GET bzz:/theswarm.eth/: error resolving theswarm.eth: no contract code at given address
Viktor Trón
@zelig
Feb 19 22:18
correct. sadly that is exactly what is happening if your blockchain is not uptodate