--- Log opened Thu Aug 15 00:00:41 2019
01:35 -!- harding [~harding@cpe-66-91-121-125.hawaii.res.rr.com] has quit [Ping timeout: 248 seconds]
01:36 -!- harding [~harding@cpe-66-91-121-125.hawaii.res.rr.com] has joined #bitcoin-builds
02:35 -!- alko [~alko@BSN-77-147-172.static.siol.net] has quit [Ping timeout: 245 seconds]
02:57 -!- alko [~alko@BSN-77-147-172.static.siol.net] has joined #bitcoin-builds
05:37 -!- elichai2 [uid212594@gateway/web/irccloud.com/x-chkqjvgluvzddzsc] has joined #bitcoin-builds
07:55 -!- hebasto [~hebasto@95.164.65.194] has joined #bitcoin-builds
11:21 < cfields> dongcarl: thank!
11:21 < cfields> *thanks. heh
11:22 < dongcarl> cfields: Some discussion in another thread on the same topic: https://github.com/bitcoin/bitcoin/pull/16611#issuecomment-521396904
12:11 -!- provoostenator [~vwDZ2BYsc@2a05:d014:5f:e100:fd30:8af7:2d6a:cbb1] has joined #bitcoin-builds
12:22 < provoostenator> MarcoFalke is there a way to pass -j20 etc to the test runner?
12:22 < MarcoFalke> MAKEJOBS="-j20", I believe
12:22 < MarcoFalke> Oh, you mean the ci?
12:22 < MarcoFalke> or the test runner?
12:22 < MarcoFalke> ./test_runner.py --help
12:23 < provoostenator> ./ci/test_run_all.sh
12:24 < provoostenator> Basically some way to push the host machine to the max
12:24 < provoostenator> Could also be running multiple docker containers in parallel.
12:24 < MarcoFalke> https://github.com/bitcoin/bitcoin/blob/367b023ae444e7d9641c4a3469f9154461e50e68/ci/test/00_setup_env.sh#L14
12:25 < MarcoFalke> so `export MAKEJOBS="-j20"` should do it
12:25 < MarcoFalke> provoostenator: Running multiple dockers would mean it needs multiple checkouts of the project
12:25 < MarcoFalke> (different dirs)
12:26 < provoostenator> Ok, that could get tedious
12:26 < MarcoFalke> I wish there was a "one-way-mount" option in docker
12:26 < provoostenator> MAKEJOBS should help, and a way to pass a bigger number to the function tests job
12:27 < provoostenator> Maybe RAM disk support too, which I find quite dramatically speeds up the functional test suite.
12:27 < provoostenator> (3 GB is enough for 100 pararallel jobs)
12:27 < MarcoFalke> Hmm, we could copy the repo into the docker instead of mounting
12:29 < provoostenator> I thought disks inside Docker are slow though.
12:29 < MarcoFalke> that makes it hard to look at compile failures as well
12:29 < provoostenator> git subtree could help
12:30 < MarcoFalke> how?
12:30 < provoostenator> Clone repo once, check out the commit / branch you want to run, create 1 subtree for each Docker instance, and mount those
12:31 < MarcoFalke> You mean worktree?
12:31 < provoostenator> Yes, sorry
12:31 < MarcoFalke> heh
12:31 < dongcarl> Worktrees are very weird without the master
12:31 < dongcarl> In particular, git operations won’t work
12:31 < dongcarl> And also, not that much better than just copying in
12:31 < MarcoFalke> dongcarl: Huh? They should IIRC
12:32 < provoostenator> dongcarl: that's fine, because you can make the worktrees after checking out the right branch
12:32 < provoostenator> And nuke them when done
12:32 < MarcoFalke> dongcarl: It should mirror everything to master, no?
12:32 < MarcoFalke> I am pretty sure on that
12:32 < provoostenator> But indeed perhaps not much benefit compared to copying
12:38 < dongcarl> MarcoFalke: Perhaps we're talking about different things, but if you bind mount a working tree that's not the main one into a Linux container, and invoke `git status` or something inside the container, it will not work
12:40 < MarcoFalke> Sure
12:40 < dongcarl> BTW, docker has read-only bind mounts: https://docs.docker.com/storage/bind-mounts/#use-a-read-only-bind-mount
12:40 < MarcoFalke> It will work if the main one is in the container as well
12:40 < dongcarl> MarcoFalke: Right.
12:59 -!- hebasto [~hebasto@95.164.65.194] has quit [Remote host closed the connection]
13:13 < provoostenator> Travis machines 5 and 8 seem to have a problem; seeing both fail in different PRs.
13:14 < provoostenator> They get completely stuck after the UniValue unit tests
13:14 < provoostenator> E.g. https://github.com/bitcoin/bitcoin/pull/16620
13:25 < ryanofsky> cfields, curious about https://pastebin.com/raw/3PS5rtdN. what should they do to avoid being anticompetitive (other than ceasing to exist)
17:21 -!- lightlike [~lightlike@2001:16b8:571f:c600:f442:737b:d7:3900] has joined #bitcoin-builds
17:22 -!- lightlike [~lightlike@2001:16b8:571f:c600:f442:737b:d7:3900] has left #bitcoin-builds ["Leaving"]
19:10 -!- hebasto [~hebasto@95.164.65.194] has joined #bitcoin-builds
19:43 -!- hebasto [~hebasto@95.164.65.194] has quit [Remote host closed the connection]
23:07 -!- elichai2 [uid212594@gateway/web/irccloud.com/x-chkqjvgluvzddzsc] has quit [Quit: Connection closed for inactivity]
23:20 -!- jb55 [~jb55@gateway/tor-sasl/jb55] has quit [Remote host closed the connection]
23:21 -!- jb55 [~jb55@gateway/tor-sasl/jb55] has joined #bitcoin-builds
--- Log closed Fri Aug 16 00:00:42 2019