--- Log opened Wed Jul 10 00:00:06 2019
01:01 -!- jonatack60 [56dec717@abayonne-654-1-56-23.w86-222.abo.wanadoo.fr] has joined #bitcoin-builds
01:27 -!- jonatack60 [56dec717@abayonne-654-1-56-23.w86-222.abo.wanadoo.fr] has quit [Remote host closed the connection]
02:43 < fanquake> dongcarl: if we're explicitly disabling something that is disabled by default, do we want to also explicitly disable all other disabled by default things that we can? i.e for zmq, we pass --without-libsodium.
03:07 < midnightmagic> oh, hey nothingmuch \o
03:40 < nothingmuch> midnightmagic: hola
03:54 -!- jonatack29 [d598a14a@213.152.161.74] has joined #bitcoin-builds
07:05 < dongcarl> fanquake: Hmmm... I wish there were ways to `--disable-all` then only say what we want enabled
07:05 < dongcarl> cuz I want to be explicit, and be somewhat forwards compatible
07:08 < fanquake> dongcarl: yea I’m still unsure. I have a branch with a bunch of fairly simple flag additions to most depends packages, including one fix. But just have to figure out how far we want to go hah.
07:33 -!- hebasto [~hebasto@95.164.65.194] has joined #bitcoin-builds
10:42 -!- hebasto [~hebasto@95.164.65.194] has quit [Remote host closed the connection]
12:32 -!- jonatack29 [d598a14a@213.152.161.74] has quit [Remote host closed the connection]
13:21 -!- ryanofsky_ is now known as ryanofsky
13:23 < ryanofsky> any suggestions for getting depends build output to be less verbose on travis? i am seeing "The job exceeded the maximum log length, and has been terminated" after adding a native_boost package in 41410
13:24 < ryanofsky> #16367 not 41410
13:25  * dongcarl looking
13:36 < dongcarl> ryanofsky: No clean way... Does modifying `test_05_before_script.sh` seem okay for testing? Or are you talking about a more long term solution?
13:37 < ryanofsky> more talking about a long term solution, since i'm able to test locally and there are other things holding up that pr
13:39 < dongcarl> ryanofsky: Okay. I'll put up an issue. Depends builds should output individual log files for each package, so in the future we can diff those log files and have better introspection on what's failing, log files would also solve your problems here...
13:39 < ryanofsky> in general would a good approach be to send build output to a file and only display it if there is an error?
13:39 < dongcarl> Right, that's what I'm thinking
13:40 < dongcarl> MarcoFalke: Travis retains build artifacts even on failure, right? (I'm kinda clueless about travis)
13:41 < ryanofsky> i think all we currently get out of failed builds is output logs
13:42 < ryanofsky> could do something fancy, but probably just using cat or tail to send errors to output should be sufficient for most cases
13:42 < dongcarl> ryanofsky: Oh I see what you're saying
13:42 < dongcarl> Yeah that would work
13:45 < ryanofsky> Thanks for the help. It's not a big issue, I was mostly curious if travis "maximum log length" issue had come up before
13:45 < dongcarl> :-)
--- Log closed Thu Jul 11 00:00:07 2019