00:14 <headius[m]> daveg_lookout: I landed the fix on 9.2 branch and master

00:32 <travis-ci> jruby/jruby (jruby-9.2:3448088 by Charles Oliver Nutter): The build is still failing. https://travis-ci.com/jruby/jruby/builds/212931826 [174 min 3 sec]

00:32 <headius[m]> Hmm

15:39 <headius[m]> good morning!

15:44 <travis-ci> jruby/jruby (jruby-9.2:3448088 by Charles Oliver Nutter): The build is still failing. https://travis-ci.com/jruby/jruby/builds/212931826 [174 min 4 sec]

15:46 <daveg_lookout[m]> good morning! we're running with monkeypatched Monitor in staging now, running some perf tests later today, and will get it to prod tomorrow. Thanks again for the fast debugging and fix!

15:47 <headius[m]> Thank you for the updated testing! This is a pretty big bug in the monitor library so it's great that we have found it

16:18 <headius[m]> daveg_lookout: were you ever able to pull a heap dump>

17:03 <travis-ci> jruby/jruby (jruby-9.2:c862b52 by Charles Oliver Nutter): The build is still failing. https://travis-ci.com/jruby/jruby/builds/213022700 [167 min 24 sec]

17:04 <headius[m]> no

17:41 <travis-ci> jruby/jruby (jruby-9.2:93e3865 by Charles Oliver Nutter): The build is still failing. https://travis-ci.com/jruby/jruby/builds/213025406 [169 min 1 sec]

17:42 <headius[m]> flaky

17:47 <travis-ci> jruby/jruby (jruby-9.2:93e3865 by Charles Oliver Nutter): The build is still failing. https://travis-ci.com/jruby/jruby/builds/213025406 [169 min 23 sec]

17:54 <travis-ci> jruby/jruby (jruby-9.2:e3404e8 by Charles Oliver Nutter): The build is still failing. https://travis-ci.com/jruby/jruby/builds/213034248 [169 min 5 sec]

19:55 <headius[m]> enebo: help a bro out and verify this branch passes your release smoke test on Windows: https://github.com/jruby/jruby/pull/6532

19:56 <headius[m]> it basically tries to unify some argument and executable handling between popen and system (JDK logic for Windows) to fix https://github.com/jruby/jruby/issues/6516

19:57 <headius[m]> I was going to try to get Windows using more of the native IO and process stuff but ran into the big blocker: jnr-enxio only supports epoll and kqueue, and Windows only has select (with its macro grossness) or WaitForSingleObject

20:05 <headius[m]> I thought I was doing it right but it tried to install the bindex C lib and crapped out

20:05 <headius[m]> everything worked up to that point though

20:05 <enebo[m]> ok I will see if that works at least

20:05 <headius[m]> I haven't looked at your script recently

20:05 <enebo[m]> Rails 5.2 I think is what I test on also because of the sassc stuff

20:05 <enebo[m]> I think it is the same script but I use a different version of Rails

20:06 <headius[m]> yeah I forced it to 6.0.0 to get the ruby sass

20:06 <headius[m]> but I guess bindex is an issue too

20:08 <enebo[m]> I think bindex was fixes after 6.0.0

20:08 <headius[m]> oh I see now it installed rails 6.0.0 but when it generates app it does not stick to that version

20:08 <headius[m]> so rails new ended up installing 6.0.3.4

20:24 <headius[m]> enebo: I did that fix against 9.2.15 but we could make it 9.3 if it seems too risky

20:25 <enebo[m]> headius: if there is no detected problems we can consider what other things may raise that message

20:26 <headius[m]> the message fix should be benign

20:26 <headius[m]> the popen changes less so

21:06 <daveg_lookout[m]> headius: no, sorry :-( After about an hour I stopped paying attention, then forgot about it when we deployed new instances with the monkeypatched Monitor. That code is in production now. We'll probably increase traffic to those instances tomorrow

21:50 <enebo[m]> rails is fine and popen is working as well

22:19 <headius[m]> daveg_lookout: well no matter, we will see if it behaves better now

22:20 <headius[m]> enebo: ok I will go ahead and merge that then

22:47 <travis-ci> jruby/jruby (jruby-9.2:cdc3c8c by Charles Oliver Nutter): The build is still failing. https://travis-ci.com/jruby/jruby/builds/213072257 [169 min 25 sec]