<JRubyGithub>
[jruby] enebo pushed 2 new commits to master: http://git.io/vq124
<JRubyGithub>
jruby/master ff331eb Thomas E. Enebo: Bump for version
<JRubyGithub>
jruby/master 452dfcf Thomas E. Enebo: Merge branch 'master' of github.com:jruby/jruby
JRubyGithub has left #jruby [#jruby]
nirvdrum has quit [Ping timeout: 276 seconds]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] egadstar opened issue #3123: private method error in jruby-9.0.0.0.rc1 http://git.io/vq1oD
JRubyGithub has left #jruby [#jruby]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] enebo closed issue #3059: Windows bin/*.bat files missing in 9.0.0.0.rc1 http://git.io/vL3rt
JRubyGithub has left #jruby [#jruby]
samphippen has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
donV has joined #jruby
anaeem1_ has quit [Remote host closed the connection]
anaeem1 has joined #jruby
dumdedum has quit [Ping timeout: 244 seconds]
lanceball is now known as lance|afk
nirvdrum has joined #jruby
jrubyuser has joined #jruby
anaeem1 has quit [Ping timeout: 246 seconds]
<jrubyuser>
hello, running jruby on tomcat (redmine) it starts a lot of ruby-fibers threads, tried to use "Djruby.thread.pooling=true -Djruby.thread.pool.enabled=true" in my tomcat startup but that didn't work, that is killing my tomcat instance, how do I fix this?
<rtyler>
enebo: bintray has been quite good for me
<rtyler>
enebo: the gradle bintray plugin is pretty helpful, I recently automated the releases to bintray for one of Lookout's projects, which was neat
<enebo>
rtyler: you pay or use OSS? freebie?
<rtyler>
freebie!
<rtyler>
the fact that it links to jcenter is also useful
<rtyler>
so jcenter is a superset of maven central
<rtyler>
and it's easy to submit packages there
<enebo>
rtyler: do you also use sonatype oss?
<rtyler>
no
<rtyler>
sonatype burned bridges with me during the departure from oracle for jenkins <_<
<enebo>
I see
<enebo>
mpapis: 9.0.0.0.rc2 buddy
<rtyler>
enebo: we use artifactory at Lookout, and suffice it to say I'm relatively happy with jfrog's support and tooling
<enebo>
rtyler: we use conatype for all our jruby-related stuff so I may continue just to keep to one tool
<rtyler>
functionally they're not really different if you're using a generic maven release plugin
<enebo>
mkristian__: to add a new groupid I need to send open an issue on sonatype? I want org.purugin
<rtyler>
bintray's fairly self-service too by the way
<enebo>
rtyler: yeah I know sonatype is in the -P magic so I suspect it is not generic
<rtyler>
rules only start to apply when you want to tie into jcenter
<rtyler>
I've got all kinds of arbitrary maven repos up there
<enebo>
rtyler: I think these requests somehow are just for pub/priv key acknowledgement but I don’t understand why if I am already pushing one group I cannot push another
<enebo>
e.g. new groupid
subbu has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] nirvdrum opened issue #3124: [Truffle] Threads may not arrive at safepoint and prevent shutdown. http://git.io/vq1pA
JRubyGithub has left #jruby [#jruby]
tvo1 has joined #jruby
shellac has quit [Quit: Computer has gone to sleep.]
<Antiarc>
Morning folks
<Antiarc>
And grats on rc2!
<headius>
jrubyuser: the fiber threads are killing your tomcat?
donV has quit [Quit: donV]
djbkd has quit [Remote host closed the connection]
djbkd has joined #jruby
samphippen has joined #jruby
x1337807x has joined #jruby
pietr0 has joined #jruby
nateberkopec has joined #jruby
shellac has joined #jruby
x1337807x has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
yfeldblum has joined #jruby
<jrubyuser>
jruby is starting unlimited amount of thread (ruby-fiber), how do I limit that?
<headius>
jrubyuser: it should only be spinning up as many threads as it needs
<Antiarc>
Do you know what code is starting all the fibers?
<jrubyuser>
i'm using redmine on tomcat
<headius>
if it's spinning through a lot of fiber threads you may be able to adjust their time to live with fiber.thread.pool.ttl
lance|afk is now known as lanceball
<headius>
default is 60 seconds
<headius>
jrubyuser: how many threads are we talking?
<headius>
and how are you deploying it? mutliple JRuby instances or a single instance?
<jrubyuser>
not sure, I believe it is a single instance
<headius>
how many threads?
<jrubyuser>
depends on server load, but sometimes it gets heavy load, i have other webapp in the same tomcat that need priority
blaines has joined #jruby
<jrubyuser>
if redmine user gets a timeout response or some other error it doesn't matter, as long as the other webapps are responding
<headius>
well you might try that ttl setting, see if it helps keep the number of threads down
<jrubyuser>
so if I use -Dfiber.thread.pool.ttl=10 it will destroy the idle fiber in 10 seconds? is there a fiber.thread.pool.maxthread or something like that?
<headius>
but the pool should grow to the max number of live fibers and that's all
<headius>
there's no maxthread because we'd need to produce a hard error when trying to start a fiber above that max
<headius>
and yes, ttl=10 will kill idle thread fibers after 10s
<headius>
my suspicion is that something in redmine is probably starting fibers and not completing them, so they have to wait for a few GC cycles to shut down
<jrubyuser>
just let me understand something, i'm not a ruby coder, fiber is not a thread?
<jrubyuser>
I see a lot of "ruby-0-fiber-???" in jvm's thread list
blaines__ has joined #jruby
blaines has quit [Ping timeout: 248 seconds]
<jrubyuser>
because of that i thought fiber was some sort of thread implementation for some specific purpose
<mkristian__>
enebo, yes you do need to submit an issue on oss.sonatype
<Antiarc>
jrubyuser: fibers are implemented as threads on jruby. In Ruby though, they're kind of like cooperatively-scheduled lightweight threads
<jrubyuser>
I see
<enebo>
mkristian__: thanks
<Antiarc>
In Ruby, they're commonly associated with Celluloid and EventMachine. I'm not sure if Redmine is using either
<jrubyuser>
-Djruby.thread.pooling=true or -Djruby.thread.pool.enabled=true should have any effect on that?
<jrubyuser>
jruby docs (about performance) show both options, seems misdirecting information
<headius>
jrubyuser: those were used for a thread pool to back "normal" threads that doesn't exist anymore
<headius>
it wasn't useful and didn't have any real benefits
<jrubyuser>
creating and destroying fiber is expensive task? is there minimun alive fibers setting?
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] mkristian pushed 1 new commit to master: http://git.io/vqMpf
<JRubyGithub>
jruby/master 6892884 Christian Meier: adjust the working directory on test_load_compiled_ruby
JRubyGithub has left #jruby [#jruby]
<jrubyuser>
thank you for all the help, I will apply that option, it should help
jrubyuser has quit [Quit: Page closed]
<enebo>
asarih: what is the name of the beer place you were at but I have not been? :)