jruby runtimes and tomcat -


we running jruby/rails app under tomcat, , can not make shift threadsafe due of components using. in interim plan set jruby.runtime.min , jruby.runtime.max prevent perm-gen out of memory errors. unfortunately throttle number of concurrent users small number, question when runtime recycled? session based or else. lowetring session timeout speed availability of pools or there more that. links better understand specifics of how jruby runtime works appreciated.

first of there's bit of confusion in q - setting min/max runtime parameters not prevent perm-gen errors ... need increase memory bit (depending on application's footprint) you're seeing esp. if you're setting min/max values higher.

secondly - sure throttle concurrency not - think of mongrel (thin) cluster: concurrent requests handled >= number of running instances

a single runtime blocked duration of single request (does not depend @ on session state), after response returned runtime returned pool , ready handle requests. if request comes in , there's no runtime in pool try wait 1 become available before returning http error (can configured using jruby.runtime.acquire.timeout).

i recommend run or @ least (integration) test min = max since booting runtime on demand might costly (yet again application dependent) , @ same time it's harder guess memory requirement when boot rails instance on demand while serving. aware might take while boot since application needs loading on deploy min-times.


Comments

Popular posts from this blog

css - Which browser returns the correct result for getBoundingClientRect of an SVG element? -

gcc - Calling fftR4() in c from assembly -

.htaccess - Matching full URL in RewriteCond -