codegl
commented
11 years ago Hi Matthias,
Unfortunately it is possible to have 'too much' stuff going on, you need to add a few -no: flags on the commandline to filter it down. I'm working on making the app less memory intensive by keeping the traces in memory in typed arrays, but right now your best bet is to just add some commandline filters.
Rik Arends
TraceGL On Apr 23, 2013, at 7:50 PM, Mathias Pettersson notifications@github.com wrote:
I have a small application where the intensive parts consist of:
express (some basic routes that more or less just respond with static JSON) uglify-js (minifying quite a bit of client side javascript) less (parses the less files into css) When I try to start the app with node trace server.js it spikes the CPU at 100%, eats around 1.2GB memory and then stays there for a while.
After the cpu usage dies down you can't access *:2000 and the app doesn't respond to http traffic.
What is sent to the console: https://gist.github.com/mape/1cb2a722e7b25d5a5555
— Reply to this email directly or view it on GitHub.
I have a small application where the intensive parts consist of:
When I try to start the app with node tracegl server.js it spikes the CPU at 100%, eats around 1.2GB memory and then stays there for a while.
After the cpu usage dies down you can't access *:2000 and the app doesn't respond to http traffic.
What is sent to the console: https://gist.github.com/mape/1cb2a722e7b25d5a5555
When running with -nolib it loads the app but still ends up with: FATAL ERROR: JS Allocation failed - process out of memory