Show HN: Real-time, top-like metrics for Nginx

Post by stefantalpalaru » Fri, Mar 21, 2014, 6:39 PM UTC

scrollaway wrote: post #18

Very cool tool. It seems to be clearing the screen every time it runs though which isn't great, especially when it crashes that hides the backtrace. Consider using ncurses? On a different note... developers, please use Python 3. Several Linux distributions are now using Py3 as their main python and those scripts with an "env python" hashbang will not work. Porting your codebase to be python 2+3 compatible is dead sim…

Post by bndr » Fri, Mar 21, 2014, 7:11 PM UTC

diziet wrote: post #20

lebinh wrote: post #16

Earlier quoted context omitted.

Thanks, I'll definitely add that, maybe put this on pypi asap. For now you can clone the repo and then run pip install -r requirements.txt to install required dependencies.

Doesn't seem to like to read the logs : ./ngxtop.py -l /opt/nginx/logs/access.log ## just pretends like the log is not being written to For some reason the access log file is defaulted to logs/access.log too~

Post by jamra » Fri, Mar 21, 2014, 7:47 PM UTC

evilduck wrote: post #12

korzun wrote: post #3

Am I the only one that bothered when tools like this do not give you some insight on resource consumption/limit? Even if it's just parsing logs via sqlite, I still want to see the limitations and X per Y usage metrics to see if it can handle monitoring high traffic domain without impacting performance.

You're seriously bothered about the robustness of a 2 hour old project?

Post by cheald » Fri, Mar 21, 2014, 9:47 PM UTC

Post by babuskov » Sat, Mar 22, 2014, 12:48 AM UTC