> Hello,
>
> Il 11/07/2012 11:20, Gelonida N ha scritto:
>> Hi,
>> I'm having a nginx / uwsgi / django server and wanted to get some
>> profiling data to identify bottle necks. (80% dynamic difficult to cache
>> contents, so don't worry too much about serving static contents)
>>
>> What would be recommended strategies to create decent reports, that can
>> help to pinpoint potentially critical code.
>>
>>
>> I posted another message in the django group for django specific
>> profiling.
>> Here I'd be mostly interested in uwsgi specic tools.
>
> I'd start finding issue in the application. You can start with django
> debug toolbar
>
>>
>> Number of held back uwsgi requests
>> --------------------------------------
>> How could I identify all requests, that are blocked due to the fact,
>> that I don't have enough uwsgi processes spawned.
>
> uwsgitop may help to track how busy are your workers,
> https://github.com/unbit/uwsgitop
>
>>
>> report of Django requests and their execution time
>> -----------------------------------------------------
>> Just use the uwsgi logs?
>
> Philip Gatt developed uwsgiFouine to extract nice formatted reports from
> uwsgi logs, i forked it to fix a couple of issues
> https://github.com/xrmx/uwsgiFouine .
>
>


I include the --profiler option too

--profiler pycall

--profiler pyline

they will write function timings in the logs. You can analize logs to find
bottleneck (not an easy task, but it should be doable investing a bit of
time in log parser)


-- 
Roberto De Ioris
http://unbit.it
_______________________________________________
uWSGI mailing list
[email protected]
http://lists.unbit.it/cgi-bin/mailman/listinfo/uwsgi

Reply via email to