2013/1/15 Marcin Deranek <[email protected]>

> 1. Hostnames with dots
> In our environment (CentOS/RedHat) hostnames are FQDN as recommended by
> RedHat/CentOS(*) This does not play nicely with carbon metric names as
> dot is a namespace separator.
>

dots as separator are carbon/graphite feature, but if uWSGI uses fqdn as
default name it should replace all dots with underscores.
But I doubt that it does, I bealive that it uses hostname not fqdn - those
are two different things, and AFAIK hostname should not contain domain,
fqdn is for whole thing.
Verify:
`hostname` - should return host name without domain (for example node1)
`hostname -f` - should return whole fqdn (for example node1.domain.com)


> 2. Different namespace
> Currently carbon plugin does not allow me to put my metrics in
> arbitrary location (besides I found default metric path a bit too long
> for my needs).
> Maybe extra option(?) to allow to override base path for metrics ?
>

We could use extra options for specificity graphs root.


> 3. Missing metrics
> When I look at stats server output I see quite some metrics there which
> are not present in carbon plugin and I find them useful eg.
>
> respawn_count
> exceptions
>
> Probably other people might find some other metrics important for them.
>

Wait for metrics subsystem, it's gonna be real easy to add anything (
https://github.com/unbit/uwsgi/issues/24)


> 4. Busy / idle workers
> Currently we only have busyness which is not exactly what busy/idle
> is eg. yesterday one of our instances locked up, so all workers were
> busy and uwsgi couldn't dispatch any requests to workers (no request
> was served by that time period). During that time busyness was 0 (most
> likely due to the fact that no request was served during that period).
>

If I remeber correctly busyness is calculated by counting time needed for
requests which were completed during last carbon interval. If there was no
*complated* requests, than we can't count it, so we get 0.
In case of carbon stats it is a minor issue, but busyness is also used in
cheaper_busyness plugin so I will try to improve calculation in such case.


> 5. Empty metrics
> Some of the metrics are generated even though their value is always 0
> (eg. rss/vsz) as those only get updated in certain circumstances
> (enabled resource limiting base don memory footprint or memory usage
> reporting). In such circumstances I don't think it makes sense to
> generate them at all.

6. Calculation for cheaped / active workers
> I see that in some cases we calculate some total metrics (regading
> whole instance) for all (even cheaped) workers and some only for active
> workers (process is running).
> I would say that eg. harakiri metric should be counted over all
> rather than active only workers as it's a counter.
>

Those are easy to fix, make a pull request or just report a issue and I
will try to fix it.

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

Reply via email to