pcp
[Top] [All Lists]

Re: [pcp] fetchgroups api - python bindings

To: Mark Goodwin <mgoodwin@xxxxxxxxxx>, "Frank Ch. Eigler" <fche@xxxxxxxxxx>
Subject: Re: [pcp] fetchgroups api - python bindings
From: Marko Myllynen <myllynen@xxxxxxxxxx>
Date: Sun, 13 Dec 2015 17:10:00 +0200
Cc: pcp@xxxxxxxxxxx
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <566A59A5.6090403@xxxxxxxxxx>
Organization: Red Hat
References: <20151206204742.GC22561@xxxxxxxxxx> <5666E4F7.4070005@xxxxxxxxxx> <y0m4mft10vf.fsf@xxxxxxxx> <566A59A5.6090403@xxxxxxxxxx>
Reply-to: myllynen@xxxxxxxxxx
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.4.0
Hi,

On 2015-12-11 07:05, Mark Goodwin wrote:
> 
> - how come no support for PM_TYPE_AGGREGATE (and events)?

I didn't pay attention to this earlier but after Lukas pointed out
pmdapipe which could be used to get e.g. perf(1) data alongside with
other metrics, it occurred to me that if pmfg would support events as
well then with pmrep(1) we could perhaps create metric set definitions
covering everything up from the application level (e.g. Apache, MySQL,
Samba, ...) down to HW/tracepoint level (as needed), that would be
pretty interesting.

But I don't consider that as a blocker, more like a wishlist item.

Thanks,

-- 
Marko Myllynen

<Prev in Thread] Current Thread [Next in Thread>