pcp
[Top] [All Lists]

Re: [pcp] pmcc.py bug?

To: Mark Goodwin <mgoodwin@xxxxxxxxxx>
Subject: Re: [pcp] pmcc.py bug?
From: Nathan Scott <nathans@xxxxxxxxxx>
Date: Thu, 5 Jun 2014 01:59:09 -0400 (EDT)
Cc: pcp <pcp@xxxxxxxxxxx>
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <5390030A.2000303@xxxxxxxxxx>
References: <5386E607.2020007@xxxxxxxxxx> <631390734.19335445.1401783604765.JavaMail.zimbra@xxxxxxxxxx> <5390030A.2000303@xxxxxxxxxx>
Reply-to: Nathan Scott <nathans@xxxxxxxxxx>
Thread-index: fTJfvwbrhIDDC6QxcZAGC4CLCTgLgQ==
Thread-topic: pmcc.py bug?
Hi Mark,

----- Original Message -----
> On 06/03/2014 06:20 PM, Nathan Scott wrote:
> [ ... ]
> >
> > ... with as much common code pushed into pmcc as I sensibly could.  I
> > think there's still a little work that the script is doing that could
> > be pulled into pmcc as well, but for now this should keep you moving.
> >
> 
> thanks, I'll use this as the basis for pcp-iostat.py and will update
> pmcc with any further commonality as it emerges (e.g. rate calculations,

OK, great.

> group fetch time deltas, etc).

I snuck that in already when you weren't looking (well, --interval support
anyway, fair bit more work to allow different groups with different deltas
but that may be overkill for the kinds of tools that'd use this?)  The run
method on the MetricGroupManager class does provide a typical fetch loop,
anyway, automating --interval/--samples/--start/--finish handling.

cheers.

--
Nathan

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