pcp
[Top] [All Lists]

Re: [pcp] hinv.*

To: Nathan Scott <nathans@xxxxxxxxxx>, Michele Baldessari <michele@xxxxxxxxxx>
Subject: Re: [pcp] hinv.*
From: Mark Goodwin <mgoodwin@xxxxxxxxxx>
Date: Tue, 28 Oct 2014 16:49:52 +1100
Cc: pcp@xxxxxxxxxxx
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <1558759660.1356853.1414474824972.JavaMail.zimbra@xxxxxxxxxx>
References: <20141021222355.GC6656@xxxxxxxxxxxxxxx> <573592010.74725901.1413999659304.JavaMail.zimbra@xxxxxxxxxx> <54483934.4060802@xxxxxxxxxx> <20141023072636.GB6158@xxxxxxxxxxxxxxx> <1558759660.1356853.1414474824972.JavaMail.zimbra@xxxxxxxxxx>
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.7.0
On 10/28/2014 04:40 PM, Nathan Scott wrote:
Hi guys,

----- Original Message -----
[...]
Thanks Mark, the patch works fine, I see the different values now.
Given how dynamic CPUs can be these days, it might be nice to
add the online/offline status (/sys/devices/system/cpu/cpu1/online)
as well.

Good idea - I'll hack that up tomorrow unless someone beats me to it.

that could be opening a nice big juicy can of worms - would we also
want the cpu instance domain to become dynamic? Or keep it how it
is with online and offline CPUs included?

The default logging here is "once" for hinv, so that might need tweaking as
well.

Yeah, hmmm, thats a bit awkward - I don't really want us to have to
explicitly list each hinv metric in the logging configs.  Not sure
what else we can do though.

The 'atop-summary' logger config already includes it :

#+ tools/atop-summary:y:default:
## metrics sampled once by the atop command
log advisory on default {
        hinv.cpu.clock
        hinv.ncpu
        mem.physmem
}

Regards
-- Mark


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