pcp
[Top] [All Lists]

Re: [pcp] Derived metric issues

To: Nathan Scott <nathans@xxxxxxxxxx>, Ken McDonell <kenj@xxxxxxxxxxxxxxxx>
Subject: Re: [pcp] Derived metric issues
From: Marko Myllynen <myllynen@xxxxxxxxxx>
Date: Wed, 10 Feb 2016 10:50:57 +0200
Cc: pcp developers <pcp@xxxxxxxxxxx>
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <1781835934.19755039.1455052728610.JavaMail.zimbra@xxxxxxxxxx>
Organization: Red Hat
References: <56B99888.2020408@xxxxxxxxxx> <56BA4445.2030404@xxxxxxxxxxxxxxxx> <56BA4880.1000607@xxxxxxxxxx> <1781835934.19755039.1455052728610.JavaMail.zimbra@xxxxxxxxxx>
Reply-to: Marko Myllynen <myllynen@xxxxxxxxxx>
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.5.0
Hi,

On 2016-02-09 23:18, Nathan Scott wrote:
> ----- Original Message -----
>>> [...]
>>> pmRegisterDerived does not apply retrospectively to any open contexts,
>>> so the normal use would be to make all calls to pmRegisterDerived
>>> (possibly via pmLoadDerivedConfig(3)) and then call pmNewContext(3).
>>
>> A-ha!
>>
>>> I'd welcome any suggestions as to how this important piece of
>>> information could be made more visible.
>>
>> After reading the above now it's of course obvious.. Perhaps a short
>> note in the first section of pmLoadDerivedConfig(3) along the lines
>> "Note that pmLoadDerivedConfig needs to be called before creating a new
>> context" or something like that would be enough.
> 
> pmReconnectContext(3) can also be used, I believe (that's what the python
> wrapper uses anyway).

Btw, FWIW, I remembered that I mentioned in the past another (unrelated)
case where the man pages could be clarified a bit wrt calling order:

http://oss.sgi.com/pipermail/pcp/2015-November/008752.html

Thanks,

-- 
Marko Myllynen

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