pcp
[Top] [All Lists]

RE: proc.psinfo.wchan_s broken?

To: "Siekas, Greg" <greg.siekas@xxxxxxxxxx>
Subject: RE: proc.psinfo.wchan_s broken?
From: Ken McDonell <kenj@xxxxxxxxxxxxxxxx>
Date: Fri, 05 Sep 2008 09:55:13 +1000
Cc: Nathan Scott <nscott@xxxxxxxxxx>, pcp@xxxxxxxxxxx
In-reply-to: <258BABDDC780384FBF0FF15CCFDFABE10673EDA3@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>
References: <258BABDDC780384FBF0FF15CCFDFABE10673E89D@xxxxxxxxxxxxxxxxxxxxxxxxxxxx> <1220511873.4544.33.camel@xxxxxxxxxxxxxxxxxx> <258BABDDC780384FBF0FF15CCFDFABE10673EDA3@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>
Reply-to: kenj@xxxxxxxxxxxxxxxx
Sender: pcp-bounce@xxxxxxxxxxx
On Thu, 2008-09-04 at 07:09 -0700, Siekas, Greg wrote:
> Nathan,
> 
...
> 
> Is proc.psinfo.wchan  also broken for newer kernels?
> 
> Is there a list somewhere that indicates which metrics do and don't work
> for various operating systems and kernels?

In general there should not need to be any such list ... the PCP agents
should be maintained in such a way that if a specific metric is not
available on a particular platform, the agent should return
PM_ERR_APPVERSION which was designed into the PCP protocols for just
this situation.

So the metacode should read like

if kernel_version == 2.6.18
        use method A
elif kernel_version < 2.6.18 && kernel_version >= 2.6.0
        use method B
else
        return PM_ERR_APPVERSION

Unfortunately, this style has not be rigorously followed by the
generations of PMDA maintainers over many years, so this works in some
places, and not elsewhere.


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