pcp
[Top] [All Lists]

Re: [pcp] problem with python version of pmclient

To: Ken McDonell <kenj@xxxxxxxxxxxxxxxx>
Subject: Re: [pcp] problem with python version of pmclient
From: Nathan Scott <nathans@xxxxxxxxxx>
Date: Sun, 31 Jan 2016 01:07:15 -0500 (EST)
Cc: PCP <pcp@xxxxxxxxxxx>
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <005901d15be5$b4ee2780$1eca7680$@internode.on.net>
References: <005901d15be5$b4ee2780$1eca7680$@internode.on.net>
Reply-to: Nathan Scott <nathans@xxxxxxxxxx>
Thread-index: AdFb5T52fNO1SnuJQzyRo9F/hjd9WqLUQtD8
Thread-topic: problem with python version of pmclient

----- Original Message -----
> On an MP machine ...
> 
> kenj@bozo-vm:~/src/pcp/qa$ python
> /usr/share/pcp/demos/pmclient/pmclient_fg.python
> Host: bozo-vm, 2 cpu(s), 2016-01-31 16:05:22.292435
>   CPU  Busy    Busy  Free Mem   Disk     Load Average
>  Util   CPU    Util  (Mbytes)   IOPS    1 Min  15 Min
>  0.00     0    0.00  3215.109      5     0.01    0.09
> 
> But on an SP machine the output is botched ...
> 
> kenj@vm01:~$ python /usr/share/pcp/demos/pmclient/pmclient_fg.python
> Host: vm01, 1 cpu(s), 2016-01-31 16:05:53.336159
>   CPU 0.00     9.160     10     0.43    0.57
>  0.00     9.168      0     0.40    0.57
>  0.00     9.168      0     0.37    0.57
> 
> Regular old pmclient produces the correct output in both cases.
> 
> This is causing my qa/056 failures (and I suspect qa/073 but I've not
> triaged that one).
> 

I've pushed through a fix, though no UP machine to test it on.

cheers.

--
Nathan

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