pcp
[Top] [All Lists]

Re: [pcp] python QA/709 failures

To: Ken McDonell <kenj@xxxxxxxxxxxxxxxx>
Subject: Re: [pcp] python QA/709 failures
From: Stan Cox <scox@xxxxxxxxxx>
Date: Thu, 04 Apr 2013 14:42:39 -0400
Cc: Nathan Scott <nathans@xxxxxxxxxx>, PCP Mailing List <pcp@xxxxxxxxxxx>
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <514F89FC.5020102@xxxxxxxxxxxxxxxx>
References: <1086370001.24034377.1364166803775.JavaMail.root@xxxxxxxxxx> <514F89FC.5020102@xxxxxxxxxxxxxxxx>
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130110 Thunderbird/17.0.2
On 03/24/2013 07:19 PM, Ken McDonell wrote:
failing cases are single CPU machines.

This hopefully catches all the uniprocessor accesses. The change is to my development version which has the pcp accesses split out into a module pmsubsys.py.

commit c6bfc687d70e67df54cb9f176c1dda84bffacb1c
Author: Stan Cox <scox@xxxxxxxxxx>
Date:   Tue Apr 2 21:45:03 2013 -0400

    Test playback.  Always fetch scalars.

    * qa/709: Add playback testing.
    * pmsubsys.py (get_scalar_value): Also fetch the value.
    * pmatop.py: Use get_scalar_value.
    * pmcollectl.py: Use get_scalar_value.
      (main): Improve unimplemented subsystem handling.

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