I've merged Marc's Python change and this allowed me to revert the band-aid fix
for qa/709.
One quick and non-scientific test suggests that the values reported by
pmcollectl and collectl are back in agreement again.
Changes committed to git://git.pcp.io/kenj/pcp master
Ken McDonell (2):
qa/152: more diags to help debug intermittent failures
qa/709: revert "field overflow" changes
Marc Ma (1):
[python-pcp] Fix a bug in pmsubsys which causes timestamp to reset.
qa/152 | 7 +++++--
qa/152.out | 2 --
qa/709 | 5 -----
src/python/pcp/pmsubsys.py | 2 +-
4 files changed, 6 insertions(+), 10 deletions(-)
Details ...
commit c16f35edd878850cebab43aca5d94d9ecf2fa0de
Author: Ken McDonell <kenj@xxxxxxxxxxxxxxxx>
Date: Fri Jul 17 07:06:34 2015 +1000
qa/709: revert "field overflow" changes
Revert the part of commit 68c462a that split long sequences of
numbers to address a perceived field width overflow issue.
As mamarc@xxxxxxxxxx has diagnosed, this is a problem in the
base PCP Python code, and with his patch in place, the band-aid
for qa/709 is no longer required.
commit c0da055a11f5e3f43e90b2385f6100f09b1275a8
Author: Marc Ma <mamarc@xxxxxxxxxx>
Date: Thu Jul 16 12:03:45 2015 -0700
[python-pcp] Fix a bug in pmsubsys which causes timestamp to reset.
commit 8591191eade21f9bb8a88fea23ede1c612bcda54
Author: Ken McDonell <kenj@xxxxxxxxxxxxxxxx>
Date: Thu Jul 16 20:13:50 2015 +1000
qa/152: more diags to help debug intermittent failures
|