Changes committed to git://git.pcp.io/kenj/pcp master
Ken McDonell (4):
qa/845: additional diagnostics
qa/src/statvfs.c: tool to report stafs() and statvfs() results
qa/src/statvfs.c: oops, add copyright.
mounts PMDA: correct a number of 32-bit unsafe calculations
qa/845 | 3 +-
qa/src/GNUlocaldefs | 6 ++++-
qa/src/statvfs.c | 54 ++++++++++++++++++++++++++++++++++++++++++++++
src/pmdas/mounts/mounts.c | 10 ++++----
4 files changed, 66 insertions(+), 7 deletions(-)
Details ...
commit 2099a1b9ba002bf3c9f944469b1513c70754c037
Author: Ken McDonell <kenj@xxxxxxxxxxxxxxxx>
Date: Thu Oct 22 11:08:08 2015 +1100
mounts PMDA: correct a number of 32-bit unsafe calculations
qa/845 has been failing for a long time on:
vm01 3.10.6 i686 Ubuntu 12.10 (quantal)
vm02 3.10.6 i686 openSUSE 13.2 (Harlequin)
vm04 3.10.6 i586 CentOS 5.11 (Final)
vm12 3.10.6 i686 Fedora 22
vm21 3.10.6 i686 Debian 7.4
vm23 3.10.6 i686 Fedora 20 (Heisenbug)
vm24 3.10.6 i686 openSUSE 13.1 (Bottle)
vm25 3.10.6 x86_64 CentOS 5.11 (Final)
All but the last are 32-bit platforms ... inspection of the code
shows several expressions that can produce overflow if performed
with 32-bit integers. This commit adds appropriate casts to
promote the arithmetic to __uint64_t (to match the PCP metric
values).
Note: failures on vm25 and vm04 are in fact Python syntax errors,
so a different fix is needed there.
commit 3097d9c7bdb0a8372dbf8cdc14d6424d51d080ab
Author: Ken McDonell <kenj@xxxxxxxxxxxxxxxx>
Date: Thu Oct 22 11:06:20 2015 +1100
qa/src/statvfs.c: oops, add copyright.
commit 65ccd609d367035cd74a6e06af691821f39db7c0
Author: Ken McDonell <kenj@xxxxxxxxxxxxxxxx>
Date: Thu Oct 22 11:04:26 2015 +1100
qa/src/statvfs.c: tool to report stafs() and statvfs() results
Used to debug mounts PMDA ... may be useful in another context,
so commit rather than toss.
commit d3f5e7539be54b1a36a85850aaeebf450e5b71e4
Author: Ken McDonell <kenj@xxxxxxxxxxxxxxxx>
Date: Thu Oct 22 11:02:12 2015 +1100
qa/845: additional diagnostics
|