Hi -
As per http://oss.sgi.com/bugzilla/show_bug.cgi?id=1093, the
pcpfans.git branch fche/pcp2graphite, this brings a supportable
version of this old ditty [1] into the tree.
[1] https://web.elastic.org/~fche/blog3/archive/2014/04/13/pcp-and-graphite.
commit a4a5d438bf356717329a2f2f8588f47bbb3e9271
Author: Frank Ch. Eigler <fche@xxxxxxxxxx>
Date: Sun Jan 18 15:54:20 2015 -0500
PCP PR/1093: add pcp2graphite
Modernized this old script from last April, after users requested its
proper inclusion in PCP. Hand-tested via fedora-21
graphite-web-0.9.12 (but beware fedora bug#1183270).
commit ff0ed079b7f875b794ab2f00539220d80931a5db
Author: Frank Ch. Eigler <fche@xxxxxxxxxx>
Date: Sun Jan 18 15:15:39 2015 -0500
pmParseUnitsStr: python binding
Add a pmapi context pmParseUnitsStr(string) -> (units,multiplier)
frontend to the python bindings. Add test case, confirming a sample
success case, an error case, and a ctypes type-error case. (The
latter is entertaining: if it were not for an isinstance() check in
the python binding, passing an integer parameter to the underlying
LIBPCP.pmParse$foo function could cast it to a char*, and cause a
SEGV. Something worth checking on other python bindings.)
|