cpu plugin: Various, random fixes and improvements:

[collectd.git] / src / collectd.conf.pod

diff --git a/src/collectd.conf.pod b/src/collectd.conf.pod
index ba99b60..d2e6843 100644 (file)
--- a/src/collectd.conf.pod
+++ b/src/collectd.conf.pod
@@ -1207,29 +1207,53 @@ F</proc/sys/net/ipv4/netfilter> instead of F</proc/sys/net/netfilter/>.
 
 =head2 Plugin C<cpu>
 
-The I<CPU plugin> collects CPU usage metrics.
+The I<CPU plugin> collects CPU usage metrics. By default, CPU usage is reported
+as Jiffies, using the C<cpu> type. Two aggregations are available:
+
+=over 4
+
+=item
+
+Sum, per-state, over all CPUs installed in the system; and
+
+=item
+
+Sum, per-CPU, over all non-idle states of a CPU, creating an "active" state.
+
+=back
+
+The two aggregations can be combined, leading to I<collectd> only emitting a
+single "active" metric for the entire system. As soon as one of these
+aggregations (or both) is enabled, the I<cpu plugin> will report a percentage,
+rather than Jiffies. In addition, you can request individual, per-state,
+per-CPU metrics to be reported as percentage.
 
 The following configuration options are available:
 
 =over 4
 
-=item B<ReportActive> B<false>|B<true>
+=item B<ReportByState> B<true>|B<false>
 
-Reports non-idle CPU usage as the "active" value. Defaults to false.
+When set to B<true>, the default, reports per-state metrics, e.g. "system",
+"user" and "idle".
+When set to B<false>, aggregates (sums) all I<non-idle> states into one
+"active" metric.
 
-=item B<ReportByCpu> B<false>|B<true>
+=item B<ReportByCpu> B<true>|B<false>
 
-When true reports usage for all cores. When false, reports cpu usage
-aggregated over all cores.
-Defaults to true.
+When set to B<true>, the default, reports per-CPU (per-core) metrics.
+When set to B<false>, instead of reporting metrics for individual CPUs, only a
+global sum of CPU states is emitted.
 
 =item B<ValuesPercentage> B<false>|B<true>
 
-When true report percentage usage instead of tick values. Defaults to false.
+This option is only considered when both, B<ReportByCpu> and B<ReportByState>
+are set to B<true>. In this case, by default, metrics will be reported as
+Jiffies. By setting this option to B<true>, you can request percentage values
+in the un-aggregated (per-CPU, per-state) mode as well.
 
 =back
 
-
 =head2 Plugin C<cpufreq>
 
 This plugin doesn't have any options. It reads
@@ -3890,6 +3914,12 @@ behavior is to let the kernel choose the appropriate interface. Be warned
 that the manual selection of an interface for unicast traffic is only
 necessary in rare cases.
 
+=item B<ResolveInterval> I<Seconds>
+
+Sets the interval at which to re-resolve the DNS for the I<Host>. This is
+useful to force a regular DNS lookup to support a high availability setup. If
+not specified, re-resolves are never attempted.
+
 =back
 
 =item B<E<lt>Listen> I<Host> [I<Port>]B<E<gt>>