1 =head1 NAME
3 rrdtune - Modify some basic properties of a Round Robin Database
5 =head1 SYNOPSIS
7 B<rrdtool> B<tune> I<filename>
8 S<[B<--heartbeat>|B<-h> I<ds-name>:I<heartbeat>]>
9 S<[B<--minimum>|B<-i> I<ds-name>:I<min>]>
10 S<[B<--maximum>|B<-a> I<ds-name>:I<max>]>
11 S<[B<--data-source-type>|B<-d> I<ds-name>:I<DST>]>
12 S<[B<--data-source-rename>|B<-r> I<old-name>:I<new-name>]>
13 S<[B<--deltapos> I<scale-value>]>
14 S<[B<--deltaneg> I<scale-value>]>
15 S<[B<--failure-threshold> I<failure-threshold>]>
16 S<[B<--window-length> I<window-length>]>
17 S<[B<--alpha> I<adaption-parameter>]>
18 S<[B<--beta> I<adaption-parameter>]>
19 S<[B<--gamma> I<adaption-parameter>]>
20 S<[B<--gamma-deviation> I<adaption-parameter>]>
21 S<[B<--smoothing-window> I<fraction-of-season>]>
22 S<[B<--smoothing-window-deviation> I<fraction-of-season>]>
23 S<[B<--aberrant-reset> I<ds-name>]>
25 =head1 DESCRIPTION
27 The tune option allows you to alter some of the basic configuration
28 values stored in the header area of a Round Robin Database (B<RRD>).
30 One application of the B<tune> function is to relax the
31 validation rules on an B<RRD>. This allows to fill a new B<RRD> with
32 data available in larger intervals than what you would normally want
33 to permit. Be very careful with tune operations for COMPUTE data sources.
34 Setting the I<min>, I<max>, and I<heartbeat> for a COMPUTE data source
35 without changing the data source type to a non-COMPUTE B<DST> WILL corrupt
36 the data source header in the B<RRD>.
38 A second application of the B<tune> function is to set or alter parameters
39 used by the specialized function B<RRAs> for aberrant behavior detection.
41 =over 8
43 =item I<filename>
45 The name of the B<RRD> you want to tune.
47 =item S<B<--heartbeat>|B<-h> I<ds-name>:I<heartbeat>>
49 modify the I<heartbeat> of a data source. By setting this to a high
50 value the RRD will accept things like one value per day.
52 =item S<B<--minimum>|B<-i> I<ds-name>:I<min>>
54 alter the minimum value acceptable as input from the data source.
55 Setting I<min> to 'U' will disable this limit.
57 =item S<B<--maximum>|B<-a> I<ds-name>:I<max>>
59 alter the maximum value acceptable as input from the data source.
60 Setting I<max> to 'U' will disable this limit.
62 =item S<B<--data-source-type>|B<-d> I<ds-name>:I<DST>>
64 alter the type B<DST> of a data source.
66 =item S<B<--data-source-rename>|B<-r> I<old-name>:I<new-name>>
68 rename a data source.
70 =item S<B<--deltapos> I<scale-value>>
72 Alter the deviation scaling factor for the upper bound of the
73 confidence band used internally to calculate violations for the
74 FAILURES B<RRA>. The default value is 2. Note that this parameter is
75 not related to graphing confidence bounds which must be specified as a
76 CDEF argument to generate a graph with confidence bounds. The graph
77 scale factor need not to agree with the value used internally by the
78 FAILURES B<RRA>.
80 =item S<B<--deltaneg> I<scale-value>>
82 Alter the deviation scaling factor for the lower bound of the confidence band
83 used internally to calculate violations for the FAILURES B<RRA>. The default
84 value is 2. As with B<--deltapos>, this argument is unrelated to the scale
85 factor chosen when graphing confidence bounds.
87 =item S<B<--failure-threshold> I<failure-threshold>>
89 Alter the number of confidence bound violations that constitute a failure for
90 purposes of the FAILURES B<RRA>. This must be an integer less than or equal to
91 the window length of the FAILURES B<RRA>. This restriction is not verified by
92 the tune option, so one can reset failure-threshold and window-length
93 simultaneously. Setting this option will reset the count of violations to 0.
95 =item S<B<--window-length> I<window-length>>
97 Alter the number of time points in the temporal window for determining
98 failures. This must be an integer greater than or equal to the window
99 length of the FAILURES B<RRA> and less than or equal to 28. Setting
100 this option will reset the count of violations to 0.
102 =item S<B<--alpha> I<adaption-parameter>>
104 Alter the intercept adaptation parameter for the Holt-Winters
105 forecasting algorithm. This parameter must be between 0 and 1.
107 =item S<B<--beta> I<adaption-parameter>>
109 Alter the slope adaptation parameter for the Holt-Winters forecasting
110 algorithm. This parameter must be between 0 and 1.
112 =item S<B<--gamma> I<adaption-parameter>>
114 Alter the seasonal coefficient adaptation parameter for the SEASONAL
115 B<RRA>. This parameter must be between 0 and 1.
117 =item S<B<--gamma-deviation> I<adaption-parameter>>
119 Alter the seasonal deviation adaptation parameter for the DEVSEASONAL
120 B<RRA>. This parameter must be between 0 and 1.
122 =item S<B<--smoothing-window> I<fraction-of-season>>
124 Alter the size of the smoothing window for the SEASONAL B<RRA>. This must
125 be between 0 and 1.
127 =item S<B<--smoothing-window-deviation> I<fraction-of-season>>
129 Alter the size of the smoothing window for the DEVSEASONAL B<RRA>. This must
130 be between 0 and 1.
132 =item S<B<--aberrant-reset> I<ds-name>>
134 This option causes the aberrant behavior detection algorithm to reset
135 for the specified data source; that is, forget all it is has learnt so far.
136 Specifically, for the HWPREDICT or MHWPREDICT B<RRA>, it sets the intercept and
137 slope coefficients to unknown. For the SEASONAL B<RRA>, it sets all seasonal
138 coefficients to unknown. For the DEVSEASONAL B<RRA>, it sets all seasonal
139 deviation coefficients to unknown. For the FAILURES B<RRA>, it erases the
140 violation history. Note that reset does not erase past predictions
141 (the values of the HWPREDICT or MHWPREDICT B<RRA>), predicted deviations (the
142 values of the DEVPREDICT B<RRA>), or failure history (the values of the
143 FAILURES B<RRA>). This option will function even if not all the listed
144 B<RRAs> are present.
146 Due to the implementation of this option, there is an indirect impact on
147 other data sources in the RRD. A smoothing algorithm is applied to
148 SEASONAL and DEVSEASONAL values on a periodic basis. During bootstrap
149 initialization this smoothing is deferred. For efficiency, the implementation
150 of smoothing is not data source specific. This means that utilizing
151 reset for one data source will delay running the smoothing algorithm
152 for all data sources in the file. This is unlikely to have serious
153 consequences, unless the data being collected for the non-reset data sources
154 is unusually volatile during the reinitialization period of the reset
155 data source.
157 Use of this tuning option is advised when the behavior of the data source
158 time series changes in a drastic and permanent manner.
160 =back
162 =head1 EXAMPLE 1
164 C<rrdtool tune data.rrd -h in:100000 -h out:100000 -h through:100000>
166 Set the minimum required heartbeat for data sources 'in', 'out'
167 and 'through' to 10'000 seconds which is a little over one day in data.rrd.
168 This would allow to feed old data from MRTG-2.0 right into
169 RRDtool without generating *UNKNOWN* entries.
171 =head1 EXAMPLE 2
173 C<rrdtool tune monitor.rrd --window-length 5 --failure-threshold 3>
175 If the FAILURES B<RRA> is implicitly created, the default
176 window-length is 9 and the default failure-threshold is 7. This
177 command now defines a failure as 3 or more violations in a temporal
178 window of 5 time points.
180 =head1 AUTHOR
182 Tobias Oetiker <tobi@oetiker.ch>