327dc0d6629813faf8ab0f71d0e026160f6f20a6
1 RRDGRAPH_RPN(1) rrdtool RRDGRAPH_RPN(1)
6 rrdgraph_rpn - About RPN Math in rrdtool graph
9 _\bR_\bP_\bN _\be_\bx_\bp_\br_\be_\bs_\bs_\bi_\bo_\bn:=_\bv_\bn_\ba_\bm_\be|_\bo_\bp_\be_\br_\ba_\bt_\bo_\br|_\bv_\ba_\bl_\bu_\be[,_\bR_\bP_\bN _\be_\bx_\bp_\br_\be_\bs_\bs_\bi_\bo_\bn]
14 stack and push your data onto this stack. Whenever you execute an oper-
15 ation, it takes as many elements from the stack as needed. Pushing is
16 done implicitly, so whenever you specify a number or a variable, it
17 gets pushed onto the stack automatically.
19 At the end of the calculation there should be one and only one value
20 left on the stack. This is the outcome of the function and this is
24 support a limited list of functions.
26 Example: "VDEF:maximum=mydata,MAXIMUM"
28 This will set variable "maximum" which you now can use in the rest of
29 your RRD script.
31 Example: "CDEF:mydatabits=mydata,8,*"
34 ator _\b*. The operator needs two elements and uses those to return one
36 guessed, this instruction means nothing more than _\bm_\by_\bd_\ba_\bt_\ba_\bb_\bi_\bt_\bs _\b= _\bm_\by_\bd_\ba_\bt_\ba _\b*
38 which order to process the input. For expressions like "a = b + 3 * 5"
40 with parentheses you could change this order: "a = (b + 3) * 5". In
44 Boolean operators
47 Pop two elements from the stack, compare them for the selected con-
54 tively to _\bp_\bo_\bs_\bi_\bt_\bi_\bv_\be _\bo_\br _\bn_\be_\bg_\ba_\bt_\bi_\bv_\be _\bi_\bn_\bf_\bi_\bn_\bi_\bt_\by. Returns 1 for true or 0
55 for false.
59 Pops three elements from the stack. If the element popped last is
60 0 (false), the value popped first is pushed back onto the stack,
61 otherwise the value popped second is pushed back. This does,
62 indeed, mean that any value other than 0 is considered to be true.
64 Example: "A,B,C,IF" should be read as "if (A) then (B) else (C)"
68 Comparing values
71 Pops two elements from the stack and returns the smaller or larger,
78 Pops two elements from the stack and uses them to define a range.
79 Then it pops another element and if it falls inside the range, it
82 The range defined includes the two boundaries (so: a number equal
83 to one of the boundaries will be pushed back). If any of the three
84 numbers involved is either _\bu_\bn_\bk_\bn_\bo_\bw_\bn or _\bi_\bn_\bf_\bi_\bn_\bi_\bt_\be this function will
88 lower than 0 or if it is higher than 100.
92 Arithmetics
95 Add, subtract, multiply, divide, modulo
99 NAN-safe addition. If one parameter is NAN/UNKNOWN it'll be treated
100 as zero. If both parameters are NAN/UNKNOWN, NAN/UNKNOWN will be
101 returned.
105 Sine and cosine (input in radians), log and exp (natural loga-
106 rithm), square root.
110 Arctangent (output in radians).
114 Arctangent of y,x components (output in radians). This pops one
115 element from the stack, the x (cosine) component, and then a sec-
116 ond, which is the y (sine) component. It then pushes the arctan-
117 gent of their ratio, resolving the ambiguity between quadrants.
119 Example: "CDEF:angle=Y,X,ATAN2,RAD2DEG" will convert "X,Y" compo-
120 nents into an angle in degrees.
124 Round down or up to the nearest integer.
128 Convert angle in degrees to radians, or radians to degrees.
132 Take the absolute value.
134 Set Operations
139 then sorted (or reversed) in place on the stack.
141 Example: "CDEF:x=v1,v2,v3,v4,v5,v6,6,SORT,POP,5,REV,POP,+,+,+,4,/"
142 will compute the average of the values v1 to v6 after removing the
143 smallest and largest.
147 Pop one element (_\bc_\bo_\bu_\bn_\bt) from the stack. Now pop _\bc_\bo_\bu_\bn_\bt elements and
148 build the average, ignoring all UNKNOWN values in the process.
150 Example: "CDEF:x=a,b,c,d,4,AVG"
154 Create a "sliding window" average of another data series.
156 Usage: CDEF:smoothed=x,1800,TREND
158 This will create a half-hour (1800 second) sliding window average
159 of x. The average is essentially computed as shown here:
161 +---!---!---!---!---!---!---!---!--->
162 now
163 delay t0
164 <--------------->
165 delay t1
166 <--------------->
167 delay t2
168 <--------------->
170 Value at sample (t0) will be the average between (t0-delay) and (t0)
171 Value at sample (t1) will be the average between (t1-delay) and (t1)
172 Value at sample (t2) will be the average between (t2-delay) and (t2)
174 TRENDNAN is - in contrast to TREND - NAN-safe. If you use TREND and
175 one source value is NAN the complete sliding window is affected.
176 The TRENDNAN operation ignores all NAN-values in a sliding window
177 and computes the average of the remaining values.
181 Create a "sliding window" average/sigma of another data series,
182 that also shifts the data series by given amounts of of time as
183 well
185 Usage - explicit stating shifts: CDEF:predict=<shift n>,...,<shift
186 1>,n,<window>,x,PREDICT CDEF:sigma=<shift n>,...,<shift 1>,n,<win-
187 dow>,x,PREDICTSIGMA
189 Usage - shifts defined as a base shift and a number of time this is
190 applied CDEF:predict=<shift multiplier>,-n,<window>,x,PREDICT
191 CDEF:sigma=<shift multiplier>,-n,<window>,x,PREDICTSIGMA
193 Example: CDEF:predict=172800,86400,2,1800,x,PREDICT
195 This will create a half-hour (1800 second) sliding window aver-
196 age/sigma of x, that average is essentially computed as shown here:
198 +---!---!---!---!---!---!---!---!---!---!---!---!---!---!---!---!---!--->
199 now
200 shift 1 t0
201 <----------------------->
202 window
203 <--------------->
204 shift 2
205 <----------------------------------------------->
206 window
207 <--------------->
208 shift 1 t1
209 <----------------------->
210 window
211 <--------------->
212 shift 2
213 <----------------------------------------------->
214 window
215 <--------------->
217 Value at sample (t0) will be the average between (t0-shift1-window) and (t0-shift1)
218 and between (t0-shift2-window) and (t0-shift2)
219 Value at sample (t1) will be the average between (t1-shift1-window) and (t1-shift1)
220 and between (t1-shift2-window) and (t1-shift2)
222 The function is by design NAN-safe. This also allows for extrapo-
223 lation into the future (say a few days) - you may need to define
224 the data series whit the optional start= parameter, so that the
225 source data series has enough data to provide prediction also at
226 the beginning of a graph...
228 Here an example, that will create a 10 day graph that also shows
229 the prediction 3 days into the future with its uncertainty value
230 (as defined by avg+-4*sigma) This also shows if the prediction is
231 exceeded at a certain point.
233 rrdtool graph image.png --imgformat=PNG \
234 --start=-7days --end=+3days --width=1000 --height=200
235 --alt-autoscale-max \
236 DEF:value=value.rrd:value:AVERAGE:start=-14days \
237 LINE1:value#ff0000:value \
238 CDEF:predict=86400,-7,1800,value,PREDICT \
239 CDEF:sigma=86400,-7,1800,value,PREDICTSIGMA \
240 CDEF:upper=predict,sigma,3,*,+ \
241 CDEF:lower=predict,sigma,3,*,- \
242 LINE1:predict#00ff00:prediction \
243 LINE1:upper#0000ff:upper\ certainty\ limit \
244 LINE1:lower#0000ff:lower\ certainty\ limit \
245 CDEF:exceeds=value,UN,0,value,lower,upper,LIMIT,UN,IF \
246 TICK:exceeds#aa000080:1
248 Note: Experience has shown that a factor between 3 and 5 to scale
249 sigma is a good discriminator to detect abnormal behaviour. This
250 obviously depends also on the type of data and how "noisy" the data
251 series is.
253 This prediction can only be used for short term extrapolations -
254 say a few days into the future-
256 Special values
259 Pushes an unknown value on the stack
263 Pushes a positive or negative infinite value on the stack. When
264 such a value is graphed, it appears at the top or bottom of the
265 graph, no matter what the actual value on the y-axis is.
271 allows you to do calculations across the data. This function can-
277 otherwise the result of the vname variable at the previous time
278 step. This allows you to do calculations across the data. This
283 Pushes the number 1 if this is the first value of the data set, the
284 number 2 if it is the second, and so on. This special value allows
285 you to make calculations based on the position of the value within
288 Time
289 Time inside RRDtool is measured in seconds since the epoch. The
290 epoch is defined to be "Thu Jan 1 00:00:00 UTC 1970".
294 Pushes the current time on the stack.
298 Pushes the time the currently processed value was taken at onto the
299 stack.
304 valid at that time including daylight saving time if your OS sup-
305 ports it, and pushes the result on the stack. There is an elabo-
306 rate example in the examples section below on how to use this.
308 Processing the stack directly
311 Duplicate the top element, remove the top element, exchange the two
312 top elements.
320 MAXIMUM, MINIMUM, AVERAGE
321 Return the corresponding value, MAXIMUM and MINIMUM also return the
322 first occurrence of that value in the time component.
324 Example: "VDEF:avg=mydata,AVERAGE"
326 STDEV
327 Returns the standard deviation of the values.
329 Example: "VDEF:stdev=mydata,STDEV"
331 LAST, FIRST
332 Return the last/first value including its time. The time for FIRST
333 is actually the start of the corresponding interval, whereas LAST
334 returns the end of the corresponding interval.
336 Example: "VDEF:first=mydata,FIRST"
338 TOTAL
339 Returns the rate from each defined time slot multiplied with the
340 step size. This can, for instance, return total bytes transfered
341 when you have logged bytes per second. The time component returns
342 the number of seconds.
344 Example: "VDEF:total=mydata,TOTAL"
346 PERCENT, PERCENTNAN
347 This should follow a D\bDE\bEF\bF or C\bCD\bDE\bEF\bF _\bv_\bn_\ba_\bm_\be. The _\bv_\bn_\ba_\bm_\be is popped,
348 another number is popped which is a certain percentage (0..100).
349 The data set is then sorted and the value returned is chosen such
356 (NaN < -INF < finite values < INF)
358 Example: "VDEF:perc95=mydata,95,PERCENT"
359 "VDEF:percnan95=mydata,95,PERCENTNAN"
361 LSLSLOPE, LSLINT, LSLCORREL
362 Return the parameters for a L\bLeast S\bSquares L\bLine _\b(_\by _\b= _\bm_\bx _\b+_\bb_\b) which
364 line related to the COUNT position of the data. LSLINT is the
366 the graph. LSLCORREL is the Correlation Coefficient (also know as
367 Pearson's Product Moment Correlation Coefficient). It will range
368 from 0 to +/-1 and represents the quality of fit for the approxima-
369 tion.
371 Example: "VDEF:slope=mydata,LSLSLOPE"
374 rrdgraph gives an overview of how r\brr\brd\bdt\bto\boo\bol\bl g\bgr\bra\bap\bph\bh works. rrdgraph_data
375 describes D\bDE\bEF\bF,C\bCD\bDE\bEF\bF and V\bVD\bDE\bEF\bF in detail. rrdgraph_rpn describes the R\bRP\bPN\bN
377 all of the graph and print functions.
379 Make sure to read rrdgraph_examples for tips&tricks.
382 Program by Tobias Oetiker <tobi@oetiker.ch>
384 This manual page by Alex van den Bogaerdt <alex@vandenbogaerdt.nl> with
385 corrections and/or additions by several people
389 1.3.99909060808 2009-02-21 RRDGRAPH_RPN(1)