1 module Database.RRDtool.Create
9 -- Database.RRDtool.Expression
18 , VariableShiftPredictOp(..)
19 , FixedShiftPredictOp(..)
25 import Data.Time.Clock
26 import Data.Time.Clock.POSIX
27 import Database.RRDtool.Expression
30 -- |A single RRD can accept input from several data sources (DS), for
31 -- example incoming and outgoing traffic on a specific communication
32 -- line. With the DS configuration option you must define some basic
33 -- properties of each data source you want to store in the RRD.
35 -- /NOTE on COUNTER vs DERIVE/
37 -- by Don Baarda <don.baarda@baesystems.com>
39 -- If you cannot tolerate ever mistaking the occasional counter reset
40 -- for a legitimate counter wrap, and would prefer \"Unknowns\" for
41 -- all legitimate counter wraps and resets, always use DERIVE with
42 -- @'dsMin' = 0@. Otherwise, using COUNTER with a suitable max will
43 -- return correct values for all legitimate counter wraps, mark some
44 -- counter resets as \"Unknown\", but can mistake some counter resets
45 -- for a legitimate counter wrap.
47 -- For a 5 minute step and 32-bit counter, the probability of
48 -- mistaking a counter reset for a legitimate wrap is arguably about
49 -- 0.8% per 1Mbps of maximum bandwidth. Note that this equates to 80%
50 -- for 100Mbps interfaces, so for high bandwidth interfaces and a
51 -- 32bit counter, DERIVE with @'dsMin' = 0@ is probably preferable. If
52 -- you are using a 64bit counter, just about any max setting will
53 -- eliminate the possibility of mistaking a reset for a counter wrap.
55 = -- |GAUGE is for things like temperatures or number of people in
56 -- a room or the value of a RedHat share.
58 -- |The name you will use to reference this particular data
59 -- source from an RRD. A ds-name must be 1 to 19 characters
60 -- long in the characters @[a-zA-Z0-9_]@.
62 -- |Defines the maximum number of seconds that may
63 -- pass between two updates of this data source before the
64 -- value of the data source is assumed to be @*UNKNOWN*@.
65 , dsHeartbeat :: !NominalDiffTime
66 -- |'dsMin' and 'dsMax' Define the expected range values for
67 -- data supplied by a data source. If 'dsMin' and\/or 'dsMax'
68 -- any value outside the defined range will be regarded as
69 -- @*UNKNOWN*@. If you do not know or care about 'dsMin' and
70 -- 'dsMax', set them to 'Nothing' for unknown. Note that
71 -- 'dsMin' and 'dsMax' always refer to the processed values of
72 -- the DS. For a traffic-'COUNTER' type DS this would be the
73 -- maximum and minimum data-rate expected from the device.
75 -- If information on minimal\/maximal expected values is
76 -- available, always set the min and\/or max properties. This
77 -- will help RRDtool in doing a simple sanity check on the
78 -- data supplied when running update.
79 , dsMin :: !(Maybe Double)
81 , dsMax :: !(Maybe Double)
83 -- |COUNTER is for continuous incrementing counters like the
84 -- ifInOctets counter in a router. The COUNTER data source assumes
85 -- that the counter never decreases, except when a counter
86 -- overflows. The update function takes the overflow into
87 -- account. The counter is stored as a per-second rate. When the
88 -- counter overflows, RRDtool checks if the overflow happened at
89 -- the 32bit or 64bit border and acts accordingly by adding an
90 -- appropriate value to the result.
93 , dsHeartbeat :: !NominalDiffTime
94 , dsMin :: !(Maybe Double)
95 , dsMax :: !(Maybe Double)
97 -- |DERIVE will store the derivative of the line going from the
98 -- last to the current value of the data source. This can be
99 -- useful for gauges, for example, to measure the rate of people
100 -- entering or leaving a room. Internally, derive works exactly
101 -- like COUNTER but without overflow checks. So if your counter
102 -- does not reset at 32 or 64 bit you might want to use DERIVE and
103 -- combine it with a 'dsMin' value of 0.
106 , dsHeartbeat :: !NominalDiffTime
107 , dsMin :: !(Maybe Double)
108 , dsMax :: !(Maybe Double)
110 -- |ABSOLUTE is for counters which get reset upon reading. This is
111 -- used for fast counters which tend to overflow. So instead of
112 -- reading them normally you reset them after every read to make
113 -- sure you have a maximum time available before the next
114 -- overflow. Another usage is for things you count like number of
115 -- messages since the last update.
118 , dsHeartbeat :: !NominalDiffTime
119 , dsMin :: !(Maybe Double)
120 , dsMax :: !(Maybe Double)
122 -- |COMPUTE is for storing the result of a formula applied to
123 -- other data sources in the RRD. This data source is not supplied
124 -- a value on update, but rather its Primary Data Points (PDPs)
125 -- are computed from the PDPs of the data sources according to the
126 -- rpn-expression that defines the formula. Consolidation
127 -- functions are then applied normally to the PDPs of the COMPUTE
128 -- data source (that is the rpn-expression is only applied to
129 -- generate PDPs). In database software, such data sets are
130 -- referred to as \"virtual\" or \"computed\" columns.
133 | forall a. IsCommonExpr a => COMPUTE {
135 -- |rpn-expression defines the formula used to compute the
136 -- PDPs of a COMPUTE data source from other data sources in
137 -- the same \<RRD\>. It is similar to defining a CDEF argument
138 -- for the graph command. For COMPUTE data sources, the
139 -- following RPN operations are not supported: COUNT, PREV,
140 -- TIME, and LTIME. In addition, in defining the RPN
141 -- expression, the COMPUTE data source may only refer to the
142 -- names of data source listed previously in the create
143 -- command. This is similar to the restriction that CDEFs must
144 -- refer only to DEFs and CDEFs previously defined in the same
154 -- , dsExpr = Previous :<: Const 100
155 -- , dsExpr = Var "foo" :<: Const 100
156 , dsExpr = AverageOf (Const 100 .*. Const 200 .*. HNil)
159 -- |The 'createRRD' function lets you set up new Round Robin Database
160 -- (RRD) files. The file is created at its final, full size and filled
161 -- with @*UNKNOWN*@ data.
163 :: FilePath -- ^The name of the RRD you want to create. RRD files
164 -- should end with the extension @.rrd@. However,
165 -- RRDtool will accept any filename.
166 -> Bool -- ^Do not clobber an existing file of the same name.
167 -> Maybe POSIXTime -- ^Specifies the time in seconds since
168 -- @1970-01-01 UTC@ when the first value should
169 -- be added to the RRD. RRDtool will not accept
170 -- any data timed before or at the time
171 -- specified. (default: @now - 10s@)
172 -> Maybe NominalDiffTime -- ^Specifies the base interval in
173 -- seconds with which data will be fed
174 -- into the RRD. (default: 300 sec)
175 -> [DataSource] -- ^Data sources to accept input from.
177 createRRD = error "FIXME"