{ "content": [ { "type": "text", "text": "# rrdcreate (man)\n\n## NAME\n\nrrdcreate - Set up a new Round Robin Database\n\n## SYNOPSIS\n\nrrdtool create filename [--start|-b start time] [--step|-s step] [--template|-t template-\nfile] [--source|-r source-file] [--no-overwrite|-O] [--daemon|-d address] [DS:ds-\nname[=mapped-ds-name[[source-index]]]:DST:dst arguments] [RRA:CF:cf arguments]\n\n## DESCRIPTION\n\nThe create function of RRDtool lets you set up new Round Robin Database (RRD) files. The\nfile is created at its final, full size and filled with *UNKNOWN* data, unless one or more\nsource RRD files have been specified and they hold suitable data to \"pre-fill\" the new RRD\nfile.\n\n## Sections\n\n- **NAME**\n- **SYNOPSIS**\n- **DESCRIPTION** (4 subsections)\n- **HOW TO MEASURE**\n- **EXAMPLE**\n- **EXAMPLE 2**\n- **EXAMPLE 3**\n- **SECURITY**\n- **AUTHORS**\n\nUse structuredContent.sections for detailed options, examples, and full documentation.\n" } ], "structuredContent": { "command": "rrdcreate", "section": "", "mode": "man", "summary": "rrdcreate - Set up a new Round Robin Database", "synopsis": "rrdtool create filename [--start|-b start time] [--step|-s step] [--template|-t template-\nfile] [--source|-r source-file] [--no-overwrite|-O] [--daemon|-d address] [DS:ds-\nname[=mapped-ds-name[[source-index]]]:DST:dst arguments] [RRA:CF:cf arguments]", "tldr_summary": null, "tldr_examples": [], "tldr_source": null, "flags": [ { "flag": "", "long": null, "arg": null, "description": "Do not clobber an existing file of the same name. --daemon|-d address Address of the rrdcached daemon. For a list of accepted formats, see the -l option in the rrdcached manual. rrdtool create --daemon unix:/var/run/rrdcached.sock /var/lib/rrd/foo.rrd I [--template|-t  template-file] Specifies a template RRD file to take step, DS and RRA definitions from. This allows one to base the structure of a new file on some existing file. The data of the template file is NOT used for pre-filling, but it is possible to specify the same file as a source file (see below). Additional DS and RRA definitions are permitted, and will be added to those taken from the template. --source|-r source-file One or more source RRD files may be named on the command line. Data from these source files will be used to prefill the created RRD file. The output file and one source file may refer to the same file name. This will effectively replace the source file with the new RRD file. While there is the danger to loose the source file because it gets replaced, there is no danger that the source and the new file may be \"garbled\" together at any point in time, because the new file will always be created as a temporary file first and will only be moved to its final destination once it has been written in its entirety. Prefilling is done by matching up DS names, RRAs and consolidation functions and choosing the best available data resolution when doing so. Prefilling may not be mathematically correct in all cases (e.g. if resolutions have to change due to changed stepping of the target RRD and old and new resolutions do not match up with old/new bin boundaries in RRAs). In other words: A best effort is made to preserve data during prefilling. Also, pre-filling of RRAs may only be possible for certain kinds of DS types. Prefilling may also have strange effects on Holt-Winters forecasting RRAs. In other words: there is no guarantee for data- correctness. When \"pre-filling\" a RRD file, the structure of the new file must be specified as usual using DS and RRA specifications as outlined below. Data will be taken from source files based on DS names and types and in the order the source files are specified in. Data sources with the same name from different source files will be combined to form a new data source. Generally, for any point in time the new RRD file will cover after its creation, data from only one source file will have been used for pre-filling. However, data from multiple sources may be combined if it refers to different times or an earlier named source file holds unknown data for a time where a later one holds known data. If this automatic data selection is not desired, the DS syntax allows one to specify a mapping of target and source data sources for prefilling. This syntax allows one to rename data sources and to restrict prefilling for a DS to only use data from a single source file. Prefilling currently only works reliably for RRAs using one of the classic consolidation functions, that is one of: AVERAGE, MIN, MAX, LAST. It might also currently have problems with COMPUTE data sources. Note that the act of prefilling during create is similar to a lot of the operations available via the tune command, but using create syntax. DS:ds-name[=mapped-ds-name[[source-index]]]:DST:dst arguments A single RRD can accept input from several data sources (DS), for example incoming and outgoing traffic on a specific communication line. With the DS configuration option you must define some basic properties of each data source you want to store in the RRD. ds-name is the name you will use to reference this particular data source from an RRD. A ds- name must be 1 to 19 characters long in the characters [a-zA-Z0-9]. DST defines the Data Source Type. The remaining arguments of a data source entry depend on the data source type. For GAUGE, COUNTER, DERIVE, DCOUNTER, DDERIVE and ABSOLUTE the format for a data source entry is: DS:ds-name:{GAUGE | COUNTER | DERIVE | DCOUNTER | DDERIVE | ABSOLUTE}:heartbeat:min:max For COMPUTE data sources, the format is: DS:ds-name:COMPUTE:rpn-expression In order to decide which data source type to use, review the definitions that follow. Also consult the section on \"HOW TO MEASURE\" for further insight. GAUGE is for things like temperatures or number of people in a room or the value of a RedHat share. COUNTER is for continuous incrementing counters like the ifInOctets counter in a router. The COUNTER data source assumes that the counter never decreases, except when a counter overflows. The update function takes the overflow into account. The counter is stored as a per-second rate. When the counter overflows, RRDtool checks if the overflow happened at the 32bit or 64bit border and acts accordingly by adding an appropriate value to the result. DCOUNTER the same as COUNTER, but for quantities expressed as double-precision floating point number. Could be used to track quantities that increment by non-integer numbers, i.e. number of seconds that some routine has taken to run, total weight processed by some technology equipment etc. The only substantial difference is that DCOUNTER can either be upward counting or downward counting, but not both at the same time. The current direction is detected automatically on the second non-undefined counter update and any further change in the direction is considered a reset. The new direction is determined and locked in by the second update after reset and its difference to the value at reset. DERIVE will store the derivative of the line going from the last to the current value of the data source. This can be useful for gauges, for example, to measure the rate of people entering or leaving a room. Internally, derive works exactly like COUNTER but without overflow checks. So if your counter does not reset at 32 or 64 bit you might want to use DERIVE and combine it with a MIN value of 0. DDERIVE the same as DERIVE, but for quantities expressed as double-precision floating point number. NOTE on COUNTER vs DERIVE by Don Baarda If you cannot tolerate ever mistaking the occasional counter reset for a legitimate counter wrap, and would prefer \"Unknowns\" for all legitimate counter wraps and resets, always use DERIVE with min=0. Otherwise, using COUNTER with a suitable max will return correct values for all legitimate counter wraps, mark some counter resets as \"Unknown\", but can mistake some counter resets for a legitimate counter wrap. For a 5 minute step and 32-bit counter, the probability of mistaking a counter reset for a legitimate wrap is arguably about 0.8% per 1Mbps of maximum bandwidth. Note that this equates to 80% for 100Mbps interfaces, so for high bandwidth interfaces and a 32bit counter, DERIVE with min=0 is probably preferable. If you are using a 64bit counter, just about any max setting will eliminate the possibility of mistaking a reset for a counter wrap. ABSOLUTE is for counters which get reset upon reading. This is used for fast counters which tend to overflow. So instead of reading them normally you reset them after every read to make sure you have a maximum time available before the next overflow. Another usage is for things you count like number of messages since the last update. COMPUTE is for storing the result of a formula applied to other data sources in the RRD. This data source is not supplied a value on update, but rather its Primary Data Points (PDPs) are computed from the PDPs of the data sources according to the rpn-expression that defines the formula. Consolidation functions are then applied normally to the PDPs of the COMPUTE data source (that is the rpn-expression is only applied to generate PDPs). In database software, such data sets are referred to as \"virtual\" or \"computed\" columns. heartbeat defines the maximum number of seconds that may pass between two updates of this data source before the value of the data source is assumed to be *UNKNOWN*. min and max define the expected range values for data supplied by a data source. If min and/or max are specified any value outside the defined range will be regarded as *UNKNOWN*. If you do not know or care about min and max, set them to U for unknown. Note that min and max always refer to the processed values of the DS. For a traffic-COUNTER type DS this would be the maximum and minimum data-rate expected from the device. If information on minimal/maximal expected values is available, always set the min and/or max properties. This will help RRDtool in doing a simple sanity check on the data supplied when running update. rpn-expression defines the formula used to compute the PDPs of a COMPUTE data source from other data sources in the same . It is similar to defining a CDEF argument for the graph command. Please refer to that manual page for a list and description of RPN operations supported. For COMPUTE data sources, the following RPN operations are not supported: COUNT, PREV, TIME, and LTIME. In addition, in defining the RPN expression, the COMPUTE data source may only refer to the names of data source listed previously in the create command. This is similar to the restriction that CDEFs must refer only to DEFs and CDEFs previously defined in the same graph command. When pre-filling the new RRD file using one or more source RRDs, the DS specification may hold an optional mapping after the DS name. This takes the form of an equal sign followed by a mapped-to DS name and an optional source index enclosed in square brackets. For example, the DS DS:a=b[2]:GAUGE:120:0:U specifies that the DS named a should be pre-filled from the DS named b in the second listed source file (source indices are 1-based). RRA:CF:cf arguments The purpose of an RRD is to store data in the round robin archives (RRA). An archive consists of a number of data values or statistics for each of the defined data-sources (DS) and is defined with an RRA line. When data is entered into an RRD, it is first fit into time slots of the length defined with the -s option, thus becoming a primary data point. The data is also processed with the consolidation function (CF) of the archive. There are several consolidation functions that consolidate primary data points via an aggregate function: AVERAGE, MIN, MAX, LAST. AVERAGE the average of the data points is stored. MIN the smallest of the data points is stored. MAX the largest of the data points is stored. LAST the last data points is used. Note that data aggregation inevitably leads to loss of precision and information. The trick is to pick the aggregate function such that the interesting properties of your data is kept across the aggregation process. The format of RRA line for these consolidation functions is: RRA:{AVERAGE | MIN | MAX | LAST}:xff:steps:rows xff The xfiles factor defines what part of a consolidation interval may be made up from *UNKNOWN* data while the consolidated value is still regarded as known. It is given as the ratio of allowed *UNKNOWN* PDPs to the number of PDPs in the interval. Thus, it ranges from 0 to 1 (exclusive). steps defines how many of these primary data points are used to build a consolidated data point which then goes into the archive. See also \"STEP, HEARTBEAT, and Rows As Durations\". rows defines how many generations of data values are kept in an RRA. Obviously, this has to be greater than zero. See also \"STEP, HEARTBEAT, and Rows As Durations\"." } ], "examples": [ "rrdtool create temperature.rrd --step 300 \\", "DS:temp:GAUGE:600:-273:5000 \\", "RRA:AVERAGE:0.5:1:1200 \\", "RRA:MIN:0.5:12:2400 \\", "RRA:MAX:0.5:12:2400 \\", "RRA:AVERAGE:0.5:12:2400", "This sets up an RRD called temperature.rrd which accepts one temperature value every 300", "seconds. If no new data is supplied for more than 600 seconds, the temperature becomes", "*UNKNOWN*. The minimum acceptable value is -273 and the maximum is 5'000.", "A few archive areas are also defined. The first stores the temperatures supplied for 100", "hours (1'200 * 300 seconds = 100 hours). The second RRA stores the minimum temperature", "recorded over every hour (12 * 300 seconds = 1 hour), for 100 days (2'400 hours). The third", "and the fourth RRA's do the same for the maximum and average temperature, respectively." ], "see_also": [], "section_outline": [ { "name": "NAME", "lines": 2, "subsections": [] }, { "name": "SYNOPSIS", "lines": 4, "subsections": [] }, { "name": "DESCRIPTION", "lines": 24, "subsections": [ { "name": "--no-overwrite|-O", "lines": 215 }, { "name": "Aberrant Behavior Detection with Holt-Winters Forecasting", "lines": 121 }, { "name": "STEP, HEARTBEAT, and Rows As Durations", "lines": 34 }, { "name": "The HEARTBEAT and the STEP", "lines": 64 } ] }, { "name": "HOW TO MEASURE", "lines": 29, "subsections": [] }, { "name": "EXAMPLE", "lines": 16, "subsections": [] }, { "name": "EXAMPLE 2", "lines": 35, "subsections": [] }, { "name": "EXAMPLE 3", "lines": 19, "subsections": [] }, { "name": "SECURITY", "lines": 4, "subsections": [] }, { "name": "AUTHORS", "lines": 5, "subsections": [] } ], "sections": { "NAME": { "content": "rrdcreate - Set up a new Round Robin Database\n", "subsections": [] }, "SYNOPSIS": { "content": "rrdtool create filename [--start|-b start time] [--step|-s step] [--template|-t template-\nfile] [--source|-r source-file] [--no-overwrite|-O] [--daemon|-d address] [DS:ds-\nname[=mapped-ds-name[[source-index]]]:DST:dst arguments] [RRA:CF:cf arguments]\n", "subsections": [] }, "DESCRIPTION": { "content": "The create function of RRDtool lets you set up new Round Robin Database (RRD) files. The\nfile is created at its final, full size and filled with *UNKNOWN* data, unless one or more\nsource RRD files have been specified and they hold suitable data to \"pre-fill\" the new RRD\nfile.\n\nfilename\nThe name of the RRD you want to create. RRD files should end with the extension .rrd.\nHowever, RRDtool will accept any filename.\n\n--start|-b start time (default: now - 10s)\nSpecifies the time in seconds since 1970-01-01 UTC when the first value should be added to\nthe RRD. RRDtool will not accept any data timed before or at the time specified.\n\nSee also \"AT-STYLE TIME SPECIFICATION\" in rrdfetch for other ways to specify time.\n\nIf one or more source files is used to pre-fill the new RRD, the --start option may be\nomitted. In that case, the latest update time among all source files will be used as the last\nupdate time of the new RRD file, effectively setting the start time.\n\n--step|-s step (default: 300 seconds)\nSpecifies the base interval in seconds with which data will be fed into the RRD. A scaling\nfactor may be present as a suffix to the integer; see \"STEP, HEARTBEAT, and Rows As\nDurations\".\n", "subsections": [ { "name": "--no-overwrite|-O", "content": "Do not clobber an existing file of the same name.\n\n--daemon|-d address\nAddress of the rrdcached daemon. For a list of accepted formats, see the -l option in the\nrrdcached manual.\n\nrrdtool create --daemon unix:/var/run/rrdcached.sock /var/lib/rrd/foo.rrd I\n\n[--template|-t  template-file]\nSpecifies a template RRD file to take step, DS and RRA definitions from. This allows one to\nbase the structure of a new file on some existing file. The data of the template file is NOT\nused for pre-filling, but it is possible to specify the same file as a source file (see\nbelow).\n\nAdditional DS and RRA definitions are permitted, and will be added to those taken from the\ntemplate.\n\n--source|-r source-file\nOne or more source RRD files may be named on the command line. Data from these source files\nwill be used to prefill the created RRD file. The output file and one source file may refer\nto the same file name. This will effectively replace the source file with the new RRD file.\nWhile there is the danger to loose the source file because it gets replaced, there is no\ndanger that the source and the new file may be \"garbled\" together at any point in time,\nbecause the new file will always be created as a temporary file first and will only be moved\nto its final destination once it has been written in its entirety.\n\nPrefilling is done by matching up DS names, RRAs and consolidation functions and choosing the\nbest available data resolution when doing so. Prefilling may not be mathematically correct in\nall cases (e.g. if resolutions have to change due to changed stepping of the target RRD and\nold and new resolutions do not match up with old/new bin boundaries in RRAs).\n\nIn other words: A best effort is made to preserve data during prefilling. Also, pre-filling\nof RRAs may only be possible for certain kinds of DS types. Prefilling may also have strange\neffects on Holt-Winters forecasting RRAs. In other words: there is no guarantee for data-\ncorrectness.\n\nWhen \"pre-filling\" a RRD file, the structure of the new file must be specified as usual using\nDS and RRA specifications as outlined below. Data will be taken from source files based on DS\nnames and types and in the order the source files are specified in. Data sources with the\nsame name from different source files will be combined to form a new data source. Generally,\nfor any point in time the new RRD file will cover after its creation, data from only one\nsource file will have been used for pre-filling. However, data from multiple sources may be\ncombined if it refers to different times or an earlier named source file holds unknown data\nfor a time where a later one holds known data.\n\nIf this automatic data selection is not desired, the DS syntax allows one to specify a\nmapping of target and source data sources for prefilling. This syntax allows one to rename\ndata sources and to restrict prefilling for a DS to only use data from a single source file.\n\nPrefilling currently only works reliably for RRAs using one of the classic consolidation\nfunctions, that is one of: AVERAGE, MIN, MAX, LAST. It might also currently have problems\nwith COMPUTE data sources.\n\nNote that the act of prefilling during create is similar to a lot of the operations available\nvia the tune command, but using create syntax.\n\nDS:ds-name[=mapped-ds-name[[source-index]]]:DST:dst arguments\nA single RRD can accept input from several data sources (DS), for example incoming and\noutgoing traffic on a specific communication line. With the DS configuration option you must\ndefine some basic properties of each data source you want to store in the RRD.\n\nds-name is the name you will use to reference this particular data source from an RRD. A ds-\nname must be 1 to 19 characters long in the characters [a-zA-Z0-9].\n\nDST defines the Data Source Type. The remaining arguments of a data source entry depend on\nthe data source type. For GAUGE, COUNTER, DERIVE, DCOUNTER, DDERIVE and ABSOLUTE the format\nfor a data source entry is:\n\nDS:ds-name:{GAUGE | COUNTER | DERIVE | DCOUNTER | DDERIVE | ABSOLUTE}:heartbeat:min:max\n\nFor COMPUTE data sources, the format is:\n\nDS:ds-name:COMPUTE:rpn-expression\n\nIn order to decide which data source type to use, review the definitions that follow. Also\nconsult the section on \"HOW TO MEASURE\" for further insight.\n\nGAUGE\nis for things like temperatures or number of people in a room or the value of a RedHat\nshare.\n\nCOUNTER\nis for continuous incrementing counters like the ifInOctets counter in a router. The\nCOUNTER data source assumes that the counter never decreases, except when a counter\noverflows. The update function takes the overflow into account. The counter is stored\nas a per-second rate. When the counter overflows, RRDtool checks if the overflow happened\nat the 32bit or 64bit border and acts accordingly by adding an appropriate value to the\nresult.\n\nDCOUNTER\nthe same as COUNTER, but for quantities expressed as double-precision floating point\nnumber. Could be used to track quantities that increment by non-integer numbers, i.e.\nnumber of seconds that some routine has taken to run, total weight processed by some\ntechnology equipment etc. The only substantial difference is that DCOUNTER can either be\nupward counting or downward counting, but not both at the same time. The current\ndirection is detected automatically on the second non-undefined counter update and any\nfurther change in the direction is considered a reset. The new direction is determined\nand locked in by the second update after reset and its difference to the value at reset.\n\nDERIVE\nwill store the derivative of the line going from the last to the current value of the\ndata source. This can be useful for gauges, for example, to measure the rate of people\nentering or leaving a room. Internally, derive works exactly like COUNTER but without\noverflow checks. So if your counter does not reset at 32 or 64 bit you might want to use\nDERIVE and combine it with a MIN value of 0.\n\nDDERIVE\nthe same as DERIVE, but for quantities expressed as double-precision floating point\nnumber.\n\nNOTE on COUNTER vs DERIVE\n\nby Don Baarda \n\nIf you cannot tolerate ever mistaking the occasional counter reset for a legitimate\ncounter wrap, and would prefer \"Unknowns\" for all legitimate counter wraps and resets,\nalways use DERIVE with min=0. Otherwise, using COUNTER with a suitable max will return\ncorrect values for all legitimate counter wraps, mark some counter resets as \"Unknown\",\nbut can mistake some counter resets for a legitimate counter wrap.\n\nFor a 5 minute step and 32-bit counter, the probability of mistaking a counter reset for\na legitimate wrap is arguably about 0.8% per 1Mbps of maximum bandwidth. Note that this\nequates to 80% for 100Mbps interfaces, so for high bandwidth interfaces and a 32bit\ncounter, DERIVE with min=0 is probably preferable. If you are using a 64bit counter, just\nabout any max setting will eliminate the possibility of mistaking a reset for a counter\nwrap.\n\nABSOLUTE\nis for counters which get reset upon reading. This is used for fast counters which tend\nto overflow. So instead of reading them normally you reset them after every read to make\nsure you have a maximum time available before the next overflow. Another usage is for\nthings you count like number of messages since the last update.\n\nCOMPUTE\nis for storing the result of a formula applied to other data sources in the RRD. This\ndata source is not supplied a value on update, but rather its Primary Data Points (PDPs)\nare computed from the PDPs of the data sources according to the rpn-expression that\ndefines the formula. Consolidation functions are then applied normally to the PDPs of the\nCOMPUTE data source (that is the rpn-expression is only applied to generate PDPs). In\ndatabase software, such data sets are referred to as \"virtual\" or \"computed\" columns.\n\nheartbeat defines the maximum number of seconds that may pass between two updates of this\ndata source before the value of the data source is assumed to be *UNKNOWN*.\n\nmin and max define the expected range values for data supplied by a data source. If min\nand/or max are specified any value outside the defined range will be regarded as *UNKNOWN*.\nIf you do not know or care about min and max, set them to U for unknown. Note that min and\nmax always refer to the processed values of the DS. For a traffic-COUNTER type DS this would\nbe the maximum and minimum data-rate expected from the device.\n\nIf information on minimal/maximal expected values is available, always set the min and/or max\nproperties. This will help RRDtool in doing a simple sanity check on the data supplied when\nrunning update.\n\nrpn-expression defines the formula used to compute the PDPs of a COMPUTE data source from\nother data sources in the same . It is similar to defining a CDEF argument for the graph\ncommand. Please refer to that manual page for a list and description of RPN operations\nsupported. For COMPUTE data sources, the following RPN operations are not supported: COUNT,\nPREV, TIME, and LTIME. In addition, in defining the RPN expression, the COMPUTE data source\nmay only refer to the names of data source listed previously in the create command. This is\nsimilar to the restriction that CDEFs must refer only to DEFs and CDEFs previously defined in\nthe same graph command.\n\nWhen pre-filling the new RRD file using one or more source RRDs, the DS specification may\nhold an optional mapping after the DS name. This takes the form of an equal sign followed by\na mapped-to DS name and an optional source index enclosed in square brackets.\n\nFor example, the DS\n\nDS:a=b[2]:GAUGE:120:0:U\n\nspecifies that the DS named a should be pre-filled from the DS named b in the second listed\nsource file (source indices are 1-based).\n\nRRA:CF:cf arguments\nThe purpose of an RRD is to store data in the round robin archives (RRA). An archive consists\nof a number of data values or statistics for each of the defined data-sources (DS) and is\ndefined with an RRA line.\n\nWhen data is entered into an RRD, it is first fit into time slots of the length defined with\nthe -s option, thus becoming a primary data point.\n\nThe data is also processed with the consolidation function (CF) of the archive. There are\nseveral consolidation functions that consolidate primary data points via an aggregate\nfunction: AVERAGE, MIN, MAX, LAST.\n\nAVERAGE\nthe average of the data points is stored.\n\nMIN the smallest of the data points is stored.\n\nMAX the largest of the data points is stored.\n\nLAST\nthe last data points is used.\n\nNote that data aggregation inevitably leads to loss of precision and information. The trick\nis to pick the aggregate function such that the interesting properties of your data is kept\nacross the aggregation process.\n\nThe format of RRA line for these consolidation functions is:\n\nRRA:{AVERAGE | MIN | MAX | LAST}:xff:steps:rows\n\nxff The xfiles factor defines what part of a consolidation interval may be made up from\n*UNKNOWN* data while the consolidated value is still regarded as known. It is given as the\nratio of allowed *UNKNOWN* PDPs to the number of PDPs in the interval. Thus, it ranges from 0\nto 1 (exclusive).\n\nsteps defines how many of these primary data points are used to build a consolidated data\npoint which then goes into the archive. See also \"STEP, HEARTBEAT, and Rows As Durations\".\n\nrows defines how many generations of data values are kept in an RRA. Obviously, this has to\nbe greater than zero. See also \"STEP, HEARTBEAT, and Rows As Durations\".\n" }, { "name": "Aberrant Behavior Detection with Holt-Winters Forecasting", "content": "In addition to the aggregate functions, there are a set of specialized functions that enable\nRRDtool to provide data smoothing (via the Holt-Winters forecasting algorithm), confidence\nbands, and the flagging aberrant behavior in the data source time series:\n\n• RRA:HWPREDICT:rows:alpha:beta:seasonal period[:rra-num]\n\n• RRA:MHWPREDICT:rows:alpha:beta:seasonal period[:rra-num]\n\n• RRA:SEASONAL:seasonal period:gamma:rra-num[:smoothing-window=fraction]\n\n• RRA:DEVSEASONAL:seasonal period:gamma:rra-num[:smoothing-window=fraction]\n\n• RRA:DEVPREDICT:rows:rra-num\n\n• RRA:FAILURES:rows:threshold:window length:rra-num\n\nThese RRAs differ from the true consolidation functions in several ways. First, each of the\nRRAs is updated once for every primary data point. Second, these RRAs are interdependent. To\ngenerate real-time confidence bounds, a matched set of SEASONAL, DEVSEASONAL, DEVPREDICT, and\neither HWPREDICT or MHWPREDICT must exist. Generating smoothed values of the primary data\npoints requires a SEASONAL RRA and either an HWPREDICT or MHWPREDICT RRA. Aberrant behavior\ndetection requires FAILURES, DEVSEASONAL, SEASONAL, and either HWPREDICT or MHWPREDICT.\n\nThe predicted, or smoothed, values are stored in the HWPREDICT or MHWPREDICT RRA. HWPREDICT\nand MHWPREDICT are actually two variations on the Holt-Winters method. They are\ninterchangeable. Both attempt to decompose data into three components: a baseline, a trend,\nand a seasonal coefficient. HWPREDICT adds its seasonal coefficient to the baseline to form\na prediction, whereas MHWPREDICT multiplies its seasonal coefficient by the baseline to form\na prediction. The difference is noticeable when the baseline changes significantly in the\ncourse of a season; HWPREDICT will predict the seasonality to stay constant as the baseline\nchanges, but MHWPREDICT will predict the seasonality to grow or shrink in proportion to the\nbaseline. The proper choice of method depends on the thing being modeled. For simplicity, the\nrest of this discussion will refer to HWPREDICT, but MHWPREDICT may be substituted in its\nplace.\n\nThe predicted deviations are stored in DEVPREDICT (think a standard deviation which can be\nscaled to yield a confidence band). The FAILURES RRA stores binary indicators. A 1 marks the\nindexed observation as failure; that is, the number of confidence bounds violations in the\npreceding window of observations met or exceeded a specified threshold. An example of using\nthese RRAs to graph confidence bounds and failures appears in rrdgraph.\n\nThe SEASONAL and DEVSEASONAL RRAs store the seasonal coefficients for the Holt-Winters\nforecasting algorithm and the seasonal deviations, respectively. There is one entry per\nobservation time point in the seasonal cycle. For example, if primary data points are\ngenerated every five minutes and the seasonal cycle is 1 day, both SEASONAL and DEVSEASONAL\nwill have 288 rows.\n\nIn order to simplify the creation for the novice user, in addition to supporting explicit\ncreation of the HWPREDICT, SEASONAL, DEVPREDICT, DEVSEASONAL, and FAILURES RRAs, the RRDtool\ncreate command supports implicit creation of the other four when HWPREDICT is specified alone\nand the final argument rra-num is omitted.\n\nrows specifies the length of the RRA prior to wrap around. Remember that there is a one-to-\none correspondence between primary data points and entries in these RRAs. For the HWPREDICT\nCF, rows should be larger than the seasonal period. If the DEVPREDICT RRA is implicitly\ncreated, the default number of rows is the same as the HWPREDICT rows argument. If the\nFAILURES RRA is implicitly created, rows will be set to the seasonal period argument of the\nHWPREDICT RRA. Of course, the RRDtool resize command is available if these defaults are not\nsufficient and the creator wishes to avoid explicit creations of the other specialized\nfunction RRAs.\n\nseasonal period specifies the number of primary data points in a seasonal cycle. If SEASONAL\nand DEVSEASONAL are implicitly created, this argument for those RRAs is set automatically to\nthe value specified by HWPREDICT. If they are explicitly created, the creator should verify\nthat all three seasonal period arguments agree.\n\nalpha is the adaption parameter of the intercept (or baseline) coefficient in the Holt-\nWinters forecasting algorithm. See rrdtool for a description of this algorithm. alpha must\nlie between 0 and 1. A value closer to 1 means that more recent observations carry greater\nweight in predicting the baseline component of the forecast. A value closer to 0 means that\npast history carries greater weight in predicting the baseline component.\n\nbeta is the adaption parameter of the slope (or linear trend) coefficient in the Holt-Winters\nforecasting algorithm. beta must lie between 0 and 1 and plays the same role as alpha with\nrespect to the predicted linear trend.\n\ngamma is the adaption parameter of the seasonal coefficients in the Holt-Winters forecasting\nalgorithm (HWPREDICT) or the adaption parameter in the exponential smoothing update of the\nseasonal deviations. It must lie between 0 and 1. If the SEASONAL and DEVSEASONAL RRAs are\ncreated implicitly, they will both have the same value for gamma: the value specified for the\nHWPREDICT alpha argument. Note that because there is one seasonal coefficient (or deviation)\nfor each time point during the seasonal cycle, the adaptation rate is much slower than the\nbaseline. Each seasonal coefficient is only updated (or adapts) when the observed value\noccurs at the offset in the seasonal cycle corresponding to that coefficient.\n\nIf SEASONAL and DEVSEASONAL RRAs are created explicitly, gamma need not be the same for both.\nNote that gamma can also be changed via the RRDtool tune command.\n\nsmoothing-window specifies the fraction of a season that should be averaged around each\npoint. By default, the value of smoothing-window is 0.05, which means each value in SEASONAL\nand DEVSEASONAL will be occasionally replaced by averaging it with its (seasonal period*0.05)\nnearest neighbors. Setting smoothing-window to zero will disable the running-average\nsmoother altogether.\n\nrra-num provides the links between related RRAs. If HWPREDICT is specified alone and the\nother RRAs are created implicitly, then there is no need to worry about this argument. If\nRRAs are created explicitly, then carefully pay attention to this argument. For each RRA\nwhich includes this argument, there is a dependency between that RRA and another RRA. The\nrra-num argument is the 1-based index in the order of RRA creation (that is, the order they\nappear in the create command). The dependent RRA for each RRA requiring the rra-num argument\nis listed here:\n\n• HWPREDICT rra-num is the index of the SEASONAL RRA.\n\n• SEASONAL rra-num is the index of the HWPREDICT RRA.\n\n• DEVPREDICT rra-num is the index of the DEVSEASONAL RRA.\n\n• DEVSEASONAL rra-num is the index of the HWPREDICT RRA.\n\n• FAILURES rra-num is the index of the DEVSEASONAL RRA.\n\nthreshold is the minimum number of violations (observed values outside the confidence bounds)\nwithin a window that constitutes a failure. If the FAILURES RRA is implicitly created, the\ndefault value is 7.\n\nwindow length is the number of time points in the window. Specify an integer greater than or\nequal to the threshold and less than or equal to 28. The time interval this window\nrepresents depends on the interval between primary data points. If the FAILURES RRA is\nimplicitly created, the default value is 9.\n" }, { "name": "STEP, HEARTBEAT, and Rows As Durations", "content": "Traditionally RRDtool specified PDP intervals in seconds, and most other values as either\nseconds or PDP counts. This made the specification for databases rather opaque; for example\n\nrrdtool create power.rrd \\\n--start now-2h --step 1 \\\nDS:watts:GAUGE:300:0:24000 \\\nRRA:AVERAGE:0.5:1:864000 \\\nRRA:AVERAGE:0.5:60:129600 \\\nRRA:AVERAGE:0.5:3600:13392 \\\nRRA:AVERAGE:0.5:86400:3660\n\ncreates a database of power values collected once per second, with a five minute (300 second)\nheartbeat, and four RRAs: ten days of one second, 90 days of one minute, 18 months of one\nhour, and ten years of one day averages.\n\nStep, heartbeat, and PDP counts and rows may also be specified as durations, which are\npositive integers with a single-character suffix that specifies a scaling factor. See\n\"rrdscaledduration\" in librrd for scale factors of the supported suffixes: \"s\" (seconds),\n\"m\" (minutes), \"h\" (hours), \"d\" (days), \"w\" (weeks), \"M\" (months), and \"y\" (years).\n\nScaled step and heartbeat values (which are natively durations in seconds) are used directly,\nwhile consolidation function row arguments are divided by their step to produce the number of\nrows.\n\nWith this feature the same specification as above can be written as:\n\nrrdtool create power.rrd \\\n--start now-2h --step 1s \\\nDS:watts:GAUGE:5m:0:24000 \\\nRRA:AVERAGE:0.5:1s:10d \\\nRRA:AVERAGE:0.5:1m:90d \\\nRRA:AVERAGE:0.5:1h:18M \\\nRRA:AVERAGE:0.5:1d:10y\n" }, { "name": "The HEARTBEAT and the STEP", "content": "Here is an explanation by Don Baarda on the inner workings of RRDtool. It may help you to\nsort out why all this *UNKNOWN* data is popping up in your databases:\n\nRRDtool gets fed samples/updates at arbitrary times. From these it builds Primary Data Points\n(PDPs) on every \"step\" interval. The PDPs are then accumulated into the RRAs.\n\nThe \"heartbeat\" defines the maximum acceptable interval between samples/updates. If the\ninterval between samples is less than \"heartbeat\", then an average rate is calculated and\napplied for that interval. If the interval between samples is longer than \"heartbeat\", then\nthat entire interval is considered \"unknown\". Note that there are other things that can make\na sample interval \"unknown\", such as the rate exceeding limits, or a sample that was\nexplicitly marked as unknown.\n\nThe known rates during a PDP's \"step\" interval are used to calculate an average rate for that\nPDP. If the total \"unknown\" time accounts for more than half the \"step\", the entire PDP is\nmarked as \"unknown\". This means that a mixture of known and \"unknown\" sample times in a\nsingle PDP \"step\" may or may not add up to enough \"known\" time to warrant a known PDP.\n\nThe \"heartbeat\" can be short (unusual) or long (typical) relative to the \"step\" interval\nbetween PDPs. A short \"heartbeat\" means you require multiple samples per PDP, and if you\ndon't get them mark the PDP unknown. A long heartbeat can span multiple \"steps\", which means\nit is acceptable to have multiple PDPs calculated from a single sample. An extreme example of\nthis might be a \"step\" of 5 minutes and a \"heartbeat\" of one day, in which case a single\nsample every day will result in all the PDPs for that entire day period being set to the same\naverage rate. -- Don Baarda \n\ntime|\naxis|\nbegin|00|\n|01|\nu|02|----* sample1, restart \"hb\"-timer\nu|03| /\nu|04| /\nu|05| /\nu|06|/ \"hbt\" expired\nu|07|\n|08|----* sample2, restart \"hb\"\n|09| /\n|10| /\nu|11|----* sample3, restart \"hb\"\nu|12| /\nu|13| /\nstep1u|14| /\nu|15|/ \"swt\" expired\nu|16|\n|17|----* sample4, restart \"hb\", create \"pdp\" for step1 =\n|18| / = unknown due to 10 \"u\" labeled secs > 0.5 * step\n|19| /\n|20| /\n|21|----* sample5, restart \"hb\"\n|22| /\n|23| /\n|24|----* sample6, restart \"hb\"\n|25| /\n|26| /\n|27|----* sample7, restart \"hb\"\nstep2|28| /\n|22| /\n|23|----* sample8, restart \"hb\", create \"pdp\" for step1, create \"cdp\"\n|24| /\n|25| /\n\ngraphics by vladimir.lavrov@desy.de.\n" } ] }, "HOW TO MEASURE": { "content": "Here are a few hints on how to measure:\n\nTemperature\nUsually you have some type of meter you can read to get the temperature. The temperature\nis not really connected with a time. The only connection is that the temperature reading\nhappened at a certain time. You can use the GAUGE data source type for this. RRDtool will\nthen record your reading together with the time.\n\nMail Messages\nAssume you have a method to count the number of messages transported by your mail server\nin a certain amount of time, giving you data like '5 messages in the last 65 seconds'. If\nyou look at the count of 5 like an ABSOLUTE data type you can simply update the RRD with\nthe number 5 and the end time of your monitoring period. RRDtool will then record the\nnumber of messages per second. If at some later stage you want to know the number of\nmessages transported in a day, you can get the average messages per second from RRDtool\nfor the day in question and multiply this number with the number of seconds in a day.\nBecause all math is run with Doubles, the precision should be acceptable.\n\nIt's always a Rate\nRRDtool stores rates in amount/second for COUNTER, DERIVE, DCOUNTER, DDERIVE and ABSOLUTE\ndata. When you plot the data, you will get on the y axis amount/second which you might\nbe tempted to convert to an absolute amount by multiplying by the delta-time between the\npoints. RRDtool plots continuous data, and as such is not appropriate for plotting\nabsolute amounts as for example \"total bytes\" sent and received in a router. What you\nprobably want is plot rates that you can scale to bytes/hour, for example, or plot\nabsolute amounts with another tool that draws bar-plots, where the delta-time is clear on\nthe plot for each point (such that when you read the graph you see for example GB on the\ny axis, days on the x axis and one bar for each day).\n", "subsections": [] }, "EXAMPLE": { "content": "rrdtool create temperature.rrd --step 300 \\\nDS:temp:GAUGE:600:-273:5000 \\\nRRA:AVERAGE:0.5:1:1200 \\\nRRA:MIN:0.5:12:2400 \\\nRRA:MAX:0.5:12:2400 \\\nRRA:AVERAGE:0.5:12:2400\n\nThis sets up an RRD called temperature.rrd which accepts one temperature value every 300\nseconds. If no new data is supplied for more than 600 seconds, the temperature becomes\n*UNKNOWN*. The minimum acceptable value is -273 and the maximum is 5'000.\n\nA few archive areas are also defined. The first stores the temperatures supplied for 100\nhours (1'200 * 300 seconds = 100 hours). The second RRA stores the minimum temperature\nrecorded over every hour (12 * 300 seconds = 1 hour), for 100 days (2'400 hours). The third\nand the fourth RRA's do the same for the maximum and average temperature, respectively.\n", "subsections": [] }, "EXAMPLE 2": { "content": "rrdtool create monitor.rrd --step 300 \\\nDS:ifOutOctets:COUNTER:1800:0:4294967295 \\\nRRA:AVERAGE:0.5:1:2016 \\\nRRA:HWPREDICT:1440:0.1:0.0035:288\n\nThis example is a monitor of a router interface. The first RRA tracks the traffic flow in\noctets; the second RRA generates the specialized functions RRAs for aberrant behavior\ndetection. Note that the rra-num argument of HWPREDICT is missing, so the other RRAs will\nimplicitly be created with default parameter values. In this example, the forecasting\nalgorithm baseline adapts quickly; in fact the most recent one hour of observations (each at\n5 minute intervals) accounts for 75% of the baseline prediction. The linear trend forecast\nadapts much more slowly. Observations made during the last day (at 288 observations per day)\naccount for only 65% of the predicted linear trend. Note: these computations rely on an\nexponential smoothing formula described in the LISA 2000 paper.\n\nThe seasonal cycle is one day (288 data points at 300 second intervals), and the seasonal\nadaption parameter will be set to 0.1. The RRD file will store 5 days (1'440 data points) of\nforecasts and deviation predictions before wrap around. The file will store 1 day (a seasonal\ncycle) of 0-1 indicators in the FAILURES RRA.\n\nThe same RRD file and RRAs are created with the following command, which explicitly creates\nall specialized function RRAs using \"STEP, HEARTBEAT, and Rows As Durations\".\n\nrrdtool create monitor.rrd --step 5m \\\nDS:ifOutOctets:COUNTER:30m:0:4294967295 \\\nRRA:AVERAGE:0.5:1:2016 \\\nRRA:HWPREDICT:5d:0.1:0.0035:1d:3 \\\nRRA:SEASONAL:1d:0.1:2 \\\nRRA:DEVSEASONAL:1d:0.1:2 \\\nRRA:DEVPREDICT:5d:5 \\\nRRA:FAILURES:1d:7:9:5\n\nOf course, explicit creation need not replicate implicit create, a number of arguments could\nbe changed.\n", "subsections": [] }, "EXAMPLE 3": { "content": "rrdtool create proxy.rrd --step 300 \\\nDS:Requests:DERIVE:1800:0:U \\\nDS:Duration:DERIVE:1800:0:U \\\nDS:AvgReqDur:COMPUTE:Duration,Requests,0,EQ,1,Requests,IF,/ \\\nRRA:AVERAGE:0.5:1:2016\n\nThis example is monitoring the average request duration during each 300 sec interval for\nrequests processed by a web proxy during the interval. In this case, the proxy exposes two\ncounters, the number of requests processed since boot and the total cumulative duration of\nall processed requests. Clearly these counters both have some rollover point, but using the\nDERIVE data source also handles the reset that occurs when the web proxy is stopped and\nrestarted.\n\nIn the RRD, the first data source stores the requests per second rate during the interval.\nThe second data source stores the total duration of all requests processed during the\ninterval divided by 300. The COMPUTE data source divides each PDP of the AccumDuration by the\ncorresponding PDP of TotalRequests and stores the average request duration. The remainder of\nthe RPN expression handles the divide by zero case.\n", "subsections": [] }, "SECURITY": { "content": "Note that new rrd files will have the permission 0644 regardless of your umask setting. If a\nfile with the same name previously exists, its permission settings will be copied to the new\nfile.\n", "subsections": [] }, "AUTHORS": { "content": "Tobias Oetiker , Peter Stamfest \n\n\n\n1.7.2 2022-03-17 RRDCREATE(1)", "subsections": [] } } } }