{
    "content": [
        {
            "type": "text",
            "text": "# rrdfetch (man)\n\n## NAME\n\nrrdfetch - Fetch data from an RRD.\n\n## SYNOPSIS\n\nrrdtool fetch filename CF [--resolution|-r resolution] [--start|-s start] [--end|-e end]\n[--align-start|-a] [--daemon|-d address]\n\n## DESCRIPTION\n\nThe fetch function is normally used internally by the graph function to get data from RRDs.\nfetch will analyze the RRD and try to retrieve the data in the resolution requested.  The\ndata fetched is printed to stdout. *UNKNOWN* data is often represented by the string \"NaN\"\ndepending on your OS's printf function.\n\n## Sections\n\n- **NAME**\n- **SYNOPSIS**\n- **DESCRIPTION**\n- **ENVIRONMENT VARIABLES**\n- **AUTHOR**\n\nUse structuredContent.sections for detailed options, examples, and full documentation.\n"
        }
    ],
    "structuredContent": {
        "command": "rrdfetch",
        "section": "",
        "mode": "man",
        "summary": "rrdfetch - Fetch data from an RRD.",
        "synopsis": "rrdtool fetch filename CF [--resolution|-r resolution] [--start|-s start] [--end|-e end]\n[--align-start|-a] [--daemon|-d address]",
        "tldr_summary": null,
        "tldr_examples": [],
        "tldr_source": null,
        "flags": [],
        "examples": [],
        "see_also": [],
        "section_outline": [
            {
                "name": "NAME",
                "lines": 2,
                "subsections": []
            },
            {
                "name": "SYNOPSIS",
                "lines": 3,
                "subsections": []
            },
            {
                "name": "DESCRIPTION",
                "lines": 212,
                "subsections": []
            },
            {
                "name": "ENVIRONMENT VARIABLES",
                "lines": 7,
                "subsections": []
            },
            {
                "name": "AUTHOR",
                "lines": 3,
                "subsections": []
            }
        ],
        "sections": {
            "NAME": {
                "content": "rrdfetch - Fetch data from an RRD.\n",
                "subsections": []
            },
            "SYNOPSIS": {
                "content": "rrdtool fetch filename CF [--resolution|-r resolution] [--start|-s start] [--end|-e end]\n[--align-start|-a] [--daemon|-d address]\n",
                "subsections": []
            },
            "DESCRIPTION": {
                "content": "The fetch function is normally used internally by the graph function to get data from RRDs.\nfetch will analyze the RRD and try to retrieve the data in the resolution requested.  The\ndata fetched is printed to stdout. *UNKNOWN* data is often represented by the string \"NaN\"\ndepending on your OS's printf function.\n\nfilename\nthe name of the RRD you want to fetch the data from.\n\nCF      the   consolidation  function  that  is  applied  to  the  data  you  want  to  fetch\n(AVERAGE,MIN,MAX,LAST)\n\n--resolution|-r resolution (default is the highest resolution)\nthe interval you want the values to have (seconds per value).  An optional suffix may\nbe used (e.g. \"5m\" instead of 300 seconds).  rrdfetch will try to match your request,\nbut it will return data even if  no  absolute  match  is  possible.  See  \"RESOLUTION\nINTERVAL\".\n\n--start|-s start (default end-1day)\nstart  of  the  time  series. A time in seconds since epoch (1970-01-01) is required.\nNegative numbers are relative to the current time. By default, one day worth of  data\nwill be fetched. See also \"AT-STYLE TIME SPECIFICATION\" for a detailed explanation on\nways to specify the start time.\n\n--end|-e end (default now)\nthe  end  of  the  time  series  in  seconds  since  epoch.  See  also \"AT-STYLE TIME\nSPECIFICATION\" for a detailed explanation of how to specify the end time.\n\n--align-start|-a\nAutomatically adjust the start time down to be aligned with the resolution.  The end-\ntime is adjusted by the same amount.  This avoids the need for external  calculations\ndescribed  in  RESOLUTION INTERVAL, though if a specific RRA is desired this will not\nensure the start and end fall within its bounds.\n\n--daemon|-d address\nAddress of the rrdcached daemon. If specified, a  \"flush\"  command  is  sent  to  the\nserver before reading the RRD files. This allows rrdtool to return fresh data even if\nthe  daemon  is  configured  to cache values for a long time.  For a list of accepted\nformats, see the -l option in the rrdcached manual.\n\nrrdtool fetch --daemon unix:/var/run/rrdcached.sock /var/lib/rrd/foo.rrd AVERAGE\n\nPlease note that due to thread-safety reasons, the time  specified  with  -s  and  -e\ncannot  use  the  complex  forms described in \"AT-STYLE TIME SPECIFICATION\". The only\naccepted arguments are \"simple integers\". Positive values are interpreted as  seconds\nsince  epoch,  negative  values  (and  zero)  are  interpreted as relative to now. So\n\"1272535035\" refers to \"09:57:15 (UTC), April 29th 2010\" and \"-3600\" means \"one  hour\nago\".\n\nRESOLUTION INTERVAL\nIn order to get RRDtool to fetch anything other than the finest resolution RRA both the start\nand  end  time  must be specified on boundaries that are multiples of the desired resolution.\nConsider the following example:\n\nrrdtool create subdata.rrd -s 10 \\\nDS:ds0:GAUGE:5m:0:U \\\nRRA:AVERAGE:0.5:5m:300h \\\nRRA:AVERAGE:0.5:15m:300h \\\nRRA:AVERAGE:0.5:1h:50d \\\nRRA:MAX:0.5:1h:50d \\\nRRA:AVERAGE:0.5:1d:600d \\\nRRA:MAX:0.5:1d:600d\n\nThis RRD collects data every 10 seconds and stores its averages over 5 minutes, 15 minutes, 1\nhour, and 1 day, as well as the maxima for 1 hour and 1 day.\n\nConsider now that you want to fetch the 15 minute average data for the last hour.  You  might\ntry\n\nrrdtool fetch subdata.rrd AVERAGE -r 15m -s -1h\n\nHowever,  this  will  almost always result in a time series that is NOT in the 15 minute RRA.\nTherefore, the highest resolution RRA, i.e. 5 minute averages, will be chosen which  in  this\ncase is not what you want.\n\nHence, make sure that\n\n1. both start and end time are a multiple of 900 (\"15m\")\n\n2. both start and end time are within the desired RRA\n\nSo, if time now is called \"t\", do\n\nend time == int(t/900)*900,\nstart time == end time - 1hour,\nresolution == 900.\n\nUsing the bash shell, this could look be:\n\nTIME=$(date +%s)\nRRDRES=900\nrrdtool fetch subdata.rrd AVERAGE -r $RRDRES \\\n-e $(($TIME/$RRDRES*$RRDRES)) -s e-1h\n\nOr in Perl:\n\nperl -e '$ctime = time; $rrdres = 900; \\\nsystem \"rrdtool fetch subdata.rrd AVERAGE \\\n-r $rrdres -e @{[int($ctime/$rrdres)*$rrdres]} -s e-1h\"'\n\nOr using the --align-start flag:\n\nrrdtool fetch subdata.rrd AVERAGE -a -r 15m -s -1h\n\nAT-STYLE TIME SPECIFICATION\nApart  from  the  traditional Seconds since epoch, RRDtool does also understand at-style time\nspecification. The specification is called \"at-style\" after the Unix command at(1)  that  has\nmoderately  complex  ways to specify time to run your job at a certain date and time. The at-\nstyle specification consists of two parts: the TIME  REFERENCE  specification  and  the  TIME\nOFFSET specification.\n\nTIME REFERENCE SPECIFICATION\nThe  time  reference specification is used, well, to establish a reference moment in time (to\nwhich the time offset is then applied to). When present, it should come first, when  omitted,\nit  defaults  to  now.  On  its  own part, time reference consists of a time-of-day reference\n(which should come first, if present) and a day reference.\n\nThe time-of-day can be specified as HH:MM, HH.MM, or just HH. You can suffix it with am or pm\nor use 24-hours clock. Some special times of day are understood as well,  including  midnight\n(00:00), noon (12:00) and British teatime (16:00).\n\nThe  day  can  be  specified as month-name day-of-the-month and optional a 2- or 4-digit year\nnumber (e.g. March 8 1999). Alternatively, you can use day-of-week-name (e.g. Monday), or one\nof the words: yesterday, today, tomorrow. You can also specify the day  as  a  full  date  in\nseveral numerical formats, including MM/DD/[YY]YY, DD.MM.[YY]YY, or YYYYMMDD.\n\nNOTE1:  this  is  different  from  the original at(1) behavior, where a single-number date is\ninterpreted as MMDD[YY]YY.\n\nNOTE2: if you specify the day in this way, the time-of-day is REQUIRED as well.\n\nFinally, you can use the words now, start, end or epoch as your time reference. Now refers to\nthe current moment (and is also the default time reference).  Start  (end)  can  be  used  to\nspecify  a  time  relative  to the start (end) time for those tools that use these categories\n(rrdfetch, rrdgraph) and epoch indicates the *IX epoch (*IX timestamp 0 = 1970-01-01 00:00:00\nUTC). epoch is useful to disambiguate between a timestamp value and some forms of abbreviated\ndate/time specifications, because it allows one  to  use  time  offset  specifications  using\nunits,  eg.  epoch+19711205s  unambiguously  denotes  timestamp  19711205  and not 1971-12-05\n00:00:00 UTC.\n\nMonth and day of the week names can be used in their naturally abbreviated  form  (e.g.,  Dec\nfor December, Sun for Sunday, etc.). The words now, start, end can be abbreviated as n, s, e.\n\nTIME OFFSET SPECIFICATION\nThe time offset specification is used to add/subtract certain time intervals to/from the time\nreference  moment. It consists of a sign (+ or -) and an amount. The following time units can\nbe used to specify the amount: years, months, weeks, days, hours, minutes, or seconds.  These\nunits can be used in singular or plural form, and abbreviated naturally or to a single letter\n(e.g.  +3days,  -1wk,  -3y).  Several  time  units  can  be  combined  (e.g.,  -5mon1w2d)  or\nconcatenated (e.g., -5h45min = -5h-45min = -6h+15min = -7h+1h30m-15min, etc.)\n\nNOTE3: If you specify time offset in days, weeks, months, or years, you  will  end  with  the\ntime offset that may vary depending on your time reference, because all those time units have\nno  single  well defined time interval value (1 year contains either 365 or 366 days, 1 month\nis 28 to 31 days long, and even 1 day may be not equal to 24 hours twice a  year,  when  DST-\nrelated  clock adjustments take place).  To cope with this, when you use days, weeks, months,\nor years as your time offset units your time reference date is adjusted  accordingly  without\ntoo  much  further  effort  to ensure anything about it (in the hope that mktime(3) will take\ncare of this later).  This may lead to some  surprising  (or  even  invalid!)  results,  e.g.\n'May 31 -1month'  =  'Apr 31' (meaningless) = 'May 1' (after mktime(3) normalization); in the\nEET timezone '3:30am Mar 29 1999 -1 day' yields '3:30am Mar 28 1999'  (Sunday)  which  is  an\ninvalid  time/date  combination  (because of 3am -> 4am DST forward clock adjustment, see the\nbelow example).\n\nIn contrast, hours, minutes, and seconds are well  defined  time  intervals,  and  these  are\nguaranteed  to  always  produce  time  offsets  exactly  as specified (e.g. for EET timezone,\n'8:00 Mar 27 1999 +2 days' = '8:00 Mar 29 1999', but since there is 1-hour DST forward  clock\nadjustment   that   occurs   around   3:00 Mar 28 1999,  the  actual  time  interval  between\n8:00 Mar 27 1999   and   8:00 Mar 29 1999   equals   47   hours;   on   the    other    hand,\n'8:00 Mar 27 1999 +48 hours' = '9:00 Mar 29 1999', as expected)\n\nNOTE4: The single-letter abbreviation for both months and minutes is m. To disambiguate them,\nthe parser tries to read your mind :) by applying the following two heuristics:\n\n1. If  m  is  used  in  context of (i.e. right after the) years, months, weeks, or days it is\nassumed to mean months, while in the context of  hours,  minutes,  and  seconds  it  means\nminutes.   (e.g., in -1y6m or +3w1m m is interpreted as months, while in -3h20m or +5s2m m\nthe parser decides for minutes).\n\n2. Out of context (i.e. right after the + or - sign) the meaning of m  is  guessed  from  the\nnumber  it  directly  follows.  Currently, if the number's absolute value is below 6 it is\nassumed that m means months, otherwise it is  treated  as  minutes.   (e.g.,  -6m  ==  -6m\nminutes, while +5m == +5 months)\n\nFinal  NOTES:  Time  specification is case-insensitive.  Whitespace can be inserted freely or\nomitted  altogether.   There  are,  however,  cases  when  whitespace  is   required   (e.g.,\n'midnight Thu').  In  this  case  you should either quote the whole phrase to prevent it from\nbeing taken apart by your shell or use '' (underscore) or ',' (comma) which  also  count  as\nwhitespace (e.g., midnightThu or midnight,Thu).\n\nTIME SPECIFICATION EXAMPLES\nOct 12 -- October 12 this year\n\n-1month  or  -1m  -- current time of day, only a month before (may yield surprises, see NOTE3\nabove).\n\nnoon yesterday -3hours -- yesterday morning; can also be specified as 9am-1day.\n\n23:59 31.12.1999 -- 1 minute to the year 2000.\n\n12/31/99 11:59pm -- 1 minute to the year 2000 for imperialists.\n\n12am 01/01/01 -- start of the new millennium\n\nend-3weeks or e-3w -- 3 weeks before end time (may be used as start time specification).\n\nstart+6hours or s+6h -- 6 hours after start time (may be used as end time specification).\n\n931200300 -- 18:45 (UTC), July 5th, 1999 (yes, seconds since 1970 are valid as well).\n\n19970703 12:45 -- 12:45  July 3th, 1997 (my favorite, and it  has  even  got  an  ISO  number\n(8601)).\n",
                "subsections": []
            },
            "ENVIRONMENT VARIABLES": {
                "content": "The following environment variables may be used to change the behavior of \"rrdtool fetch\":\n\nRRDCACHEDADDRESS\nIf  this  environment  variable  is  set  it  will have the same effect as specifying the\n\"--daemon\" option on the command line. If both are present,  the  command  line  argument\ntakes precedence.\n",
                "subsections": []
            },
            "AUTHOR": {
                "content": "Tobias Oetiker <tobi@oetiker.ch>\n\n1.7.2                                        2024-03-31                                  RRDFETCH(1)",
                "subsections": []
            }
        }
    }
}