_R_R_D_C_A_C_H_E_D(1)                         rrdtool                        _R_R_D_C_A_C_H_E_D(1)

NNAAMMEE
     rrdcached - Data caching daemon for rrdtool

SSYYNNOOPPSSIISS
     rrrrddccaacchheedd  [--aa _a_l_l_o_c___s_i_z_e] [--bb _b_a_s_e___d_i_r [--BB]] [--FF] [--ff _t_i_m_e_o_u_t] [--GG _g_r_o_u_p]]
     [--gg]  [--jj _j_o_u_r_n_a_l___d_i_r]  [--LL]  [--ll _a_d_d_r_e_s_s]  [--mm _m_o_d_e]  [--OO]   [--oo _l_o_g___f_i_l_e]
     [--PP _p_e_r_m_i_s_s_i_o_n_s]    [--pp _p_i_d___f_i_l_e]    [--RR]   [--ss _g_r_o_u_p]   [--tt _w_r_i_t_e___t_h_r_e_a_d_s]
     [--UU _u_s_e_r]] [--VV _l_o_g___l_e_v_e_l] [--ww _t_i_m_e_o_u_t] [--zz _d_e_l_a_y]

DDEESSCCRRIIPPTTIIOONN
     rrrrddccaacchheedd is a daemon that receives updates to existing RRD files,  accumu‐
     lates  them and, if enough have been received or a defined time has passed,
     writes the updates to the RRD file. A _f_l_u_s_h command may be  used  to  force
     writing of values to disk, so that graphing facilities and similar can work
     with up-to-date data.

     The  daemon  was  written with big setups in mind. Those setups usually run
     into IO related problems sooner or later for reasons that  are  beyond  the
     scope of this document. Check the wiki at the RRDtool homepage for details.
     Also  check "SECURITY CONSIDERATIONS" below before using this daemon! A de‐
     tailed description of how the daemon operates can be found in the  "HOW  IT
     WORKS" section below.

OOPPTTIIOONNSS
     --ll _a_d_d_r_e_s_s
         Tells the daemon to bind to _a_d_d_r_e_s_s and accept incoming TCP connections
         on  that  socket.  If _a_d_d_r_e_s_s begins with "unix:", everything following
         that prefix is interpreted as the path to a UNIX domain socket.  Other‐
         wise the address or node name are resolved using getaddrinfo().

         For  network  sockets,  a port may be specified by using the form "[[_a_d_‐
         _d_r_e_s_s]]::_p_o_r_t". If the address is an IPv4 address or  a  fully  qualified
         domain  name  (i. e.  the address contains at least one dot (".")), the
         square brackets  can  be  omitted,  resulting  in  the  (simpler)  "_a_d_‐
         _d_r_e_s_s::_p_o_r_t"  pattern.  The default port is 4422221177. If you specify a net‐
         work socket, it is mandatory to read the "SECURITY CONSIDERATIONS" sec‐
         tion.

         The following formats are accepted. Please note that the address of the
         UNIX domain socket mmuusstt start with a slash in the second case!

            unix:</path/to/unix.sock>
            /<path/to/unix.sock>
            <hostname-or-ip>
            [<hostname-or-ip>]:<port>
            <hostname-or-ipv4>:<port>

         Given a port without a host (e.g. "-l :42217") the daemon  will  listen
         on  that port on all network interfaces.  Use "-L" to avoid the need to
         explicitly provide the port if the default port is desired.

         If no --ll option is not specified the default  address,  "unix:/tmp/rrd‐
         cached.sock", will be used.  Multiple --ll options may be provided.

     --LL  Tells  the  daemon to bind to the default TCP port on all available in‐
         terfaces.  It is equivalent to "-l ''" without  the  confusion  of  the
         empty string parameter.

     --ss _g_r_o_u_p___n_a_m_e|_g_i_d
         Set  the  group permissions of a UNIX domain socket. The option accepts
         either a numeric group id or group name. That group will then have both
         read and write permissions (the socket will have file permissions 0760)
         for the socket and, therefore, is able to send commands to the  daemon.
         This  may  be  useful  in  cases  where  you  cannot easily run all RRD
         processes with the same user  privileges  (e.g.  graph  generating  CGI
         scripts  that  typically  run  in  the  permission  context  of the web
         server).

         This option affects the _f_o_l_l_o_w_i_n_g UNIX socket addresses (the  following
         --ll  options)  or  the default socket (if no --ll options have been speci‐
         fied), i.e., you may specify different settings for different sockets.

         The default is not to change ownership or  permissions  of  the  socket
         and, thus, use the system default.

     --mm _m_o_d_e
         Set the file permissions of a UNIX domain socket. The option accepts an
         octal  number  representing  the bit pattern for the mode (see cchhmmoodd(1)
         for details).

         Please note  that  not  all  systems  honor  this  setting.  On  Linux,
         read/write  permissions  are required to connect to a UNIX socket. How‐
         ever, many BSD-derived systems ignore permissions for UNIX sockets. See
         uunniixx(7) for details.

         This option affects the _f_o_l_l_o_w_i_n_g UNIX socket addresses (the  following
         --ll  options)  or  the default socket (if no --ll options have been speci‐
         fied), i.e., you may specify different settings for different sockets.

         The default is not to change ownership or  permissions  of  the  socket
         and, thus, use the system default.

     --PP _c_o_m_m_a_n_d[,_c_o_m_m_a_n_d[,...]]
         Specifies  the  commands accepted via both a network and a UNIX socket.
         This allows administrators of _R_R_D_C_a_c_h_e_D to control the actions accepted
         from various sources.

         The arguments given to the --PP option is a comma separated list of  com‐
         mands.   For  example,  to allow one the "FLUSH" and "PENDING" commands
         one could specify:

           rrdcached -P FLUSH,PENDING $MORE_ARGUMENTS

         The --PP option affects the _f_o_l_l_o_w_i_n_g socket addresses (the following  --ll
         options)  or the default socket (if no --ll options have been specified).
         In the  following  example,  only  the  IPv4  network  socket  (address
         10.0.0.1) will be restricted to the "FLUSH" and "PENDING" commands:

           rrdcached -l unix:/some/path -P FLUSH,PENDING -l 10.0.0.1

         A  complete  list  of  available  commands  can be found in the section
         "Valid Commands" below. There are two minor special exceptions:

         •   The "HELP" and "QUIT" commands are always allowed.

         •   If the "BATCH" command is accepted, the  .. command  will  automati‐
             cally be accepted, too.

         Please also read "SECURITY CONSIDERATIONS" below.

     --VV _l_o_g___l_e_v_e_l
         rrdcached under load can severely flood the logs. This command line op‐
         tion specifies the maximum log_level to be used, meaning that a message
         with verbosity _h_i_g_h_e_r than log_level is muted (LOG_EMERG being the low‐
         est and LOG_DEBUG highest).

         Accepted   values   for  "log_level"  (lowest  to  highest  verbosity):
         LOG_EMERG,  LOG_ALERT,  LOG_CRIT,  LOG_ERR,  LOG_WARNING,   LOG_NOTICE,
         LOG_INFO, LOG_DEBUG

         Default log level when this flag is _N_O_T present: LLOOGG__EERRRR

         See also: _s_y_s_l_o_g_._h

     --oo _l_o_g___f_i_l_e
         Log to the given file instead of syslog.

     --ww _t_i_m_e_o_u_t
         Data  is written to disk every _t_i_m_e_o_u_t seconds.  An optional suffix may
         be used (e.g. "5m" instead of 300 seconds).   If  this  option  is  not
         specified the default interval of 300 seconds will be used.

     --zz _d_e_l_a_y
         If  specified,  rrdcached  will  delay writing of each RRD for a random
         number of seconds in the range [0,_d_e_l_a_y).  This  will  avoid  too  many
         writes  being  queued  simultaneously.  This value should be no greater
         than the value specified in --ww.  An optional suffix may be  used  (e.g.
         "3m" instead of 180 seconds).  By default, there is no delay.

     --ff _t_i_m_e_o_u_t
         Every _t_i_m_e_o_u_t seconds the entire cache is searched for old values which
         are  written  to  disk.  This only concerns files to which updates have
         stopped, so setting this to a high value, such as 3600 seconds, is  ac‐
         ceptable  in most cases.  An optional suffix may be used (e.g. "1h" in‐
         stead of 3600 seconds).  This timeout defaults to 3600 seconds.

     --pp _f_i_l_e
         Sets the name and location of the PID-file. If not specified,  the  de‐
         fault, "_$_l_o_c_a_l_s_t_a_t_e_d_i_r/run/rrdcached.pid" will be used.

     --tt _w_r_i_t_e___t_h_r_e_a_d_s
         Specifies  the  number  of threads used for writing RRD files.  The de‐
         fault is 4.  Increasing this number will allow rrdcached to  have  more
         simultaneous  I/O  requests into the kernel.  This may allow the kernel
         to re-order disk writes, resulting in better disk throughput.

     --jj _d_i_r
         Write updates to a journal in _d_i_r.  In the event of a program or system
         crash, this will allow the daemon to write any updates that were  pend‐
         ing at the time of the crash.

         On  startup, the daemon will check for journal files in this directory.
         If found, all updates therein will be read into memory before the  dae‐
         mon starts accepting new connections.

         The  journal will be rotated with the same frequency as the flush timer
         given by --ff.

         When journaling is enabled, the daemon will use a fast shutdown  proce‐
         dure.   Rather  than  flushing all files to disk, it will make sure the
         journal is properly written and exit  immediately.   Although  the  RRD
         data  files are not fully up-to-date, no information is lost; all pend‐
         ing updates will be replayed from the  journal  next  time  the  daemon
         starts up.

         To disable fast shutdown, use the --FF option.

     --FF  ALWAYS  flush all updates to the RRD data files when the daemon is shut
         down, regardless of journal setting.

     --gg  Run in the foreground.  The daemon will not ffoorrkk(()).

     --bb _d_i_r
         The daemon will change into a specific directory at startup. All  files
         passed  to  the  daemon, that are specified by a rreellaattiivvee path, will be
         interpreted to be relative to this directory. If not given the default,
         "/tmp", will be used.

           +------------------------+------------------------+
           ! Command line           ! File updated           !
           +------------------------+------------------------+
           ! foo.rrd                ! /tmp/foo.rrd           !
           ! foo/bar.rrd            ! /tmp/foo/bar.rrd       !
           ! /var/lib/rrd/foo.rrd   ! /var/lib/rrd/foo.rrd   !
           +------------------------+------------------------+
           Paths given on the command  line and paths actually
           updated by the daemon,  assuming the base directory
           "/tmp".

         WWAARRNNIINNGG:: The paths up to and including the base directory MMUUSSTT  NNOOTT  BBEE
         symbolic links.  In other words, if the base directory is specified as:

             -b /base/dir/somewhere

         ... then NNOONNEE of the following should be symbolic links:

             /base
             /base/dir
             /base/dir/somewhere

     --BB  Only  permit  writes  into  the base directory specified in --bb (and any
         sub-directories).  This does NNOOTT detect symbolic links.  Paths contain‐
         ing "../" will also be blocked.

     --RR  Permit recursive subdirectory creation in the base directory  specified
         in --bb (and any sub-directories). Can only be used when --BB is also set.

     --aa _a_l_l_o_c___s_i_z_e
         Allocate  value pointers in chunks of _a_l_l_o_c___s_i_z_e.  This may improve CPU
         utilization on machines with slow  realloc()  implementations,  in  ex‐
         change  for  slightly higher memory utilization.  The default is 1.  Do
         not set this more than the --ww value divided by your  average  RRD  step
         size.

     --OO  Prevent  the  CREATE command from overwriting existing files, even when
         it is instructed to do so.  This is for added security.

     --GG -_g_r_o_u_p
         When running as daemon and invoked from  a  privileged  account,  reset
         group  privileges  to  those of _g_r_o_u_p.  The group may be specified as a
         name or as a group ID.  The daemon will exit with a  diagnostic  if  it
         cannot successfully transition to the specified group.

     --UU -_u_s_e_r
         When  running  as  daemon  and invoked from a privileged account, reset
         user privileges to those of _u_s_e_r.  The user may be specified as a  name
         or  as  a user ID.  The daemon will exit with a diagnostic if it cannot
         successfully transition to the specified user.

AAFFFFEECCTTEEDD RRRRDDTTOOOOLL CCOOMMMMAANNDDSS
     The following commands may be made aware of the rrrrddccaacchheedd using the command
     line argument ----ddaaeemmoonn or the environment variable RRRRDDCCAACCHHEEDD__AADDDDRREESSSS:

     •   dump

     •   fetch

     •   flush

     •   graph

     •   graphv

     •   info

     •   first

     •   last

     •   lastupdate

     •   update

     •   xport

     •   create

     •   list

     The uuppddaattee command can send values to the daemon instead of writing them to
     the disk itself. All other commands can send a FFLLUUSSHH command (see below) to
     the daemon before accessing the files, so they work  with  up-to-date  data
     even if the cache timeout is large.

EERRRROORR RREEPPOORRTTIINNGG
     The  daemon  reports  errors in one of two ways: During startup, error mes‐
     sages are printed to "STDERR". One of the steps when starting up is to fork
     to the background and closing "STDERR" - after this writing directly to the
     user is no longer possible. Once this has happened, the  daemon  will  send
     log  messages  to  the  system logging daemon using ssyysslloogg(3). The facility
     used is "LOG_DAEMON".

HHOOWW IITT WWOORRKKSS
     When receiving an update, rrrrddccaacchheedd does not write to disk but looks for an
     entry for that file in its internal tree. If not found, an entry is created
     including the current time (called "First" in the diagram below). This time
     is nnoott the time specified on the command line but the  time  the  operating
     system  considers  to  be  "now".  The  value and time of the value (called
     "Time" in the diagram below) are appended to the tree node.

     When appending a value to a tree node, it is checked whether it's  time  to
     write   the   values   to   disk.   Values   are   written   to   disk   if
     "now() - First >= timeout", where "timeout" is the timeout specified  using
     the  --ww  option, see "OPTIONS". If the values are "old enough" they will be
     enqueued in the "update queue", i. e. they will be appended to  the  linked
     list  shown  below.   Because the tree nodes and the elements of the linked
     list are the same data structures in memory, any update to a file that  has
     already  been enqueued will be written with the next write to the RRD file,
     too.

     A separate "update thread" constantly dequeues the first element in the up‐
     date queue and writes all its values to the appropriate file. So as long as
     the update queue is not empty files are written  at  the  highest  possible
     rate.

     Since the timeout of files is checked only when new values are added to the
     file,  "dead"  files, i. e. files that are not updated anymore, would never
     be written to disk. Therefore, every now and then, controlled by the --ff op‐
     tion, the entire tree is walked and all "old" values  are  enqueued.  Since
     this  only  affects  "dead" files and walking the tree is relatively expen‐
     sive, you should set the "flush interval" to a reasonably high  value.  The
     default is 3600 seconds (one hour).

     The  downside of caching values is that they won't show up in graphs gener‐
     ated from the RRD files. To get around this, the daemon provides the "flush
     command" to flush specific files. This means that the file is  inserted  at
     the  hheeaadd of the update queue or moved there if it is already enqueued. The
     flush command will return only after the file's pending updates  have  been
     written to disk.

      +------+   +------+                               +------+
      ! head !   ! root !                               ! tail !
      +---+--+   +---+--+                               +---+--+
          !         /\                                      !
          !        /  \                                     !
          !       /\  /\                                    !
          !      /\/\ \ `----------------- ... --------,    !
          V     /      `-------,                       !    V
      +---+----+---+    +------+-----+             +---+----+---+
      ! File:  foo !    ! File:  bar !             ! File:  qux !
      ! First: 101 !    ! First: 119 !             ! First: 180 !
      ! Next:&bar -+--->! Next:&... -+---> ... --->! Next:NULL  !
      | Prev:NULL  !<---+-Prev:&foo  !<--- ... ----+-Prev: &... !
      +============+    +============+             +============+
      ! Time:  100 !    ! Time:  120 !             ! Time:  180 !
      ! Value:  10 !    ! Value: 0.1 !             ! Value: 2,2 !
      +------------+    +------------+             +------------+
      ! Time:  110 !    ! Time:  130 !             ! Time:  190 !
      ! Value:  26 !    ! Value: 0.1 !             ! Value: 7,3 !
      +------------+    +------------+             +------------+
      :            :    :            :             :            :
      +------------+    +------------+             +------------+
      ! Time:  230 !    ! Time:  250 !             ! Time:  310 !
      ! Value:  42 !    ! Value: 0.2 !             ! Value: 1,2 !
      +------------+    +------------+             +------------+

     The above diagram demonstrates:

     •   Files/values are stored in a (balanced) tree.

     •   Tree nodes and entries in the update queue are the same data structure.

     •   The local time ("First") and the time specified in updates ("Time") may
         differ.

     •   Timed out values are inserted at the "tail".

     •   Explicitly flushed values are inserted at the "head".

     •   ASCII art rocks.

SSEECCUURRIITTYY CCOONNSSIIDDEERRAATTIIOONNSS
   AAuutthheennttiiccaattiioonn
     If  your rrdtool installation was built without libwrap there is no form of
     authentication for clients connecting to the rrdcache daemon!

     If your rrdtool installation was  built  with  libwrap  then  you  can  use
     hosts_access  to restrict client access to the rrdcache daemon (rrdcached).
     For more information on how to use hosts_access to restrict access  to  the
     rrdcache daemon you should read the hhoossttss__aacccceessss(5) man pages.

     It is still highly recommended to install a packet filter or similar mecha‐
     nism  to prevent unauthorized connections. Unless you have a dedicated VLAN
     or VPN for this, using network sockets is probably a bad idea!

   AAuutthhoorriizzaattiioonn
     There is minimal per-socket authorization.

     Authorization is currently done on a  per-socket  basis.  That  means  each
     socket  has  a  list of commands it will accept and it will accept. It will
     accept only those commands explicitly listed but it will (currently) accept
     these commands from anyone reaching the socket.

     If the networking sockets are to be used, it is necessary to  restrict  the
     accepted commands to those needed by external clients. If, for example, ex‐
     ternal  clients want to draw graphs of the cached data, they should only be
     allowed to use the "FLUSH" command.

     Authorization does not work when rrdcached is socket-activated by systemd.

   EEnnccrryyppttiioonn
     There is no encryption.

     Again, this may be added in the future, but for the time being it  is  your
     job to keep your private data private. Install a VPN or an encrypted tunnel
     if you statistics are confidential!

   SSaanniittyy cchheecckkiinngg
     There is no sanity checking.

     The  daemon  will  blindly  write  to  any file it gets told, so you really
     should create a separate user just for this daemon. Also it does not do any
     sanity checks, so if it gets told to write values for a time far in the fu‐
     ture, your files will be messed up good!

   CCoonncclluussiioonn
     •   Security is the job of the administrator.

     •   We recommend to allow write access via UNIX domain sockets only.

     •   You have been warned.

PPRROOTTOOCCOOLL
     The daemon communicates with clients using  a  line  based  ASCII  protocol
     which  is  easy to read and easy to type. This makes it easy for scripts to
     implement the protocol and possible for users to use telnet to  connect  to
     the daemon and test stuff "by hand".

     The  protocol is line based, this means that each record consists of one or
     more lines. A line is terminated by the line feed character 0x0A,  commonly
     written  as  "\n". In the examples below, this character will be written as
     "<LF>" ("line feed").

     After the connection has been established, the client is expected to send a
     "command". A command consists of the command keyword, possibly  some  argu‐
     ments,  and  a  terminating  newline character. For a list of commands, see
     "Valid Commands" below.

     Example:

       FLUSH /tmp/foo.rrd<LF>

     The daemon answers with a line consisting of a status code and a short sta‐
     tus message, separated by one or more space characters. A  negative  status
     code  signals  an  error, a positive status code or zero signal success. If
     the status code is greater than zero, it indicates the number of lines that
     follow the status line.

     Examples:

      0 Success<LF>

      2 Two lines follow<LF>
      This is the first line<LF>
      And this is the second line<LF>

   VVaalliidd CCoommmmaannddss
     The following commands are understood by the daemon:

     FFLLUUSSHH _f_i_l_e_n_a_m_e
         Causes the daemon to put _f_i_l_e_n_a_m_e to the hheeaadd of the update queue (pos‐
         sibly moving it there if the node is already enqueued). The answer will
         be sent aafftteerr the node has been dequeued.

     FFLLUUSSHHAALLLL
         Causes the daemon to start flushing ALL pending values to  disk.   This
         returns immediately, even though the writes may take a long time.

     PPEENNDDIINNGG _f_i_l_e_n_a_m_e
         Shows  any  "pending"  updates for a file, in order.  The updates shown
         have not yet been written to the underlying RRD file.

     FFEETTCCHH _f_i_l_e_n_a_m_e _C_F [_s_t_a_r_t [_e_n_d] [_d_s ...]]
         Calls "rrd_fetch" with the specified arguments and returns  the  result
         in  text  form.  If  necessary,  the file is flushed to disk first. The
         client side function "rrdc_fetch" (declared in  "rrd_client.h")  parses
         the  output and behaves just like "rrd_fetch_r" for easy integration of
         remote queries.  ds defines the columns to dump -  if  none  are  given
         then all are returned

     FFEETTCCHHBBIINN _f_i_l_e_n_a_m_e _C_F [_s_t_a_r_t [_e_n_d] [_d_s ...]]
         Calls  "rrd_fetch"  with the specified arguments and returns the result
         in text/binary form to avoid unnecessary un/marshalling  overhead.   If
         necessary,  the file is flushed to disk first. The client side function
         "rrdc_fetch" (declared in "rrd_client.h") parses the output and behaves
         just like "rrd_fetch_r" for easy integration of remote queries.  ds de‐
         fines the columns to dump - if none are given then all are returned

     FFOORRGGEETT _f_i_l_e_n_a_m_e
         Removes _f_i_l_e_n_a_m_e from the cache.  Any pending updates WWIILLLL BBEE LLOOSSTT.

     QQUUEEUUEE
         Shows the files that are on the output queue.   Returns  zero  or  more
         lines in the following format, where <num_vals> is the number of values
         to be written for the <file>:

             <num_vals> <file>

     HHEELLPP [_c_o_m_m_a_n_d]
         Returns  a  short  usage message. If no command is given, or _c_o_m_m_a_n_d is
         HHEELLPP, a list of commands supported by the daemon is returned. Otherwise
         a short description, possibly containing a pointer to a manual page, is
         returned.  Obviously, this is meant for interactive usage and the  for‐
         mat  in which the commands and usage summaries are returned is not well
         defined.

     SSTTAATTSS
         Returns a list of metrics which can be used to measure the daemons per‐
         formance and check its status. For a  description  of  the  values  re‐
         turned, see "Performance Values" below.

         The  format  in  which the values are returned is similar to many other
         line based protocols: Each value is printed on a  separate  line,  each
         consisting  of  the  name of the value, a colon, one or more spaces and
         the actual value.

         Example:

          9 Statistics follow
          QueueLength: 0
          UpdatesReceived: 30
          FlushesReceived: 2
          UpdatesWritten: 13
          DataSetsWritten: 390
          TreeNodesNumber: 13
          TreeDepth: 4
          JournalBytes: 190
          JournalRotate: 0

     PPIINNGGPING-PONG, this is very useful when using connection pool between  user
         client and RRDCACHED.

         Example:

          0 PONG

     UUPPDDAATTEE _f_i_l_e_n_a_m_e _v_a_l_u_e_s [_v_a_l_u_e_s ...]
         Adds  more data to a filename. This is tthhee operation the daemon was de‐
         signed for, so describing the mechanism again is unnecessary. Read "HOW
         IT WORKS" above for a detailed explanation.

         Note that rrdcached only accepts absolute timestamps in the update val‐
         ues.  Updates strings like "N:1:2:3" are automatically converted to ab‐
         solute time by the RRD client library before sending to rrdcached.

     WWRROOTTEE _f_i_l_e_n_a_m_e
         This command is written to the journal after  a  file  is  successfully
         written  out  to  disk.   It is used during journal replay to determine
         which updates have already been applied.  It is _o_n_l_y valid in the jour‐
         nal; it is not accepted from the other command channels.

     FFIIRRSSTT _f_i_l_e_n_a_m_e [_r_r_a_n_u_m]
         Return the timestamp for the first CDP in the specified  RRA.   Default
         is to use RRA zero if none is specified.

     LLAASSTT _f_i_l_e_n_a_m_e
         Return  the  timestamp  for  the last update to the specified RRD. Note
         that the cache is _n_o_t flushed before checking, as  the  client  is  ex‐
         pected to request this separately if it is required.

     IINNFFOO _f_i_l_e_n_a_m_e
         Return  the  configuration information for the specified RRD. Note that
         the cache is _n_o_t flushed before checking, as the client is expected  to
         request this separately if it is required.

         The information is returned, one item per line, with the format:

          I<keyname> I<type> I<value>

     CCRREEAATTEE _f_i_l_e_n_a_m_e [-s _s_t_e_p_s_i_z_e] [-b _b_e_g_i_n_t_i_m_e] [-r _s_o_u_r_c_e_f_i_l_e ...] [-t _t_e_m_‐
     _p_l_a_t_e_f_i_l_e] [-O] _D_S_d_e_f_i_n_i_t_i_o_n_s ... _R_R_A_d_e_f_i_n_i_t_i_o_n_s ...
         This  will  create  the  RRD file according to the supplied parameters,
         provided the parameters are valid, and (if the -O option is given or if
         the rrdcached was started with the -O flag) the specified _f_i_l_e_n_a_m_e does
         not already exist.

     BBAATTCCHH
         This command initiates the bulk load of multiple commands.  This is de‐
         signed for installations with extremely high  update  rates,  since  it
         permits more than one command to be issued per rreeaadd(()) and wwrriittee(()).

         All  commands are executed just as they would be if given individually,
         except for output to the user.  Messages indicating  success  are  sup‐
         pressed, and error messages are delayed until the client is finished.

         Command processing is finished when the client sends a dot (".") on its
         own  line.   After the client has finished, the server responds with an
         error count and the list of error messages (if any).  Each  error  mes‐
         sages  indicates the number of the command to which it corresponds, and
         the error message itself.  The first user command after BBAATTCCHH  is  com‐
         mand number one.

             client:  BATCH
             server:  0 Go ahead.  End with dot '.' on its own line.
             client:  UPDATE x.rrd 1223661439:1:2:3            <--- command #1
             client:  UPDATE y.rrd 1223661440:3:4:5            <--- command #2
             client:  and so on...
             client:  .
             server:  2 Errors
             server:  1 message for command 1
             server:  12 message for command 12

     LLIISSTT [RECURSIVE] I/<path>
         This  command  allows  to list directories and rrd databases as seen by
         the daemon.  The root "directory" is the base_dir (see '-b dir').  When
         invoked  with  'LIST RECURSIVE /<path>' it will behave similarly to 'ls
         -R' but limited to rrd files (listing all the rrd bases in the  subtree
         of <path>, skipping empty directories).

     SSUUSSPPEENNDD _f_i_l_e_n_a_m_e
         Suspend  writing to an RRD file. While a file is suspended, all metrics
         for it are cached in memory until RREESSUUMMEE is called for that file or RREE‐‐
         SSUUMMEEAALLLL is called.

     RREESSUUMMEE _f_i_l_e_n_a_m_e
         Resume writing to an RRD file previously suspended by SSUUSSPPEENNDD  or  SSUUSS‐‐
         PPEENNDDAALLLL.

     SSUUSSPPEENNDDAALLLL
         Suspend  writing  to all RRD files. While a file is suspended, all met‐
         rics for it are cached in memory until RREESSUUMMEE is called for  that  file
         or RREESSUUMMEEAALLLL is called.

     RREESSUUMMEEAALLLL
         Resume writing to all RRD files previously suspended by SSUUSSPPEENNDD or SSUUSS‐‐
         PPEENNDDAALLLL.

     QQUUIITTDisconnect from rrdcached.

   PPeerrffoorrmmaannccee VVaalluueess
     The following counters are returned by the SSTTAATTSS command:

     QQuueeuueeLLeennggtthh _(_u_n_s_i_g_n_e_d _6_4_b_i_t _i_n_t_e_g_e_r_)
         Number of nodes currently enqueued in the update queue.

     UUppddaatteessRReecceeiivveedd _(_u_n_s_i_g_n_e_d _6_4_b_i_t _i_n_t_e_g_e_r_)
         Number of UPDATE commands received.

     FFlluusshheessRReecceeiivveedd _(_u_n_s_i_g_n_e_d _6_4_b_i_t _i_n_t_e_g_e_r_)
         Number of FLUSH commands received.

     UUppddaatteessWWrriitttteenn _(_u_n_s_i_g_n_e_d _6_4_b_i_t _i_n_t_e_g_e_r_)
         Total  number of updates, i. e. calls to "rrd_update_r", since the dae‐
         mon was started.

     DDaattaaSSeettssWWrriitttteenn _(_u_n_s_i_g_n_e_d _6_4_b_i_t _i_n_t_e_g_e_r_)
         Total number of "data sets"  written  to  disk  since  the  daemon  was
         started. A data set is one or more values passed to the UUPPDDAATTEE command.
         For  example: "1223661439:123:456" is one data set with two values. The
         term "data set" is used to prevent confusion whether individual  values
         or groups of values are counted.

     TTrreeeeNNooddeessNNuummbbeerr _(_u_n_s_i_g_n_e_d _6_4_b_i_t _i_n_t_e_g_e_r_)
         Number of nodes in the cache.

     TTrreeeeDDeepptthh _(_u_n_s_i_g_n_e_d _6_4_b_i_t _i_n_t_e_g_e_r_)
         Depth of the tree used for fast key lookup.

     JJoouurrnnaallBByytteess _(_u_n_s_i_g_n_e_d _6_4_b_i_t _i_n_t_e_g_e_r_)
         Total number of bytes written to the journal since startup.

     JJoouurrnnaallRRoottaattee _(_u_n_s_i_g_n_e_d _6_4_b_i_t _i_n_t_e_g_e_r_)
         Number of times the journal has been rotated since startup.

SSIIGGNNAALLSS
     SIGINT and SIGTERM
         The daemon exits normally on receipt of either of these signals.  Pend‐
         ing updates are handled in accordance with the --jj and --FF options.

     SIGUSR1
         The daemon exits AFTER flushing all updates out to disk.  This may take
         a while.

     SIGUSR2
         The  daemon  exits  immediately,  without flushing updates out to disk.
         Pending updates will be replayed  from  the  journal  when  the  daemon
         starts up again.  WWAARRNNIINNGG:: iiff jjoouurrnnaalliinngg ((--jj)) iiss NNOOTT eennaabblleedd,, aannyy ppeenndd‐‐
         iinngg uuppddaatteess WWIILLLL BBEE LLOOSSTT.

BBUUGGSS
     No known bugs at the moment.

SSEEEE AALLSSOO
     rrdtool, rrdgraph

AAUUTTHHOORR
     Florian Forster <octo at verplant.org>

     Both rrrrddccaacchheedd and this manual page have been written by Florian.

CCOONNTTRRIIBBUUTTOORRSS
     kevin  brintnall  <kbrint@rufus.net> Steve Shipway <steve@steveshipway.org>
     Martin Sperl <rrdtool@martin.sperl.org>

1.10.0                             2026-05-23                       _R_R_D_C_A_C_H_E_D(1)
