path: root/lib/kernel/doc/src/disk_log.xml

<?xml version="1.0" encoding="utf-8" ?>
<!DOCTYPE erlref SYSTEM "erlref.dtd">

<erlref>
  <header>
    <copyright>
      <year>1997</year>
      <year>2016</year>
      <holder>Ericsson AB, All Rights Reserved</holder>
    </copyright>
    <legalnotice>
  Licensed under the Apache License, Version 2.0 (the "License");
  you may not use this file except in compliance with the License.
  You may obtain a copy of the License at
 
      http://www.apache.org/licenses/LICENSE-2.0

  Unless required by applicable law or agreed to in writing, software
  distributed under the License is distributed on an "AS IS" BASIS,
  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  See the License for the specific language governing permissions and
  limitations under the License.

  The Initial Developer of the Original Code is Ericsson AB.
    </legalnotice>

    <title>disk_log</title>
    <prepared>Claes Wikstr&ouml;m</prepared>
    <responsible>Claes Wikstr&ouml;m</responsible>
    <docno></docno>
    <approved>nobody</approved>
    <checked>no</checked>
    <date>1999-10-10</date>
    <rev>D</rev>
    <file>disk_log.sgml</file>
  </header>
  <module>disk_log</module>
  <modulesummary>A disk-based term logging facility.</modulesummary>
  <description>
    <p><c>disk_log</c> is a disk-based term logger that enables
      efficient logging of items on files.</p>
    <p>Two types of logs are supported:</p>
      <taglist>
	<tag>halt logs</tag>
	<item><p>Appends items to a single file, which size can
	be limited by the disk log module.</p></item>
	<tag>wrap logs</tag>
	<item><p>Uses a sequence of wrap log files of limited size. As a
	wrap log file is filled up, further items are logged on to the next
	file in the sequence, starting all over with the first file when
	the last file is filled up.</p></item>
      </taglist>
     <p>For efficiency reasons, items are always written to files as binaries.</p>

    <p>Two formats of the log files are supported:</p>
      <taglist>
	<tag>internal format</tag>
	<item><p>Supports automatic repair of log files that are not
	properly closed and enables efficient reading of logged items in
	<em>chunks</em> using a set of functions defined in this module.
	This is the only way to read internally formatted logs.
	An item logged to an internally formatted log must not occupy more
	than 4 GB of disk space (the size must fit in 4 bytes).</p></item>
	<tag>external format</tag>
	<item><p>Leaves it up to the user to read the logged deep byte lists.
	The disk log module cannot repair externally formatted logs.</p></item>
      </taglist>

    <p>For each open disk log, one process handles requests
      made to the disk log. This process is created when
      <seealso marker="#open/1"><c>open/1</c></seealso>
      is called, provided there exists no process handling the disk log.
      A process that opens a disk log can be an <em>owner</em>
      or an anonymous <em>user</em> of the disk log. Each owner is 
      linked to the disk log process, and an owner can close the disk log
      either explicitly (by calling <c>close/1</c> or <c>lclose/1,2</c>)
      or by terminating.</p>
      <p>Owners can subscribe to <em>notifications</em>,
      messages of the form <c>{disk_log, Node, Log, Info}</c>, which are sent
      from the disk log process when certain events occur, see 
      the functions and in particular the <c>open/1</c> option
      <seealso marker="#notify"><c>notify</c></seealso>.
      A log can have many owners, but a process cannot own a
      log more than once. However, the same process can open the log
      as a user more than once.</p>
      <p>For a disk log process to close its file properly and terminate,
      it must be closed by its owners and once by some non-owner process
      for each time the log was used anonymously. The users are counted
      and there must not be any users left when the disk log process terminates.
      </p>
    <p>Items can be logged <em>synchronously</em> by using functions
      <seealso marker="#log/2"><c>log/2</c></seealso>,
      <seealso marker="#blog/2"><c>blog/2</c></seealso>,
      <seealso marker="#log_terms/2"><c>log_terms/2</c></seealso>, and
      <seealso marker="#blog_terms/2"><c>blog_terms/2</c></seealso>.
      For each of these functions, the caller is put
      on hold until the items are logged (but not necessarily
      written, use <c>sync/1</c> to ensure that). By adding an <c>a</c>
      to each of the mentioned function names, we get functions that log
      items <em>asynchronously</em>. Asynchronous functions do not wait for
      the disk log process to write the items to the file, but
      return the control to the caller more or less immediately.
      </p>
    <p>When using the internal format for logs, use functions
      <seealso marker="#log/2"><c>log/2</c></seealso>,
      <seealso marker="#log_terms/2"><c>log_terms/2</c></seealso>,
      <seealso marker="#alog/2"><c>alog/2</c></seealso>, and
      <seealso marker="#alog_terms/2"><c>alog_terms/2</c></seealso>.
      These functions log one or more Erlang terms.
      By prefixing each of the functions with a <c>b</c> (for "binary"),
      we get the corresponding <c>blog()</c> functions for the external format.
      These functions log one or more deep lists of bytes or, alternatively,
      binaries of deep lists of bytes.
      For example, to log the string <c>"hello"</c> in ASCII format, you
      can use <c>disk_log:blog(Log, "hello")</c>, or
      <c>disk_log:blog(Log, list_to_binary("hello"))</c>. The two
      alternatives are equally efficient.</p>
      <p>The <c>blog()</c> functions can also be used for internally formatted
      logs, but in this case they must be called with binaries constructed
      with calls to
      <seealso marker="erts:erlang#term_to_binary/1"><c>term_to_binary/1</c></seealso>.
      There is no check to ensure
      this, it is entirely the responsibility of the caller. If these
      functions are called with binaries that do not correspond to
      Erlang terms, the
      <seealso marker="#chunk/2"><c>chunk/2,3</c></seealso>
      and automatic repair
      functions fail. The corresponding terms (not the binaries)
      are returned when <c>chunk/2,3</c> is called.
      </p>
    <p>A collection of open disk logs with the same name running on
      different nodes is said to be a <em>distributed disk log</em>
      if requests made to any of the logs are automatically made to
      the other logs as well. The members of such a collection are
      called individual distributed disk logs, or just distributed
      disk logs if there is no risk of confusion. There is no order
      between the members of such a collection. For example, logged
      terms are not necessarily written to the node where the
      request was made before written to the other nodes. However,
      a few functions do not make requests to all
      members of distributed disk logs, namely
      <seealso marker="#info/1"><c>info/1</c></seealso>,
      <seealso marker="#chunk/2"><c>chunk/2,3</c></seealso>,
      <seealso marker="#bchunk/2"><c>bchunk/2,3</c></seealso>,
      <seealso marker="#chunk_step/3"><c>chunk_step/3</c></seealso>, and
      <seealso marker="#lclose/1"><c>lclose/1,2</c></seealso>.</p>
     <p>An open disk log that is not a distributed disk
      log is said to be a <em>local disk log</em>. A local disk log is
      only accessible from the node where the disk log process runs,
      whereas a distributed disk log is accessible from all nodes in
      the Erlang system, except for those nodes where a local
      disk log with the same name as the distributed disk log exists.
      All processes on nodes that have access to a local or
      distributed disk log can log items or otherwise change, inspect,
      or close the log.
      </p>
    <p>It is not guaranteed that all log files of a distributed disk log 
      contain the same log items. No attempt is made to synchronize
      the contents of the files. However, as long as at least one of 
      the involved nodes is alive at each time, all items are logged.
      When logging items to a distributed log, or otherwise trying to
      change the log, the replies from individual logs are
      ignored. If all nodes are down, the disk log functions
      reply with a <c>nonode</c> error.
      </p>
    <note>
      <p>In some applications, it can be unacceptable that
        replies from individual logs are ignored. An alternative in such
        situations is to use many local disk logs instead of one
        distributed disk log, and implement the distribution without use
        of the <c>disk_log</c> module.</p>
    </note>
    <p>Errors are reported differently for asynchronous log attempts
      and other uses of the <c>disk_log</c> module. When used synchronously,
      this module replies with an error message, but when called
      asynchronously, this module does not know where to send
      the error message. Instead, owners subscribing to notifications
      receive an <c>error_status</c> message. 
      </p>
    <p>The <c>disk_log</c> module does not report errors to the
      <seealso marker="error_logger"><c>error_logger</c></seealso>
      module. It is up to the caller to decide
      whether to employ the error logger. Function
      <seealso marker="#format_error/1"><c>format_error/1</c></seealso>
      can be used to produce readable messages from error replies.
      However, information events are sent to the error logger in two
      situations, namely when a log is repaired, or when a file is missing
      while reading chunks.
      </p>
    <p>Error message <c>no_such_log</c> means that the specified
      disk log is not open. Nothing is said about whether the disk log
      files exist or not.
      </p>
    <note>
      <p>If an attempt to reopen or truncate a log fails (see
        <seealso marker="#reopen/2"><c>reopen/2,3</c></seealso>
	and
	<seealso marker ="#truncate/1"><c>truncate/1,2</c></seealso>)
	the disk log process terminates immediately. Before the process
	terminates, links to owners and blocking processes (see
	<seealso marker="#block/1"><c>block/1,2</c></seealso>) are removed.
        The effect is that the links work in one direction only. Any
        process using a disk log must check for error message
        <c>no_such_log</c> if some other process truncates or
        reopens the log simultaneously.</p>
    </note>
  </description>
  <datatypes>
    <datatype>
      <name name="log"/>
    </datatype>
    <datatype>
      <name name="dlog_size"/>
    </datatype>
    <datatype>
      <name name="dlog_format"/>
    </datatype>
    <datatype>
      <name name="dlog_head_opt"/>
    </datatype>
    <datatype>
      <name name="dlog_byte"/>
    </datatype>
    <datatype>
      <name name="dlog_mode"/>
    </datatype>
    <datatype>
      <name name="dlog_type"/>
    </datatype>
    <datatype>
      <name name="continuation"/>
      <desc><p>Chunk continuation returned by
         <c>chunk/2,3</c>, <c>bchunk/2,3</c>, or <c>chunk_step/3</c>.</p>
      </desc>
    </datatype>
    <datatype>
      <name name="bytes"/>
    </datatype>
    <datatype>
      <name name="invalid_header"/>
    </datatype>
    <datatype>
      <name name="file_error"/>
    </datatype>
  </datatypes>
  <funcs>
    <func>
      <name name="accessible_logs" arity="0"/>
      <fsummary>Return the accessible disk logs on the current node.</fsummary>
      <desc>
        <p>Returns the names of the disk logs accessible on the current node.
          The first list contains local disk logs and the
          second list contains distributed disk logs.
          </p>
      </desc>
    </func>
    <func>
      <name name="alog" arity="2"/>
      <name name="balog" arity="2"/>
      <fsummary>Asynchronously log an item on to a disk log.</fsummary>
      <type variable="Log"/>
      <type variable="Term" name_i="1"/>
      <type variable="Bytes"/>
      <type name="notify_ret"/>
      <desc>
        <p>Asynchronously append an item to a disk log. <c>alog/2</c> is
          used for internally formatted logs and <c>balog/2</c>
          for externally formatted logs. <c>balog/2</c> can also be used
          for internally formatted logs if the binary is
          constructed with a call to
	  <seealso marker="erts:erlang#term_to_binary/1"><c>term_to_binary/1</c></seealso>.
          </p>
        <p>Owners subscribing to notifications receive
          message <c>read_only</c>, <c>blocked_log</c>,
          or <c>format_external</c> if the item cannot be written
          on the log, and possibly one of the messages <c>wrap</c>,
          <c>full</c>, or <c>error_status</c> if an item is written
          on the log. Message <c>error_status</c> is sent if
          something is wrong with the header function or if a file error
          occurs.
          </p>
      </desc>
    </func>
    <func>
      <name name="alog_terms" arity="2"/>
      <name name="balog_terms" arity="2"/>
      <fsummary>Asynchronously log many items on to a disk log.</fsummary>
      <type variable="Log"/>
      <type variable="TermList" name_i="1"/>
      <type variable="ByteList"/>
      <type name="notify_ret"/>
      <desc>
        <p>Asynchronously append a list of items to a disk log.
          <c>alog_terms/2</c> is used for internally
          formatted logs and <c>balog_terms/2</c>
          for externally formatted logs. <c>balog_terms/2</c> can also be used
          for internally formatted logs if the binaries are
          constructed with calls to
	  <seealso marker="erts:erlang#term_to_binary/1"><c>term_to_binary/1</c></seealso>.
          </p>
        <p>Owners subscribing to notifications receive
          message <c>read_only</c>, <c>blocked_log</c>,
          or <c>format_external</c> if the items cannot be written
          on the log, and possibly one or more of the messages <c>wrap</c>,
          <c>full</c>, and <c>error_status</c> if items are written
          on the log. Message <c>error_status</c> is sent if
          something is wrong with the header function or if a file error
          occurs.
          </p>
      </desc>
    </func>
    <func>
      <name name="block" arity="1"/>
      <name name="block" arity="2"/>
      <fsummary>Block a disk log.</fsummary>
      <type name="block_error_rsn"/>
      <desc>
        <p>With a call to <c>block/1,2</c> a process can block a log.
          If the blocking process is not an owner of the log, a temporary
          link is created between the disk log process and the blocking
          process. The link ensures that the disk log is
          unblocked if the blocking process terminates without
          first closing or unblocking the log.
          </p>
        <p>Any process can probe a blocked log with <c>info/1</c> or
          close it with <c>close/1</c>. The blocking process can also
          use functions <c>chunk/2,3</c>, <c>bchunk/2,3</c>,
          <c>chunk_step/3</c>, and <c>unblock/1</c> without being
          affected by the block. Any other attempt than those
          mentioned so far to update or read a blocked log suspends the
          calling process until the log is unblocked or returns
          error message <c>{blocked_log, <anno>Log</anno>}</c>, depending on
          whether the value of <c><anno>QueueLogRecords</anno></c> is <c>true</c>
          or <c>false</c>. <c><anno>QueueLogRecords</anno></c> defaults to
          <c>true</c>, which is used by <c>block/1</c>.
          </p>
      </desc>
    </func>
    <func>
      <name name="change_header" arity="2"/>
      <fsummary>Change option head or head_func for an owner of a disk log.</fsummary>
      <desc>
        <p>Changes the value of option <c>head</c> or <c>head_func</c> for an owner of a disk log.</p>
      </desc>
    </func>
    <func>
      <name name="change_notify" arity="3"/>
      <fsummary>Change option notify for an owner of a disk log.</fsummary>
      <desc>
        <p>Changes the value of option <c>notify</c> for an owner of a disk log. </p>
      </desc>
    </func>
    <func>
      <name name="change_size" arity="2"/>
      <fsummary>Change the size of an open disk log.</fsummary>
      <desc>
        <p>Changes the size of an open log.
          For a halt log, the size can always be increased,
          but it cannot be decreased to something less than
          the current file size.
          </p>
        <p>For a wrap log, both the size and the number of files can always
	  be increased, as long as the number of files does not
          exceed 65000. If the maximum number of files is decreased, the
          change is not valid until the current file is full and the
          log wraps to the next file.
          The redundant files are removed the next time the log wraps around,
          that is, starts to log to file number 1.
          </p>
        <p>As an example, assume that the old maximum number of files 
          is 10 and that the new maximum number of files is 6. If 
          the current file number is not greater than the new maximum number 
          of files, files 7-10 are removed when file 6
          is full and the log starts to write to file number 1 again. 
          Otherwise, the files greater than the current
          file are removed when the current file is full (for example, if
          the current file is 8, files 9 and 10 are removed). The files between
          the new maximum number of files and the current
          file (that is, files 7 and 8) are removed the next time file 6
          is full.
          </p>
        <p>If the size of the files is decreased, the change immediately
          affects the current log. It does not change the
          size of log files already full until the next time they are used.
          </p>
        <p>If the log size is decreased, for example, to save space,
          function
	  <seealso marker="#inc_wrap_file/1"><c>inc_wrap_file/1</c></seealso>
	  can be used to force the log to wrap.
          </p>
      </desc>
    </func>
    <func>
      <name name="chunk" arity="2"/>
      <name name="chunk" arity="3"/>
      <name name="bchunk" arity="2"/>
      <name name="bchunk" arity="3"/>
      <fsummary>Read a chunk of items written to a disk log.</fsummary>
      <type variable="Log"/>
      <type variable="Continuation"/>
      <type variable="N"/>
      <type name="chunk_ret"/>
      <type name="bchunk_ret"/>
      <type name="chunk_error_rsn"/>
      <desc>
        <p>Efficiently reads the terms that are appended
	  to an internally formatted log. It minimizes disk
          I/O by reading 64 kilobyte chunks from the file. Functions
          <c>bchunk/2,3</c> return the binaries read from
          the file, they do not call <c>binary_to_term()</c>. Apart from that,
          they work just like <c>chunk/2,3</c>.
          </p>
        <p>The first time <c>chunk()</c> (or <c>bchunk()</c>) is called,
          an initial continuation, the atom <c>start</c>, must be
          provided. If a disk log process is running on the
          current node, terms are read from that log. Otherwise, an
          individual distributed log on some other node is chosen, if
          such a log exists.
          </p>
        <p>When <c>chunk/3</c> is called, <c><anno>N</anno></c> controls the
          maximum number of terms that are read from the log in each
          chunk. Defaults to <c>infinity</c>, which means that all the
          terms contained in the 64 kilobyte chunk are read. If less than
          <c><anno>N</anno></c> terms are returned, this does not necessarily mean
          that the end of the file is reached.
          </p>
        <p><c>chunk()</c> returns a tuple
          <c>{<anno>Continuation2</anno>, <anno>Terms</anno>}</c>, where
	  <c><anno>Terms</anno></c> is a list
          of terms found in the log. <c><anno>Continuation2</anno></c> is yet
          another continuation, which must be passed on to any
          subsequent calls to <c>chunk()</c>. With a series of calls to
          <c>chunk()</c>, all terms from a log can be extracted.
          </p>
        <p><c>chunk()</c> returns a tuple
          <c>{<anno>Continuation2</anno>, <anno>Terms</anno>, <anno>Badbytes</anno>}</c>
	  if the log is opened in read-only mode and the read chunk is corrupt.
	  <c><anno>Badbytes</anno></c> is the number of bytes in the file found not to be
          Erlang terms in the chunk. Notice that the log is not repaired.
          When trying to read chunks from a log opened in read-write mode,
          tuple <c>{corrupt_log_file, <anno>FileName</anno>}</c> is returned if the
          read chunk is corrupt.
          </p>
        <p><c>chunk()</c> returns <c>eof</c> when the end of the log is
          reached, or <c>{error, <anno>Reason</anno>}</c> if an error occurs. If
          a wrap log file is missing, a message is output on the error log.
          </p>
        <p>When <c>chunk/2,3</c> is used with wrap logs, the returned
          continuation might not be valid in the next call to
          <c>chunk()</c>. This is because the log can wrap and delete
          the file into which the continuation points. To prevent this,
          the log can be blocked during the search.
          </p>
      </desc>
    </func>
    <func>
      <name name="chunk_info" arity="1"/>
      <fsummary>Return information about a chunk continuation of a disk log.</fsummary>
      <desc>
        <p>Returns the pair <c>{node, <anno>Node</anno>}</c>,
          describing the chunk continuation returned by
          <c>chunk/2,3</c>, <c>bchunk/2,3</c>, or <c>chunk_step/3</c>.</p>
	  <p>Terms are read from the disk log running on <c><anno>Node</anno></c>.</p>
      </desc>
    </func>
    <func>
      <name name="chunk_step" arity="3"/>
      <fsummary>Step forward or backward among the wrap log files of a disk log.</fsummary>
      <desc>
        <p>Can be used with <c>chunk/2,3</c> and <c>bchunk/2,3</c>
	  to search through an internally formatted wrap log. It takes as
          argument a continuation as returned by <c>chunk/2,3</c>,
          <c>bchunk/2,3</c>, or <c>chunk_step/3</c>, and steps forward
          (or backward) <c><anno>Step</anno></c> files in the wrap log. The
          continuation returned, points to the first log item in the
          new current file.
          </p>
        <p>If atom <c>start</c> is specified as continuation, a disk log
          to read terms from is chosen. A local or distributed disk log
          on the current node is preferred to an
          individual distributed log on some other node.
          </p>
        <p>If the wrap log is not full because all files are not yet
          used, <c>{error, end_of_log}</c> is returned if trying to
          step outside the log.
          </p>
      </desc>
    </func>
    <func>
      <name name="close" arity="1"/>
      <fsummary>Close a disk log.</fsummary>
      <type name="close_error_rsn"/>
      <desc>
        <p><marker id="close_1"></marker>Closes a
          local or distributed disk log properly. An internally
          formatted log must be closed before the Erlang system is
          stopped. Otherwise, the log is regarded as unclosed and the
          automatic repair procedure is activated next time the
          log is opened.
          </p>
        <p>The disk log process is not terminated as long as there are
          owners or users of the log. All owners must close the log,
	  possibly by terminating. Also, any other process, not only the processes
	  that have opened the log anonymously, can decrement the <c>users</c>
          counter by closing the log. 
          Attempts to close a log by a process that is 
          not an owner are ignored if there are no users.
          </p>
        <p>If the log is blocked by the closing process, the log is also
          unblocked.
          </p>
      </desc>
    </func>
    <func>
      <name name="format_error" arity="1"/>
      <fsummary>Return an English description of a disk log error reply.</fsummary>
      <desc>
        <p>Given the error returned by any function in this module, 
          this function returns a descriptive string
          of the error in English. For file errors, function
          <c>format_error/1</c> in module
	  <seealso marker="file#format_error/1"><c>file</c></seealso>
	  is called.</p>
      </desc>
    </func>
    <func>
      <name name="inc_wrap_file" arity="1"/>
      <fsummary>Change to the next wrap log file of a disk log.</fsummary>
      <type name="inc_wrap_error_rsn"/>
      <type name="invalid_header"/>
      <desc>
        <p>Forces the internally formatted disk log to start logging to the
          next log file. It can be used, for example, with
          <c>change_size/2</c> to reduce the amount of disk space allocated
          by the disk log.
          </p>
        <p>Owners subscribing to notifications normally
          receive a <c>wrap</c> message, but if
          an error occurs with a reason tag of <c>invalid_header</c> or
          <c>file_error</c>, an <c>error_status</c> message is sent.</p>
      </desc>
    </func>
    <func>
      <name name="info" arity="1"/>
      <fsummary>Return information about a disk log.</fsummary>
      <type name="dlog_info"/>
      <desc>
        <p>Returns a list of <c>{Tag, Value}</c> pairs describing the log.
	  If a disk log process is running on the current node,
	  that log is used as source of information, otherwise an individual
	  distributed log on some other node is chosen, if such a log exists.
          </p>
        <p>The following pairs are returned for all logs:
          </p>
        <taglist>
	  <tag><c>{name, <anno>Log</anno>}</c></tag>
          <item>
            <p><c><anno>Log</anno></c> is the log name
              as specified by the <c>open/1</c> option <c>name</c>.</p>
          </item>
	  <tag><c>{file, <anno>File</anno>}</c></tag>
          <item>
            <p>For halt logs <c><anno>File</anno></c> is the
              filename, and for wrap logs <c><anno>File</anno></c> is the base name.</p>
          </item>
	  <tag><c>{type, <anno>Type</anno>}</c></tag>
          <item>
            <p><c><anno>Type</anno></c> is the log type
              as specified by the <c>open/1</c> option <c>type</c>.</p>
          </item>
	  <tag><c>{format, <anno>Format</anno>}</c></tag>
          <item>
            <p><c><anno>Format</anno></c> is the log format
              as specified by the <c>open/1</c> option <c>format</c>.</p>
          </item>
	  <tag><c>{size, <anno>Size</anno>}</c></tag>
          <item>
            <p><c><anno>Size</anno></c> is the log size
              as specified by the <c>open/1</c> option <c>size</c>,
              or the size set by <c>change_size/2</c>. The value set by 
              <c>change_size/2</c> is reflected immediately.</p>
          </item>
	  <tag><c>{mode, <anno>Mode</anno>}</c></tag>
          <item>
            <p><c><anno>Mode</anno></c> is the log mode
              as specified by the <c>open/1</c> option <c>mode</c>.</p>
          </item>
	  <tag><c>{owners, [{pid(), <anno>Notify</anno>}]}</c></tag>
          <item>
            <p><c><anno>Notify</anno></c>
              is the value set by the <c>open/1</c> option <c>notify</c>
              or function <c>change_notify/3</c> for the owners of
              the log.</p>
          </item>
	  <tag><c>{users, <anno>Users</anno>}</c></tag>
          <item>
            <p><c><anno>Users</anno></c> is the number
              of anonymous users of the log, see the <c>open/1</c> option
              <seealso marker="#linkto"><c>linkto</c></seealso>.</p>
          </item>
	  <tag><c>{status, <anno>Status</anno>}</c></tag>
          <item>
            <p><c><anno>Status</anno></c> is <c>ok</c>
              or <c>{blocked, <anno>QueueLogRecords</anno>}</c> as set by functions
              <c>block/1,2</c> and <c>unblock/1</c>.</p>
          </item>
	  <tag><c>{node, <anno>Node</anno>}</c></tag>
          <item>
            <p>The information returned by the
              current invocation of function <c>info/1</c> is
              gathered from the disk log process running on <c><anno>Node</anno></c>.</p>
          </item>
	  <tag><c>{distributed, <anno>Dist</anno>}</c></tag>
          <item>
            <p>If the log is local on
              the current node, <c><anno>Dist</anno></c> has the value <c>local</c>,
              otherwise all nodes where the log is distributed
              are returned as a list.</p>
          </item>
        </taglist>
        <p>The following pairs are returned for all logs opened in 
          <c>read_write</c> mode:
          </p>
        <taglist>
	  <tag><c>{head, <anno>Head</anno>}</c></tag>
          <item>
            <p>Depending on the value of
              the <c>open/1</c> options <c>head</c> and <c>head_func</c>,
              or set by function <c>change_header/2</c>, the value
              of <c><anno>Head</anno></c> is <c>none</c> (default),
              <c>{head, H}</c> (<c>head</c> option), or <c>{M,F,A}</c>
              (<c>head_func</c> option).</p>
          </item>
	  <tag><c>{no_written_items, <anno>NoWrittenItems</anno>}</c></tag>
          <item>
            <p><c><anno>NoWrittenItems</anno></c> is the number of items
              written to the log since the disk log process was created.</p>
          </item>
        </taglist>
        <p>The following pair is returned for halt logs opened in 
          <c>read_write</c> mode:
          </p>
        <taglist>
	  <tag><c>{full, <anno>Full</anno>}</c></tag>
          <item>
            <p><c><anno>Full</anno></c> is <c>true</c> or
              <c>false</c> depending on whether the halt log is full or not.</p>
          </item>
        </taglist>
        <p>The following pairs are returned for wrap logs opened in 
          <c>read_write</c> mode:
          </p>
        <taglist>
	  <tag><c>{no_current_bytes, integer() >= 0}</c></tag>
          <item>
            <p>The number
              of bytes written to the current wrap log file.</p>
          </item>
	  <tag><c>{no_current_items, integer() >= 0}</c></tag>
          <item>
            <p>The number
              of items written to the current wrap log file, header
              inclusive.</p>
          </item>
	  <tag><c>{no_items, integer() >= 0}</c></tag>
          <item>
            <p>The total number
              of items in all wrap log files.</p>
          </item>
	  <tag><c>{current_file, integer()}</c></tag>
          <item>
            <p>The ordinal for
              the current wrap log file in the range <c>1..MaxNoFiles</c>,
              where <c>MaxNoFiles</c> is specified by the <c>open/1</c> option
              <c>size</c> or set by <c>change_size/2</c>.</p>
          </item>
	  <tag><c>{no_overflows, {<anno>SinceLogWasOpened</anno>, <anno>SinceLastInfo</anno>}}</c></tag>
          <item>
            <p><c><anno>SinceLogWasOpened</anno></c> (<c><anno>SinceLastInfo</anno></c>)
	      is the number of times a wrap log file has been filled up and a
              new one is opened or <c>inc_wrap_file/1</c> has been called since
              the disk log was last opened (<c>info/1</c> 
              was last called). The first time <c>info/2</c> is called 
              after a log was (re)opened or truncated, the two values 
              are equal.</p>
          </item>
        </taglist>
        <p>Notice that functions <c>chunk/2,3</c>, <c>bchunk/2,3</c>, and
          <c>chunk_step/3</c> do not affect any value
          returned by <c>info/1</c>.
          </p>
      </desc>
    </func>
    <func>
      <name name="lclose" arity="1"/>
      <name name="lclose" arity="2"/>
      <fsummary>Close a disk log on one node.</fsummary>
      <type name="lclose_error_rsn"/>
      <desc>
        <p><c>lclose/1</c> closes a local log or an individual distributed
	log on the current node.</p>
        <p><c>lclose/2</c> closes an individual distributed log on the
	specified node if the node is not the current one.</p>
        <p><c>lclose(<anno>Log</anno>)</c> is equivalent to
          <c>lclose(<anno>Log</anno>,&nbsp;node())</c>.
          See also <seealso marker="#close_1"><c>close/1</c></seealso>.
          </p>
        <p>If no log with the specified name exist on the specified node,
	<c>no_such_log</c> is returned.
          </p>
      </desc>
    </func>
    <func>
      <name name="log" arity="2"/>
      <name name="blog" arity="2"/>
      <fsummary>Log an item onto a disk log.</fsummary>
      <type variable="Log"/>
      <type variable="Term" name_i="1"/>
      <type variable="Bytes"/>
      <type name="log_error_rsn"/>
      <desc>
        <p>Synchronously
          appends a term to a disk log. Returns <c>ok</c> or
          <c>{error, <anno>Reason</anno>}</c> when the term is written to
          disk. If the log is distributed, <c>ok</c> is returned,
	  unless all nodes are down. Terms are written by
          the ordinary <c>write()</c> function of the
          operating system. Hence, it is not guaranteed that the term
          is written to disk, it can linger in
          the operating system kernel for a while. To ensure that the
          item is written to disk, function
	  <seealso marker="#sync/1"><c>sync/1</c></seealso>
          must be called.
          </p>
        <p><c>log/2</c> is used for internally formatted logs,
          and <c>blog/2</c> for externally formatted logs. 
          <c>blog/2</c> can also be used
          for internally formatted logs if the binary is
          constructed with a call to
	  <seealso marker="erts:erlang#term_to_binary/1">
	  <c>term_to_binary/1</c></seealso>.</p>
        <p>Owners subscribing to notifications are notified
          of an error with an <c>error_status</c> message if the error 
          reason tag is <c>invalid_header</c> or <c>file_error</c>.
          </p>
      </desc>
    </func>
    <func>
      <name name="log_terms" arity="2"/>
      <name name="blog_terms" arity="2"/>
      <fsummary>Log many items onto a disk log.</fsummary>
      <type variable="Log"/>
      <type variable="TermList" name_i="1"/>
      <type variable="BytesList"/>
      <type name="log_error_rsn"/>
      <desc>
        <p>Synchronously appends a list of items to the log. It is more
          efficient to use these functions instead of functions <c>log/2</c>
	  and <c>blog/2</c>. The specified list is split into as large
	  sublists as possible (limited by the size of wrap log files),
	  and each sublist is logged as one single item, which reduces
	  the overhead.
          </p>
        <p><c>log_terms/2</c> is used for internally formatted
          logs, and <c>blog_terms/2</c> for externally formatted logs.
          <c>blog_terms/2</c> can also be used
          for internally formatted logs if the binaries are
          constructed with calls to
	  <seealso marker="erts:erlang#term_to_binary/1">
	  <c>term_to_binary/1</c></seealso>.</p>
        <p>Owners subscribing to notifications are notified
          of an error with an <c>error_status</c> message if the error 
          reason tag is <c>invalid_header</c> or <c>file_error</c>.
          </p>
      </desc>
    </func>
    <func>
      <name name="open" arity="1"/>
      <fsummary>Open a disk log file.</fsummary>
      <type name="dlog_options"/>
      <type name="dlog_option"/>
      <type name="open_ret"/>
      <type name="ret"/>
      <type name="dist_open_ret"/>
      <type name="dist_error_rsn"/>
      <type name="open_error_rsn"/>
      <type name="dlog_optattr"/>
      <type name="dlog_size"/>
      <desc>
        <p>Parameter <c><anno>ArgL</anno></c> is a list of the following
	  options:</p>
        <taglist>
	  <tag><c>{name, <anno>Log</anno>}</c></tag>
          <item>
            <p>Specifies the log name.
              This name must be passed on as a parameter in
              all subsequent logging operations. A name must always
              be supplied.
              </p>
          </item>
	  <tag><c>{file, <anno>FileName</anno>}</c></tag>
          <item>
            <p>Specifies the name of the
              file to be used for logged terms. If this value is
              omitted and the log name is an atom or a string,
              the filename defaults to <c>lists:concat([<anno>Log</anno>, ".LOG"])</c>
	      for halt logs.</p>
	    <p>For wrap logs, this is the base name of the files. Each file in
	      a wrap log is called <c><![CDATA[<base_name>.N]]></c>, where <c>N</c>
	      is an integer. Each wrap log also has two files called
              <c><![CDATA[<base_name>.idx]]></c> and <c><![CDATA[<base_name>.siz]]></c>.
              </p>
          </item>
	  <tag><c>{linkto, <anno>LinkTo</anno>}</c><marker id="linkto"></marker></tag>
          <item>
            <p>If <c><anno>LinkTo</anno></c> is a pid, it becomes an owner of the
              log. If <c><anno>LinkTo</anno></c> is <c>none</c>, the log records
              that it is used anonymously by some process by
              incrementing the <c>users</c> counter. By default, the
              process that calls <c>open/1</c> owns the log.
              </p>
          </item>
	  <tag><c>{repair, <anno>Repair</anno>}</c></tag>
          <item>
            <p>If <c><anno>Repair</anno></c> is <c>true</c>,
              the current log file is repaired, if needed. As the
              restoration is initiated, a message is output on the error log.
              If <c>false</c> is specified,
              no automatic repair is attempted. Instead, the
              tuple <c>{error, {need_repair, <anno>Log</anno>}}</c> is returned if an
              attempt is made to open a corrupt log file. 
              If <c>truncate</c> is specified, the log file becomes
              truncated, creating an empty log. Defaults to
              <c>true</c>, which has no effect on logs opened in 
              read-only mode.
              </p>
          </item>
	  <tag><c>{type, <anno>Type</anno>}</c></tag>
          <item>
            <p>The log type. Defaults to <c>halt</c>.
              </p>
          </item>
	  <tag><c>{format, <anno>Format</anno>}</c></tag>
          <item>
            <p>Disk log format. Defaults to <c>internal</c>.
              </p>
          </item>
	  <tag><c>{size, <anno>Size</anno>}</c></tag>
          <item>
            <p>Log size.</p>
              <p>When a halt log has reached its maximum size, all attempts to
              log more items are rejected. Defaults to
              <c>infinity</c>, which for halt implies that there is no
              maximum size.</p>
	    <p>For wrap logs, parameter <c><anno>Size</anno></c>
              can be a pair
              <c>{<anno>MaxNoBytes</anno>, <anno>MaxNoFiles</anno>}</c> or
	      <c>infinity</c>.
	      In the latter case, if the files of an existing wrap log
              with the same name can be found, the size is read
              from the existing wrap log, otherwise an error is returned.</p>
            <p>Wrap logs write at most <c><anno>MaxNoBytes</anno></c>
              bytes on each file and use <c><anno>MaxNoFiles</anno></c>
              files before starting all over with the first wrap log
              file. Regardless of <c><anno>MaxNoBytes</anno></c>,
              at least the header (if there is one) and one
              item are written on each wrap log file before
              wrapping to the next file.</p>
            <p>When opening an existing wrap log, it is not
              necessary to supply a value for option <c>Size</c>, but any
              supplied value must equal the current log size, otherwise
              the tuple <c>{error, {size_mismatch, <anno>CurrentSize</anno>,
	      <anno>NewSize</anno>}}</c> is returned.</p>
          </item>
	  <tag><c>{distributed, <anno>Nodes</anno>}</c></tag>
          <item>
            <p>This option can be used for
              adding members to a distributed disk log.
              Defaults to <c>[]</c>, which means that
              the log is local on the current node.
              </p>
          </item>
	  <tag><c>{notify, boolean()}</c><marker id="notify"></marker></tag>
          <item>
            <p>If <c>true</c>, the log owners
              are notified when certain log events occur.
              Defaults to <c>false</c>. The owners are sent one of the
              following messages when an event occurs:
              </p>
            <taglist>
	      <tag><c>{disk_log, Node, Log, {wrap, NoLostItems}}</c></tag>
              <item>
                <p>Sent when a wrap log has
                  filled up one of its files and a new file is
                  opened. <c>NoLostItems</c> is the number of
                  previously logged items that were lost when
                  truncating existing files.
                  </p>
              </item>
	      <tag><c>{disk_log, Node, Log, {truncated, NoLostItems}}</c></tag>
              <item>
                <p>Sent when a log is
                  truncated or reopened. For halt logs <c>NoLostItems</c>
                  is the number of items written on the log since the 
                  disk log process was created. For wrap logs 
                  <c>NoLostItems</c> is the number of items on all 
                  wrap log files.
                  </p>
              </item>
	      <tag><c>{disk_log, Node, Log, {read_only, Items}}</c></tag>
              <item>
                <p>Sent when an asynchronous log attempt is made to
                  a log file opened in read-only mode. 
                  <c>Items</c> is the items from the log attempt.
                  </p>
              </item>
	      <tag><c>{disk_log, Node, Log, {blocked_log, Items}}</c></tag>
              <item>
                <p>Sent when an asynchronous log attempt is made to
                  a blocked log that does not queue log attempts.
                  <c>Items</c> is the items from the log attempt.
                  </p>
              </item>
	      <tag><c>{disk_log, Node, Log, {format_external, Items}}</c></tag>
              <item>
                <p>Sent when function <c>alog/2</c> or <c>alog_terms/2</c> is
                  used for internally formatted logs. <c>Items</c> is the 
                  items from the log attempt.
                  </p>
              </item>
	      <tag><c>{disk_log, Node, Log, full}</c></tag>
              <item>
                <p>Sent when
                  an attempt to log items to a wrap log would write more
                  bytes than the limit set by option <c>size</c>.
                  </p>
              </item>
	      <tag><c>{disk_log, Node, Log, {error_status, Status}}</c></tag>
              <item>
                <p>Sent when the error status changes. The error status
                  is defined by the outcome of the last attempt to log
                  items to the log, or to truncate the log, or the last
                  use of function <c>sync/1</c>, <c>inc_wrap_file/1</c>, or
                  <c>change_size/2</c>. <c>Status</c> is either <c>ok</c> or
                  <c>{error, Error}</c>, the former is the initial value.
                  </p>
              </item>
            </taglist>
          </item>
	  <tag><c>{head, <anno>Head</anno>}</c></tag>
          <item>
            <p>Specifies a header to be
              written first on the log file. If the log is a wrap
              log, the item <c><anno>Head</anno></c> is written first in each new file.
              <c><anno>Head</anno></c> is to be a term if the format is
              <c>internal</c>, otherwise a deep list of bytes (or a binary).
              Defaults to <c>none</c>, which means that
              no header is written first on the file.
              </p>
          </item>
	  <tag><c>{head_func, {M,F,A}}</c></tag>
          <item>
            <p>Specifies a function
              to be called each time a new log file is opened.
              The call <c>M:F(A)</c> is assumed to return <c>{ok, Head}</c>.
              The item <c>Head</c> is written first in each file.
              <c>Head</c> is to be a term if the format is
              <c>internal</c>, otherwise a deep list of bytes (or a binary).
              </p>
          </item>
	  <tag><c>{mode, <anno>Mode</anno>}</c></tag>
          <item>
            <p>Specifies if the log is to be
              opened in read-only or read-write mode. Defaults to
              <c>read_write</c>.
              </p>
          </item>
        </taglist>
        <p><c>open/1</c> returns <c>{ok, <anno>Log</anno>}</c> if the
          log file is successfully opened. If the file is
          successfully repaired, the tuple <c>{repaired, <anno>Log</anno>,
	  {recovered, <anno>Rec</anno>}, {badbytes, <anno>Bad</anno>}}</c>
	  is returned, where <c><anno>Rec</anno></c> is the number of
	  whole Erlang terms found in the file and <c><anno>Bad</anno></c>
	  is the number of bytes in the file that
          are non-Erlang terms. If the parameter <c>distributed</c>
          is specified, <c>open/1</c> returns a list of
          successful replies and a list of erroneous replies. Each
          reply is tagged with the node name.
          </p>
        <p>When a disk log is opened in read-write mode, any existing
          log file is checked for. If there is none, a new empty
          log is created, otherwise the existing file is opened at the
          position after the last logged item, and the logging of items
          starts from there. If the format is <c>internal</c>
          and the existing file is not recognized as an internally
          formatted log, a tuple
	  <c>{error, {not_a_log_file, <anno>FileName</anno>}}</c>
          is returned.
          </p>
        <p><c>open/1</c> cannot be used for changing the
          values of options of an open log. When there are prior
          owners or users of a log, all option values except <c>name</c>, 
          <c>linkto</c>, and <c>notify</c> are only checked against
          the values supplied before as option values
          to function <c>open/1</c>, <c>change_header/2</c>, <c>change_notify/3</c>,
          or <c>change_size/2</c>. Thus,
          none of the options except <c>name</c> is mandatory. If some
          specified value differs from the current value, a tuple
          <c>{error, {arg_mismatch, <anno>OptionName</anno>, <anno>CurrentValue</anno>, <anno>Value</anno>}}</c>
          is returned.</p>
	<note><p>If an owner attempts to open a log
          as owner once again, it is acknowledged with the return value
          <c>{ok, <anno>Log</anno>}</c>, but the state of the disk log is not
          affected.</p></note>
        <p>If a log with a specified name is local on some node,
          and one tries to open the log distributed on the same node,
          the tuple <c>{error, {node_already_open, <anno>Log</anno>}}</c> is
          returned. The same tuple is returned if the log is distributed on
          some node, and one tries to open the log locally on the same node.
          Opening individual distributed disk logs for the first time
          adds those logs to a (possibly empty) distributed disk log.
          The supplied option values are used
          on all nodes mentioned by option <c>distributed</c>.
          Individual distributed logs know nothing
          about each other's option values, so each node can be
          given unique option values by creating a distributed
          log with many calls to <c>open/1</c>.
          </p>
        <p>A log file can be opened more than once by giving
          different values to option <c>name</c> or by using the
          same file when distributing a log on different nodes.
          It is up to the user of module <c>disk_log</c>
          to ensure that not more than one disk log process has write
	  access to any file, otherwise the file can be corrupted.
          </p>
        <p>If an attempt to open a log file for the first time fails, 
          the disk log process terminates with the EXIT message 
          <c>{{failed,Reason},[{disk_log,open,1}]}</c>.
          The function returns <c>{error, Reason}</c> for all other errors.
          </p>
      </desc>
    </func>
    <func>
      <name name="pid2name" arity="1"/>
      <fsummary>Return the name of the disk log handled by a pid.</fsummary>
      <desc>
        <p>Returns the log name
          given the pid of a disk log process on the current node, or
          <c>undefined</c> if the specified pid is not a disk log process.
          </p>
        <p>This function is meant to be used for debugging only.
          </p>
      </desc>
    </func>
    <func>
      <name name="reopen" arity="2"/>
      <name name="reopen" arity="3"/>
      <name name="breopen" arity="3"/>
      <fsummary>Reopen a disk log and save the old log.</fsummary>
      <type variable="Log"/>
      <type variable="File" name_i="1"/>
      <type variable="Head" name_i="2"/>
      <type variable="BHead"/>
      <type name="reopen_error_rsn"/>
      <desc>
        <p>Renames the log file
          to <c><anno>File</anno></c> and then recreates a new log file.
          If a wrap log exists, <c><anno>File</anno></c> is used as the base name
          of the renamed files.
          By default the header given to <c>open/1</c> is written first in
          the newly opened log file, but if argument <c><anno>Head</anno></c> or
          <c><anno>BHead</anno></c> is specified, this item is used instead.
          The header argument is used only once. Next time a wrap log file
          is opened, the header given to <c>open/1</c> is used.
          </p>
        <p><c>reopen/2,3</c> are used for internally formatted
          logs, and <c>breopen/3</c> for externally formatted logs.
          </p>
        <p>Owners subscribing to notifications receive
          a <c>truncate</c> message.
          </p>
        <p>Upon failure to reopen the log, the disk log process terminates
          with the EXIT message <c>{{failed,Error},[{disk_log,Fun,Arity}]}</c>.
          Other processes having requests queued receive the message
          <c>{disk_log, Node, {error, disk_log_stopped}}</c>.
          </p>
      </desc>
    </func>
    <func>
      <name name="sync" arity="1"/>
      <fsummary>Flush the contents of a disk log to the disk.</fsummary>
      <type name="sync_error_rsn"/>
      <desc>
        <p>Ensures that the contents of the log are written to the disk.
          This is usually a rather expensive operation.
          </p>
      </desc>
    </func>
    <func>
      <name name="truncate" arity="1"/>
      <name name="truncate" arity="2"/>
      <name name="btruncate" arity="2"/>
      <fsummary>Truncate a disk log.</fsummary>
      <type variable="Log"/>
      <type variable="Head" name_i="2"/>
      <type variable="BHead"/>
      <type name="trunc_error_rsn"/>
      <desc>
        <p>Removes all items from a disk log.
          If argument <c><anno>Head</anno></c> or <c><anno>BHead</anno></c> is
          specified, this item is written first in the newly truncated
          log, otherwise the header given to <c>open/1</c> is used.
          The header argument is used only once. Next time a wrap log file
          is opened, the header given to <c>open/1</c> is used.
          </p>
        <p><c>truncate/1,2</c> are used for internally
          formatted logs, and <c>btruncate/2</c> for externally formatted
          logs.
          </p>
        <p>Owners subscribing to notifications receive
          a <c>truncate</c> message.
          </p>
        <p>If the attempt to truncate the log fails, the disk log process
          terminates with the EXIT message 
          <c>{{failed,Reason},[{disk_log,Fun,Arity}]}</c>.
          Other processes having requests queued receive the message
          <c>{disk_log, Node, {error, disk_log_stopped}}</c>.
          </p>
      </desc>
    </func>
    <func>
      <name name="unblock" arity="1"/>
      <fsummary>Unblock a disk log.</fsummary>
      <type name="unblock_error_rsn"/>
      <desc>
        <p>Unblocks a log.
          A log can only be unblocked by the blocking process.
          </p>
      </desc>
    </func>
  </funcs>

  <section>
    <title>See Also</title>
    <p><seealso marker="file"><c>file(3)</c></seealso>,
      <seealso marker="pg2"><c>pg2(3)</c></seealso>,
      <seealso marker="wrap_log_reader"><c>wrap_log_reader(3)</c></seealso></p>
  </section>
</erlref>