path: root/ctdb/doc/ctdb-tunables.7.xml

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE refentry
	PUBLIC "-//OASIS//DTD DocBook XML V4.5//EN"
	"http://www.oasis-open.org/docbook/xml/4.5/docbookx.dtd">

<refentry id="ctdb-tunables.7">

  <refmeta>
    <refentrytitle>ctdb-tunables</refentrytitle>
    <manvolnum>7</manvolnum>
    <refmiscinfo class="source">ctdb</refmiscinfo>
    <refmiscinfo class="manual">CTDB - clustered TDB database</refmiscinfo>
  </refmeta>

  <refnamediv>
    <refname>ctdb-tunables</refname>
    <refpurpose>CTDB tunable configuration variables</refpurpose>
  </refnamediv>

  <refsect1>
    <title>DESCRIPTION</title>

    <para>
      CTDB's behaviour can be configured by setting run-time tunable
      variables.  This lists and describes all tunables.  See the
      <citerefentry><refentrytitle>ctdb</refentrytitle>
      <manvolnum>1</manvolnum></citerefentry>
      <command>listvars</command>, <command>setvar</command> and
      <command>getvar</command> commands for more details.
    </para>

    <para>
      Unless otherwise stated, tunables should be set to the same
      value on all nodes.  Setting tunables to different values across
      nodes may produce unexpected results.  Future releases may set
      (some or most) tunables globally across the cluster but doing so
      is currently a manual process.
    </para>

    <para>
      Tunables can be set at startup from the
      <filename>/usr/local/etc/ctdb/ctdb.tunables</filename>
      configuration file.

      <literallayout>
<replaceable>TUNABLE</replaceable>=<replaceable>VALUE</replaceable>
      </literallayout>

      Comment lines beginning with '#' are permitted.  Whitespace may
      be used for formatting/alignment.  VALUE must be a non-negative
      integer and must be the last thing on a line (i.e. no trailing
      garbage, trailing comments are not permitted).
    </para>

    <para>
      For example:

      <screen format="linespecific">
MonitorInterval=20
      </screen>
    </para>

    <para>
      The available tunable variables are listed alphabetically below.
    </para>

    <refsect2>
      <title>AllowClientDBAttach</title>
      <para>Default: 1</para>
      <para>
	When set to 0, clients are not allowed to attach to any databases.
	This can be used to temporarily block any new processes from
	attaching to and accessing the databases.  This is mainly used
	for detaching a volatile database using 'ctdb detach'.
      </para>
    </refsect2>

    <refsect2>
      <title>AllowMixedVersions</title>
      <para>Default: 0</para>
      <para>
	CTDB will not allow incompatible versions to co-exist in
	a cluster.  If a version mismatch is found, then losing CTDB
	will shutdown.	To disable the incompatible version check,
	set this tunable to 1.
      </para>
      <para>
	For version checking, CTDB uses major and minor version.
	For example, CTDB 4.6.1 and CTDB 4.6.2 are matching versions;
	CTDB 4.5.x and CTDB 4.6.y do not match.
      </para>
      <para>
	CTDB with version check support will lose to CTDB without
	version check support.	Between two different CTDB versions with
	version check support, one running for less time will lose.
	If the running time for both CTDB versions with version check
	support is equal (to seconds), then the older version will lose.
	The losing CTDB daemon will shutdown.
      </para>
    </refsect2>

    <refsect2>
      <title>AllowUnhealthyDBRead</title>
      <para>Default: 0</para>
      <para>
	When set to 1, ctdb allows database traverses to read unhealthy
	databases.  By default, ctdb does not allow reading records from
	unhealthy databases.
      </para>
    </refsect2>

    <refsect2>
      <title>ControlTimeout</title>
      <para>Default: 60</para>
      <para>
	This is the default setting for timeout for when sending a
	control message to either the local or a remote ctdb daemon.
      </para>
    </refsect2>

    <refsect2>
      <title>DatabaseHashSize</title>
      <para>Default: 100001</para>
      <para>
	Number of the hash chains for the local store of the tdbs that
	ctdb manages.
      </para>
    </refsect2>

    <refsect2>
      <title>DatabaseMaxDead</title>
      <para>Default: 5</para>
      <para>
	Maximum number of dead records per hash chain for the tdb databases
	managed by ctdb.
      </para>
    </refsect2>

    <refsect2>
      <title>DBRecordCountWarn</title>
      <para>Default: 100000</para>
      <para>
	When set to non-zero, ctdb will log a warning during recovery if
	a database has more than this many records. This will produce a
	warning if a database grows uncontrollably with orphaned records.
      </para>
    </refsect2>

    <refsect2>
      <title>DBRecordSizeWarn</title>
      <para>Default: 10000000</para>
      <para>
	When set to non-zero, ctdb will log a warning during recovery
	if a single record is bigger than this size. This will produce
	a warning if a database record grows uncontrollably.
      </para>
    </refsect2>

    <refsect2>
      <title>DBSizeWarn</title>
      <para>Default: 1000000000</para>
      <para>
	When set to non-zero, ctdb will log a warning during recovery if
	a database size is bigger than this. This will produce a warning
	if a database grows uncontrollably.
      </para>
    </refsect2>

    <refsect2>
      <title>DeferredAttachTO</title>
      <para>Default: 120</para>
      <para>
	When databases are frozen we do not allow clients to attach to
	the databases. Instead of returning an error immediately to the
	client, the attach request from the client is deferred until
	the database becomes available again at which stage we respond
	to the client.
      </para>
      <para>
	This timeout controls how long we will defer the request from the
	client before timing it out and returning an error to the client.
      </para>
    </refsect2>

    <refsect2>
      <title>ElectionTimeout</title>
      <para>Default: 3</para>
      <para>
	The number of seconds to wait for the election of recovery
	master to complete. If the election is not completed during this
	interval, then that round of election fails and ctdb starts a
	new election.
      </para>
    </refsect2>

    <refsect2>
      <title>EnableBans</title>
      <para>Default: 1</para>
      <para>
	This parameter allows ctdb to ban a node if the node is misbehaving.
      </para>
      <para>
	When set to 0, this disables banning completely in the cluster
	and thus nodes can not get banned, even it they break. Don't
	set to 0 unless you know what you are doing.
      </para>
    </refsect2>

    <refsect2>
      <title>EventScriptTimeout</title>
      <para>Default: 30</para>
      <para>
	Maximum time in seconds to allow an event to run before timing
	out.  This is the total time for all enabled scripts that are
	run for an event, not just a single event script.
      </para>
      <para>
	Note that timeouts are ignored for some events ("takeip",
	"releaseip", "startrecovery", "recovered") and converted to
	success.  The logic here is that the callers of these events
	implement their own additional timeout.
      </para>
    </refsect2>

    <refsect2>
      <title>FetchCollapse</title>
      <para>Default: 1</para>
      <para>
       This parameter is used to avoid multiple migration requests for
       the same record from a single node. All the record requests for
       the same record are queued up and processed when the record is
       migrated to the current node.
      </para>
      <para>
	When many clients across many nodes try to access the same record
	at the same time this can lead to a fetch storm where the record
	becomes very active and bounces between nodes very fast. This
	leads to high CPU utilization of the ctdbd daemon, trying to
	bounce that record around very fast, and poor performance.
	This can improve performance and reduce CPU utilization for
	certain workloads.
      </para>
    </refsect2>

    <refsect2>
      <title>HopcountMakeSticky</title>
      <para>Default: 50</para>
      <para>
	For database(s) marked STICKY (using 'ctdb setdbsticky'),
	any record that is migrating so fast that hopcount
	exceeds this limit is marked as STICKY record for
	<varname>StickyDuration</varname> seconds. This means that
	after each migration the sticky record will be kept on the node
	<varname>StickyPindown</varname>milliseconds and prevented from
	being migrated off the node.
       </para>
       <para>
	This will improve performance for certain workloads, such as
	locking.tdb if many clients are opening/closing the same file
	concurrently.
      </para>
    </refsect2>

    <refsect2>
      <title>IPAllocAlgorithm</title>
      <para>Default: 2</para>
      <para>
	Selects the algorithm that CTDB should use when doing public
	IP address allocation.  Meaningful values are:
      </para>
      <variablelist>
	<varlistentry>
	  <term>0</term>
	  <listitem>
	    <para>
	      Deterministic IP address allocation.
	    </para>
	    <para>
	      This is a simple and fast option.  However, it can cause
	      unnecessary address movement during fail-over because
	      each address has a "home" node.  Works badly when some
	      nodes do not have any addresses defined.  Should be used
	      with care when addresses are defined across multiple
	      networks.
	    </para>
	    <para>
	      You can override the automatic "home" node allocation by
	      creating a file "home_nodes" next to the
	      "public_addresses" file. As an example the following
	      "home_nodes" file assigns the address 192.168.1.1 to
	      node 0 and 192.168.1.2 to node 2:
	    </para>
	    <screen format="linespecific">
	      192.168.1.1 0
	      192.168.1.2 2
            </screen>
	  </listitem>
	</varlistentry>
	<varlistentry>
	  <term>1</term>
	  <listitem>
	    <para>
	      Non-deterministic IP address allocation.
	    </para>
	    <para>
	      This is a relatively fast option that attempts to do a
	      minimise unnecessary address movements.  Addresses do
	      not have a "home" node.  Rebalancing is limited but it
	      usually adequate.  Works badly when addresses are
	      defined across multiple networks.
	    </para>
	  </listitem>
	</varlistentry>
	<varlistentry>
	  <term>2</term>
	  <listitem>
	    <para>
	      LCP2 IP address allocation.
	    </para>
	    <para>
	      Uses a heuristic to assign addresses defined across
	      multiple networks, usually balancing addresses on each
	      network evenly across nodes.  Addresses do not have a
	      "home" node.  Minimises unnecessary address movements.
	      The algorithm is complex, so is slower than other
	      choices for a large number of addresses.  However, it
	      can calculate an optimal assignment of 900 addresses in
	      under 10 seconds on modern hardware.
	    </para>
	  </listitem>
	</varlistentry>
      </variablelist>
      <para>
	If the specified value is not one of these then the default
	will be used.
      </para>
    </refsect2>

    <refsect2>
      <title>KeepaliveInterval</title>
      <para>Default: 5</para>
      <para>
	How often in seconds should the nodes send keep-alive packets to
	each other.
      </para>
    </refsect2>

    <refsect2>
      <title>KeepaliveLimit</title>
      <para>Default: 5</para>
      <para>
	After how many keepalive intervals without any traffic should
	a node wait until marking the peer as DISCONNECTED.
       </para>
       <para>
	If a node has hung, it can take
	<varname>KeepaliveInterval</varname> *
	(<varname>KeepaliveLimit</varname> + 1) seconds before
	ctdb determines that the node is DISCONNECTED and performs
	a recovery. This limit should not be set too high to enable
	early detection and avoid any application timeouts (e.g. SMB1)
	to kick in before the fail over is completed.
      </para>
    </refsect2>

    <refsect2>
      <title>LockProcessesPerDB</title>
      <para>Default: 200</para>
      <para>
	This is the maximum number of lock helper processes ctdb will
	create for obtaining record locks.  When ctdb cannot get a record
	lock without blocking, it creates a helper process that waits
	for the lock to be obtained.
      </para>
    </refsect2>

    <refsect2>
      <title>LogLatencyMs</title>
      <para>Default: 0</para>
      <para>
	When set to non-zero, ctdb will log if certains operations
	take longer than this value, in milliseconds, to complete.
	These operations include "process a record request from client",
	"take a record or database lock", "update a persistent database
	record" and "vacuum a database".
      </para>
    </refsect2>

    <refsect2>
      <title>MaxQueueDropMsg</title>
      <para>Default: 1000000</para>
      <para>
	This is the maximum number of messages to be queued up for
	a client before ctdb will treat the client as hung and will
	terminate the client connection.
      </para>
    </refsect2>

    <refsect2>
      <title>MonitorInterval</title>
      <para>Default: 15</para>
      <para>
	How often should ctdb run the 'monitor' event in seconds to check
	for a node's health.
      </para>
    </refsect2>

    <refsect2>
      <title>MonitorTimeoutCount</title>
      <para>Default: 20</para>
      <para>
	How many 'monitor' events in a row need to timeout before a node
	is flagged as UNHEALTHY.  This setting is useful if scripts can
	not be written so that they do not hang for benign reasons.
      </para>
    </refsect2>

    <refsect2>
      <title>NoIPFailback</title>
      <para>Default: 0</para>
      <para>
	When set to 1, ctdb will not perform failback of IP addresses
	when a node becomes healthy. When a node becomes UNHEALTHY,
	ctdb WILL perform failover of public IP addresses, but when the
	node becomes HEALTHY again, ctdb will not fail the addresses back.
      </para>
      <para>
	Use with caution! Normally when a node becomes available to the
	cluster ctdb will try to reassign public IP addresses onto the
	new node as a way to distribute the workload evenly across the
	clusternode. Ctdb tries to make sure that all running nodes have
	approximately the same number of public addresses it hosts.
      </para>
      <para>
	When you enable this tunable, ctdb will no longer attempt to
	rebalance the cluster by failing IP addresses back to the new
	nodes. An unbalanced cluster will therefore remain unbalanced
	until there is manual intervention from the administrator. When
	this parameter is set, you can manually fail public IP addresses
	over to the new node(s) using the 'ctdb moveip' command.
      </para>
    </refsect2>

    <refsect2>
      <title>NoIPTakeover</title>
      <para>Default: 0</para>
      <para>
	When set to 1, ctdb will not allow IP addresses to be failed
	over to other nodes.  Any IP addresses already hosted on
	healthy nodes will remain.  Any IP addresses hosted on
	unhealthy nodes will be released by unhealthy nodes and will
	become un-hosted.
      </para>
    </refsect2>

    <refsect2>
      <title>PullDBPreallocation</title>
      <para>Default: 10*1024*1024</para>
      <para>
	This is the size of a record buffer to pre-allocate for sending
	reply to PULLDB control. Usually record buffer starts with size
	of the first record and gets reallocated every time a new record
	is added to the record buffer. For a large number of records,
	this can be very inefficient to grow the record buffer one record
	at a time.
      </para>
    </refsect2>

    <refsect2>
      <title>QueueBufferSize</title>
      <para>Default: 1024</para>
      <para>
	This is the maximum amount of data (in bytes) ctdb will read
	from a socket at a time.
      </para>
      <para>
	For a busy setup, if ctdb is not able to process the TCP sockets
	fast enough (large amount of data in Recv-Q for tcp sockets),
	then this tunable value should be increased.  However, large
	values can keep ctdb busy processing packets and prevent ctdb
	from handling other events.
      </para>
    </refsect2>

    <refsect2>
      <title>RecBufferSizeLimit</title>
      <para>Default: 1000000</para>
      <para>
        This is the limit on the size of the record buffer to be sent
        in various controls.  This limit is used by new controls used
        for recovery and controls used in vacuuming.
      </para>
    </refsect2>

    <refsect2>
      <title>RecdFailCount</title>
      <para>Default: 10</para>
      <para>
	If the recovery daemon has failed to ping the main daemon for
	this many consecutive intervals, the main daemon will consider
	the recovery daemon as hung and will try to restart it to recover.
      </para>
    </refsect2>

    <refsect2>
      <title>RecdPingTimeout</title>
      <para>Default: 60</para>
      <para>
	If the main daemon has not heard a "ping" from the recovery daemon
	for this many seconds, the main daemon will log a message that
	the recovery daemon is potentially hung.  This also increments a
	counter which is checked against <varname>RecdFailCount</varname>
	for detection of hung recovery daemon.
      </para>
    </refsect2>

    <refsect2>
      <title>RecLockLatencyMs</title>
      <para>Default: 1000</para>
      <para>
	When using a reclock file for split brain prevention, if set
	to non-zero this tunable will make the recovery daemon log a
	message if the fcntl() call to lock/testlock the recovery file
	takes longer than this number of milliseconds.
      </para>
    </refsect2>

    <refsect2>
      <title>RecoverInterval</title>
      <para>Default: 1</para>
      <para>
	How frequently in seconds should the recovery daemon perform the
	consistency checks to determine if it should perform a recovery.
      </para>
    </refsect2>

    <refsect2>
      <title>RecoverTimeout</title>
      <para>Default: 120</para>
      <para>
	This is the default setting for timeouts for controls when sent
	from the recovery daemon. We allow longer control timeouts from
	the recovery daemon than from normal use since the recovery
	daemon often use controls that can take a lot longer than normal
	controls.
      </para>
    </refsect2>

    <refsect2>
      <title>RecoveryBanPeriod</title>
      <para>Default: 300</para>
      <para>
       The duration in seconds for which a node is banned if the node
       fails during recovery.  After this time has elapsed the node will
       automatically get unbanned and will attempt to rejoin the cluster.
      </para>
      <para>
       A node usually gets banned due to real problems with the node.
       Don't set this value too small.  Otherwise, a problematic node
       will try to re-join cluster too soon causing unnecessary recoveries.
      </para>
    </refsect2>

    <refsect2>
      <title>RecoveryDropAllIPs</title>
      <para>Default: 120</para>
      <para>
	If a node is stuck in recovery, or stopped, or banned, for this
	many seconds, then ctdb will release all public addresses on
	that node.
      </para>
    </refsect2>

    <refsect2>
      <title>RecoveryGracePeriod</title>
      <para>Default: 120</para>
      <para>
       During recoveries, if a node has not caused recovery failures
       during the last grace period in seconds, any records of
       transgressions that the node has caused recovery failures will be
       forgiven. This resets the ban-counter back to zero for that node.
      </para>
    </refsect2>

    <refsect2>
      <title>RepackLimit</title>
      <para>Default: 10000</para>
      <para>
        During vacuuming, if the number of freelist records are more than
        <varname>RepackLimit</varname>, then the database is repacked
        to get rid of the freelist records to avoid fragmentation.
      </para>
    </refsect2>

    <refsect2>
      <title>RerecoveryTimeout</title>
      <para>Default: 10</para>
      <para>
	Once a recovery has completed, no additional recoveries are
	permitted until this timeout in seconds has expired.
      </para>
    </refsect2>

    <refsect2>
      <title>SeqnumInterval</title>
      <para>Default: 1000</para>
      <para>
	Some databases have seqnum tracking enabled, so that samba will
	be able to detect asynchronously when there has been updates
	to the database.  Every time a database is updated its sequence
	number is increased.
      </para>
      <para>
	This tunable is used to specify in milliseconds how frequently
	ctdb will send out updates to remote nodes to inform them that
	the sequence number is increased.
      </para>
    </refsect2>

    <refsect2>
      <title>StatHistoryInterval</title>
      <para>Default: 1</para>
      <para>
	Granularity of the statistics collected in the statistics
	history. This is reported by 'ctdb stats' command.
      </para>
    </refsect2>

    <refsect2>
      <title>StickyDuration</title>
      <para>Default: 600</para>
      <para>
	Once a record has been marked STICKY, this is the duration in
	seconds, the record will be flagged as a STICKY record.
      </para>
    </refsect2>

    <refsect2>
      <title>StickyPindown</title>
      <para>Default: 200</para>
      <para>
	Once a STICKY record has been migrated onto a node, it will be
	pinned down on that node for this number of milliseconds. Any
	request from other nodes to migrate the record off the node will
	be deferred.
      </para>
    </refsect2>

    <refsect2>
      <title>TakeoverTimeout</title>
      <para>Default: 9</para>
      <para>
	This is the duration in seconds in which ctdb tries to complete IP
	failover.
      </para>
    </refsect2>

    <refsect2>
      <title>TickleUpdateInterval</title>
      <para>Default: 20</para>
      <para>
	Every <varname>TickleUpdateInterval</varname> seconds, ctdb
	synchronizes the client connection information across nodes.
      </para>
    </refsect2>

    <refsect2>
      <title>TraverseTimeout</title>
      <para>Default: 20</para>
      <para>
	This is the duration in seconds for which a database traverse
	is allowed to run.  If the traverse does not complete during
	this interval, ctdb will abort the traverse.
      </para>
    </refsect2>

    <refsect2>
      <title>VacuumFastPathCount</title>
      <para>Default: 60</para>
      <para>
       During a vacuuming run, ctdb usually processes only the records
       marked for deletion also called the fast path vacuuming. After
       finishing <varname>VacuumFastPathCount</varname> number of fast
       path vacuuming runs, ctdb will trigger a scan of complete database
       for any empty records that need to be deleted.
      </para>
    </refsect2>

    <refsect2>
      <title>VacuumInterval</title>
      <para>Default: 10</para>
      <para>
        Periodic interval in seconds when vacuuming is triggered for
        volatile databases.
      </para>
    </refsect2>

    <refsect2>
      <title>VacuumMaxRunTime</title>
      <para>Default: 120</para>
      <para>
        The maximum time in seconds for which the vacuuming process is
        allowed to run.  If vacuuming process takes longer than this
        value, then the vacuuming process is terminated.
      </para>
    </refsect2>

    <refsect2>
      <title>VerboseMemoryNames</title>
      <para>Default: 0</para>
      <para>
	When set to non-zero, ctdb assigns verbose names for some of
	the talloc allocated memory objects.  These names are visible
	in the talloc memory report generated by 'ctdb dumpmemory'.
      </para>
    </refsect2>

  </refsect1>

  <refsect1>
    <title>FILES></title>

    <simplelist>
      <member><filename>/usr/local/etc/ctdb/ctdb.tunables</filename></member>
    </simplelist>
  </refsect1>

  <refsect1>
    <title>SEE ALSO</title>
    <para>
      <citerefentry><refentrytitle>ctdb</refentrytitle>
      <manvolnum>1</manvolnum></citerefentry>,

      <citerefentry><refentrytitle>ctdbd</refentrytitle>
      <manvolnum>1</manvolnum></citerefentry>,

      <citerefentry><refentrytitle>ctdb.conf</refentrytitle>
      <manvolnum>5</manvolnum></citerefentry>,

      <citerefentry><refentrytitle>ctdb</refentrytitle>
      <manvolnum>7</manvolnum></citerefentry>,

      <ulink url="http://ctdb.samba.org/"/>
    </para>
  </refsect1>

  <refentryinfo>
    <author>
      <contrib>
	This documentation was written by
	Ronnie Sahlberg,
	Amitay Isaacs,
	Martin Schwenke
      </contrib>
    </author>

    <copyright>
      <year>2007</year>
      <holder>Andrew Tridgell</holder>
      <holder>Ronnie Sahlberg</holder>
    </copyright>
    <legalnotice>
      <para>
	This program is free software; you can redistribute it and/or
	modify it under the terms of the GNU General Public License as
	published by the Free Software Foundation; either version 3 of
	the License, or (at your option) any later version.
      </para>
      <para>
	This program is distributed in the hope that it will be
	useful, but WITHOUT ANY WARRANTY; without even the implied
	warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR
	PURPOSE.  See the GNU General Public License for more details.
      </para>
      <para>
	You should have received a copy of the GNU General Public
	License along with this program; if not, see
	<ulink url="http://www.gnu.org/licenses"/>.
      </para>
    </legalnotice>
  </refentryinfo>

</refentry>