transactions-production-consideration.txt

.. _production-considerations:

=========================
Production Considerations
=========================

.. default-domain:: mongodb

.. contents:: On this page
   :local:
   :backlinks: none
   :depth: 1
   :class: twocols

The following page lists some production considerations for running
transactions. These apply whether you run transactions on replica sets
or sharded clusters. For running transactions on sharded clusters, see
also the :doc:`/core/transactions-sharded-clusters` for additional
considerations that are specific to sharded clusters.

Availability
------------

- MongoDB supports multi-document transactions on replica sets.

- Distributed transactions add support for multi-document transactions on 
  sharded clusters and incorporates the existing support for
  multi-document transactions on replica sets.

.. note:: Distributed Transactions and Multi-Document Transactions

   The two terms are synonymous. Distributed transactions refer to 
   multi-document transactions on sharded clusters and replica sets. 
   Multi-document transactions (whether on sharded clusters or replica sets) are 
   also known as distributed transactions.

Feature Compatibility
---------------------

To use transactions, the :ref:`featureCompatibilityVersion <view-fcv>`
for all members of the deployment must be at least:

.. list-table::
   :header-rows: 1

   * - Deployment
     - Minimum ``featureCompatibilityVersion``
     
   * - Replica Set
     - ``4.0`` 
     
   * - Sharded Cluster
     - ``4.2``

To check the fCV for a member, connect to the member and run the
following command:

.. code-block:: javascript

   db.adminCommand( { getParameter: 1, featureCompatibilityVersion: 1 } )

For more information, see the
:dbcommand:`setFeatureCompatibilityVersion` reference page.

.. _transaction-limit:

Runtime Limit
-------------

.. note::
   
   To configure maximum transaction lifetimes in {+atlas+},
   see :atlas:`Set Transaction Lifetime </cluster-additional-settings/#set-transaction-lifetime>` 
   in the Atlas documentation.

By default, a transaction must have a runtime of less than one minute.
You can modify this limit using
:parameter:`transactionLifetimeLimitSeconds` for the
:binary:`~bin.mongod` instances. For sharded clusters, the parameter
must be modified for all shard replica set members. Transactions that
exceeds this limit are considered expired and will be aborted by a
periodic cleanup process.

For sharded clusters, you can also specify a ``maxTimeMS`` limit on
``commitTransaction``. For more information, see :ref:`Sharded Clusters
Transactions Time Limit <transactions-sharded-clusters-time-limit>`.

.. _txn-oplog-size-limit:

Oplog Size Limit
----------------

MongoDB creates as many oplog entries as necessary to the encapsulate all write 
operations in a transaction, instead of a single entry for all write operations 
in the transaction. This removes the 16MB total size limit for a transaction 
imposed by the single oplog entry for all its write operations. Although the 
total size limit is removed, each oplog entry still must be within the
BSON document size limit of 16MB.

WiredTiger Cache
----------------

To prevent storage cache pressure from negatively impacting the
performance:

- When you abandon a transaction, abort the transaction.

- When you encounter an error during individual operation in the
  transaction, abort and retry the transaction.

The :parameter:`transactionLifetimeLimitSeconds` also ensures that
expired transactions are aborted periodically to relieve storage cache
pressure.

.. note::

   If you have an uncommitted transaction that causes excessive pressure
   on the :ref:`WiredTiger cache <storage-wiredtiger>`, the transaction
   aborts and returns a :term:`write conflict` error.

   If a transaction is too large to ever fit in the WiredTiger cache,
   the transaction aborts and returns a ``TransactionTooLargeForCache``
   error.

Transactions and Security
-------------------------

- If running with :ref:`access control <authorization>`, you must
  have :ref:`privileges <built-in-roles>` for the
  :ref:`operations in the transaction <transactions-operations>`.

- If running with :ref:`auditing <auditing>`, operations in an
  aborted transaction are still audited. However, there is no audit
  event that indicates that the transaction aborted.

.. _transactions-disabled-wcmajority:

Shard Configuration Restriction
-------------------------------

.. include:: /includes/extracts/transactions-shards-wcmajority-disabled.rst

Sharded Clusters and Arbiters
-----------------------------

.. include:: /includes/extracts/transactions-arbiters.rst

.. _txns-locks:

Acquiring Locks
---------------

By default, transactions wait up to ``5`` milliseconds to acquire locks
required by the operations in the transaction. If the transaction
cannot acquire its required locks within the ``5`` milliseconds, the
transaction aborts.

Transactions release all locks upon abort or commit.

.. include:: /includes/extracts/transactions-operations-catalog-tip.rst

Lock Request Timeout
~~~~~~~~~~~~~~~~~~~~

.. note::
   
   {+atlas+} clusters restrict the use of the :dbcommand:`setParameter` 
   command. For more information, see :ref:`<unsupported-commands>` in 
   the Atlas documentation.

   To modify your Atlas cluster parameters, contact 
   :atlas:`Atlas Support </support>`.

You can use the :parameter:`maxTransactionLockRequestTimeoutMillis`
parameter to adjust how long transactions wait to acquire locks.
Increasing :parameter:`maxTransactionLockRequestTimeoutMillis` allows
operations in the transactions to wait the specified time to acquire
the required locks. This can help obviate transaction aborts on
momentary concurrent lock acquisitions, like fast-running metadata
operations. However, this could possibly delay the abort of deadlocked
transaction operations.

You can also use operation-specific timeout by setting
:parameter:`maxTransactionLockRequestTimeoutMillis` to ``-1``.

.. _txn-prod-considerations-ddl:

Pending DDL Operations and Transactions
---------------------------------------

If a multi-document transaction is in progress, new DDL operations that
affect the same database(s) or collection(s) wait behind the
transaction. While these pending DDL operations exist, new transactions
that access the same database(s) or collection(s) as the pending DDL
operations cannot obtain the required locks and and will abort after
waiting :parameter:`maxTransactionLockRequestTimeoutMillis`. In
addition, new non-transaction operations that access the same 
database(s) or collection(s) will block until they reach their 
``maxTimeMS`` limit.

Consider the following scenarios:

DDL Operation That Requires a Collection Lock
  While an in-progress transaction is performing various CRUD operations
  on the ``employees`` collection in the ``hr`` database, an
  administrator issues the :method:`db.collection.createIndex()` DDL
  operation against the ``employees`` collection.
  :method:`~db.collection.createIndex()` requires an exclusive
  collection lock on the collection.

  Until the in-progress transaction completes, the
  :method:`~db.collection.createIndex()` operation must wait to obtain
  the lock. Any new transaction that affects the ``employees``
  collection and starts while the :method:`~db.collection.createIndex()`
  is pending must wait until after 
  :method:`~db.collection.createIndex()` completes. 

  The pending :method:`~db.collection.createIndex()` DDL operation does
  not affect transactions on other collections in the ``hr`` database.
  For example, a new transaction on the ``contractors`` collection in
  the ``hr`` database can start and complete as normal.

DDL Operation That Requires a Database Lock  
  While an in-progress transaction is performing various CRUD operations
  on the ``employees`` collection in the ``hr`` database, an
  administrator issues the :dbcommand:`renameCollection` DDL operation 
  to rename the ``vendors.contractors`` collection to 
  ``hr.contractors``. ``renameCollection`` requires a database lock on 
  the target database (``hr``) when it differs from the source database 
  (``vendors``).

  Until the in-progress transaction completes, the 
  ``renameCollection`` operation must wait to obtain the 
  lock. Any new transaction that affects the ``hr`` database or *any* 
  of its collections and starts while the ``renameCollection`` is 
  pending must wait until after ``renameCollection`` completes.

In either scenario, if the DDL operation remains pending for more than
:parameter:`maxTransactionLockRequestTimeoutMillis`, pending
transactions waiting behind that operation abort. That is, the value of 
:parameter:`maxTransactionLockRequestTimeoutMillis` must at least cover
the time required for the in-progress transaction *and* the pending DDL
operation to complete.

.. seealso::

   - :ref:`transactions-write-conflicts`
   
   - :ref:`transactions-stale-reads`

   - :ref:`faq-concurrency-database-lock`

   - :ref:`faq-concurrency-collection-lock`

.. _transactions-write-conflicts:

In-progress Transactions and Write Conflicts
--------------------------------------------

.. include:: /includes/extracts/transactions-write-conflict.rst

.. seealso::

   - :ref:`txns-locks`
   - :ref:`txn-prod-considerations-ddl`
   - :data:`$currentOp output <$currentOp.prepareReadConflicts>`

.. _transactions-stale-reads:

In-progress Transactions and Stale Reads
----------------------------------------

.. include:: /includes/extracts/transactions-stale-reads.rst

In-progress Transactions and Chunk Migration
--------------------------------------------

.. include:: /includes/extracts/transactions-chunk-migration.rst

.. seealso::

   :serverstatus:`shardingStatistics.countDonorMoveChunkLockTimeout`


.. _transactions-prod-consideration-outside-reads:

Outside Reads During Commit
---------------------------

.. include:: /includes/extracts/transactions-multi-shard-block-external-reads.rst

Additional Information
----------------------

.. seealso::

   :doc:`/core/transactions-sharded-clusters`