|
5 | 5 | * :ref:`wal_dir_rescan_delay <cfg_binary_logging_snapshots-wal_dir_rescan_delay>`
|
6 | 6 | * :ref:`wal_queue_max_size <cfg_binary_logging_snapshots-wal_queue_max_size>`
|
7 | 7 | * :ref:`wal_cleanup_delay <cfg_binary_logging_snapshots-wal_cleanup_delay>`
|
| 8 | +* :ref:`wal_ext <cfg_binary_logging_snapshots-wal_ext>` |
| 9 | +* :ref:`secure_erasing <cfg_binary_logging_secure_erasing>` |
8 | 10 |
|
9 | 11 | .. _cfg_binary_logging_snapshots-force_recovery:
|
10 | 12 |
|
11 | 13 | .. confval:: force_recovery
|
12 | 14 |
|
13 | 15 | Since version 1.7.4.
|
| 16 | + |
14 | 17 | If ``force_recovery`` equals true, Tarantool tries to continue if there is
|
15 | 18 | an error while reading a :ref:`snapshot file<index-box_persistence>`
|
16 | 19 | (at server instance start) or a :ref:`write-ahead log file<internals-wal>`
|
|
31 | 34 | .. confval:: wal_max_size
|
32 | 35 |
|
33 | 36 | Since version 1.7.4.
|
| 37 | + |
34 | 38 | The maximum number of bytes in a single write-ahead log file.
|
35 | 39 | When a request would cause an .xlog file to become larger than
|
36 | 40 | ``wal_max_size``, Tarantool creates another WAL file.
|
|
45 | 49 | .. confval:: snap_io_rate_limit
|
46 | 50 |
|
47 | 51 | Since version 1.4.9.
|
| 52 | + |
48 | 53 | Reduce the throttling effect of :doc:`box.snapshot() </reference/reference_lua/box_snapshot>` on
|
49 | 54 | INSERT/UPDATE/DELETE performance by setting a limit on how many
|
50 | 55 | megabytes per second it can write to disk. The same can be
|
|
64 | 69 |
|
65 | 70 | .. confval:: wal_mode
|
66 | 71 |
|
67 |
| - Since version 1.6.2. Specify fiber-WAL-disk synchronization mode as: |
| 72 | + Since version 1.6.2. |
| 73 | + |
| 74 | + Specify fiber-WAL-disk synchronization mode as: |
68 | 75 |
|
69 | 76 | * ``none``: write-ahead log is not maintained.
|
70 | 77 | A node with ``wal_mode = none`` can't be replication master;
|
|
83 | 90 | .. confval:: wal_dir_rescan_delay
|
84 | 91 |
|
85 | 92 | Since version 1.6.2.
|
| 93 | + |
86 | 94 | Number of seconds between periodic scans of the write-ahead-log
|
87 | 95 | file directory, when checking for changes to write-ahead-log
|
88 | 96 | files for the sake of :ref:`replication <replication>` or :ref:`hot standby <index-hot_standby>`.
|
|
97 | 105 | .. confval:: wal_queue_max_size
|
98 | 106 |
|
99 | 107 | Since version :doc:`2.8.1 </release/2.8.1>`.
|
| 108 | + |
100 | 109 | The size of the queue (in bytes) used by a :ref:`replica <replication-roles>` to submit
|
101 | 110 | new transactions to a :ref:`write-ahead log<internals-wal>` (WAL).
|
102 | 111 | This option helps limit the rate at which a replica submits transactions to the WAL.
|
|
118 | 127 | .. confval:: wal_cleanup_delay
|
119 | 128 |
|
120 | 129 | Since version :doc:`2.6.3 </release/2.6.3>`.
|
| 130 | + |
121 | 131 | The delay (in seconds) used to prevent the :ref:`Tarantool garbage collector <cfg_checkpoint_daemon-garbage-collector>`
|
122 | 132 | from immediately removing :ref:`write-ahead log<internals-wal>` files after a node restart.
|
123 | 133 | This delay eliminates possible erroneous situations when the master deletes WALs
|
|
136 | 146 | | Type: number
|
137 | 147 | | Default: 14400 seconds
|
138 | 148 | | Environment variable: TT_WAL_CLEANUP_DELAY
|
139 |
| - | Dynamic: **yes** |
| 149 | + | Dynamic: **yes** |
| 150 | +
|
| 151 | + |
| 152 | +.. _cfg_binary_logging_snapshots-wal_ext: |
| 153 | + |
| 154 | +.. confval:: wal_ext |
| 155 | + |
| 156 | + Since version :doc:`2.11.0 </release/2.11.0>`. |
| 157 | + |
| 158 | + (Enterprise Edition only) Allows you to add auxiliary information to each :ref:`write-ahead log <internals-wal>` record. |
| 159 | + For example, you can enable storing an old and new tuple for each CRUD operation performed. |
| 160 | + This information might be helpful for implementing a CDC (Change Data Capture) utility that transforms a data replication stream. |
| 161 | + |
| 162 | + You can enable storing old and new tuples as follows: |
| 163 | + |
| 164 | + * Set the ``old`` and ``new`` options to ``true`` to store old and new tuples in a write-ahead log for all spaces. |
| 165 | + |
| 166 | + .. code-block:: lua |
| 167 | +
|
| 168 | + box.cfg { |
| 169 | + wal_ext = { old = true, new = true } |
| 170 | + } |
| 171 | +
|
| 172 | + * To adjust these options for specific spaces, use the ``spaces`` option. |
| 173 | + |
| 174 | + .. code-block:: lua |
| 175 | +
|
| 176 | + box.cfg { |
| 177 | + wal_ext = { |
| 178 | + old = true, new = true, |
| 179 | + spaces = { |
| 180 | + space1 = { old = false }, |
| 181 | + space2 = { new = false } |
| 182 | + } |
| 183 | + } |
| 184 | + } |
| 185 | +
|
| 186 | +
|
| 187 | + The configuration for specific spaces has priority over the global configuration, |
| 188 | + so only new tuples are added to the log for ``space1`` and only old tuples for ``space2``. |
| 189 | + |
| 190 | + Note that records with additional fields are :ref:`replicated <replication-architecture>` as follows: |
| 191 | + |
| 192 | + * If a replica doesn't support the extended format configured on a master, auxiliary fields are skipped. |
| 193 | + * If a replica and master have different configurations for WAL records, a master's configuration is ignored. |
| 194 | + |
| 195 | + | Type: map |
| 196 | + | Default: nil |
| 197 | + | Environment variable: TT_WAL_EXT |
| 198 | +
|
| 199 | + |
| 200 | +.. _cfg_binary_logging_secure_erasing: |
| 201 | + |
| 202 | +.. confval:: secure_erasing |
| 203 | + |
| 204 | + Since version :doc:`3.0.0 </release/3.0.0>`. |
| 205 | + |
| 206 | + (Enterprise Edition only) If **true**, forces Tarantool to overwrite a data file a few times before deletion to render recovery of a deleted file impossible. |
| 207 | + The option applies to both ``.xlog`` and ``.snap`` files as well as Vinyl data files. |
| 208 | + |
| 209 | + | Type: boolean |
| 210 | + | Default: false |
| 211 | + | Environment variable: TT_SECURE_ERASING |
| 212 | + | Dynamic: **yes** |
0 commit comments