All instance options
Configuration file path
Environment variable:MEILI_CONFIG_FILE_PATHCLI option:
--config-file-pathDefault:
./config.tomlExpected value: a filepath Designates the location of the configuration file to load at launch.
Database path
Environment variable:MEILI_DB_PATHCLI option:
--db-pathDefault value:
"data.ms/"Expected value: a filepath Designates the location where database files will be created and retrieved.
Environment
Environment variable:MEILI_ENVCLI option:
--envDefault value:
developmentExpected value:
production or development
Configures the instance’s environment. Value must be either production or development.
production:
- Setting a master key of at least 16 bytes is mandatory. If no master key is provided or if it is under 16 bytes, Meilisearch will suggest a secure autogenerated master key
- The search preview interface is disabled
development:
- Setting a master key is optional. If no master key is provided or if it is under 16 bytes, Meilisearch will suggest a secure autogenerated master key
- Search preview is enabled
HTTP address & port binding
Environment variable:MEILI_HTTP_ADDRCLI option:
--http-addrDefault value:
"localhost:7700"Expected value: an HTTP address and port Sets the HTTP address and port Meilisearch will use.
Master key
Environment variable:MEILI_MASTER_KEYCLI option:
--master-keyDefault value:
NoneExpected value: a UTF-8 string of at least 16 bytes Sets the instance’s master key, automatically protecting all routes except
GET /health. This means you will need a valid API key to access all other endpoints.
When --env is set to production, providing a master key is mandatory. If none is given, or it is under 16 bytes, Meilisearch will throw an error and refuse to launch.
When --env is set to development, providing a master key is optional. If none is given, all routes will be unprotected and publicly accessible.
If you do not supply a master key in production or development environments or it is under 16 bytes, Meilisearch will suggest a secure autogenerated master key you can use when restarting your instance.
Learn more about Meilisearch’s use of security keys.
Disable analytics
Environment variable:MEILI_NO_ANALYTICSCLI option:
--no-analytics
Deactivates Meilisearch’s built-in telemetry when provided.
Meilisearch automatically collects data from all instances that do not opt out using this flag. All gathered data is used solely for the purpose of improving Meilisearch, and can be deleted at any time.
Read more about our policy on data collection, or take a look at the comprehensive list of all data points we collect.
Dumpless upgrade
Environment variable:MEILI_EXPERIMENTAL_DUMPLESS_UPGRADECLI option:
--experimental-dumpless-upgradeDefault value: None
Expected value: None Migrates the database to a new Meilisearch version after you have manually updated the binary. Learn more about updating Meilisearch to a new release.
Dump directory
Environment variable:MEILI_DUMP_DIRCLI option:
--dump-dirDefault value:
dumps/Expected value: a filepath pointing to a valid directory Sets the directory where Meilisearch will create dump files. Learn more about creating dumps.
Import dump
Environment variable:MEILI_IMPORT_DUMPCLI option:
--import-dumpDefault value: none
Expected value: a filepath pointing to a
.dump file
Imports the dump file located at the specified path. Path must point to a .dump file. If a database already exists, Meilisearch will throw an error and abort launch.
Meilisearch will only launch once the dump data has been fully indexed. The time this takes depends on the size of the dump file.
Ignore missing dump
Environment variable:MEILI_IGNORE_MISSING_DUMPCLI option:
--ignore-missing-dump
Prevents Meilisearch from throwing an error when --import-dump does not point to a valid dump file. Instead, Meilisearch will start normally without importing any dump.
This option will trigger an error if --import-dump is not defined.
Ignore dump if DB exists
Environment variable:MEILI_IGNORE_DUMP_IF_DB_EXISTSCLI option:
--ignore-dump-if-db-existsExpected value: a boolean (
true or false)
Set this option to true to prevent a Meilisearch instance with an existing database from throwing an error when using --import-dump. When enabled, the dump will be ignored and Meilisearch will launch using the existing database.
For the environment variable, set MEILI_IGNORE_DUMP_IF_DB_EXISTS=true. For the CLI option, pass --ignore-dump-if-db-exists.
This option will trigger an error if --import-dump is not defined.
Log level
Environment variable:MEILI_LOG_LEVELCLI option:
--log-levelDefault value:
'INFO'Expected value: one of
ERROR, WARN, INFO, DEBUG, TRACE, OR OFF
Defines how much detail should be present in Meilisearch’s logs.
Meilisearch currently supports five log levels, listed in order of increasing verbosity:
'ERROR': only log unexpected events indicating Meilisearch is not functioning as expected'WARN': log all unexpected events, regardless of their severity'INFO': log all events. This is the default value of--log-level'DEBUG': log all events and include detailed information on Meilisearch’s internal processes. Useful when diagnosing issues and debugging'TRACE': log all events and include even more detailed information on Meilisearch’s internal processes. We do not advise using this level as it is extremely verbose. Use'DEBUG'before considering'TRACE'.'OFF': disable logging
Customize log output
Environment variable:MEILI_EXPERIMENTAL_LOGS_MODECLI option:
--experimental-logs-modeDefault value:
'human'Expected value: one of
human or json
Defines whether logs should output a human-readable text or JSON data.
Max indexing memory
Environment variable:MEILI_MAX_INDEXING_MEMORYCLI option:
--max-indexing-memoryDefault value: 2/3 of the available RAM
Expected value: an integer (
104857600) or a human readable size ('100Mb')
Sets the maximum amount of RAM Meilisearch can use when indexing. By default, Meilisearch uses no more than two thirds of available memory.
The value must either be given in bytes or explicitly state a base unit: 107374182400, '107.7Gb', or '107374 Mb'.
It is possible that Meilisearch goes over the exact RAM limit during indexing. In most contexts and machines, this should be a negligible amount with little to no impact on stability and performance.
Reduce indexing memory usage
Environment variable:MEILI_EXPERIMENTAL_REDUCE_INDEXING_MEMORY_USAGECLI option:
--experimental-reduce-indexing-memory-usageDefault value:
NoneEnables
MDB_WRITEMAP, an LMDB option. Activating this option may reduce RAM usage in some UNIX and UNIX-like setups. However, it may also negatively impact write speeds and overall performance.
Max indexing threads
Environment variable:MEILI_MAX_INDEXING_THREADSCLI option:
--max-indexing-threadsDefault value: half of the available threads
Expected value: an integer Sets the maximum number of threads Meilisearch can use during indexing. By default, the indexer avoids using more than half of a machine’s total processing units. This ensures Meilisearch is always ready to perform searches, even while you are updating an index. If
--max-indexing-threads is higher than the real number of cores available in the machine, Meilisearch uses the maximum number of available cores.
In single-core machines, Meilisearch has no choice but to use the only core available for indexing. This may lead to a degraded search experience during indexing.
Payload limit size
Environment variable:MEILI_HTTP_PAYLOAD_SIZE_LIMITCLI option:
--http-payload-size-limitDefault value:
100000000 (~100MB)Expected value: an integer Sets the maximum size of accepted payloads. Value must be given in bytes or explicitly stating a base unit. For example, the default value can be written as
100000000, '100Mb', or '100 MB'.
Search queue size
Environment variable:MEILI_EXPERIMENTAL_SEARCH_QUEUE_SIZECLI option:
--experimental-search-queue-sizeDefault value:
1000Expected value: an integer Configure the maximum amount of simultaneous search requests. By default, Meilisearch queues up to 1000 search requests at any given moment. This limit exists to prevent Meilisearch from consuming an unbounded amount of RAM.
Search query embedding cache
Environment variable:MEILI_EXPERIMENTAL_EMBEDDING_CACHE_ENTRIESCLI option:
--experimental-embedding-cache-entriesDefault value:
0Expected value: an integer Sets the size of the search query embedding cache. By default, Meilisearch generates an embedding for every new search query. When this option is set to an integer bigger than 0, Meilisearch returns a previously generated embedding if it recently performed the same query. The least recently used entries are evicted first. Embedders with the same configuration share the same cache, even if they were declared in distinct indexes.
Schedule snapshot creation
Environment variable:MEILI_SCHEDULE_SNAPSHOTCLI option:
--schedule-snapshotDefault value: disabled if not present,
86400 if present without a valueExpected value:
None or an integer
Activates scheduled snapshots. Snapshots are disabled by default.
It is possible to use --schedule-snapshot without a value. If --schedule-snapshot is present when launching an instance but has not been assigned a value, Meilisearch takes a new snapshot every 24 hours.
For more control over snapshot scheduling, pass an integer representing the interval in seconds between each snapshot. When --schedule-snapshot=3600, Meilisearch takes a new snapshot every hour. Meilisearch waits for the configured interval before creating the first scheduled snapshot.
When using the configuration file, it is also possible to explicitly pass a boolean value to
schedule_snapshot. Meilisearch takes a new snapshot every 24 hours when schedule_snapshot=true, and takes no snapshots when schedule_snapshot=false.Snapshot destination
Environment variable:MEILI_SNAPSHOT_DIRCLI option:
--snapshot-dirDefault value:
snapshots/Expected value: a filepath pointing to a valid directory Sets the directory where Meilisearch will store snapshots.
Uncompressed snapshots
Environment variable:MEILI_EXPERIMENTAL_NO_SNAPSHOT_COMPACTIONCLI option:
--experimental-no-snapshot-compactionDisables snapshot compression. This may significantly speed up snapshot creation at the cost of bigger snapshot files.
Import snapshot
Environment variable:MEILI_IMPORT_SNAPSHOTCLI option:
--import-snapshotDefault value:
NoneExpected value: a filepath pointing to a snapshot file Launches Meilisearch after importing a previously-generated snapshot at the given filepath. This command will throw an error if:
- A database already exists
- No valid snapshot can be found in the specified path
--ignore-snapshot-if-db-exists and --ignore-missing-snapshot options, respectively.
Ignore missing snapshot
Environment variable:MEILI_IGNORE_MISSING_SNAPSHOTCLI option:
--ignore-missing-snapshot
Prevents a Meilisearch instance from throwing an error when --import-snapshot does not point to a valid snapshot file.
This command will throw an error if --import-snapshot is not defined.
Ignore snapshot if DB exists
Environment variable:MEILI_IGNORE_SNAPSHOT_IF_DB_EXISTSCLI option:
--ignore-snapshot-if-db-exists
Prevents a Meilisearch instance with an existing database from throwing an error when using --import-snapshot. Instead, the snapshot will be ignored and Meilisearch will launch using the existing database.
This command will throw an error if --import-snapshot is not defined.
Task webhook URL
Environment variable:MEILI_TASK_WEBHOOK_URLCLI option:
--task-webhook-urlDefault value:
NoneExpected value: a URL string Notifies the configured URL whenever Meilisearch finishes processing a task or batch of tasks. Meilisearch uses the URL as given, retaining any specified query parameters. The webhook payload contains the list of finished tasks in ndjson. For more information, consult the dedicated task webhook guide. The task webhook option requires having access to a command-line interface. If you are using Meilisearch Cloud, use the
/webhooks API route instead.
Task webhook authorization header
Environment variable:MEILI_TASK_WEBHOOK_AUTHORIZATION_HEADERCLI option:
--task-webhook-authorization-headerDefault value:
NoneExpected value: an authentication token string Includes an authentication token in the authorization header when notifying the webhook URL.
Maximum number of batched tasks
Environment variable:MEILI_EXPERIMENTAL_MAX_NUMBER_OF_BATCHED_TASKSCLI option:
--experimental-max-number-of-batched-tasksDefault value: unlimited
Expected value: an integer Limit the number of tasks Meilisearch performs in a single batch. May improve stability in systems handling a large queue of resource-intensive tasks.
Maximum batch payload size
Environment variable:MEILI_EXPERIMENTAL_LIMIT_BATCHED_TASKS_TOTAL_SIZECLI option:
--experimental-limit-batched-tasks-total-sizeDefault value: Half of total available memory, up to a maximum of 10 GiB
Expected value: an integer Sets a maximum payload size for batches in bytes. Smaller batches are less efficient, but consume less RAM and reduce immediate latency.
Replication parameters
Environment variable:MEILI_EXPERIMENTAL_REPLICATION_PARAMETERSCLI option:
--experimental-replication-parametersDefault value:
NoneHelps running Meilisearch in cluster environments. It does this by modifying task handling in three ways:
- Task auto-deletion is disabled
- Allows you to manually set task uids by adding a custom
TaskIdheader to your API requests - Allows you to dry register tasks by specifying a
DryRun: trueheader in your request
Disable new indexer
Environment variable:MEILI_EXPERIMENTAL_NO_EDITION_2024_FOR_SETTINGSCLI option:
--experimental-no-edition-2024-for-settingsDefault value:
NoneFalls back to previous settings indexer.
Disable new indexer for dumps
Environment variable:MEILI_EXPERIMENTAL_NO_EDITION_2024_FOR_DUMPSCLI option:
--experimental-no-edition-2024-for-dumpsDefault value:
NoneFalls back to previous indexer when importing dumps.
Enable logs route
Environment variable:MEILI_EXPERIMENTAL_ENABLE_LOGS_ROUTECLI option:
--experimental-enable-logs-routeDefault value:
NoneEnables the
/logs/stream, /logs/stderr and DELETE /logs/stream routes for log streaming and configuration.
Enable metrics
Environment variable:MEILI_EXPERIMENTAL_ENABLE_METRICSCLI option:
--experimental-enable-metricsDefault value:
NoneEnables the Prometheus
/metrics endpoint for monitoring.
CONTAINS filter operator
Environment variable:MEILI_EXPERIMENTAL_CONTAINS_FILTERCLI option:
--experimental-contains-filterDefault value:
NoneEnables the
CONTAINS filter operator at launch. It can also be toggled at runtime via the /experimental-features API route.
Drop search after
Environment variable:MEILI_EXPERIMENTAL_DROP_SEARCH_AFTERCLI option:
--experimental-drop-search-afterDefault value:
60Expected value: an integer (seconds) Sets the maximum time in seconds a search request can take before being dropped. Helps prevent slow searches from blocking resources.
Searches per core
Environment variable:MEILI_EXPERIMENTAL_NB_SEARCHES_PER_CORECLI option:
--experimental-nb-searches-per-coreDefault value:
4Expected value: an integer Configures the number of concurrent search requests each CPU core can handle.
Search personalization
Environment variable:MEILI_EXPERIMENTAL_PERSONALIZATION_API_KEYCLI option:
--experimental-personalization-api-keyDefault value:
NoneExpected value: a Cohere API key Enables search personalization. Must be a valid Cohere API key in string format.
Allow requests to private networks
Environment variable:MEILI_EXPERIMENTAL_ALLOWED_IP_NETWORKSCLI option:
--experimental-allowed-ip-networksDefault value:
NoneExpected value: a list of comma-separated CIDR networks Allow Meilisearch to query services running on private networks. By default, Meilisearch will prevent any requests to a host resolving to a non-global IP, in the sense of the IANA IPv4 Special-Purpose Address Registry or the IANA IPv6 Special-Purpose Address Registry. This is done to prevent potential firewall bypasses (see our blog post on SSRF vulnerability). You may want to allow requests to private networks to query internal services during webhooks or use local embedding services. To do so, specify a list of comma-separated CIDR networks (e.g.
192.168.0.0/16,10.0.0.0/8). You may specify any to allow all
requests regardless of target IP (use only in controlled environments, this is not recommended for production).
Remote search timeout
Environment variable:MEILI_EXPERIMENTAL_REMOTE_SEARCH_TIMEOUT_SECONDSDefault value:
30Expected value: a positive integer (seconds) Sets the maximum time in seconds a remote federated search request can take before timing out. This configuration is only available via environment variable; no CLI flag is available.
Disable FID-based database cleanup
Environment variable:MEILI_EXPERIMENTAL_DISABLE_FID_BASED_DATABASES_CLEANUPDefault value:
falseExpected value: a boolean Allows you to opt out of the field ID-based database cleanup when upgrading from Meilisearch versions prior to v1.32. Set this to
true if you experience issues during the upgrade process. This configuration is only available via environment variable; no CLI flag is available.
S3 options
S3 snapshot storage requires the Meilisearch Enterprise Edition. See Enterprise and Community editions for details.
Bucket URL
Environment variable:MEILI_S3_BUCKET_URLCLI option:
--s3-bucket-urlDefault value:
NoneThe URL for your S3 bucket. The URL must follow the format
https://s3.REGION.amazonaws.com.
Bucket region
Environment variable:MEILI_S3_BUCKET_REGIONCLI option:
--s3-bucket-regionDefault value:
NoneThe region of your S3 bucket. Must be a valid AWS region, such as
us-east-1.
Bucket name
Environment variable:MEILI_S3_BUCKET_NAMECLI option:
--s3-bucket-nameDefault value:
NoneThe name of your S3 bucket.
Snapshot prefix
Environment variable:MEILI_S3_SNAPSHOT_PREFIXCLI option:
--s3-snapshot-prefixDefault value:
NoneThe path leading to the snapshot directory in your S3 bucket. Uses normal slashes.
Access key
Environment variable:MEILI_S3_ACCESS_KEYCLI option:
--s3-access-keyDefault value:
NoneYour S3 bucket’s access key.
Secret key
Environment variable:MEILI_S3_SECRET_KEYCLI option:
--s3-secret-keyDefault value:
NoneYour S3 bucket’s secret key.
Role ARN
Environment variable:MEILI_EXPERIMENTAL_S3_ROLE_ARNCLI option:
--experimental-s3-role-arnDefault value:
NoneIAM role ARN for web identity federation. Use this instead of access key and secret key for authentication. Cannot be combined with
--s3-access-key and --s3-secret-key.
Web identity token file
Environment variable:MEILI_EXPERIMENTAL_S3_WEB_IDENTITY_TOKEN_FILECLI option:
--experimental-s3-web-identity-token-fileDefault value:
NonePath to the web identity token file for S3 authentication via web identity federation. Cannot be combined with
--s3-access-key and --s3-secret-key.
Maximum parallel in-flight requests
Environment variable:MEILI_EXPERIMENTAL_S3_MAX_IN_FLIGHT_PARTSCLI option:
--experimental-s3-max-in-flight-partsDefault value:
10The maximum number of in-flight multipart requests Meilisearch should send to S3 in parallel.
Compression level
Environment variable:MEILI_EXPERIMENTAL_S3_COMPRESSION_LEVELCLI option:
--experimental-s3-compression-levelDefault value:
0The compression level to use for the snapshot tarball. Defaults to 0, no compression.
Signature duration
Environment variable:MEILI_EXPERIMENTAL_S3_SIGNATURE_DURATION_SECONDSCLI option:
--experimental-s3-signature-duration-secondsDefault value:
28800The maximum duration processing a snapshot can take. Defaults to 8 hours.
Multipart section size
Environment variable:MEILI_EXPERIMENTAL_S3_MULTIPART_PART_SIZECLI option:
--experimental-s3-multipart-part-sizeDefault value:
NoneThe size of each multipart section. Must be >10MiB and <8GiB. Defaults to 375MiB, which enables databases of up to 3.5TiB.
SSL options
SSL authentication path
Environment variable:MEILI_SSL_AUTH_PATHCLI option:
--ssl-auth-pathDefault value:
NoneExpected value: a filepath Enables client authentication in the specified path.
SSL certificates path
Environment variable:MEILI_SSL_CERT_PATHCLI option:
--ssl-cert-pathDefault value:
NoneExpected value: a filepath pointing to a valid SSL certificate Sets the server’s SSL certificates. Value must be a path to PEM-formatted certificates. The first certificate should certify the KEYFILE supplied by
--ssl-key-path. The last certificate should be a root CA.
SSL key path
Environment variable:MEILI_SSL_KEY_PATHCLI option:
--ssl-key-pathDefault value:
NoneExpected value: a filepath pointing to a valid SSL key file Sets the server’s SSL key files. Value must be a path to an RSA private key or PKCS8-encoded private key, both in PEM format.
SSL OCSP path
Environment variable:MEILI_SSL_OCSP_PATHCLI option:
--ssl-ocsp-pathDefault value:
NoneExpected value: a filepath pointing to a valid OCSP certificate Sets the server’s OCSP file. Optional Reads DER-encoded OCSP response from OCSPFILE and staple to certificate.
SSL require auth
Environment variable:MEILI_SSL_REQUIRE_AUTHCLI option:
--ssl-require-authDefault value:
None
Makes SSL authentication mandatory.
Sends a fatal alert if the client does not complete client authentication.
SSL resumption
Environment variable:MEILI_SSL_RESUMPTIONCLI option:
--ssl-resumptionDefault value:
None
Activates SSL session resumption.
SSL tickets
Environment variable:MEILI_SSL_TICKETSCLI option:
--ssl-ticketsDefault value:
None
Activates SSL tickets.