rladmin tune
Configures parameters for databases, proxies, nodes, and clusters.
tune cluster
Configures cluster parameters.
rladmin tune cluster
[ repl_diskless { enabled | disabled } ]
[ redis_provision_node_threshold <size> ]
[ redis_migrate_node_threshold <size> ]
[ redis_provision_node_threshold_percent <percent> ]
[ redis_migrate_node_threshold_percent <percent> ]
[ max_simultaneous_backups <number> ]
[ failure_detection_sensitivity { high | low } ]
[ watchdog_profile { cloud | local-network } ]
[ slave_ha { enabled | disabled } ]
[ slave_ha_grace_period <seconds> ]
[ slave_ha_cooldown_period <seconds> ]
[ slave_ha_bdb_cooldown_period <seconds> ]
[ max_saved_events_per_type <value> ]
[ parallel_shards_upgrade <value> ]
[ default_concurrent_restore_actions <value> ]
[ show_internals { enabled | disabled } ]
[ expose_hostnames_for_all_suffixes { enabled | disabled } ]
[ redis_upgrade_policy { latest | major } ]
[ default_redis_version <value> ]
[ default_non_sharded_proxy_policy { single | all-master-shards | all-nodes } ]
[ default_sharded_proxy_policy { single | all-master-shards | all-nodes } ]
[ default_shards_placement { dense | sparse } ]
[ data_internode_encryption { enabled | disabled } ]
[ db_conns_auditing { enabled | disabled } ]
[ acl_pubsub_default { resetchannels | allchannels } ]
[ resp3_default { enabled | disabled } ]
Parameters
| Parameters | Type/Value | Description |
|---|---|---|
| acl_pubsub_default | resetchannelsallchannels | Default pub/sub ACL rule for all databases in the cluster: • resetchannels blocks access to all channels (restrictive)• allchannels allows access to all channels (permissive) |
| data_internode_encryption | enableddisabled | Activates or deactivates internode encryption for new databases |
| db_conns_auditing | enableddisabled | Activates or deactivates connection auditing by default for new databases of a cluster |
| default_concurrent_restore_actions | integerall | Default number of concurrent actions when restoring a node from a snapshot (positive integer or “all”) |
| default_non_sharded_proxy_policy | singleall-master-shardsall-nodes | Default proxy policy for newly created non-sharded databases’ endpoints |
| default_redis_version | version number | The default Redis database compatibility version used to create new databases. The value parameter should be a version number in the form of “x.y” where x represents the major version number and y represents the minor version number. The final value corresponds to the desired version of Redis. You cannot set default_redis_version to a value higher than that supported by the current redis_upgrade_policy value. |
| default_sharded_proxy_policy | singleall-master-shardsall-nodes | Default proxy policy for newly created sharded databases’ endpoints |
| default_shards_placement | densesparse | New databases place shards according to the default shard placement policy |
| expose_hostnames_for_all_suffixes | enableddisabled | Exposes hostnames for all DNS suffixes |
| failure_detection_sensitivity | highlow | Predefined thresholds and timeouts for failure detection (previously known as watchdog_profile)• high (previously local-network) – high failure detection sensitivity, lower thresholds, faster failure detection and failover• low (previously cloud) – low failure detection sensitivity, higher tolerance for latency variance (also called network jitter) |
| login_lockout_counter_reset_after | time in seconds | Time after failed login attempt before the counter resets to 0 |
| login_lockout_duration | time in seconds | Time a locked account remains locked ( “0” means only an admin can unlock the account) |
| login_lockout_threshold | integer | Number of failed sign-in attempts to trigger locking a user account (“0” means never lock the account) |
| max_saved_events_per_type | integer | Maximum number of events each type saved in CCS per object type |
| max_simultaneous_backups | integer | Number of database backups allowed to run at the same time. Combines with max_redis_forks (set by tune node) to determine the number of shard backups allowed to run simultaneously. |
| parallel_shards_upgrade | integerall | Number of shards upgraded in parallel during DB upgrade (positive integer or “all”) |
| redis_migrate_node_threshold | size in MB | Memory (in MBs by default or can be specified) needed to migrate a database between nodes |
| redis_migrate_node_threshold_percent | percentage | Memory (in percentage) needed to migrate a database between nodes |
| redis_provision_node_threshold | size in MB | Memory (in MBs by default or can be specified) needed to provision a new database |
| redis_provision_node_threshold_percent | percentage | Memory (in percentage) needed to provision a new database |
| redis_upgrade_policy | latestmajor | When you upgrade or create a new Redis database, this policy determines which version of Redis database compatibility is used. Supported values are:
|
| repl_diskless | enableddisabled | Activates or deactivates diskless replication (can be overridden per database) |
| resp3_default | enableddisabled | Determines the default value of the resp3 option upon upgrading a database to version 7.2 (defaults to enabled) |
| show_internals | enableddisabled | Controls the visibility of internal databases that are only used for the cluster’s management |
| slave_ha | enableddisabled | Activates or deactivates replica high availability in the cluster (enabled by default; use rladmin tune db to change slave_ha for a specific database)Deprecated as of Redis Enterprise Software v7.2.4. |
| slave_ha_bdb_cooldown_period | time in seconds (default: 7200) | Time (in seconds) a database must wait after its shards are relocated by replica high availability before it can go through another shard migration if another node fails (default is 2 hours) |
| slave_ha_cooldown_period | time in seconds (default: 3600) | Time (in seconds) replica high availability must wait after relocating shards due to node failure before performing another shard migration for any database in the cluster (default is 1 hour) |
| slave_ha_grace_period | time in seconds (default: 600) | Time (in seconds) between when a node fails and when replica high availability starts relocating shards to another node |
| watchdog_profile | cloudlocal-network | Watchdog profiles with preconfigured thresholds and timeouts (deprecated as of Redis Enterprise Software v6.4.2-69; use failure_detection_sensitivity instead)• cloud is suitable for common cloud environments and has a higher tolerance for latency variance (also called network jitter).• local-network is suitable for dedicated LANs and has better failure detection and failover times. |
Returns
Returns Finished successfully if the cluster configuration was changed. Otherwise, it returns an error.
Use rladmin info cluster to verify the cluster configuration was changed.
Example
$ rladmin tune cluster slave_ha enabled
Finished successfully
$ rladmin info cluster | grep slave_ha
slave_ha: enabled
tune db
Configures database parameters.
rladmin tune db { db:<id> | <name> }
[ slave_buffer <valueMG | hard:soft:time> ]
[ client_buffer <value> ]
[ repl_backlog <valueMB | auto> ]
[ crdt_repl_backlog <valueMB | auto> ]
[ repl_timeout <seconds> ]
[ repl_diskless { enabled | disabled | default } ]
[ master_persistence { enabled | disabled } ]
[ maxclients <value> ]
[ schedpolicy { cmp | mru | spread | mnp } ]
[ max_shard_pipeline <value> ]
[ conns <value> ]
[ conns_type <value> ]
[ max_client_pipeline <value> ]
[ max_connections <value> ]
[ max_aof_file_size <size> ]
[ max_aof_load_time <seconds> ]
[ oss_cluster { enabled | disabled } ]
[ oss_cluster_api_preferred_ip_type <value> ]
[ slave_ha { enabled | disabled } ]
[ slave_ha_priority <value> ]
[ skip_import_analyze { enabled | disabled } ]
[ mkms { enabled | disabled } ]
[ continue_on_error ]
[ gradual_src_mode { enabled | disabled } ]
[ gradual_sync_mode { enabled | disabled | auto } ]
[ gradual_sync_max_shards_per_source <value> ]
[ module_name <value> ] [ module_config_params <value> ]
[ crdt_xadd_id_uniqueness_mode { liberal | semi-strict | strict } ]
[ metrics_export_all { enabled | disabled } ]
[ syncer_mode { distributed | centralized }]
[ syncer_monitoring { enabled | disabled } ]
[ mtls_allow_weak_hashing { enabled | disabled } ]
[ mtls_allow_outdated_cert { enabled | disabled } ]
[ data_internode_encryption { enabled | disabled } ]
[ db_conns_auditing { enabled | disabled } ]
[ resp3 { enabled | disabled } ]
Parameters
| Parameter | Type/Value | Description |
|---|---|---|
| db:id | integer | ID of the specified database |
| name | string | Name of the specified database |
| client_buffer | value in MB hard:soft:time | Redis client output buffer limits |
| conns | integer | Size of internal connection pool, specified per-thread or per-shard depending on conns_type |
| conns_type | per-threadper-shard | Specifies connection pool size as either per-thread or per-shard |
| continue_on_error | Flag that skips tuning shards that can’t be reached | |
| crdt_repl_backlog | value in MBauto | Size of the Active-Active replication buffer |
| crdt_xadd_id_uniqueness_mode | liberalsemi-strictstrict | XADD’s behavior in an Active-Active database, defined as liberal, semi-strict, or strict (see descriptions below) |
| data_internode_encryption | enableddisabled | Activates or deactivates internode encryption for the database |
| db_conns_auditing | enableddisabled | Activates or deactivates database connection auditing for a database |
| gradual_src_mode | enableddisabled | Activates or deactivates gradual sync of sources |
| gradual_sync_max_shards_per_source | integer | Number of shards per sync source that can be replicated in parallel (positive integer) |
| gradual_sync_mode | enableddisabledauto | Activates, deactivates, or automatically determines gradual sync of source shards |
| master_persistence | enableddisabled | Activates or deactivates persistence of the primary shard |
| max_aof_file_size | size in MB | Maximum size (in MB, if not specified) of AoF file (minimum value is 10 GB) |
| max_aof_load_time | time in seconds | Time limit in seconds to load a shard from an append-only file (AOF). If exceeded, an AOF rewrite is initiated to decrease future load time. Minimum: 2700 seconds (45 minutes) Default: 3600 seconds (1 hour) |
| max_client_pipeline | integer | Maximum commands in the proxy’s pipeline per client connection (max value is 2047, default value is 200) |
| max_connections | integer | Maximum client connections to the database’s endpoint (default value is 0, which is unlimited) |
| max_shard_pipeline | integer | Maximum commands in the proxy’s pipeline per shard connection (default value is 200) |
| maxclients | integer | Controls the maximum client connections between the proxy and shards (default value is 10000) |
| metrics_export_all | enableddisabled | Activates the exporter to expose all shard metrics |
| mkms | enableddisabled | Activates multi-key multi-slot commands |
| module_config_params | string | Configures module arguments at runtime. Enclose module_config_params within quotation marks. |
| module_name | searchReJSONgraphtimeseriesbfrg | The module to configure with module_config_params |
| mtls_allow_outdated_cert | enableddisabled | Activates outdated certificates in mTLS connections |
| mtls_allow_weak_hashing | enableddisabled | Activates weak hashing (less than 2048 bits) in mTLS connections |
| oss_cluster | enableddisabled | Activates OSS cluster API |
| oss_cluster_api_preferred_ip_type | internalexternal | IP type for the endpoint and database in the OSS cluster API (default is internal) |
| repl_backlog | size in MBauto | Size of the replication buffer |
| repl_diskless | enableddisableddefault | Activates or deactivates diskless replication (defaults to the cluster setting) |
| repl_timeout | time in seconds | Replication timeout (in seconds) |
| resp3 | enableddisabled | Enables or deactivates RESP3 support (defaults to enabled) |
| schedpolicy | cmpmruspreadmnp | Controls how server-side connections are used when forwarding traffic to shards |
| skip_import_analyze | enableddisabled | Skips the analyzing step when importing a database |
| slave_buffer | value in MB hard:soft:time | Redis replica output buffer limits |
| slave_ha | enableddisabled | Activates or deactivates replica high availability (defaults to the cluster setting) |
| slave_ha_priority | integer | Priority of the database in the replica high-availability mechanism |
| syncer_mode | distributedcentralized | Configures syncer to run in distributed or centralized mode. For distributed syncer, the DMC policy must be all-nodes or all-master-nodes |
| syncer_monitoring | enableddisabled | Activates syncer monitoring |
| XADD behavior mode | Description |
|---|---|
| liberal | XADD succeeds with any valid ID (not recommended, allows duplicate IDs) |
| semi-strict | Allows a full ID. Partial IDs are completed with the unique database instance ID (not recommended, allows duplicate IDs). |
| strict | XADD fails if a full ID is given. Partial IDs are completed using the unique database instance ID. |
Returns
Returns Finished successfully if the database configuration was changed. Otherwise, it returns an error.
Use rladmin info db to verify the database configuration was changed.
Example
$ rladmin tune db db:4 repl_timeout 300
Tuning database: o
Finished successfully
$ rladmin info db db:4 | grep repl_timeout
repl_timeout: 300 seconds
tune node
Configures node parameters.
tune node { <id> | all }
[ max_listeners <value> ]
[ max_redis_forks <value> ]
[ max_redis_servers <value> ]
[ max_slave_full_syncs <value> ]
[ quorum_only { enabled | disabled } ]
Parameters
| Parameter | Type/Value | Description |
|---|---|---|
| id | integer | ID of the specified node |
| all | Configures settings for all nodes | |
| max_listeners | integer | Maximum number of endpoints that may be bound to the node |
| max_redis_forks | integer | Maximum number of background processes forked from shards that may exist on the node at any given time |
| max_redis_servers | integer | Maximum number of shards allowed to reside on the node |
| max_slave_full_syncs | integer | Maximum number of simultaneous replica full-syncs that may be running at any given time (0: Unlimited, -1: Use cluster settings) |
| quorum_only | enableddisabled | If activated, configures the node as a quorum-only node |
Returns
Returns Finished successfully if the node configuration was changed. Otherwise, it returns an error.
Use rladmin info node to verify the node configuration was changed.
Example
$ rladmin tune node 3 max_redis_servers 120
Finished successfully
$ rladmin info node 3 | grep "max redis servers"
max redis servers: 120
tune proxy
Configures proxy parameters.
rladmin tune proxy { <id> | all }
[ mode { static | dynamic } ]
[ threads <value> ]
[ max_threads <value> ]
[ scale_threshold <value> ]
[ scale_duration <seconds> ]
Parameters
| Parameter | Type/Value | Description |
|---|---|---|
| id | integer | ID of the specified proxy |
| all | Configures settings for all proxies | |
| max_threads | integer, (range: 1-255) | Maximum number of threads allowed |
| mode | staticdynamic | Determines if the proxy automatically adjusts the number of threads based on load size |
| scale_duration | time in seconds, (range: 10-300) | Time of scale_threshold CPU utilization before the automatic proxy automatically scales |
| scale_threshold | percentage, (range: 50-99) | CPU utilization threshold that triggers spawning new threads |
| threads | integer, (range: 1-255) | Initial number of threads created at startup |
Returns
Returns OK if the proxy configuration was changed. Otherwise, it returns an error.
Use rladmin info proxy to verify the proxy configuration was changed.
Example
$ rladmin tune proxy 2 scale_threshold 75
Configuring proxies:
- proxy:2: ok
$ rladmin info proxy 2 | grep scale_threshold
scale_threshold: 75 (%)