@datastax/astra-mongoose/bin/dse.yaml

Astra's NodeJS Mongoose compatibility client

# DSE Config Version: 6.9.13

# Memory limit for DSE In-Memory tables as a fraction of system memory. When not set,
# the default is 0.2 (20% of system memory).
# Specify max_memory_to_lock_fraction or max_memory_to_lock_mb, not both.

# max_memory_to_lock_fraction: 0.20

# Memory limit for DSE In-Memory tables as a maximum in MB.  When not set,
# max_memory_to_lock_fraction is used.  The max_memory_to_lock_fraction
# value is ignored if max_memory_to_lock_mb is set to a non-zero value.
# Specify max_memory_to_lock_fraction or max_memory_to_lock_mb, not both.

# max_memory_to_lock_mb: 10240

##########################
# Authentication options
#
# These options are used if the authenticator option in cassandra.yaml is set to
# com.datastax.bdp.cassandra.auth.DseAuthenticator
#
# The enabled option controls whether the DseAuthenticator will authenticate users. If
# set to true users will be authenticated, if set to false they will not.
# When not set enabled is false.
#
# DseAuthenticator allows multiple authentication schemes to be used at the same time.
# The schemes to be used are controlled by the default_scheme and other_schemes options.
# A driver can select the scheme to use during authentication.
#
# The default_scheme option selects which authentication scheme will be used if the driver
# does not request a specific scheme. This can be one of the following values:
#   internal - plain text authentication using the internal password authenticator
#   ldap     - plain text authentication using the passthrough LDAP authenticator
#   kerberos - GSSAPI authentication using the Kerberos authenticator
# The other_schemes option is a list of schemes that can also be selected for use by a
# driver and can be a list of the above schemes.
#
# The scheme_permissions option controls whether roles need to have permission granted to
# them in order to use specific authentication schemes. These permissions can be granted
# only when the DseAuthorizer is used.
#
# The allow_digest_with_kerberos option controls whether Digest-MD5 authentication is also
# allowed when Kerberos is one of the authentication schemes. If set to false, it will not
# be allowed. You must set allow_digest_with_kerberos to true in analytics clusters to use Hadoop
# inter-node authentication with Hadoop and Spark jobs.
#
# The plain_text_without_ssl controls how the DseAuthenticator reacts to plain text
# authentication requests over unencrypted client connections. It can be one of:
#   block  - block the request with an authentication error
#   warn   - log a warning about the request but allow it to continue
#   allow  - allow the request without any warning
#
# The transitional_mode option allows the DseAuthenticator to operate in a transitional
# mode during setup of authentication in a cluster. This can be one of the following values:
#   disabled   - transitional mode is disabled
#   permissive - Only super users are authenticated and logged in, all other
#                authentication attempts will be logged in as the anonymous user
#   normal     - If credentials are passed they are authenticated. If the
#                authentication is successful then the user is logged in, otherwise
#                the user is logged in as anonymous. If no credentials are passed,
#                then the user is logged in as anonymous
#   strict     - If credentials are passed they are authenticated. If the
#                authentication is successful, the user is logged in. If the
#                authentication fails, an authentication error is returned. If no
#                credentials are passed, the user is logged in as anonymous
authentication_options:
  enabled: true
  default_scheme: internal
  other_schemes:
  scheme_permissions: false
  allow_digest_with_kerberos: true
  plain_text_without_ssl: warn
  transitional_mode: disabled

##########################
# Role management options
#
# These options are used when the role_manager option in cassandra.yaml is set to
# com.datastax.bdp.cassandra.auth.DseRoleManager
#
# mode can be one of:
#   internal - the granting and revoking of roles is managed internally
#              using the GRANT ROLE and REVOKE ROLE statements
#   ldap - the granting and revoking of roles is managed by an external
#          LDAP server configured using the ldap_options.
#
# stats is a boolean option (false by default). If set to true, the DSE server will keep
# track of role creation and password change timestamps in the dse_security.role_stats table.
# This data will be updated based newly executed CREATE / ALTER / DROP role CQL statements.
#
# mode_by_authentication allows and overrides the defined mode based on the authentication scheme.
# The allow values are either internal or ldap. If not set, the defined mode will be used.
#
# role_management_options:
#     mode: internal
#     stats: false
#     mode_by_authentication:
#         internal:
#         ldap:
#         kerberos:

##########################
# Authorization options
#
# These options are used if the authorization option in cassandra.yaml is set to
# com.datastax.bdp.cassandra.auth.DseAuthorizer
#
# The enabled option controls whether the DseAuthorizer will perform authorization. If
# set to true authorization is performed, if set to false it is not.
# When not set, enabled is false.
#
# The transitional_mode option allows the DseAuthorizer to operate in a transitional
# mode during setup of authorization in a cluster. This can be one of the following values:
#   disabled   - transitional mode is disabled, all connections must provide valid credentials and
#                map to a login-enabled role
#   normal     - allow all connections that provide credentials, permissions can be granted to
#                resources but are not enforced
#   strict     - permissions can be granted to resources and are enforced on
#                authenticated users. They are not enforced against anonymous
#                users
#
# allow_row_level_security - To use row level security, set to true for the entire system.
#                            Use the same setting on all nodes.
# authorization_options:
#     enabled: false
#     transitional_mode: disabled
#     allow_row_level_security: false

##########################
# Kerberos options
#
# keytab is <path_to_keytab>/dse.keytab
# The keytab file must contain the credentials for both of the fully resolved principal names, which
# replace _HOST with the fully qualified domain name (FQDN) of the host in the service_principal and
# http_principal settings. The UNIX user running DSE must also have read permissions on the keytab.
#
# The service_principal is the DataStax Enterprise process runs under must use the form
# <dse_user>/_HOST@<REALM>
#
# The http_principal is used by the Tomcat application container to run DSE Search.
#
# The qop is the Quality of Protection (QOP) values that clients and servers
# can use for each connection.  Valid values are:
#   auth      - (default) authentication only
#   auth-int  - authentication plus integrity protection of all transmitted data
#   auth-conf - authentication plus integrity protection and encryption of all
#               transmitted data
#
# Warning - Encryption using auth-conf is separate and completely independent
#           of whether encryption is done using SSL.  If auth-conf is selected here
#           and SSL is enabled, the transmitted data is encrypted twice.
kerberos_options:
  keytab: resources/dse/conf/dse.keytab
  service_principal: dse/_HOST@REALM
  http_principal: HTTP/_HOST@REALM
  qop: auth

##########################
# LDAP options
#
# These are options are only used when the com.datastax.bdp.cassandra.auth.DseAuthenticator
# is configured as the authenticator in cassandra.yaml and 'ldap' scheme is selected in
# authentication_options and/or role_management_options above.

# ldap_options:
#     # LDAP server address or comma separated list of alternative addresses. Each address may include port after a colon.
#     # If port is not defined per single address, server_port will be used by default. All servers are equivalent and
#     # the one which is chosen for a certain request depends on the current state (recent failures, number of idle
#     # connections is the pool).
#     server_host:
#
#     # The port on which the LDAP server listens, usually port 389 for unencrypted
#     # connections and port 636 for SSL-encrypted connections. If use_tls is set to true, use the
#     # unencrypted port
#     server_port: 389
#
#     # The distinguished name (DN) of an account that is used to search for other users on the
#     # LDAP server. This user should have only the necessary permissions to do the search
#     # If not present then an anonymous bind is used for the search
#     search_dn:
#
#     # Password of the search_dn account
#     search_password:
#
#     # Set to true to use an SSL encrypted connection. In this case the server_port needs
#     # to be set to the LDAP port for the server
#     use_ssl: false
#
#     # Set to true to initiate a TLS encrypted connection on the default ldap port
#     use_tls: false
#
#     truststore_path:
#     truststore_password:
#     truststore_type: jks
#
#     ssl_protocol: TLS
#
#     user_search_base:
#     extra_user_search_bases: []
#     user_search_filter: (uid={0})
#
#     # Set to true to enable hostname verification. This is only effective if one of use_ssl or use_tls is true
#     # and truststore_path has a valid trust store.
#     hostname_verification: false
#
#     # Set to the attribute on the user entry containing group membership information.
#     user_memberof_attribute: memberof
#
#     # The group_search_type defines how group membership will be determined for a user. It can be one of:
#     #     directory_search - each group entry has the attribute whose values refer to that group's members;
#     #                        the search is performed by finding those groups whose member is the given user;
#     #                        for that purpose, group_search_base and group_search_filter are used;
#     #
#     #     memberof_search  - each entry has the attribute (indicated by user_memberof_attribute parameter) whose
#     #                        values refer to the groups that entry is a member of; that is, the membership information
#     #                        is attached directly to the user entry;
#     #
#     # Note that group_search_base and group_search_filter are also used to find the group entry by its name; we need
#     # that to discover group hierarchy - when we have a role, which is mapped to some LDAP group, and we need to find
#     # all roles that role belongs to, we need to find that role first - if that role was a user, we would use
#     # user_search_base and user_search_filter; when it is a group, we use group_search_base and group_search_filter;
#     # Conclusion is that in case of memberof_search, group_search_filter must be able to find a group by that group's
#     # name, and in case of directory_search, group_search_filter must be able to find a group by that group's name or
#     # any of that group's members DN, for example: (|(member={0})(cn={0}))
#     group_search_type: directory_search
#     group_search_base:
#     extra_group_search_bases: []
#     group_search_filter: (uniquemember={0})
#
#     # The attribute in the group entry that holds the group name.
#     group_name_attribute: cn
#
#     # The following options are for optimized searching all parent groups, including inherited ones if the LDAP server
#     # supports such queries. Such a query can retrieve all parent groups with a single request with no need to
#     # traverse group hierarchy level by level;
#     # Similarly to group_search_type, we can use either memberof_search or directory_search, depending on the LDAP
#     # server which is used, or leave it empty to disable this mechanism if LDAP server does not support it;
#     # When we use memberof_search, we only need to define all_parent_groups_memberof_attribute, which is a special attribute
#     # that includes information about all parent groups, including transitive ones, for example Oracle LDAP supports
#     # this mechanism with attribute named isMemberOf
#     # When we use directory_search, we only need to define all_parent_groups_search_filter, which should allow to find all
#     # the groups the role belongs to; for example ActiveDirectory server supports so called
#     # LDAP_MATCHING_RULE_IN_CHAIN search operator which can be defined as (member:1.2.840.113556.1.4.1941:={0})
#     all_parent_groups_search_type:
#     all_parent_groups_memberof_attribute: isMemberOf
#     all_parent_groups_search_filter: (member:1.2.840.113556.1.4.1941:={0})
#
#     # Validity period for the credentials cache in milli-seconds (remote bind is an expensive
#     # operation). Defaults to 0, set to 0 to disable.
#     credentials_validity_in_ms: 0
#
#     # Validity period for the search cache in seconds. Defaults to 0, set to 0 to disable.
#     search_validity_in_seconds: 0
#
#     # Validity period for the groups cache in milliseconds. The groups cache holds ldap group entries
#     # and is used as a source for the roles cache.
#     # Defaults to the search cache validity period (in milliseconds) for backward compatibility.
#     # Set to 0 to disable.
#     groups_validity_in_ms: 0
#
#     # Refresh interval for groups cache (if enabled). After this interval, cache entries become eligible
#     # for refresh. Upon next access, an async reload is scheduled and the old value returned until it completes.
#     # If groups_validity_in_ms is non-zero, then this must be also.
#     # Defaults to the same value as groups_validity_in_ms.
#     groups_update_interval_in_ms: 0
#
#     # Connection pool settings - not that a separate connection pool will be created for each provided server address
#     connection_pool:
#         max_active: 8
#         max_idle: 8
#
#     # DNS service discovery configuration. By default, when it is disabled (it is disabled when fqdn is empty), LDAP
#     # servers list is taken from server_host param. When fqdn is defined, we will try to retrieve LDAP servers from
#     # the SRV records of the provided domain. In this case, the servers provided in server_host are used as a fallback
#     # list of servers, when DNS is not available or returns no SRV records.
#     dns_service_discovery:
#         # fully qualified domain name to get the SRV records from; leave empty to disable DNS service discovery.
#         # example: _ldap._tcp.example.com
#         fqdn:
#
#         # timeout in ms for querying DNS; it has be be between 0 and 1 hour equivalent
#         lookup_timeout_ms: 5000
#
#         # for how long the old results should be retained/cached (value between 0 and 10 days equivalent);
#         # note that it is applicable only if automatic polling is enabled (see polling_interval_ms)
#         retention_duration_ms: 600000
#
#         # how often we should try to refresh the list of obtained servers, leave 0 to disable periodical refreshing
#         # (value between 0 and 10 days equivalent). Note that if automatic polling is enabled, it does not make
#         # sense to set retention_duration_ms to a value lower than the polling_interval_ms as our cached results
#         # would be expired by the time we try to refresh the results.
#         polling_interval_ms: 0

# To ensure that records with TTLs are purged from DSE Search indexes when they expire, DSE
# periodically checks all indexes for expired documents and deletes them. These settings
# control the scheduling and execution of those checks.
ttl_index_rebuild_options:

  # By default, schedule a check every 300 seconds:
  fixed_rate_period: 300

  # The number of seconds to delay the first check to speed up startup time:
  initial_delay: 20

  # All documents determined to be expired are deleted from the index during each check, but
  # to avoid memory pressure, their unique keys are retrieved and deletes issued in batches.
  # This determines the maximum number of documents per batch:
  max_docs_per_batch: 4096

  # Maximum number of search indexes that can execute TTL cleanup concurrently:
  thread_pool_size: 1

# DSE Search resource upload size limit in MB. A value of '0' disables resource uploading.
solr_resource_upload_limit_mb: 10

# Transport options for inter-node communication between DSE Search nodes.
shard_transport_options:
  # The cumulative shard request timeout, in milliseconds, defines the internal timeout for all
  # search queries to prevent long running queries. Default is 60000 (1 minute).
  netty_client_request_timeout: 60000

# ---- DSE Search index encryption options

# solr_encryption_options:
#     # Whether to allocate shared index decryption cache off JVM heap.
#     # Default is off heap allocation (true).
#     decryption_cache_offheap_allocation: true

#     # The maximum size of shared DSE Search decryption cache, in MB.
#     # Default is 256 MB.
#     decryption_cache_size_in_mb: 256

# ---- DSE Search indexing settings

# # The maximum number of queued partitions during search index rebuilding. (This serves primarily
# # as a safeguard against excessive heap usage by the indexing queue.) If set lower than the
# # number of TPC threads, not all TPC threads can be actively indexing.
# #
# # Default: 1024
# back_pressure_threshold_per_core: 1024
#
# # The max time to wait for flushing of index updates during re-index.
# # Flushing should always complete successfully, in order to fully sync search indexes
# # with DSE data. DataStax recommends to always set at a reasonably high value.
# #
# # Default: 5 minutes
# flush_max_time_per_core: 5
#
# # The maximum time to wait for each search index to load on startup and create/reload search index operations.
# # Only change this advanced option if any exceptions happen during search index loading.
# #
# # Default: 5 minutes
# load_max_time_per_core: 5
#
# # Applies the configured Cassandra disk failure policy to index write failures.
# # Default is disabled (false).
# enable_index_disk_failure_policy: false

# # The directory to store search index data. Each DSE Search index is stored under
# # a solrconfig_data_dir/keyspace.table directory.
# # Default is a solr.data directory inside Cassandra data directory, or as specified
# # by the dse.solr.data.dir system property.
# solr_data_dir: /MyDir

# # The Lucene field cache has been deprecated. Instead set docValues="true" on the field
# # in the schema.xml file.  After changing the schema, reload and reindex the search index.
# # Default: false
# solr_field_cache_enabled: false

# # Global Lucene RAM buffer usage thresholds (separate for heap and off-heap) at which DSE will force segment flush.
# # Setting this too low may induce a state of constant flushing during periods of ongoing write activity. For
# # NRT, these forced segment flushes will also de-schedule pending auto-soft commits to avoid potentially
# # flushing too many small segments.
# # Default: 1024
# ram_buffer_heap_space_in_mb: 1024
# # Default: 1024
# ram_buffer_offheap_space_in_mb: 1024

# ---- DSE Search CQL query options

# # Maximum time in milliseconds to wait for all rows
# # to be read from the database during CQL Solr queries.
# # Default is 10000 (10 seconds).
# cql_solr_query_row_timeout: 10000

##########################
# Global performance service options

# # Number of background threads used by the performance service under normal conditions.
# # Defaults to 4.
# performance_core_threads: 4
# # Maximum number of background threads used by the performance service.
# # Defaults to concurrent_writes specified in cassandra.yaml.
# performance_max_threads: 32
#
# # The number of queued tasks in the backlog when the number of performance_max_threads are busy (minimum 0).
# performance_queue_capacity: 32000
#
# # If the performance service requests more tasks than (performance_max_threads + performance_queue_capacity),
# # a dropped task warning will be issued. This warning indicates that collected statistics may not be up to date
# # because the server couldn't keep up under the current load.
#
# # You can disable some services, reconfigure some services, or increase the queue size.

##########################
# Performance service options

graph_events:
  ttl_seconds: 600

# cql_slow_log_options:
#     enabled: true
#
#     #  When t > 1, log queries taking longer than t milliseconds.
#     #      0 <= t <= 1,  log queries above t percentile
#     threshold: 200.0
#
#     # Initial number of queries before percentile filter becomes active
#     minimum_samples: 100
#
#     ttl_seconds: 259200
#
#     # Keeps slow queries in-memory only and doesn't write data to the database.
#     # WARNING - if this is set to 'false' then set threshold >= 2000, otherwise there will be a
#     # high load on the database.
#     skip_writing_to_db: true
#
#     # The number of slow queries to keep in-memory
#     num_slowest_queries: 5

cql_system_info_options:
  enabled: false
  refresh_rate_ms: 10000

resource_level_latency_tracking_options:
  enabled: false
  refresh_rate_ms: 10000

db_summary_stats_options:
  enabled: false
  refresh_rate_ms: 10000

cluster_summary_stats_options:
  enabled: false
  refresh_rate_ms: 10000

spark_cluster_info_options:
  enabled: false
  refresh_rate_ms: 10000

# ---- Spark application stats options
spark_application_info_options:
  enabled: false
  refresh_rate_ms: 10000

  driver:
    # enables or disables writing of the metrics collected at Spark Driver to Cassandra
    sink: false

    # enables or disables Spark Cassandra Connector metrics at Spark Driver
    connectorSource: false

    # enables or disables JVM heap and GC metrics at Spark Driver
    jvmSource: false

    # enables or disables application state metrics
    stateSource: false

  executor:
    # enables or disables writing of the metrics collected at executors to Cassandra
    sink: false

    # enables or disables Spark Cassandra Connector metrics at executors
    connectorSource: false

    # enables or disables JVM heap and GC metrics at executors
    jvmSource: false

# Table Histogram data tables options
histogram_data_options:
  enabled: false
  refresh_rate_ms: 10000
  retention_count: 3

# User/Resource latency tracking settings
user_level_latency_tracking_options:
  enabled: false
  refresh_rate_ms: 10000
  top_stats_limit: 100
  quantiles: false

# ---- DSE Search Performance Objects

solr_slow_sub_query_log_options:
  enabled: false
  ttl_seconds: 604800
  async_writers: 1
  threshold_ms: 3000

solr_update_handler_metrics_options:
  enabled: false
  ttl_seconds: 604800
  refresh_rate_ms: 60000

solr_request_handler_metrics_options:
  enabled: false
  ttl_seconds: 604800
  refresh_rate_ms: 60000

solr_index_stats_options:
  enabled: false
  ttl_seconds: 604800
  refresh_rate_ms: 60000

solr_cache_stats_options:
  enabled: false
  ttl_seconds: 604800
  refresh_rate_ms: 60000

solr_latency_snapshot_options:
  enabled: false
  ttl_seconds: 604800
  refresh_rate_ms: 60000

# Node health is a score-based representation of how fit a node is to handle queries. The score is a
# function of how long a node has been up and the rate of dropped mutations in the recent past.
node_health_options:
  refresh_rate_ms: 60000
  # The amount of continuous uptime required for the node to reach the maximum uptime score. If you
  # are concerned with consistency during repair after a period of downtime, you may want to
  # temporarily increase this time to the expected time it will take to complete repair.
  #
  # Default - 10800 seconds (3 hours)
  uptime_ramp_up_period_seconds: 10800
  # The time window in the past over which the rate of dropped mutations affects the node health score.
  # Default - 30 minutes
  dropped_mutation_window_minutes: 30

# If enabled (true), replica selection for distributed DSE Search queries takes node health into account
# when multiple candidates exist for a particular token range. Set to false to ignore
# node health when choosing replicas.
#
# Health-based routing allows us to make a trade-off between index consistency and query throughput. If
# the primary concern is query performance, it may make sense to set this to "false".
#
# Default is enabled (true).
enable_health_based_routing: true

# If enabled (true), DSE Search reindexing of bootstrapped data will happen asynchronously, and the node will join the ring straight
# after bootstrap.
#
# Default is disabled (false). The node will wait for reindexing of bootstrapped data to finish before joining the ring.
async_bootstrap_reindex: false

# Lease metrics. Enable these metrics to help monitor the performance of the lease subsystem.
# ttl_seconds controls how long the log of lease holder changes persists.
lease_metrics_options:
  enabled: false
  ttl_seconds: 604800

# The directory where system keys are kept.
#
# Keys used for SSTable encryption must be distributed to all nodes.
# DSE must be able to read and write to the directory.
#
# This directory should have 700 permissions and belong to the dse user.
system_key_directory: /etc/dse/conf

# If this is set to true, DSE requires the following config values to be encrypted:
#     resources/cassandra/conf/cassandra.yaml:
#         server_encryption_options.keystore_password
#         server_encryption_options.truststore_password
#         client_encryption_options.keystore_password
#         client_encryption_options.truststore_password
#    resources/dse/conf/dse.yaml:
#         ldap_options.search_password
#         ldap_options.truststore_password
#
# It's an error if the passwords aren't encrypted.
# Config values can be encrypted with "dsetool encryptconfigvalue"
config_encryption_active: false

# The name of the system key used to encrypt / decrypt passwords stored
# in configuration files.
#
# If config_encryption_active is true, it's an error if a valid key with
# this name isn't in the system key directory keyfiles, and KMIP managed
# keys can be created with "dsetool createsystemkey"
config_encryption_key_name: system_key

##########################
# Spark-related settings

# The length of a shared secret used to authenticate Spark components and encrypt the connections between them.
# Note that this is not the strength of the cipher used for encrypting connections.
spark_shared_secret_bit_length: 256

# Enables Spark security based on shared secret infrastructure. Enables mutual authentication between Spark master
# and worker nodes. If DSE authentication is enabled, spark security is forced to be enabled and this parameter is ignored.
spark_security_enabled: false

# Enables encryption between Spark master and worker nodes, except Web UI. The connection uses the
# Digest-MD5 SASL-based encryption mechanism. This option applies only if spark_security_enabled is true.
# If DSE authentication is enabled, spark security encryption is forced to be enabled and this parameter is ignored.
spark_security_encryption_enabled: false

# # How often Spark plugin should check for Spark Master / Spark Worker readiness to start. The value is
# # a time (in ms) between subsequent retries.
# spark_daemon_readiness_assertion_interval: 1000

#
# Legacy Resource Manager options
#
# Controls the physical resources that can be used by Spark applications on this node.
# cores_total is the number of cores and and memory_total is total system memory that you can assign to all executors
# that are run by the work pools on this node. The values can be absolute (exact number of cores) or the
# memory size (use metric suffixes like M for mega, and G for giga) or a fraction of physical cores reported by the OS,
# and fraction of available memory, where available memory is calculated as: total physical memory - DSE max heap size.
# cores_total and memory_total replace initial_spark_worker_resources option which was used in earlier DSE versions.
# The default 0.7 for cores and memory corresponds to the default value of initial_spark_worker_resources 0.7.
# DSE does not support setting Spark Worker cores and memory through environment variables SPARK_WORKER_CORES
# and SPARK_WORKER_MEMORY. cores_total and memory_total can be set from environment variables SPARK_WORKER_TOTAL_CORES and
# SPARK_WORKER_TOTAL_MEMORY
# resource_manager_options:
#     worker_options:
#         cores_total: 0.7
#         memory_total: 0.6
#
#         workpools:
#             - name: alwayson_sql
#               cores: 0.25
#               memory: 0.25

# In DSE 5.1 and later: Communication between Spark applications and the resource manager are routed through
# the CQL native protocol. Enabling client encryption in cassandra.yaml will also enable encryption for
# the communication with the DSE Spark Master. To secure the communication between Spark Driver and Spark Executors,
# enable Spark authentication and encryption for that application.
# In contrast, mutual authentication and encryption of communication between DSE Spark Master and Workers are
# managed by spark_security_enabled and spark_security_encryption_enabled in dse.yaml.

# Spark UI options apply to Spark Master and Spark Worker UIs and to Spark daemon UIs in general. Spark UI options do NOT
# apply to user applications even if they run in cluster mode.
spark_ui_options:
  # Valid values are:
  # inherit - SSL settings are inherited from DSE client encryption options
  # custom - SSL settings from encryption_options below
  encryption: inherit

  encryption_options:
    enabled: false
    keystore: resources/dse/conf/.ui-keystore
    keystore_password: cassandra
    # require_client_auth: false
    # Set truststore and truststore_password if require_client_auth is true
    # truststore: resources/dse/conf/.ui-truststore
    # truststore_password: cassandra
    # More advanced defaults:
    # protocol: TLS
    # algorithm: SunX509
    #
    # Set keystore_type for keystore, valid types can be JKS, JCEKS, PKCS12 or PKCS11
    # for file based keystores prefer PKCS12
    # keystore_type: JKS
    #
    # Set truststore_type for truststore, valid types can be JKS, JCEKS or PKCS12
    # for file based truststores prefer PKCS12
    # truststore_type: JKS
    #
    # cipher_suites: [TLS_RSA_WITH_AES_128_CBC_SHA,TLS_RSA_WITH_AES_256_CBC_SHA,TLS_DHE_RSA_WITH_AES_128_CBC_SHA,TLS_DHE_RSA_WITH_AES_256_CBC_SHA,TLS_ECDHE_RSA_WITH_AES_128_CBC_SHA,TLS_ECDHE_RSA_WITH_AES_256_CBC_SHA]

# Configure how the driver and executor processes are created and managed.
spark_process_runner:
  # Valid options are: default, run_as
  runner_type: default

  # DSE uses sudo to run Spark application components (drivers and executors) as specific OS users.
  # A set of predefined users, called slot users, is used for this purpose. All drivers and executors
  # owned by some DSE user are run as some slot user x. Drivers and executors of any other DSE user
  # use different slots.
  # Setting up slots:
  # 1. Create n users (n = number of slots), call them slot1, slot2, ..., slotn, with no login. Each user
  #    should have primary group the same as its name, so for example slot1:slot1, slot2:slot2, ...
  # 2. Add DSE service user (the user who runs DSE server) to the slot user groups; the DSE service user must be
  #    in all slot user groups.
  # 3. Modify the sudoers files so that:
  #    a) DSE service user can execute any command as any slot user without providing a password
  #    b) umask is overridden to 007 for those commands so that files created by sub-processes will not be accessible
  #       by anyone by default,
  #    For example, if we have two slot users slot1, slot2, and DSE service user dse, add these slot users to sudoers:
  #    Runas_Alias     SLOTS = slot1, slot2
  #    Defaults>SLOTS  umask=007
  #    Defaults>SLOTS  umask_override
  #    dse             ALL=(SLOTS) NOPASSWD: ALL
  run_as_runner_options:
    user_slots:
      - slot1
      - slot2

# AlwaysOn SQL options have dependence on workpool setting of resource_manager_options. Set workpool configuration if you
# enable alwayson_sql_options.
# alwayson_sql_options:
#     # Set to true to enable the node for AlwaysOn SQL. Only an Analytics node
#     # can be enabled as an AlwaysOn SQL node.
#     enabled: false
#
#     # AlwaysOn SQL Thrift port
#     thrift_port: 10000
#
#     # AlwaysOn SQL WebUI port
#     web_ui_port: 9077
#
#     # The waiting time to reserve the Thrift port if it's not available
#     reserve_port_wait_time_ms: 100
#
#     # The waiting time to check AlwaysOn SQL health status
#     alwayson_sql_status_check_wait_time_ms: 500
#
#     # The work pool name used by AlwaysOn SQL
#     workpool: alwayson_sql
#
#     # Location in DSEFS of the log files
#     log_dsefs_dir: /spark/log/alwayson_sql
#
#     # The role to use for internal communication by AlwaysOn SQL if authentication is enabled
#     auth_user: alwayson_sql
#
#     # The maximum number of errors that can occur during AlwaysOn SQL service runner thread
#     # runs before stopping the service. A service stop requires a manual restart.
#     runner_max_errors: 10
#
#     # The interval in seconds to update heartbeat of AlwaysOn SQL. If heartbeat is not updated
#     # for more than the period of three times of the interval, AlwaysOn SQL malfunctions.
#     # AlwaysOn SQL automatically restarts.
#     heartbeat_update_interval_seconds: 30

##########################
# DSE File System (DSEFS) options
# dsefs_options:
#
#     # Whether to enable DSEFS on this node.
#     # If not set, DSEFS is enabled only on the nodes that run a Spark workload.
#     enabled: true
#
#     # The keyspace where the DSEFS metadata is stored. Optionally configure multiple DSEFS file systems
#     # within a cluster by specifying a different keyspace name for each datacenter.
#     keyspace_name: dsefs
#
#     # The local directory for storing the local node metadata, including the node identifier.
#     # The amount of data stored is nominal, and does not require configuration for throughput, latency, or capacity.
#     # This directory must not be shared by DSEFS nodes.
#     work_dir: /var/lib/dsefs
#
#     # The public port on which DSEFS listens for clients. The service on this port is bound to
#     # native_transport address.
#     public_port: 5598
#
#     # Port for inter-node communication, must be not visible from outside of the cluster.
#     # It is bound to listen address. Do not open this port to firewalls.
#     private_port: 5599
#
#     # Mandatory attribute to identify the set of directories. DataStax recommends segregating these data directories
#     # on physical devices that are different from the devices that are used for the DSE database.
#     # Using multiple directories on JBOD improves performance and capacity.
#     data_directories:
#         - dir: /var/lib/dsefs/data
#
#           # The weighting factor for this location specifies how much data to place in this directory, relative to
#           # other directories in the cluster. This soft constraint determines how DSEFS distributes the data.
#           storage_weight: 1.0
#
#           # Reserved space (in bytes) that is not going to be used for storing blocks
#           min_free_space: 268435456
#
#     # More advanced settings:
#
#     # Wait time before the DSEFS server times out while waiting for services to bootstrap.
#     service_startup_timeout_ms: 600000
#
#     # Wait time before the DSEFS server times out while waiting for services to close.
#     service_close_timeout_ms: 600000
#
#     # Wait time that the DSEFS server waits during shutdown before closing all pending connections.
#     server_close_timeout_ms: 2147483647 # Integer.MAX_VALUE
#
#     # The maximum accepted size of a compression frame defined during file upload.
#     compression_frame_max_size: 1048576
#
#     # Maximum number of elements in a single DSEFS Server query cache. DSEFS reuses this value for every cache that
#     # stores database query results.
#     query_cache_size: 2048
#
#     # The time to retain the DSEFS Server query cache element in cache. The cache element expires
#     # when this time is exceeded.
#     query_cache_expire_after_ms: 2000
#
#     internode_authentication:
#         # If enabled, the servers are obliged to authenticate all messages passed between them on private_port.
#         # The authentication protocol is based on HMAC used with a pre-shared secret available only to DSE cluster
#         # members (nodes).
#         # The actual key is never passed between the nodes.
#         # Typically there is no need to turn this authentication off and it doesn't incur any performance overhead.
#         # Disabling internode authentication is not recommended, but may be used for debugging purposes
#         # to issue internode requests manually with curl.
#         # Limitations:
#         # Beware that enabling internode authentication does not encrypt the internode traffic.
#         # Only HTTP headers are protected with HMAC, so MITM attacks are still possible on the message data.
#         # It is also possible to bypass the authentication if the DSE messaging subsystem was not
#         # properly secured and the attacker could fake being a part of the DSE cluster in order to obtain
#         # the secret key. If you need stronger security, please configure SSL.
#         enabled: true
#
#         # Algorithm used for key encryption:
#         algorithm: HmacSHA256
#
#     gossip_options:
#         # The delay between gossip rounds
#         round_delay_ms: 2000
#
#         # How long to wait after registering the Location and reading back all other Locations from the database
#         startup_delay_ms: 5000
#
#         # How long to wait after announcing shutdown before shutting down the node
#         shutdown_delay_ms: 10000
#
#     rest_options:
#         # How long RestClient is going to wait for a response corresponding to a given request
#         request_timeout_ms: 330000
#
#         # How long RestClient is going to wait for establishing a new connection
#         connection_open_timeout_ms: 10000
#
#         # How long RestClient is going to wait until all pending transfers are complete before closing
#         client_close_timeout_ms: 60000
#
#         # How long to wait for the server rest call to complete
#         server_request_timeout_ms: 300000
#
#         # Wait time, in milliseconds, before closing idle RestClient - server connection. 0 if disabled.
#         # If RestClient does not close connection after this timeout, the server closes the connection after
#         # 2 * idle_connection_timeout_ms milliseconds.
#         idle_connection_timeout_ms: 60000
#
#         # Wait time, in milliseconds, before closing idle internode connection. The internode connections are
#         # mainly used to exchange data during replication. Do not set lower than the default value for heavily
#         # utilized DSEFS clusters.
#         internode_idle_connection_timeout_ms: 120000
#
#         # Maximum number of connections to a given host per single CPU core. DSEFS keeps a connection pool for
#         # each CPU core.
#         core_max_concurrent_connections_per_host: 8
#
#     transaction_options:
#         # How long to allow a transaction to run before considering it for timing out and rollback
#         transaction_timeout_ms: 60000
#
#         # How long to wait before retrying a transaction aborted due to a conflict
#         conflict_retry_delay_ms: 10
#
#         # How many times the transaction is retried in case of a conflict before giving up
#         conflict_retry_count: 40
#
#         # How long to wait before retrying a failed transaction payload execution
#         execution_retry_delay_ms: 1000
#
#         # How many times to retry executing the payload before signaling the error to the application
#         execution_retry_count: 3
#
#     block_allocator_options:
#         # The overflow_margin_mb and overflow_factor options control how much additional data can be placed
#         # on the local (coordinator) before the local node overflows to the other nodes.
#         # A local node is preferred for a new block allocation, if
#         # used_size_on_the_local_node < average_used_size_per_node * overflow_factor + overflow_margin.
#         # The trade-off is between data locality of writes and balancing the cluster.
#         # To disable the preference for allocating blocks on the coordinator node, set these values to 0 MB and 1.0.
#         overflow_margin_mb: 1024
#         overflow_factor: 1.05

# Insightful Monitoring(Insights) Options
# enable insights_options.
# insights_options:
#     # Directory to store insights
#     data_dir: /var/lib/cassandra/insights_data
#
#     # Directory to store insight logs
#     log_dir: /var/log/cassandra/

##########################
# Audit logging options
audit_logging_options:
  enabled: false

  # The logger used for logging audit information
  # Available loggers are:
  #   CassandraAuditWriter - logs audit info to a cassandra table. This logger can be run synchronously or
  #                          asynchronously. Audit logs are stored in the dse_audit.audit_log table.
  #                          When run synchronously, a query will not execute until it has been written
  #                          to the audit log table successfully. If a failure occurs before an audit event is
  #                          written, and it's query is executed, the audit logs might contain queries that were never
  #                          executed.
  #   SLF4JAuditWriter -     logs audit info to an SLF4J logger. The logger name is `SLF4JAuditWriter`,
  #                          and can be configured in the logback.xml file.
  logger: SLF4JAuditWriter

  #     # Comma-separated list of audit event categories to be included or excluded from the audit log.
  #     # When not set, the default includes all categories.
  #     # Categories are: QUERY, DML, DDL, DCL, AUTH, ADMIN, ERROR.
  #     # Specify either included or excluded categories. Specifying both is an error.
  #     included_categories:
  #     excluded_categories:

  #     # Comma-separated list of keyspaces to be included or excluded from the audit log.
  #     # When not set, the default includes all keyspaces.
  #     # Specify either included or excluded keyspaces. Specifying both is an error.
  #     included_keyspaces:
  #     excluded_keyspaces:

  #     # Comma separated list of the roles to be audited or not.
  #     # Specify either included or excluded roles. Specifying both is an error
  #     included_roles:
  #     excluded_roles:

  # The amount of time, in hours, audit events are retained by supporting loggers.
  # Only the CassandraAuditWriter supports retention time.
  # Values of 0 or less retain events forever.
  retention_time: 0

  #     # Whether to render bound variables as CQL literals
  #     # - when disabled, primitive values are printed as they are, collections, tuples and UDT values are
  #     #   printed as hexadecimal sequences
  #     # - when enabled (default since 6.8.2), all values are printed in the form as they would be provided in a CQL statement, including
  #     #   quotation marks and escaping (in this case, the content of collections, tuples and UDT values is in human
  #     #   readable form)
  #     render_cql_literals: true

  cassandra_audit_writer_options:
    # Sets the mode the audit writer runs in.
    #
    # When run synchronously, a query is not executed until the audit event is successfully written.
    #
    # When run asynchronously, audit events are queued for writing to the audit table, but are
    # not necessarily logged before the query executes. A pool of writer threads consumes the
    # audit events from the queue, and writes them to the audit table in batch queries. While
    # this substantially improves performance under load, if there is a failure between when
    # a query is executed, and it's audit event is written to the table, the audit table may
    # be missing entries for queries that were executed.
    # valid options are 'sync' and 'async'
    mode: sync

    # The maximum number of events the writer will dequeue before writing them out to the table.
    # If you're seeing warnings in your logs about batches being too large, decrease this value.
    # Increasing guardrails.batch_size_warn_threshold_in_kb in cassandra.yaml is also an option, but make sure you understand
    # the implications before doing so.
    #
    # Only used in async mode. Must be >0
    batch_size: 50

    # The maximum amount of time in milliseconds an event will be dequeued by a writer before being written out. This
    # prevents events from waiting too long before being written to the table when there's not a lot of queries happening.
    #
    # Only used in async mode. Must be >0
    flush_time: 250

    # The size of the queue feeding the asynchronous audit log writer threads. When there are more events being
    # produced than the writers can write out, the queue will fill up, and newer queries will block until there
    # is space on the queue.
    # If a value of 0 is used, the queue size will be unbounded, which can lead to resource exhaustion under
    # heavy query load.
    queue_size: 30000

    # the consistency level used to write audit events
    write_consistency: QUORUM

#         # Where dropped events are logged
#         dropped_event_log: /var/log/cassandra/dropped_audit_events.log

#         # Partition days into hours by default
#         day_partition_millis: 3600000

##########################
# System information encryption settings
#
# If enabled, system tables that might contain sensitive information (system.batchlog,
# system.paxos), hints files, and Cassandra commit logs are encrypted with these
# encryption settings.
#
# If DSE Search index encryption is enabled, DSE Search index files are also encrypted with these settings.
# If backing C* table encryption is enabled, DSE Search commit log is encrypted with these settings.
#
# When enabling system table encryption on a node with existing data, run
# `nodetool upgradesstables -a` on the listed tables to encrypt existing data.
#
# When tracing is enabled, sensitive information is written to the tables in the
# system_traces keyspace. Configure encryption on the tables to encrypt their data
# on disk by using an encrypting compressor.
#
# DataStax recommends using remote encryption keys from a KMIP server when using Transparent Data Encryption (TDE) features.
# Local key support is provided when a KMIP server is not available.
system_info_encryption:
  enabled: false
  cipher_algorithm: AES
  secret_key_strength: 128
  chunk_length_kb: 64

#     # The encryptor will use a KMIP key server to manage its encryption keys. Specify only to use a KMIP key server,
#     # otherwise omit this entry. The default is to use local key encryption.
#     key_provider: KmipKeyProviderFactory

#     # If KmipKeyProviderFactory is used for system_info_encryption, this specifies the kmip host to be used.
#     kmip_host: kmip_host_name

##########################
# KMIP hosts options
#
# Connection settings for key servers supporting the KMIP protocol
# allow DSE encryption features to use encryption and decryption keys that are not stored
# on the same machine running DSE.
#
# Hosts are configured as <kmip_host_name>: {connection_settings}, which maps a user-defined
# name to a set of KMIP hosts and KMIP-defined credentials (keystores and truststores) that are used with a particular
# key server. This name is then used when referring to KMIP hosts. DSE supports multiple KMIP hosts.

# kmip_hosts:
#     # The unique name of this KMIP host/cluster which is specified in the table schema.
#     host.yourdomain.com:
#
#         # Comma-separated list of KMIP hosts host[:port]
#         # The current implementation of KMIP connection management supports only failover, so all requests will
#         # go through a single KMIP server. There is no load balancing. This is because there aren't many known KMIP servers
#         # that support read replication, or other strategies for availability.
#         #
#         # Hosts are tried in the order they appear, so add KMIP hosts in the intended failover sequence.
#         hosts: kmip1.yourdomain.com, kmip2.yourdomain.com
#
#         # keystore/truststore info
#         keystore_path: /path/to/keystore.jks
#         keystore_type: jks
#         keystore_password: password
#
#         truststore_path: /path/to/truststore.jks,
#         truststore_type: jks
#         truststore_password: password
#
#         # Keys read from the KMIP hosts are cached locally for the period of time specified below.
#         # The longer keys are cached, the fewer requests are made to the key server, but the longer
#         # it takes for changes (ie: revocation) to propagate to the DSE node.
#         key_cache_millis: 300000
#
#         # Refresh interval for the KMIP host key cache. After this interval, cache entries become eligible
#         # for refresh. Upon next access, an async reload is scheduled and the old value returned until it completes.
#         # If key_cache_millis is non-zero, then this must be also.
#         # Defaults to the same value as key_cache_millis.
#         key_cache_update_millis: 300000
#
#         # Socket timeout in milliseconds.
#         timeout: 1000

# # driver - DSE Search will use Solr cursor paging (deep paging) when pagination is enabled by the CQL driver.
# #
# # off - DSE Search will ignore the driver's pagination settings and use normal Solr paging unless:
# #   - The current workload is an analytics workload (ex. SearchAnalytics).
# #   - The query parameter 'paging' is set to