Configuration Index

This page lists the default configuration shipped with SyncTV. Effective values combine config files, environment variables, CLI overrides, path resolution, and timezone fallback.

See Full Configuration Example for YAML templates. Run synctv config show --output yaml to inspect the effective configuration for the current binary with secrets redacted.

OAuth2 provider instances are Runtime Settings. Remote media provider instances are persisted through the management API/CLI and are not static config-file fields.

Configuration Decisions

Most fields can keep their defaults. Production launch should prioritize secrets, database, entrypoints, security boundaries, and backups.

Setting	When to change it	Common mistake	After change	Production guidance
jwt.secret	Any production deployment	Keeping change-me-in-production or committing it	Restart required; old tokens may become invalid	Inject through Secret Manager or secret file
security.opaque_server_setup_secret	Local password/OPAQUE login	Regenerating it on every deployment	Restart required; casual changes can break password records	Generate once and back it up
security.credential_encryption_key	Storing encrypted provider credentials	Not 64 hex chars, or losing the key	Restart required; losing it breaks encrypted credential reads	Use openssl rand -hex 32 and back it up
security.email_outbox_encryption_key	Every production deployment	Different replicas use different keys	Restart required; queued email depends on this key	Generate once and back it up with the database
database.url / split fields	Production PostgreSQL	Pointing at local dev DB or oversizing the pool	Restart required	PostgreSQL must be persistent and backed up
redis.url / split fields	Production, OAuth2, multi-replica, rate limits, L2 cache	Different replicas use different Redis or key prefixes collide	Restart required	Recommended in production, required for multi-replica mode
server.cors_allowed_origins	Web frontend and API use different origins	Using * or adding a path	Restart required	Use exact origins such as https://app.example.com
server.trusted_proxies	A reverse proxy or Ingress sits in front	Trusting the whole internet	Restart required	Trust only proxy networks you control
management.*	CLI or operations control plane is needed	Exposing TCP management publicly without a token	Restart required	Prefer Unix socket; TCP must be authenticated
metrics.*	Prometheus or alerting is used	Exposing /metrics publicly	Restart required	Use auth and network isolation
webauthn.*	Passkeys are enabled	rp_origin does not match the real origin	Restart required	Use only production HTTPS origins
email.*	Email verification, password reset, email MFA, notifications	TLS/port/from address mismatch	Restart required	Verify with synctv settings test-email
cluster.* / cluster.secret	Multi-replica cluster	Cluster enabled without Redis or cluster secret	Restart required	All replicas share PostgreSQL, Redis, and secrets
livestream.*	RTMP/FLV/HLS livestreaming	HLS storage model differs between replicas	Restart required	Use shared filesystem or S3 for high traffic
proxy_slice_cache.*	Proxy Range slice cache is needed	Expecting full-file cache	Restart required	Mount enough storage; only Range slices are cached

Run these checks after changing configuration:

synctv config validate
synctv config show --output yaml
curl -fsS http://localhost:8081/health/ready

Single-node Production Focus

Production Compose uses .env.postgres, .env.redis, and .env.synctv; it does not require a hand-written full YAML file. Before launch, confirm:

Goal	Compose/env focus	Config field
Initial administrator	SYNCTV_BOOTSTRAP_ROOT_PASSWORD	bootstrap.root_password
Token signing	SYNCTV_JWT_SECRET	jwt.secret
OPAQUE password login	SYNCTV_SECURITY_OPAQUE_SERVER_SETUP_SECRET	security.opaque_server_setup_secret
Provider credential encryption	SYNCTV_SECURITY_CREDENTIAL_ENCRYPTION_KEY	security.credential_encryption_key
Email queue encryption	SYNCTV_SECURITY_EMAIL_OUTBOX_ENCRYPTION_KEY	security.email_outbox_encryption_key
Web frontend CORS	SYNCTV_SERVER_CORS_ALLOWED_ORIGINS='["https://app.example.com"]'	server.cors_allowed_origins
PostgreSQL	.env.postgres and Compose volume	database.*
Redis	.env.redis, Compose Redis service, and key prefix	redis.*
Runtime files	Compose /data volume	data_dir

Use the field index below for bare-metal, Kubernetes, secret files, or custom YAML.

Top-Level Configuration

Field	Type	Default	Details
server	object	see below	Server Listener and Runtime Paths
time	object	see below	Server Listener and Runtime Paths
public_ids	object	see below	Public IDs
security	object	see below	Security and Secrets
data_dir	string path	platform-specific	Server Listener and Runtime Paths
metrics	object	see below	Metrics Monitoring
management	object	see below	Server Listener and Runtime Paths
database	object	see below	Database and Redis
redis	object	see below	Database and Redis
jwt	object	see below	Security and Secrets
logging	object	see below	Server Listener and Runtime Paths
livestream	object	see below	Livestream Configuration
chat	object	see below	chat field table on this page
webauthn	object	see below	WebAuthn and Passkeys
email	object	see below	Email and OAuth2
media_providers	object	see below	Media Providers
webrtc	object	see below	WebRTC Configuration
connection_limits	object	see below	Rate Limits and Connection Limits
bootstrap	object	see below	Bootstrap Root User
cluster	object	see below	Cluster Configuration
password_complexity	object	see below	Security and Secrets
buffer_sizes	object	see below	Internal Buffers
cache	object	see below	Business Cache
proxy_slice_cache	object	see below	Proxy Slice Cache
messaging_rate_limits	object	see below	Rate Limits and Connection Limits
request_rate_limits	object	see below	Rate Limits and Connection Limits
request_rate_limits	object	see below	Rate Limits and Connection Limits

server

Field	Type / values	Default	Meaning
server.host	IP/listen address	0.0.0.0	Main API listen address
server.port	port	8080	HTTP REST and public gRPC listen port
server.enable_reflection	boolean	false	Enable public gRPC reflection
server.trusted_proxies	string array	[]	Trusted reverse proxy IPs/CIDRs
server.cors_allowed_origins	string array	[]	Frontend origins allowed by startup CORS
server.advertise_host	string	""	Address other nodes use to reach this node
server.shutdown_drain_timeout_seconds	seconds	30	Graceful shutdown drain timeout
server.grpc_max_message_size_bytes	bytes	16777216	Maximum gRPC message size
server.grpc_compression_enabled	boolean	true	Allow gRPC gzip compression negotiation

health

Field	Type / values	Default	Meaning
health.enabled	boolean	true	Enable the dedicated health listener
health.host	IP/listen address	127.0.0.1	Health listener address
health.port	port	8081	Health listener port for liveness and readiness endpoints

time, public_ids, and security

Field	Type / values	Default	Meaning
time.timezone	IANA timezone	""	Timezone for human-readable formatting and local datetime parsing
public_ids.sqids	object or null	null	Enable sqids public IDs when configured
public_ids.sqids.alphabet	string or null	null	sqids alphabet
public_ids.sqids.min_length	integer	12	Minimum sqids length
security.credential_encryption_key	64-character hex secret	""	Provider credential encryption key
security.totp_encryption_key	64-character hex secret	""	Dedicated TOTP encryption key; required
security.email_outbox_encryption_key	64-character hex secret	""	Dedicated email outbox encryption key; required in production
security.opaque_server_setup_secret	secret string	""	Stable OPAQUE server setup secret
security.proxy_signing_key	at least 32 characters	""	Dedicated media proxy URL signing key; required
security.media_swarm_signing_key	at least 32 characters	""	Dedicated WebRTC media swarm announcement signing key; required
security.provider_session_encryption_key	at least 32 characters	""	Dedicated provider login session key; required
security.login_discovery_key	at least 32 characters	""	Login discovery enumeration-protection key; required
security.webauthn_enumeration_key	at least 32 characters	""	WebAuthn enumeration-protection key; required
security.ssrf.enabled	boolean	false	Enable global SSRF protection for server-side outbound requests
security.ssrf.allow_private_network_targets	boolean	false	Allow server-side outbound requests to private/loopback/link-local/reserved/metadata targets
security.ssrf.allowed_hosts	string array	[]	Additional hostnames allowed by the global SSRF policy
security.ssrf.allowed_ip_ranges	IP/CIDR string array	[]	Additional IPs or CIDR ranges allowed by the global SSRF policy

metrics

Field	Type / values	Default	Meaning
metrics.enabled	boolean	false	Start the dedicated metrics listener
metrics.host	IP/listen address	0.0.0.0	Metrics listen address
metrics.port	port	9090	Metrics listen port
metrics.tls.enabled	boolean	false	Enable TLS on the metrics listener
metrics.tls.cert_path	file path	""	Metrics TLS certificate
metrics.tls.key_path	file path	""	Metrics TLS private key
metrics.auth.mode	bearer_token, basic, or kubernetes	bearer_token	Metrics authentication mode
metrics.auth.bearer_token	secret string	""	Bearer token for bearer-token mode
metrics.auth.basic_username	string	""	Basic-auth username
metrics.auth.basic_password	secret string	""	Basic-auth password
metrics.auth.kubernetes.audience	string	""	Kubernetes TokenReview audience
metrics.auth.kubernetes.authentication_cache_ttl_seconds	seconds	60	TokenReview result cache TTL
metrics.auth.kubernetes.authorization_cache_ttl_seconds	seconds	60	SubjectAccessReview result cache TTL

management

Field	Type / values	Default	Meaning
management.enabled	boolean	true	Enable the management endpoint
management.transport	unix or tcp	unix on Unix, tcp elsewhere	Management endpoint transport
management.port	port	50052	TCP management port
management.unix_socket_path	file path	platform-specific	Unix socket path
management.auth_token	secret string	""	TCP management bearer token
management.enable_reflection	boolean	false	Management gRPC reflection

database

Field	Type / values	Default	Meaning
database.url	PostgreSQL URL	postgresql://synctv:synctv@localhost:5432/synctv	Complete database connection URL
database.host	string	""	Database host for split configuration
database.port	port	0	Database port for split configuration
database.username	string	""	Database username
database.password	secret string	""	Database password
database.name	string	""	Database name
database.max_connections	integer	20	Maximum pool connections
database.min_connections	integer	5	Minimum idle pool connections
database.connect_timeout_seconds	seconds	10	Connection timeout
database.idle_timeout_seconds	seconds	600	Idle connection timeout
database.max_lifetime_seconds	seconds	1800	Maximum connection lifetime

redis

Field	Type / values	Default	Meaning
redis.url	Redis URL	""	Complete Redis connection URL
redis.host	string	""	Redis host for split configuration
redis.port	port	0	Redis port for split configuration
redis.username	string	""	Redis username
redis.password	secret string	""	Redis password
redis.database	integer	0	Redis database index
redis.connect_timeout_seconds	seconds	5	Redis connection timeout
redis.response_timeout_seconds	seconds	5	Redis command response timeout
redis.pipeline_buffer_size	integer	512	Redis connection manager pipeline buffer size
redis.key_prefix	string	synctv:	Redis key prefix
redis.deployment_mode	standalone or sentinel	standalone	Redis deployment mode
redis.sentinel_master_name	string or null	null	Sentinel master name
redis.sentinel_addresses	string array	[]	Sentinel node addresses

jwt and logging

Top-level logging receives logs from the database, Redis, caches, proxies, realtime, outbox, migrations, startup, third-party dependencies, and all remaining targets. <component> is one of server, health, metrics, management, cluster, livestream, or webrtc; each component has an independent output.

Field	Type / values	Default	Meaning
jwt.secret	secret string	change-me-in-production	Token signing secret
jwt.access_token_duration_hours	hours	1	Access-token lifetime
jwt.refresh_token_duration_days	days	30	Refresh-token lifetime
jwt.guest_token_duration_hours	hours	4	Guest-token lifetime
jwt.clock_skew_leeway_secs	seconds	60	Token timestamp skew allowance
logging.level	error, warn, info, debug, or trace	info	Global log level
logging.format	text or json	text	Global log format
logging.output	stdout, stderr, or a file object	stdout	Global log output
logging.output.path	file path	""	Global log file path; relative paths resolve under data_dir
logging.output.rotation.strategy	daily, hourly, or never	daily	Global log file rotation period
logging.output.rotation.max_files	positive integer	30	Global log files retained
logging.color	auto, always, or never	auto	Global color output
<component>.logging.level	error, warn, info, debug, or trace	info (warn for metrics/cluster)	Component log level
<component>.logging.format	text or json	text	Component log format
<component>.logging.output	stdout, stderr, or a file object	stdout	Independent component output
<component>.logging.output.path	file path	""	Component log file path; relative paths resolve under data_dir
<component>.logging.output.rotation.strategy	daily, hourly, or never	daily	File rotation period
<component>.logging.output.rotation.max_files	positive integer	30	Files retained per component
<component>.logging.color	auto, always, or never	auto	Component color output

livestream

Field	Type / values	Default	Meaning
livestream.rtmp_port	port	1935	RTMP ingest port
livestream.public_rtmp_host	string	""	Public RTMP host returned to publishers
livestream.public_webrtc_base_url	HTTP(S) origin or empty	""	Public WHIP origin in publish-key responses; empty returns a relative URL
livestream.gop_cache_size	integer	2	Number of GOPs cached per stream
livestream.stream_timeout_seconds	seconds	300	Pull-stream idle timeout
livestream.cleanup_check_interval_seconds	seconds	60	Cleanup check interval
livestream.pull_max_retries	integer	10	Maximum pull-stream retries
livestream.pull_initial_backoff_ms	milliseconds	1000	Initial pull retry backoff
livestream.pull_max_backoff_ms	milliseconds	30000	Maximum pull retry backoff
livestream.max_flv_tag_size_bytes	bytes	10485760	Maximum accepted FLV tag size
livestream.gop_cache_max_memory_mb	MB	100	GOP cache memory limit
livestream.hls_storage.type	memory, file, shared_file, or s3	memory	HLS segment storage backend
livestream.hls_storage.memory_max_mb	MB	0	HLS storage limit for memory; 0 uses the built-in default
livestream.hls_storage.path	path	""	HLS file path for the file / shared_file backend
livestream.hls_storage.endpoint	string	""	S3-compatible endpoint for the s3 backend; required when s3 is enabled
livestream.hls_storage.access_key_id	secret string	""	Access key ID for the s3 backend; required when s3 is enabled
livestream.hls_storage.secret_access_key	secret string	""	Secret access key for the s3 backend; required when s3 is enabled
livestream.hls_storage.bucket	string	""	Bucket for the s3 backend; required when s3 is enabled
livestream.hls_storage.region	string or null	null	Region for the s3 backend
livestream.hls_storage.base_path	string	hls/	Object key prefix inside the bucket
livestream.flv_max_connection_duration_seconds	seconds	86400	Maximum HTTP-FLV connection duration
livestream.flv_write_timeout_seconds	seconds	30	HTTP-FLV write timeout
livestream.webrtc.enabled	boolean	true	Enable WHIP/WHEP livestream sessions
livestream.webrtc.ice_servers	ICE server list	[]	STUN/TURN servers used by livestream PeerConnections
livestream.webrtc.ice_gathering_timeout_seconds	seconds	10	Time allowed to gather local ICE candidates
livestream.webrtc.max_sdp_bytes	bytes	262144	Limit for WHIP/WHEP SDP requests and upstream WHEP SDP responses
livestream.webrtc.max_sessions	integer	1000	Total WHIP/WHEP session limit on one node
livestream.webrtc.max_session_duration_seconds	seconds	86400	Maximum lifetime of one WHIP/WHEP session

file_storage

Field	Type / values	Default	Meaning
file_storage.default_backend	backend name	disabled	Fallback backend used by product features
file_storage.chat_attachments_backend	backend name or empty	""	Backend for chat attachments; empty inherits default_backend
file_storage.user_avatars_backend	backend name or empty	""	Backend for user avatars; empty inherits default_backend
file_storage.media_covers_backend	backend name or empty	""	Backend for media covers; empty inherits default_backend
file_storage.room_covers_backend	backend name or empty	""	Backend for room covers; empty inherits default_backend
file_storage.playlist_covers_backend	backend name or empty	""	Backend for playlist covers; empty inherits default_backend
file_storage.upload_token_secret	at least 32 characters	""	Dedicated signing key for file upload/read tokens; required
file_storage.unreferenced_object_retention_seconds	seconds	86400	Grace period before uploaded objects with no active product reference are cleaned; 0 disables orphan cleanup
file_storage.backends.<name>.type	disabled, database, or s3	disabled	Registered backend implementation type
file_storage.backends.<name>.compression	none, lz4, or zstd	zstd	Compression algorithm used by type: "database" backends for persisted file_blob_parts segments
file_storage.backends.<name>.compression_min_size_bytes	bytes	4096	Minimum original payload size before compression is attempted
file_storage.backends.<name>.compression_min_savings_percent	0 to 100	10	Minimum saved percentage required before compressed bytes are stored
file_storage.backends.<name>.<s3-field>	S3 config	default S3 fields	S3-compatible settings for backends with type: "s3"; public_base_url is required for readable file URLs after upload or ownership proof validation

Upload sessions use a server-provided part plan: clients first request FileUploadPlan with empty parts, hash every planned part with SHA-256, then submit FileUploadManifestPart[]. SyncTV computes content_manifest_sha256 from the canonical manifest and uses it for instant-upload object lookup and unfinished resumable-session lookup. Database backends persist fixed file_blob_parts segments and serve HTTP Range requests directly from those segments, decompressing only the parts that overlap the requested byte range. S3 backends use native multipart uploads with presigned part URLs, sign x-amz-checksum-sha256 into each part URL, and validate object identity from the recorded part manifest at completion without reading GB-scale objects back from S3. Duplicate objects return upload_required=false with an ownership proof challenge; readable URLs are signed after proof validation and business-reference persistence.

File Storage Examples

Disable file storage:

file_storage:
  default_backend: "disabled"
  backends: {}

Store file bytes in PostgreSQL:

file_storage:
  default_backend: "database"
  chat_attachments_backend: "database"
  user_avatars_backend: "database"
  media_covers_backend: "database"
  room_covers_backend: "database"
  playlist_covers_backend: "database"
  upload_token_secret_file: "/run/secrets/file_upload_token_secret"
  backends:
    database:
      type: "database"
      compression: "zstd"

Use S3-compatible object storage:

file_storage:
  default_backend: "s3_public"
  chat_attachments_backend: "s3_public"
  user_avatars_backend: "s3_public"
  media_covers_backend: "s3_public"
  room_covers_backend: "s3_public"
  playlist_covers_backend: "s3_public"
  upload_token_secret_file: "/run/secrets/file_upload_token_secret"
  backends:
    s3_public:
      type: "s3"
      endpoint: "https://s3.example.com"
      access_key_id_file: "/run/secrets/file_storage_s3_access_key_id"
      secret_access_key_file: "/run/secrets/file_storage_s3_secret_access_key"
      bucket: "synctv-files"
      region: "auto"
      base_path: "files/"
      public_base_url: "https://cdn.example.com/files"
      upload_expires_seconds: 900

webauthn and email

Field	Type / values	Default	Meaning
webauthn.enabled	boolean	false	Enable passkey endpoints
webauthn.rp_id	domain	""	WebAuthn relying-party ID
webauthn.rp_origin	origin	""	Primary WebAuthn origin
webauthn.rp_name	string	SyncTV	Authenticator display name
webauthn.allowed_origins	origin array	[]	Additional accepted origins
webauthn.apple_app_ids	Apple application identifier array	[]	Web Credentials associations for native Apple apps
webauthn.android_apps	Android app association array	[]	Package names and signing-certificate SHA-256 fingerprints
webauthn.allow_subdomains	boolean	false	Allow subdomains of configured origins
webauthn.allow_any_port	boolean	false	Ignore origin ports
webauthn.timeout_seconds	seconds	300	Challenge timeout

media_providers and webrtc

Field	Type / values	Default	Meaning
media_providers.alist.request_timeout_seconds	seconds	30	Local Alist provider request timeout
media_providers.alist.connect_timeout_seconds	seconds	10	Local Alist provider connection timeout
media_providers.bilibili.request_timeout_seconds	seconds	30	Local Bilibili provider request timeout
media_providers.bilibili.connect_timeout_seconds	seconds	10	Local Bilibili provider connection timeout
media_providers.emby.request_timeout_seconds	seconds	30	Local Emby/Jellyfin provider request timeout
media_providers.emby.connect_timeout_seconds	seconds	10	Local Emby/Jellyfin provider connection timeout
webrtc.mode	signaling_only or peer_to_peer	peer_to_peer	WebRTC operating mode
webrtc.enable_builtin_stun	boolean	false	Enable built-in STUN
webrtc.stun_port	port	3478	STUN port
webrtc.stun_host	IP/listen address	0.0.0.0	STUN listen address
webrtc.stun_external_addr	string	""	STUN external address announced to clients
webrtc.filter_private_ice_candidates	boolean	false	Filter private ICE candidates

connection_limits, bootstrap, and cluster

Field	Type / values	Default	Meaning
connection_limits.max_per_user	integer	20	Maximum connections per user
connection_limits.max_per_room	integer	2000	Maximum connections per room
connection_limits.max_total	integer	100000	Maximum service-wide connections
connection_limits.idle_timeout_seconds	seconds	300	Idle connection timeout
connection_limits.max_duration_seconds	seconds	86400	Maximum connection lifetime
connection_limits.ws_message_rate_limit_per_second	per-second count	50	WebSocket message rate per connection
bootstrap.create_root_user	boolean	false	Create the initial root user automatically
bootstrap.root_username	string	root	Initial root username
bootstrap.root_password	secret string	""	Initial root password
cluster.enabled	boolean	false	Enable cluster mode
cluster.host	IP/listen address	0.0.0.0	Dedicated internal gRPC listen address
cluster.port	port	50051	Dedicated internal gRPC listen port
cluster.advertise_host	hostname or IP	""	Internal address advertised to peers; empty uses server.advertise_host
cluster.advertise_port	port or 0	0	Internal port advertised to peers; 0 uses cluster.port
cluster.secret	secret string	""	Authentication secret for internal cluster gRPC
cluster.secret_file	file path or empty	""	File containing the internal cluster request authentication secret
cluster.critical_channel_capacity	integer	10000	High-priority cluster event queue capacity
cluster.publish_channel_capacity	integer	100000	Normal cluster publish queue capacity
cluster.discovery_mode	redis, static, or k8s_dns	redis	Node discovery mode
cluster.leader_election_mode	redis or k8s_lease	redis	Leader election mode
cluster.peers	string array	[]	Static peer addresses
cluster.catchup_window_secs	seconds	300	Redis Stream catch-up replay window
cluster.stream_max_length	integer	100000	Approximate Redis Stream maximum length

password_complexity, buffer_sizes, and cache

Field	Type / values	Default	Meaning
password_complexity.min_length	integer	8	Minimum password length
password_complexity.require_uppercase	boolean	true	Require uppercase letters
password_complexity.require_lowercase	boolean	true	Require lowercase letters
password_complexity.require_digit	boolean	true	Require digits
password_complexity.require_special	boolean	false	Require special characters
password_complexity.max_repeated_chars	integer	3	Maximum repeated consecutive characters; 0 disables this check
password_complexity.zxcvbn_enabled	boolean	false	Enable zxcvbn entropy scoring for account passwords
password_complexity.zxcvbn_min_score	integer 0-4	3	Minimum zxcvbn score when enabled
buffer_sizes.websocket_outbound	integer	256	Per-connection WebSocket outbound queue size
buffer_sizes.audit_buffer	integer	10000	Audit event buffer capacity
cache.l1_capacity	integer	5000	L1 in-memory cache capacity
cache.l1_ttl_seconds	seconds	300	L1 in-memory cache TTL
cache.l2_ttl_seconds	seconds	300	Redis L2 cache TTL
cache.username_cache_capacity	integer	10000	Username cache capacity
cache.username_cache_ttl_seconds	seconds	3600	Username cache TTL
proxy_slice_cache.enabled	boolean	true	Enable proxy slice cache at startup
proxy_slice_cache.slice_size_bytes	integer	2097152	Individual Range slice size
proxy_slice_cache.max_cache_size_bytes	integer	536870912	Maximum total slice cache capacity
proxy_slice_cache.segment_ttl_seconds	seconds	300	Fresh slice TTL
proxy_slice_cache.stale_max_age_seconds	seconds	60	Time expired slices may be served as stale
proxy_slice_cache.stale_while_revalidate	boolean	true	Refresh expired hits in the background
proxy_slice_cache.file_backend_enabled	boolean	false	Enable file backend for proxy slice cache
proxy_slice_cache.file_cache_dir	path	""	Proxy slice cache file directory
proxy_slice_cache.eviction_interval_seconds	seconds	60	Background eviction check interval
proxy_slice_cache.watermark_ratio	float	0.875	Target usage ratio after eviction

messaging_rate_limits

Field	Type / values	Default	Meaning
messaging_rate_limits.chat_per_second	integer	10	Maximum chat messages per window
messaging_rate_limits.window_seconds	seconds	1	Chat rate-limit window

request_rate_limits

Field	Type / values	Default	Meaning
request_rate_limits.auth_max_requests	integer	5	Maximum authentication requests per window
request_rate_limits.auth_window_seconds	seconds	60	Authentication request window
request_rate_limits.write_max_requests	integer	120	Maximum write requests per window
request_rate_limits.write_window_seconds	seconds	60	Write request window
request_rate_limits.read_max_requests	integer	600	Maximum read requests per window
request_rate_limits.read_window_seconds	seconds	60	Read request window
request_rate_limits.media_max_requests	integer	120	Maximum media requests per window
request_rate_limits.media_window_seconds	seconds	60	Media request window
request_rate_limits.admin_max_requests	integer	180	Maximum admin requests per window
request_rate_limits.admin_window_seconds	seconds	60	Admin request window
request_rate_limits.streaming_max_requests	integer	1200	Maximum streaming requests per window
request_rate_limits.streaming_window_seconds	seconds	60	Streaming request window
request_rate_limits.websocket_max_requests	integer	60	Maximum WebSocket connection attempts per window
request_rate_limits.websocket_window_seconds	seconds	60	WebSocket connection window