Helm Values Reference

This page documents all configurable values in the Cardano Rosetta Java Helm charts. For each value, the equivalent Docker Compose environment variable is shown where applicable.

Global Values

These values are shared across all subcharts via global.*.

Value	Default	Docker Compose equivalent	Description
global.network	mainnet	NETWORK	Blockchain network: mainnet, preprod, preview
global.protocolMagic	"764824073"	PROTOCOL_MAGIC	Cardano protocol magic number (always a quoted string to prevent scientific notation)
global.releaseVersion	"2.1.0"	RELEASE_VERSION	Docker image tag for API and indexer
global.cardanoNodeVersion	"10.5.4"	CARDANO_NODE_VERSION	Cardano node image tag
global.pgVersionTag	REL_18_0	PG_VERSION_TAG	PostgreSQL image tag
global.mithrilVersion	2543.1-hotfix	MITHRIL_VERSION	Mithril client image tag
global.profile	mid	—	Hardware profile: entry, mid, advanced
global.sync	true	SYNC	Set false for offline (API-only) mode

Database (global.db)

Value	Default	Docker Compose equivalent	Description
global.db.name	rosetta-java	DB_NAME	PostgreSQL database name
global.db.user	rosetta_db_admin	DB_USER	PostgreSQL user
global.db.password	"" (required)	DB_SECRET	PostgreSQL password — pass via --set or Sealed Secret
global.db.schema	public	DB_SCHEMA	PostgreSQL schema
global.db.port	5432	DB_PORT	PostgreSQL port
global.db.host	""	DB_HOST	Override to use external PostgreSQL. Empty = use in-cluster service

Mithril (global.mithril)

Value	Default	Docker Compose equivalent	Description
global.mithril.enabled	true	MITHRIL_SYNC	Download Mithril snapshot on first install
global.mithril.snapshotDigest	latest	SNAPSHOT_DIGEST	Specific snapshot digest or latest
global.mithril.aggregatorEndpoint	""	AGGREGATOR_ENDPOINT	Custom aggregator URL. Empty = use network default
global.mithril.genesisVerificationKey	""	GENESIS_VERIFICATION_KEY	Custom genesis verification key
global.mithril.ancillaryVerificationKey	""	ANCILLARY_VERIFICATION_KEY	Custom ancillary verification key

Storage (global.storage)

Value	Default	Docker Compose equivalent	Description
global.storage.cardanoNode.size	500Gi	CARDANO_NODE_DIR volume	PVC size for blockchain data
global.storage.cardanoNode.storageClass	""	—	StorageClass name. Empty = cluster default
global.storage.postgresql.size	200Gi	DB_PATH volume	PVC size for PostgreSQL data
global.storage.postgresql.storageClass	""	—	StorageClass name. Empty = cluster default

---

Hardware Profiles

Set global.profile to one of: entry, mid, advanced.

Resource Requests / Limits

Profile	cardano-node	postgresql	yaci-indexer	rosetta-api
entry req	1 CPU / 6 Gi	1 CPU / 6 Gi	500m / 2 Gi	250m / 1 Gi
entry lim	4 CPU / 12 Gi	4 CPU / 12 Gi	2 CPU / 4 Gi	1 CPU / 2 Gi
mid req	2 CPU / 12 Gi	2 CPU / 12 Gi	1 CPU / 4 Gi	500m / 2 Gi
mid lim	8 CPU / 24 Gi	8 CPU / 24 Gi	4 CPU / 8 Gi	2 CPU / 4 Gi
advanced req	4 CPU / 24 Gi	4 CPU / 24 Gi	2 CPU / 8 Gi	1 CPU / 4 Gi
advanced lim	16 CPU / 48 Gi	16 CPU / 48 Gi	8 CPU / 16 Gi	4 CPU / 8 Gi

PostgreSQL Tuning (per profile)

Parameter	entry	mid	advanced	Docker Compose env
maxConnections	120	300	600	DB_POSTGRES_MAX_CONNECTIONS
sharedBuffers	1GB	4GB	32GB	DB_POSTGRES_SHARED_BUFFERS
effectiveCacheSize	2GB	8GB	32GB	DB_POSTGRES_EFFECTIVE_CACHE_SIZE
workMem	16MB	64MB	96MB	DB_POSTGRES_WORK_MEM
API poolMin	12	150	100	API_DB_POOL_MIN_COUNT
API poolMax	12	150	550	API_DB_POOL_MAX_COUNT

---

yaci-indexer Values

Value	Default	Docker Compose equivalent	Description
yaci-indexer.env.removeSpentUtxos	true	REMOVE_SPENT_UTXOS	Prune spent UTXOs to save disk
yaci-indexer.env.removeSpentUtxosLastBlocksGraceCount	129600	REMOVE_SPENT_UTXOS_LAST_BLOCKS_GRACE_COUNT	Blocks to retain (~30 days)
yaci-indexer.env.removeSpentUtxosBatchSize	3000	REMOVE_SPENT_UTXOS_BATCH_SIZE	Pruning batch size
yaci-indexer.env.blockTransactionApiTimeoutSecs	120	BLOCK_TRANSACTION_API_TIMEOUT_SECS	Timeout for block/tx API calls
yaci-indexer.env.searchLimit	100	SEARCH_LIMIT	Maximum search results
yaci-indexer.env.continueParsingOnError	true	CONTINUE_PARSING_ON_ERROR	Continue syncing on parse errors
yaci-indexer.env.peerDiscovery	false	PEER_DISCOVERY	Enable peer discovery
yaci-indexer.env.logLevel	error	LOG	Log level: error, info, debug

---

rosetta-api Values

Value	Default	Docker Compose equivalent	Description
rosetta-api.replicaCount	1	—	Number of API replicas
rosetta-api.env.httpConnectTimeoutSeconds	5	HTTP_CONNECT_TIMEOUT_SECONDS	HTTP client connect timeout
rosetta-api.env.httpRequestTimeoutSeconds	5	HTTP_REQUEST_TIMEOUT_SECONDS	HTTP client request timeout
rosetta-api.env.syncGraceSlotsCount	100	SYNC_GRACE_SLOTS_COUNT	Slots behind tip before reporting out-of-sync
rosetta-api.env.removeSpentUtxos	true	REMOVE_SPENT_UTXOS	Must match yaci-indexer setting
rosetta-api.env.removeSpentUtxosLastBlocksGraceCount	129600	REMOVE_SPENT_UTXOS_LAST_BLOCKS_GRACE_COUNT	Must match yaci-indexer setting
rosetta-api.env.blockTransactionApiTimeoutSecs	120	BLOCK_TRANSACTION_API_TIMEOUT_SECS	Timeout for block queries
rosetta-api.env.devkitEnabled	false	DEVKIT_ENABLED	Enable DevKit integration
rosetta-api.env.tokenRegistryEnabled	false	TOKEN_REGISTRY_ENABLED	Enable Cardano token registry
rosetta-api.env.tokenRegistryBaseUrl	""	TOKEN_REGISTRY_BASE_URL	Token registry URL (leave empty = disabled)
rosetta-api.env.tokenRegistryCacheTtlHours	12	TOKEN_REGISTRY_CACHE_TTL_HOURS	Token metadata cache TTL
rosetta-api.env.tokenRegistryLogoFetch	false	TOKEN_REGISTRY_LOGO_FETCH	Fetch token logos (bandwidth-intensive)
rosetta-api.env.tokenRegistryRequestTimeoutSeconds	2	TOKEN_REGISTRY_REQUEST_TIMEOUT_SECONDS	Token registry request timeout

Ingress

rosetta-api:
  ingress:
    enabled: true
    className: nginx
    host: rosetta.example.com
    tls:
      - secretName: rosetta-tls
        hosts:
          - rosetta.example.com

---

Index Applier

Value	Default	Description
indexApplier.enabled	true	Deploy the index-applier Job
indexApplier.mode	automatic	automatic: plain Job (GitOps-friendly, no hooks). hook: legacy Helm post-install/post-upgrade hook.
indexApplier.pollInterval	60	Seconds between API readiness polls

In automatic mode the Job runs as part of the release (compatible with ArgoCD and --no-hooks). The Job is cleaned up 24 hours after completion via ttlSecondsAfterFinished.

In hook mode the Job is a Helm post-install/post-upgrade hook. Monitor it independently and never use --wait-for-jobs as it can run for up to 18 hours on mainnet.

---

Subchart Toggles

Value	Default	Description
cardano-node.enabled	true	Deploy the Cardano node
postgresql.enabled	true	Deploy in-cluster PostgreSQL
yaci-indexer.enabled	true	Deploy the Yaci indexer
rosetta-api.enabled	true	Deploy the Rosetta API

---

Using an External PostgreSQL

To use an external managed database (RDS, Cloud SQL, etc.):

# values-external-db.yaml
postgresql:
  enabled: false

global:
  db:
    host: "my-rds-instance.cluster-xyz.us-east-1.rds.amazonaws.com"
    name: "rosetta-java"
    user: "rosetta_db_admin"
    password: ""   # pass via --set global.db.password=...

helm upgrade --install rosetta helm/cardano-rosetta-java \
  -f helm/cardano-rosetta-java/values.yaml \
  -f values-external-db.yaml \
  --set global.db.password="${DB_PASSWORD}"

---

Example: Preprod on K3s (entry profile)

# values-preprod.yaml (already included in the chart)
global:
  network: preprod
  protocolMagic: "1"
  profile: entry
  storage:
    cardanoNode:
      size: 100Gi
      storageClass: local-path
    postgresql:
      size: 50Gi
      storageClass: local-path

yaci-indexer:
  env:
    removeSpentUtxos: false
    peerDiscovery: true

rosetta-api:
  env:
    removeSpentUtxos: false

helm upgrade --install rosetta helm/cardano-rosetta-java \
  -f helm/cardano-rosetta-java/values-k3s.yaml \
  -f helm/cardano-rosetta-java/values-preprod.yaml \
  --set global.db.password="${DB_PASSWORD}" \
  -n cardano