Skip to main content
Version: 0.3

Config Reference


keyspaces: Vec<KeyspaceConfig>

Multiple keyspaces can be configured in order to filter incoming messages. If the filter feature is not used, only the first configured keyspace will be considered or the default (chronicle) if none is provided.

In addition to the keyspace name, each requires a map of datacenters (name -> replication factor). See here for more information about datacenters in ScyllaDB.

listen_address: String

The dashboard listen address, where it accepts requests to manage the Scylla cluster.

thread_count: Enum

The number of threads Scylla will use. Can be one of Count(usize) (a simple scalar count) or CoreMultiple(usize) (a multiple of the number of cores the system has).

reporter_count: u8

The number of reporters Scylla will spawn.

local_datacenter: String

The Scylla local datacenter.


Specifies the number of partitions to use in the database, as well as the number of milestones to use as chunks.

NOTICE: You can't change partition_config in future without migration.


Nothing at the moment, please refer to .env.


websocket_address: String

The Broker dashboard listen address, where it accepts requests to manage the broker topology.

mqtt_brokers: Vec<Url>

  • Messages: mqtt topic used to receive incoming IOTA messages;
  • MessagesReferenced: mqtt topic used to receive incoming metadata;

NOTICE: You should at least have one of each.

api_endpoints: Vec<Url>

IOTA node-endpoints used by chronicle to fill gaps.

retries_per_endpoint: u8

Max number of retries to retrieve something from api_endpoints.

retries_per_query: usize

Max number of retries to fetch/insert something from/toscylla, before declaring an outage, which will force the broker application to pause and await for scylla cluster to recover.

collector_count: u8

The number of concurrent collectors which collect data from feed sources also it's used as solidifier_count.

requester_count: u8

The number of concurrent requesters per collector.

NOTE: requesters are used by collector to fetch missing data from api_endpoint

request_timeout_secs: u64

The api_endpoint request timeout.

parallelism: u8

The max number of concurrent solidify requests.

sync_range: Option<SyncRange>

Identiy the milestone data sync range from/to.

complete_gaps_interval_secs: u64

Interval used by syncer to check if there are some gaps to fill/complete.

logs_dir: Option<String>

If provided, it will archive the milestone data in ordered fashion.

max_log_size: Option<u64>

The upper limit of the log_file_size.

NOTE: Ensure to use a limit within your filesystem range.