Lignum system configuration

server config

Root key : server

• server.host

  Type : Ip address Desription: Host address of the lignum service where it should bind itself. It can be localhsot or bind to 0.0.0.0

• server.port

  Type : Number Value : 8080 Description: Port to which lignum should bind and listen on for incoming request.

• service-key

  Type : String Value : “service/lignum/key/master” Description: A unique service key which is used for setting up a distributed lock for lignum cluster leader. Must be same for all node in a cluster.

---

Cluster management config

Root key: consul

• consul.host

  Type : Ip address Desription: Host address of the consul service.

• consul.port

  Type : Number Desription: Port of the consul service.

• consul.session-ttl-in-seconds

  Refer consul session ttl

  Type : Number Desription: consul session ttl. Value : 15

• consul.session-renewal-ttl-in-seconds

Type      : Number
Desription: consul session renewal ttl, client holding session must renew its session periodically before the `session-ttl-in-seconds` expires.
Value     : 13

• consul.lock-delay-in-ms: 10

  Refer consul session lockdelay

Type      : Number
Desription: consul lock delay, when session is invalidated, it will not allow any other session to acquire repviously held lock fo specified time period.
Value     : 15

• consul.leader-election-interval-in-ms

  Type : Number Desription: Lignum periodically tries to elect itself as a leader, in case of leader is down and lock is released on the session key, it will acquire and elect itself as a leader, this key will control the frequency of attempt. Value : 15

for more refer detailed explanation in https://www.consul.io/docs/dynamic-app-config/sessions

Message config

Root key: message

• message.initial-size-per-topic

  Type : Number Description : Topic buffer size, the number of message should be stored in a buffer, this also control number of messages store in wal and log files.

• message.data-dir

  Type : String Description : Data directory where messages should be persisted. Value : “topics” —-

Follower config

Root key : follower

• follower.healthcheck-interval-in-seconds: 1

  Type : Number Description : Interval to check if all followers are healthy Value : 1

• follower.healthcheck-ping-timeout-in-ms

  Type : Number Description : Timeout for healthcheck on each follower Value : 100

• follower.registration-leader-check-interval-in-seconds

  Type : Number Desription : Interval to check if there is a change in cluster leader by querying consul, if changed connect to the new leader as follower.

---

Replication config

Root key : replication

• replication.internal-queue-size

  Type : Number Desription : Size of internal buffer/channel which holds messages to be sent over to follower Value : 1024

• replication.client-timeout-in-ms

  Type : Number Description : Timeout for replication write on each follower Value : 25

• replication.wal-replication-port

  Type : Number Description : WAL service port on which it recieves replication messages from the leader Value : 8124

• replication.wal-sync-interval-in-sec

  Type : Number (Duration) Description : Interval between the wal replication routine run checks on wal files and replicate to follower Value : 5

Refer the default config here