Configuring data-sources: Advanced settings

The following advanced data-source settings applies to all data-sources.

To change them, see how to configure Linkurious Enterprise.

General settings

• dataSourceConnectionTimeout (default: 30): The maximum time in seconds before the connection attempt to each data-source times out.

• dataSourceAutoReconnectInterval (default: 300): Number of seconds after which Linkurious will try to reconnect automatically to all offline data-sources. This value is internally converted in minutes; if it is not a multiple of 60, it will be rounded to the nearest number of minutes. Explicitly setting this parameter to 0 disable the offline data-sources auto-reconnection feature.

• pollInterval (default: 10): Check if the data-sources and search engine are connected at each interval (in seconds).

• timeline (default: false): Whether the visualization timeline is available in the workspace UI

Search engine settings

• indexingChunkSize (default: 5000): The number of nodes and edges retrieved at each batch when indexing the graph database.

• searchAddAllThreshold (default: 500): The maximum number of search results that the user can add to a Visualization at once.

• minSearchQueryLength (default: 3): The number of characters needed to trigger a search query. Set 1 to provide live results from the first character typed by the user.

• searchPrefixExpansion (default: true): Whether Linkurious should return nodes and edges that contain the search query as a prefix, e.g.: Searching Link will return node Linkurious.

Graph exploration settings

• supernodeThreshold (default: 10000): Number of adjacent edges after which a node is considered a supernode.

• edgesBetweenSupernodes (default: false): Whether Linkurious should retrieve edges between 2 supernodes. Note: Linkurious always return edges between a supernode and a regular node.

• rawQueryTimeout (default: 60000): Milliseconds after which a query to the database will time out.

• defaultFuzziness (default: 0.1): Default value to search fuzziness between 0 and 1. A value of 0 means exact matching of the search query.

• expandThreshold (default: 50): When the user expands a node with too many neighbors, Linkurious Enterprise will ask to refine the query so that fewer neighbors are returned.

• rawQueryLimit (default: 500): The maximum number of results returned by Linkurious Enterprise when executing a query or query template.

• sampledItemsPerType (default: 1000): The number of nodes/edges per category/type to read from the graph for the schema sampling

• sampledVisualizationItems (default: 1000): Number of nodes/edges to read from existing visualizations for the schema sampling

• showBuiltinQueries (default: true): Whether built-in queries like Shortest Path should appear in the list of Graph Query templates.

• slowQueryThreshold (default: 500): Milliseconds after which a query is logged in the log file as a slow query.

• defaultLayout (default: { "algorithm": "force", "mode": "best" }): The default layout to be applied to visualizations.

  • defaultLayout.algorithm (default: force): Can be force or hierarchical.
  • defaultLayout.mode (default: best): (best|fast) for force algorithm and (LR | RL | TB | BT) for hierarchical algorithm. See the layout section in the user manual for more information.

• searchRetryMultiplier (default: 5): During search, update the index by removing the missing items from the index server. The higher the multiplier is, the less chance search will return non-existing items but search might take more time.

Worker pool settings

The worker pool is an internal thread pool used to offload computational tasks from the main thread of the Linkurious Enterprise process. It is used to run alerts, custom queries and compute server-side visualization layouts, so that these do not affect the application responsiveness.

• workerPool (default: { "enabled": true, "size": 1, "memoryLimitMb": 1024}):

  • workerPool.enabled (default: true): Whether to enable this feature. When it is disabled, any task that would have been executed by the worker pool is instead executed directly by the main thread.

  • workerPool.size (default: 1): The fixed number of threads to spawn in the worker pool. We recommend this setting to be at least 2 if alerts are enabled. The maximum value for this setting should be number of available CPU cores on the server running Linkurious Enterprise, minus one.

  • workerPool.memoryLimitMb (default: 1024): The maximum size of the main heap in megabytes, for each thread in the worker pool.

If the worker pool is disabled, all the tasks will be executed by the main thread of the Linkurious Enterprise process.

Additional Certificate Authorities

• extraCertificateAuthorities: The relative or absolute path to a single PEM file. When set, the well known "root" CAs (like LetsEncrypt) will be extended with the extra certificates in the file. The file should consist of one trusted certificate or multiple ones concatenated one after the other in the PEM format. If in your chain of certificates you have some Intermediate certificates, adding the Root CA will be enough to trust the whole chain. If the path is relative, the file should be located within the linkurious/data folder.

If Linkurious Enterprise is installed as a service, the service needs to be uninstalled and re-installed for the change to be taken into account.

If Linkurious Enterprise is installed using Docker, then extraCertificateAuthorities will not work and the NODE_EXTRA_CA_CERTS environment variable should be used instead. The variable must contain the absolute path to a PEM file (see details on the Node.js documentation).

Password obfuscation

• obfuscation (default: false): Set to true if you want all the passwords in the configuration to be obfuscated at the next restart of Linkurious Enterprise.