Running the STH server
To run the STH server, please execute the following command from the directory where the STH component was installed:
The STH component provides the user with 2 mechanisms to configure the component to the concrete needs of the user:
- Environment variables, which can be set assigning values to them or using the
sth_default.conffile if a packaged version of the STH component is used.
config.jsfile located at the root of the STH component code, a JSON formatted file including the configuration properties.
It is important to note that environment variables, if set, take precedence over the properties defined in the
The environment variables accepted by the script (for which there exists counterpart entries in the already mentioned
config.js are the following ones:
STH_HOST: The host where the STH server will be started. Optional. Default value: "localhost".
STH_PORT: The port where the STH server will be listening. Optional. Default value: "8666".
FILTER_OUT_EMPTY: A flag indicating if the empty results should be removed from the response. Optional. Default value: "true".
TEMPORAL_DIR: A relative path from the STH home directory to a directory where the temporary files generated by the STH component are stored. These files are generated before returning them when the
filetypeis included in any data retrieval request. Default value: "temp".
MAX_PAGE_SIZE: Max page size returned by a query about raw data. Default value: "100"
DEFAULT_SERVICE: The service to be used if not sent in the Orion Context Broker notifications. Optional. Default value: "testservice".
DEFAULT_SERVICE_PATH: The service path to be used if not sent in the Orion Context Broker notifications. Optional. Default value: "/testservicepath".
DATA_MODEL: The STH component supports 3 alternative data models when storing the raw and aggregated data into the database: 1) one collection per attribute, 2) one collection per entity and 3) one collection per service path. The possible values are: "collection-per-attribute", "collection-per-entity" and "collection-per-service-path" respectively. Default value: "collection-per-entity".
DB_USERNAME: The username to use for the database connection. Optional. Default value: "".
DB_PASSWORD: The password to use for the database connection. Optional. Default value: "".
DB_URI: The URI to use for the database connection. This does not include the 'mongo://' protocol part (see a couple of examples below). Optional. Default value: "localhost:27017".
DB_AUTH_SOURCE: The auth source to use for the database connection. Optional. Default value "".
DB_RECONNECT_TRIES: MongoDB server attempts to reconnect. Default value 30.
DB_RECONNECT_INTERVAL: time to wait in milliseconds between MongoDB reconnection attemps. Default value 1000.
REPLICA_SET: The name of the replica set to connect to, if any. Default value: "".
DB_PREFIX: The prefix to be added to the service for the creation of the databases. More information below. Optional. Default value: "sth_".
COLLECTION\PREFIX: The prefix to be added to the collections in the databases. More information below. Optional. Default value: "sth_".
POOL_SIZE: The default MongoDB pool size of database connections. Optional. Default value: "5".
WRITE_CONCERN: The write concern policy to apply when writing data to the MongoDB database. Default value: "1".
SHOULD_STORE: Flag indicating if the raw and/or aggregated data should be persisted. Valid values are: "only-raw", "only-aggregated" and "both". Default value: "both".
TRUNCATION_EXPIRE_AFTER_SECONDS: Data from the raw and aggregated data collections will be removed if older than the value specified in seconds. In case of raw data the reference time is the one stored in the
recvTimeproperty whereas in the case of the aggregated data the reference of time is the one stored in the
_id.originproperty. Set the value to 0 not to apply this time-based truncation policy. Default value: "0".
TRUNCATION_SIZE: The oldest raw data (according to insertion time) will be removed if the size of the raw data collection gets bigger than the value specified in bytes. Set the value to 0 not to apply this truncation policy. Take into consideration than the "size" configuration parameter is mandatory in case size collection truncation is desired as required by MongoDB. Default value: "0". Notice that this configuration parameter does not affect the aggregated data collections since MongoDB does not currently support updating documents in capped collections which increase the size of the documents. Notice also that in case of the raw data, the size-based truncation policy takes precedence over the TTL one. More concretely, if a size limitation is set, the previous time expiration is ignored for the raw data collections since currently MongoDB does not support TTL in capped collections. Default value: "0".
TRUNCATION_MAX: The oldest raw data (according to insertion time) will be removed if the number of documents in the raw data collections goes beyond the specified value. Set the value to 0 not to apply this truncation policy. Notice that this configuration parameter does not affect the aggregated data collections since MongoDB does not currently support updating documents in capped collections which increase the size of the documents. Default value: "0".
IGNORE_BLANK_SPACES: Attribute values to one or more blank spaces should be ignored and not processed either as raw data or for the aggregated computations. Default value: "true".
NAME_MAPPING: Database and collection names are generated from the service, service path, entity ID and type and attribute names. Consequently and to avoid the restrictions imposed by MongoDB and stated at limits, it may be mapped to database and collection names which bypass those limitations. The mapping mechanism provided consists on a 1 to 1 mapping between services, service paths, entity IDs and types and attribute names via a mapping configuration file. The
NAME_MAPPINGvalue is an object including 2 properties: 1)
enabledwhich, as its name states, enables or disables the mapping mechanism (default value: "true") and 2)
configFilewhich is a relative or absolute path to the mapping configuration file (default value: "./name-mapping.json"). More information about the mapping configuration file in the "Database and collection name encoding and decoding" section of the "Installation & Administration Manual".
NAME_ENCODING: Database and collection names may be encoded to avoid the restrictions imposed by MongoDB and stated at limits. The encoding criteria is the following one: 1) encode the forbidden characters using an escaping character (
x) and the numerical Unicode code for each character (for instance, the
$character will be encoded as
x0024), 2) database and collection names already using the above encoding must be escaped prepending another
x(for instance, the text
x002awill be encoded as
xx002a, 3) the uppercase characters included in database names will be encoded using the mechanism stated in 1), 4) collection names starting with
system.will be encoded as
system.myDatawill be encoded as
xsystem.myData) and 5) the name separator character (
xffff) is not decoded. It is important to note that the encoding mechanism also applies in case a mapping has been accomplished over the resulting or new element. Default value: "true".
LOGOPS_LEVEL: The log level to use. Possible values are: "DEBUG", "INFO", "WARN", "ERROR" and "FATAL". Since the STH component uses the logops package for logging, for further information check out the logops npm package information online. Default value: "INFO".
LOGOPS_FORMAT: The log format to use. Possible values are: "json" (writes logs as JSON), "dev" (for development, used when the NODE_ENV variable is set to 'development'). Since the STH component uses the logops package for logging, for further information please check out the logops npm package information online. Default value: "json".
PROOF_OF_LIFE_INTERVAL: The time in seconds between proof of life logging messages informing that the server is up and running normally. Default value: "60".
PROCESSED_REQUEST_LOG_STATISTICS_INTERVAL: The time in seconds between processed requests statistics appear in the logs. Default value: "60".
For example, to start the STH server listening on port 7777, connecting to a MongoDB instance listening on mymongo.com:27777 and without filtering out the empty results, use:
STH_PORT=7777 DB_URI=mymongo.com:27777 FILTER_OUT_EMPTY=false ./bin/sth
On the other hand, in case of connecting to a MongoDB replica set composed of 3 machines with IPs addresses 22.214.171.124, 126.96.36.199, 188.8.131.52 listening on ports 27771, 27772 and 27773, respectively, use:
The STH component creates a new database for each
service. The name of these databases will
be the concatenation of the
DB_PREFIX environment variable and the service name, using an underscore (
_) as the
As already mentioned, all these configuration parameters can also be adjusted using the
config.js file whose contents
It is important to note that there is a limitation of 120 bytes for the namespaces (concatenation of the database name and collection names) in MongoDB. Related to this, the STH generates the collection names using the following mechanism:
- The collection names are generated as a concatenation of the
COLLECTION_PREFIXconfiguration parameter, the service path, the entity id, the entity type and the suffix ".aggr" for the collections storing the aggregated data. The length of the collection name plus the
DB_PREFIXplus the database name (or service) should not be more than 120 bytes using UTF-8 format or MongoDB will complain and will not create the collection, and consequently no data would be stored by the STH. A warning message is logged in case this happens.
Configuration for CORS in config.js
The STH component provides the CORS functionality to user in order to configure the CORS policy which can be used to process cross origin request:
enabled: This parameter is used to set the CORS policy true or false. Default value is 'false'.
origin: This parameter is used to allow only trusted domains to access the application.
headers: This parameter is used to set headers like 'Access-Control-Allow-Origin','Access-Control-Allow-Headers' which must be returned from the server in the request headers.
additionalHeaders: This parameter is used to set additionalHeaders like "fiware-servicepath,fiware-service" which must be returned from the server in the request headers.
credentials: This parameter is set as 'true' to allow "Access-Control-Allow-Credentials" in the request headers.