RecallGraph exports two settings - snapshot-intervals and sampling-probability. These are explained in the sections below.
This is a JSON object that stores the snapshot interval for each collection. Its keys are collection names and values are non-negative integers. If the value is 0, then snapshots are disabled for that collection. For a number greater than 0, RecallGraph interprets the number as the minimum number of events that must occur on a document until a snapshot is generated for that document. A snapshot is created only if the latest event is not a DELETE.
There is also a global setting for snapshot intervals which applies to all collections which are not explicitly listed in this object. This global setting is stored in the key _default in the same object. The default value for this setting is:
A sample object with some collection-specific values might look like:
"airports": 0 //snapshots disabled
If this setting is changed globally, or for a collection, after some snapshots have already been created, the existing snapshots are left untouched and the new value(s) only apply to snapshots created thereafter.
This setting is used for determining whether to collect a tracing sample. The default is 0. Leave it at 0 if you're not going to use tracing in your deployment.
If set to a value between 0 and 1 (inclusive), it is interpreted as the probability with which a tracing sample may be collected for an incoming HTTP request. Regardless of the value set here, a trace may be force-collected or force-suppressed using designated HTTP headers. These headers and other tracing parameters are covered in more detail in the tracing section.