INX Syncing
The primary function of Chronicle is to gather confirmed tangle data via an INX connection. This connection will most likely be to an IOTA Node, which should be on the same machine as the Chronicle instance.
When running with the inx feature (default using the docker image), Chronicle will stream data from the INX_URL and store it in the configured MongoDB instance.
Singleton Writer
Chronicle assumes that it is the only instance writing to the configured MongoDB database. Otherwise it will exit due to write conflicts.
Network Name Link
When Chronicle starts for the first time, it stores the latest network protocol parameters. It uses these to check that the same network is used across the lifetime of the dataset. In particular, the network name must not change, or Chronicle will fail to start.
Ledger State
When Chronicle starts syncing, it will get the current Ledger State from the INX source. Though Chronicle can sync back to the earliest data the INX connection can provide, the data may not be valid until it catches up to the ledger index of that initial state.
Analytics
The starting index determines when Analytics can begin. Analytics are calculated incrementally per-milestone, but require an initial ledger state which is only valid once Chronicle has caught up to the starting index.
CLI Analytics
Analytics can also be synced manually using the fill-analytics command-line tool. It should not be run until Chronicle reports that it is healthy (see the /health endpoint) or else the results may not be correct. However, if this happens, re-running the analytics when Chronicle is healthy will overwrite the previous values.
The --num-tasks flag can be used to parallelize these processes, but be warned: There is currently no feature to resume halted analytics. If the process is canceled without completing, it may be very difficult to restart it without re-calculating analytics.
For the list of supported analytics of each kind and more information, refer to the CLI documentation (via chronicle --help).