Basic metrics and a health status check can be enabled with the --metrics flag (default port 8001, use --metricsPort to specify).
Prometheus metrics are served on /metrics. For each chain that exists, this provides:
<chain>_blocks_processed: the number of blocks processed by the chains listener.<chain>_latest_processed_block: most recent block that has been processed by the listener.<chain>_latest_known_block: most recent block that exists on the chain.<chain>_votes_submitted: number of votes submitted by the relayer.
The endpoint /health will return the current known block height, and a timestamp of when it was first seen for every chain:
{
"chains": [
{
"chainId": "Number",
"height": "Number",
"lastUpdated": "Date"
}
]
} If the timestamp is at least 120 seconds old an error will be returned instead:
{
"error": "String"
}