The relayer sometimes can't submit blocks, which results in state root updates with no NEAR DA data. This is not common at all, but should be explored anyway (first by metrics, if this is done after #225, then by execution logs) to see if there's anything we can do to improve that - increasing retries, having a more reliable NEAR RPC...
An easy way to catch these periods is by isolating near_da: false submissions by operators in the aggregator grafana dashboards.
The relayer sometimes can't submit blocks, which results in state root updates with no NEAR DA data. This is not common at all, but should be explored anyway (first by metrics, if this is done after #225, then by execution logs) to see if there's anything we can do to improve that - increasing retries, having a more reliable NEAR RPC...
An easy way to catch these periods is by isolating
near_da: falsesubmissions by operators in the aggregator grafana dashboards.