Production Operations Reference

Operational controls and incident interfaces for live Rivellum dapps.

Runtime Checklist

Control	Verification
RPC health	`GET /health`, `GET /v1/chain`
Submit path	`POST /v1/submit` acceptance + reason metrics
Contract path	receipt + replay artifacts available
Event path	stream connected + query backfill healthy
Alerting	lag/rejection/failure thresholds wired

Deployment Gate

deterministic replay pass for critical flows
fee envelope policy enforced in submit path
indexer checkpoint recovery test passed
rollback image/tag and runbook prepared

Incident Artifact Set

Artifact	Where
receipt	`/v1/mist/receipts/:id`
pvi	`/v1/mist/pvi/:id`
replay input	`/v1/mist/replay_inputs/:id`
event window	`/v1/events/query`

SLO/SLI Suggestions

SLI	Target
submit success ratio	>= 99.9%
receipt availability latency	<= 2s p95
stream lag	<= 1 batch
replay reproducibility	100% for sampled incidents

Runbook Snippets

# node chain status
curl https://rpc.rivellum.network/v1/chain

# recent events
curl https://rpc.rivellum.network/v1/events/recent

# mist receipt
curl https://rpc.rivellum.network/v1/mist/receipts/RECEIPT_ID

Rollback Pattern

freeze writes at app ingress
pin prior portal/backend image tag
validate chain/read surfaces
replay failed receipts against prior release assumptions
reopen writes after invariant checks

Related References

/docs/dev/debugging-replay
/docs/dev/indexer-frontend-patterns
/docs/ops/disaster-recovery