Troubleshooting
Replacement Transaction Underpriced: Fix Stuck EVM Transactions
Diagnose and resolve the 'replacement transaction underpriced' error by understanding mempool replacement rules and executing a correct fee bump or cancel.
Read articleOnFinality Learn
Practical knowledge for Web3 builders
Technical explainers, troubleshooting playbooks, and infrastructure guidance shaped by the questions developers search for.
All topicsIntegration & DevelopmentInfrastructure & OperationsNetwork & Protocol GuidesRPC TroubleshootingReliability & ConsistencyPerformance & OptimizationTroubleshootingIntegration & API DesignSecurity & Operations
OnFinality Learn
237Latest articles
Integration & API Design
JSON-RPC Idempotency: Preventing Duplicate Writes on Retry
Why a timed-out JSON-RPC write may already have executed, and how to build an at-most-once client using nonces, transaction hashes, and check-then-resend.
Read articleSecurity & Operations
RPC API Key Rotation: Zero-Downtime Credential Lifecycle
A mechanism-level playbook for rotating RPC API keys, JWTs, and project IDs without dropping a single request.
Read articleReliability & Consistency
Multi-Endpoint RPC Consistency: Head Lag and Conflicting Reads
Why the same JSON-RPC query returns different data from different endpoints, and the pinning, monotonicity, and quorum discipline that makes multi-endpoint reads reproducible.
Read articlePerformance & Optimization
Reading RPC Rate-Limit Headers: X-RateLimit and Retry-After
Learn how to read RateLimit and Retry-After headers on every RPC response, pace requests before a 429, and measure your endpoint's real ceiling.
Read articlePerformance & Optimization
RPC Connection Reuse and HTTP/2 Keep-Alive: Why New Connections Cost Latency
Learn how TCP/TLS handshakes, HTTP/1.1 keep-alive pools, and HTTP/2 multiplexing affect JSON-RPC latency, reliability, and error patterns.
Read articlePerformance & Optimization
Multi-Region RPC Failover: Latency-Aware Routing and Health-Based Failover
Design multi-region, multi-endpoint RPC failover so a provider, region, or node failure degrades latency instead of breaking your app.
Read articlePerformance & Optimization
RPC Request Hedging: Cut Tail Latency Across Endpoints
Learn how request hedging collapses p99 latency for read-only JSON-RPC calls by racing duplicate requests across independent endpoints and cancelling the loser.
Read articleTroubleshooting
RPC WebSocket Reconnect Without Data Loss: Gap Detection and Backfill
Reconnecting a WebSocket is easy; proving you missed nothing is the hard part. Learn cursor-based gap detection and bounded backfill for Ethereum, Solana, Sui, and Substrate.
Read articleNetwork & Protocol Guides
Solana Commitment Levels: processed vs confirmed vs finalized
How Solana's processed, confirmed and finalized commitment levels change what every RPC read returns, and how to choose one deliberately.
Read article
Ethereum
Solana
Polygon
Base
Arbitrum
BNB Chain
Polkadot
Optimism
