eth_getLogs failing with 429 or "block range too large": pay with x402, no signup

A free public RPC fails a log scan in two ways: 429 Too Many Requests once you go faster than its shared limit, and “block range too large” once a query spans more blocks than that one provider accepts. The fix that finishes today: point the script at an endpoint that spreads requests across several providers, with a rate limit of your own, and split ranges when it tells you to. Lasso sells that endpoint for a few cents over x402 or MPP, with no account to create.

1. Buy a funded key in one paid request

curl -i -X POST https://lasso.sh/api/v1/credit \
  -H 'content-type: application/json' -d '{"usd":"1"}'

The first response is 402 Payment Required, with an x402 challenge in PAYMENT-REQUIRED and an MPP challenge in WWW-Authenticate. Pay it with any x402 client (USDC on Base) or MPP client (Tempo) by repeating the same request with the signed payment. One purchase buys $0.05 to $100. The paid response returns key.rpc_url_template. Replace {chain} with load-balanced/ethereum, the strategy meant for bulk reads, and keep the URL secret: it carries the key. Resubmitting the same signed payment shows the same key again, so a lost response loses nothing.

2. Point the script at it

export RPC_URL='<rpc_url_template with {chain} = load-balanced/ethereum>'

A funded key gets its own rate limit of 100 requests per second, with bursts to 200, instead of a public endpoint’s shared one. Lasso routes each request across several Ethereum providers by measured latency and health, and moves a failed or rate-limited read to the next provider within the same request.

3. Split ranges on Lasso’s range error

Each provider caps how many blocks one eth_getLogs may span, and Lasso learns those caps from live traffic. When a range is still too large, Lasso answers with a standard JSON-RPC error your loop can act on:

code: -32602, error.data.code: "log_range_limit";
error.data.max_block_range with a size that fits, or null when unknown.

async function getLogs(from: bigint, to: bigint, filter: object): Promise<unknown[]> {
  const res = await fetch(process.env.RPC_URL!, {
    method: "POST",
    headers: { "content-type": "application/json" },
    body: JSON.stringify({
      jsonrpc: "2.0", id: 1, method: "eth_getLogs",
      params: [{ ...filter, fromBlock: `0x${from.toString(16)}`, toBlock: `0x${to.toString(16)}` }],
    }),
  });
  const body = await res.json();
  if (!body.error) return body.result;
  if (body.error.data?.code !== "log_range_limit" || from === to) throw new Error(body.error.message);

  const fits = body.error.data.max_block_range;
  const mid = fits && BigInt(fits) <= to - from ? from + BigInt(fits) - 1n : from + (to - from) / 2n;
  return [...(await getLogs(from, mid, filter)), ...(await getLogs(mid + 1n, to, filter))];
}

Add an address to the filter when you can: when every provider refuses an address-less query, Lasso reports log_filter_refused. Each successful response is one provider’s answer; if you need every log, reconcile against a second source.

4. What it costs

eth_getLogs is $0.000075 a call on load-balanced. $1 covers about 13,000 calls. An address-less query that every provider refuses is free. If the balance runs out mid-scan, the error carries error.data.credit_url; pay it the same way to top up, and the key keeps working.

Get started

Concepts

Streams

Guides

Reference

Updates

eth_getLogs failing with 429 or "block range too large": pay with x402, no signup

1. Buy a funded key in one paid request

2. Point the script at it

3. Split ranges on Lasso’s range error

4. What it costs

​eth_getLogs failing with 429 or “block range too large”: pay with x402, no signup

​1. Buy a funded key in one paid request

​2. Point the script at it

​3. Split ranges on Lasso’s range error

​4. What it costs

eth_getLogs failing with 429 or “block range too large”: pay with x402, no signup

1. Buy a funded key in one paid request

2. Point the script at it

3. Split ranges on Lasso’s range error

4. What it costs