Skip to main content

RPC nodes

RPC node monitoring that confirms before it alerts

Pingara sends real JSON-RPC requests to your nodes from outside your network, checks the answer, and alerts you when an endpoint stops responding or starts returning errors.

RPC nodes

Why RPC, validator and bridge endpoints fail quietly

  • The node process is up but the HTTP endpoint returns an error body with status 200.
  • The provider rate-limits, or a load balancer drops one region.
  • The WebSocket endpoint for subscriptions dies while HTTP still works.
  • A certificate or DNS change breaks access for everyone except you.

Checks

What Pingara checks

  • HTTP POST with a JSON body

    Custom method, headers and body on every plan. The body can be any JSON-RPC method you are allowed to call.
  • Keyword on the response

    The response must contain a string, such as "result", or must NOT contain one, such as "error". It is a substring check.
  • WebSocket

    Pingara verifies the handshake, can then send a message you choose (for example a JSON-RPC subscribe) and fail the check if the reply does not contain the text you expect.
  • Request size

    Headers up to 4 KB, request body up to 8 KB. A JSON-RPC call fits.
  • TCP, ping and DNS

    TCP for a P2P or node port. Ping and DNS when the thing you care about is not an HTTP service.
  • Regions and how often

    All 4 regions on Pro, 1 region on Free. Every 30 seconds on Pro, every 5 minutes on Free. US East (N. Virginia), US West (Oregon), Europe (Ireland) and Asia Pacific (Singapore).
  • Quorum on Pro

    On Pro, a monitor goes down only when enough regions agree.
  • Timing and Degraded

    Each check records DNS, connect, TLS, time to first byte and total time. A monitor can be marked Degraded when responses stay slow.
  • Status page and summaries

    A public status page for the RPC service, and AI-written incident summaries.

Not offered: block-height or chain-head lag, or sync checks.

Examples

Example monitor setups

Illustrations. The hosts below are examples, not monitors Pingara runs for you.
  • EVM RPC health

    POST https://rpc.example.com. Header Content-Type: application/json. Body {"jsonrpc":"2.0","method":"eth_chainId","params":[],"id":1}. Expected status 200. Keyword must contain "result". Mark the monitor down. All 4 regions on Pro. Every 30 seconds on Pro, every 5 minutes on Free.

  • Error guard

    The same request. Keyword must NOT contain "error". Mark the monitor degraded.

  • WebSocket subscriptions

    WebSocket monitor on wss://rpc.example.com/ws. Send {"jsonrpc":"2.0","id":1,"method":"eth_subscribe","params":["newHeads"]}. Expect a reply containing "result".

  • Validator or P2P port

    TCP monitor on node.example.com port 30303. The port is an example.

  • Bridge API

    GET https://bridge.example.com/health. Keyword must contain ok.

  • A status page

    Publish the endpoints you want users to see on a public status page.

FAQ

Questions

Can Pingara send JSON-RPC requests?

Yes. HTTP monitors support POST with custom headers and a JSON body on every plan.

Does Pingara check block height or sync status?

No. It checks that the endpoint responds correctly and in time, not chain state.

Can I monitor WebSocket RPC endpoints?

Yes. A WebSocket monitor checks the handshake and can send a subscribe message and check the reply contains the text you expect.

My provider URL or header contains an API key. Is that safe?

Header values and URLs are part of the monitor settings, which everyone in your organisation can see. Use a restricted or read-only key.

How often can I check?

Every 30 seconds on Pro, every 5 minutes on Free.

Where are checks run from?

All 4 regions on Pro, 1 region on Free: US East (N. Virginia), US West (Oregon), Europe (Ireland) and Asia Pacific (Singapore). Free uses one region you choose.

What does it cost?

Free for one monitor; Pro is $9/month for 50 monitors.

RPC nodes

Start with one endpoint

Create the monitor on the free plan, or scan a site you already run.