Raze.bot Developers

Every failure is {"error": {"code", "message"}} with a matching HTTP status. The message says what to fix.

HTTP	`error.code`	Meaning	Retry
400	`bad_request`	A parameter is missing or wrong: unknown chain, invalid address (on Solana, one that is not a 32-byte key; see Addresses), interval out of the list, more than 50 ids, a cursor the server can no longer honour	No: fix the request, or start again from the first page
401	`unauthorized`	Missing or unknown key	No: fix the key
404	`not_found`	Unknown route, unknown board, or a token no source knows	No for a bad path; later for a token that may not be indexed yet
429	`rate_limited`	Over the per-minute cap of the key	Yes, after the minute turns
502	`upstream_unavailable`	A data source did not answer	Yes, with backoff
503	`upstream_unavailable`	A Solana trades, candles or wallet-trades request cannot be served right now	Yes, after `Retry-After` (2 s): the same request, same cursor
503	`server_busy`	A new WebSocket or SSE connection, while the server is at capacity	Yes, after `Retry-After` (30 s)

{ "error": { "code": "bad_request", "message": "'abc' is not a Solana address" } }

{ "error": { "code": "not_found", "message": "token not found" } }

404 on GET /v1/tokens/{chain}/{address} means no source knows the token. 502 means a source did not answer this time: retry, and do not treat it as “does not exist”. When a source fails, the API keeps answering from its last good copy for a couple of minutes before it returns 502.

Both 503s above carry a Retry-After header, in seconds. Wait that long, then send the same request again: for a page, with the same cursor. A cursor that keeps failing (a 400, or a 502 again and again) while the first page loads will not heal: start the walk again from the first page. See Pagination.

Live errors

On the WebSocket, a wrong subscribe or op answers a frame of type error, with the id of the subscription:

{ "type": "error", "id": "bad", "error": { "code": "bad_request", "message": "token is required" } }

Message	Cause
`channel is required`, `unknown channel '…'`	Missing or wrong `channel`
`chain is required`, `token is required`, `wallet is required`	A parameter the channel needs
`'…' is not a Solana address` (or another chain’s), `wallet is not a Solana address`	`token` or `wallet` is not an address of that chain (on Solana, it must be a 32-byte key)
`board must be new, graduating or graduated`	`launchpad` without a valid `board`
`the wallet channel is Solana only for now`	`wallet` on an EVM chain
`id already in use`	Subscribing twice with the same `id`; unsubscribe it first
`at most 200 subscriptions per socket`	Open a second socket
`op must be subscribe, unsubscribe or ping`, `frames must be JSON`	Malformed op

A channel that cannot load its first data (for example a token channel on an unknown token) sends an error frame with the same code the REST route would give. On SSE, wrong parameters answer 400 JSON before the stream starts. While the server is at capacity, it refuses new connections, before the WebSocket upgrade or before the SSE stream starts, with 503 and Retry-After: 30:

{ "error": { "code": "server_busy", "message": "the server is at capacity, retry in 30 s" } }

Sockets and streams already open keep running, and an open socket can still subscribe. A browser WebSocket cannot read the status: it only sees the connection fail. An EventSource does not retry a 503 on its own. Reconnect after 30 s, with backoff.

Changelog

Errors

Live errors

​Live errors

Live errors