Running a subgraph and hitting rate limits.

[Logic-Beach]

Hi team, I am running an indexer which grabs the latest block number, then hits my RPC with an eth_getTransactionReceipt request for every tx in the block. This is causing a bunch of 429 errors since the 1000s of requests hit the RPC all at once. Is there any way around this?

[matthewdarwin]

Graph-node will slow down when it gets a 429 response.

The solution to do fast indexing is to run your own RPC node. Or use firehose.

[fordN]

Graph-node now supports fetching all receipts in a single request using eth_getBlockReceipts (graphprotocol/graph-node#5341). That should help with the request volume.