Strange parsing error when storing code

[0xForerunner]

    let client = TendermintRPC::new(&chain_info.rpc_endpoint.clone().unwrap()).unwrap();
    let cosm_tome = CosmTome::new(chain_info, client);
    let wasm_data = std::fs::read(path)?;
    let tx_options = TxOptions { timeout_height: None, fee: None, memo: "wasm_deploy".into() };
    let req = StoreCodeRequest { wasm_data, instantiate_perms: None };
    let response = cosm_tome.wasm_store(req, &key, &tx_options).await?; // <== error occurs in this line

full error message

serde parse error

Caused by:
    missing field `id` at line 8 column 1

Location:
    /Users/ewoolsey/.cargo/registry/src/github.com-1ecc6299db9ec823/flex-error-0.4.4/src/tracer_impl/eyre.rs:10:9

and the trace logs

[2022-12-23T22:30:49Z TRACE hyper::proto::h1::conn] maybe_notify; read_from_io blocked
[2022-12-23T22:30:49Z TRACE want] signal: Want
[2022-12-23T22:30:49Z TRACE hyper::proto::h1::conn] flushed({role=client}): State { reading: Init, writing: Init, keep_alive: Idle }
[2022-12-23T22:30:49Z TRACE want] signal: Want
[2022-12-23T22:30:49Z TRACE hyper::proto::h1::conn] flushed({role=client}): State { reading: Init, writing: Init, keep_alive: Idle }
[2022-12-23T22:30:49Z TRACE hyper::client::pool] put; add idle connection for ("http", 167.99.177.244:26657)
[2022-12-23T22:30:49Z DEBUG hyper::client::pool] pooling idle connection for ("http", 167.99.177.244:26657)
[2022-12-23T22:30:49Z DEBUG tendermint_rpc::client::transport::http::sealed] Incoming response: {
      "jsonrpc": "2.0",
      "error": {
        "code": -32600,
        "message": "Invalid Request",
        "data": "error reading request body: http: request body too large"
      }
    }
serde parse error

Caused by:
    missing field `id` at line 8 column 1

Location:
    /Users/ewoolsey/.cargo/registry/src/github.com-1ecc6299db9ec823/flex-error-0.4.4/src/tracer_impl/eyre.rs:10:9

Everything works fine with small .wasm files < 300kB

Large ones > 530kB cause this error. The exact same message is working on a different build using tendermint_rpc 0.27

[de-husk]

This is a tricky problem, because this is controlled by a tendermint node config variable: max_body_bytes. So it's up to each RPC node to configure.

If you have really large optimized wasms and you cant configure this value, then you should try to use the CosmosgRPC client instead of the tendermint RPC, since it looks like the default is 10MB for gRPC.

This is a really bad error though.

[0xForerunner]

That’s surprising to me, since the exact same transaction using the HttpClient with tendermint_rpc v0.27 does not yield this issue. (Same node as well) are you certain an upgrade wouldn’t fix the problem?-- Cheers, Eric Woolsey

[de-husk]

Yeah, I reproed this issue locally and "fixed it" by increasing max_body_bytes and restarting the local juno node.

since the exact same transaction using the HttpClient with tendermint_rpc v0.27 does not yield this issue.

Interesting, well that's cool that an upgrade works for you, we def want to bump cosm-rs versions soon either way. I suppose they must have added some changes that decreased the body size of your store msg under the limit of the RPC node you are using.