eth_call RPC call hangs

[ruuda]

SUMMARY

The following call hangs indefinitely. This happens both on the http RPC port, and when using a websocket.

$ curl -v http://«address-removed»:8545 -H 'Content-Type: application/json' -d '{"jsonrpc":"2.0","id":2,"method":"eth_call","params":[{"data":"0x1cfe7951","from":"0x0000000000000000000000000000000000000000","to":"0xfe8cd454b4a1ca468b57d79c0cc77ef5b6f64585"},"latest"]}' --speed-time 30
*   Trying «address-removed»...
* Connected to «address-removed» port 8545 (#0)
> POST / HTTP/1.1
> Host: «address-removed»:8545
> User-Agent: curl/7.82.0
> Accept: */*
> Content-Type: application/json
> Content-Length: 188
> 
* Operation too slow. Less than 1 bytes/sec transferred the last 30 seconds
* Closing connection 0
curl: (28) Operation too slow. Less than 1 bytes/sec transferred the last 30 seconds

In the logs, we can see this immediately after initiating the call:

{"args":{"from":"0x0000000000000000000000000000000000000000","to":"0xfe8cd454b4a1ca468b57d79c0cc77ef5b6f64585","gas":null,"gasPrice":null,"maxFeePerGas":null,"maxPriorityFeePerGas":null,"value":null,"nonce":null,"data":"0x1cfe7951","input":null},"block_or_hash":{"blockNumber":"latest"},"caller":"api.go:327","level":"debug","method":"eth_call","module":"eth_rpc","msg":"request","ts":"2022-03-23T13:51:19.575280494Z"}

Afterwards it also logs this, but I’m not sure it is related:

{"caller":"indexer.go:150","latest_indexed":642146,"latest_round":809552,"level":"debug","module":"gateway/indexer","msg":"checking health","runtime_id":"000000000000000000000000000000000000000000000000e2eaa99fc008f87f","ts":"2022-03-23T13:51:20.34231545Z"}
{"caller":"indexer.go:150","latest_indexed":642146,"latest_round":809554,"level":"debug","module":"gateway/indexer","msg":"checking health","runtime_id":"000000000000000000000000000000000000000000000000e2eaa99fc008f87f","ts":"2022-03-23T13:51:30.368776278Z"}

Our log level is set to debug.

ISSUE TYPE

• Bug Report

STEPS TO REPRODUCE

For us this reproduces 100% reliably, but we know of others who are running this gateway where the call does work; it’s unclear what the problem is, but I don’t know how to narrow this down any further.

This could be related to #161 as well, which looks like a similar hang.

ACTUAL RESULTS

Nothing happens. With a websocket, there is still no response after 5 minutes. With Curl without --speed-time (which specifies a timeout in seconds), it also just hangs indefinitely, until I kill it after ~10 minutes.

Describe what actually happened. If possible run with extra verbosity (-vvvv)

If I run emerald-web3-gateway with -vvvv, it just prints its version. But we do run with

log:
  level: debug
  format: json

in the config.

EXPECTED RESULTS

We get a response to the RPC call within a few seconds. Or at least some log message about what is going on.

[ptrus]

This is likely an issue on the oasis-node side. Can you provide the output of running:

oasis-node control status -a unix:<path-to-oasis-node-internal.sock>

and logs from the oasis-node around the time of the eth_Call request.