elsirion

> You are referring to this integration test right? https://github.com/fedimint/minimint/blob/0a2b472440e9226c15d7828f069f68c290a74506/scripts/integrationtest.sh Correct, the other one is really new, ideally both would test this. > https://nigiri.vulpem.com/ I think we should make use...

Cannot reproduce. Are you sure you didn't generate the `federation_client.json` with a different version of MiniMint than when you ran `gw_configgen`? That could explain the error (mismatching config structs between...

I think we should separate the two components: * **Federation server**: requires `fedimintd` from our side and `bitcoind` (should probably be supplied optionally, some users might already have one running)....

> > Yes, although in the future it might be problem for generating TxOutProofs for users if the tx is old. > > Doesn't the client always generate the `TxOutProof`?...

> The integrations tests fail when I run them locally on Apple M1 https://gist.github.com/justinmoon/80d044c54935038966c35e8111ca07b9 I don't think the test actually fails (see line 2068), but the cleanup afterwards fails (line...

Maybe what we actually want is a single trait function `call(path: &str, arg: serde_json::Value) -> serde_json::Value` that has to be implemented and then just impl the remaining functions as default...

We should also indicate that the payment succeeded.

* `'DB error'` needs a separate effort imo, see #374 * Sanity checks for configs could arguable be handled better using errors and printing them. `expect`ing was just a 80/20...

> Why can't the fedimintd endpoint be directly fuzzed instead of the handlers? Wouldn't that increase latency while possibly reducing the capability of the fuzzer to learn because now there's...

Was there any more context given for the error? Maybe in the previous lines?