EthProvider unreachable in simulation tests

[sangaman]

I don't know how my local simulation tests got in this state exactly, I did a docker system prune -a and then npm run test:sim:build and had some early tests fail due to bugs in my new test cases, now after fixing those bugs I always get errors like below when running simulation tests. Another system prune or clean might fix it but I wanted to open an issue for this so we can track it or maybe prevent it from happening again, or at least figure out what happened here.

Any thoughts @erkarl @LePremierHomme ?

~/Documents/GitHub/xud/test/simulation/temp/indra ~/Documents/GitHub/xud/test/simulation
bash ops/start-indra.sh
n3qd1y1ll8hdu0atrfbeuubgh
Using docker images indra_name:latest 
Proxy configured
Node configured
Created secret called indra_database_dev with id q2re0q8d2m09bj84po7et3qby
Database configured
Nats configured
No $INDRA_CHAIN_PROVIDERS provided, spinning up local testnets & using those.
A secret called indra_mnemonic_dev already exists, skipping secret setup
Starting testnet_1337 & testnet_1338..
Running dev-mode image for testnet 1338: indra_builder
Unable to find image 'indra_builder:latest' locally
docker: Error response from daemon: pull access denied for indra_builder, repository does not exist or may require 'docker login': denied: requested access to the resource is denied.
See 'docker run --help'.
make: *** [Makefile:48: start] Error 125
~/Documents/GitHub/xud/test/simulation
Creating simulation_gomod_1   ... done
Creating simulation_xud_1     ... done
Creating simulation_btcd_1    ... done
Creating simulation_lnd_1     ... done
Creating simulation_connext_1 ... done
=== RUN   TestIntegration
2020/11/03 16:01:59 EthProvider reachability failure: Post http://172.17.0.1:8545: dial tcp 172.17.0.1:8545: connect: connection refused
exit status 1
FAIL	github.com/ExchangeUnion/xud-simulation	0.059s

docker-run.sh: test failed, re-running... (1/3)

Starting simulation_xud_1     ... done
Starting simulation_connext_1 ... done
Starting simulation_gomod_1   ... done
Starting simulation_lnd_1     ... done
Starting simulation_btcd_1    ... done
^Cgo build github.com/ethereum/go-ethereum/crypto/secp256k1: /usr/local/go/pkg/tool/linux_amd64/cgo: signal: interrupt

docker-run.sh: test failed, re-running... (2/3)

Starting simulation_lnd_1     ... done
Starting simulation_connext_1 ... done
Starting simulation_btcd_1    ... done
Starting simulation_xud_1     ... done
Starting simulation_gomod_1   ... done
=== RUN   TestIntegration
2020/11/03 16:02:55 EthProvider reachability failure: Post http://172.17.0.1:8545: dial tcp 172.17.0.1:8545: connect: connection refused
exit status 1
FAIL	github.com/ExchangeUnion/xud-simulation	0.082s

[ghost]

Just curious - do you also have indra cloned/previously running locally?

[sangaman]

Previously yes but I haven't run it or done anything with it in a while, and it should've been cleaned by pruning docker.