This document compiles a number of errors that appear in the network and how to solve them.
When mounting a volume for the p2p, you have to delete the local node_modules folder and run npm install inside the container. Otherwise some of the node libraries/dependencies are compiled against the host shared libraries that are not available inside the container or have a mismatched version. When that occurs you will see an error like the following:
p2p_1 | internal/modules/cjs/loader.js:846
p2p_1 | return process.dlopen(module, path.toNamespacedPath(filename));
p2p_1 | ^
p2p_1 |
p2p_1 | Error: The module '/root/enigma-p2p/node_modules/scrypt/build/Release/scrypt.node'
p2p_1 | was compiled against a different Node.js version using
p2p_1 | NODE_MODULE_VERSION 64. This version of Node.js requires
p2p_1 | NODE_MODULE_VERSION 67. Please try re-compiling or re-installing
p2p_1 | the module (for instance, using `npm rebuild` or `npm install`).
p2p_1 | at Object.Module._extensions..node (internal/modules/cjs/loader.js:846:18)
p2p_1 | at Module.load (internal/modules/cjs/loader.js:672:32)
p2p_1 | at tryModuleLoad (internal/modules/cjs/loader.js:612:12)
p2p_1 | at Function.Module._load (internal/modules/cjs/loader.js:604:3)
p2p_1 | at Module.require (internal/modules/cjs/loader.js:711:19)
p2p_1 | at require (internal/modules/cjs/helpers.js:14:16)
p2p_1 | at Object.<anonymous> (/root/enigma-p2p/node_modules/scrypt/index.js:3:20)
p2p_1 | at Module._compile (internal/modules/cjs/loader.js:805:30)
p2p_1 | at Object.Module._extensions..js (internal/modules/cjs/loader.js:816:10)
p2p_1 | at Module.load (internal/modules/cjs/loader.js:672:32)
One of the initial tests in 01_init.spec.js fails for no apparent reason, with a nondescriptive error such as a timeout, like the one included below:
● Init tests › should distribute ENG tokens
Timeout - Async callback was not invoked within the 5000ms timeout specified by jest.setTimeout.
57 | });
58 |
> 59 | it('should distribute ENG tokens', async () => {
| ^
60 | const tokenContract = enigma.tokenContract;
61 | let promises = [];
62 | const allowance = utils.toGrains(1000);
at Spec (node_modules/jest-jasmine2/build/jasmine/Spec.js:85:20)
at Suite.it (test/integrationTests/01_init.spec.js:59:3)
at Object.describe (test/integrationTests/01_init.spec.js:22:1)
This is usually caused by the compiled contracts being out of sync between the containers. These three containers: contract, client and p2p share a volume on the docker network named built_contracts that persists between network runs. The solution is to enter one of these containers and wipe out the contents of that shared folder and restart the network again, for example:
$ docker-compose exec client /bin/bash
root@client:~# cd enigma-contract/build/contracts/
root@client:~/enigma-contract/build/contracts# rm -rf *
When mounting the volume for either core or the key management (km) node, you may see an error like the following in the network output:
core_1 | ./start_core.bash: line 6: ./target/debug/enigma-core-app: No such file or directory
This means that you need to enter the container in question (either core or km) and trigger a build as follows, and then restart the network again:
core:
$ docker-compose exec core /bin/bash
root@core:~# cd enigma-core/enigma-core
root@core:~/enigma-core/enigma-core# make full-clean && make DEBUG=1
key managementnode:
$ docker-compose exec km /bin/bash
root@km:~# cd enigma-core/enigma-principal
root@km:~/enigma-core/enigma-principal# make full-clean && make DEBUG=1
You will see this error when all of the following occur:
- You are mounting a volume for
core - AND you are running in simulation mode (
SGX_MODE=SW) - AND you enter the container to build the binaries in HW mode (instead of simulation mode)
core_1 | 16:05:33 [INFO] LOG DERIVE: Err(SGX_ERROR_NO_DEVICE)
core_1 | thread 'main' panicked at '[-] Init Enclave Failed: SGX_ERROR_NO_DEVICE', libcore/result.rs:1009:5
core_1 | stack backtrace:
core_1 | 0: std::sys::unix::backtrace::tracing::imp::unwind_backtrace
core_1 | at libstd/sys/unix/backtrace/tracing/gcc_s.rs:49
core_1 | 1: std::sys_common::backtrace::print
core_1 | at libstd/sys_common/backtrace.rs:71
core_1 | at libstd/sys_common/backtrace.rs:59
core_1 | 2: std::panicking::default_hook::{{closure}}
core_1 | at libstd/panicking.rs:211
core_1 | 3: std::panicking::default_hook
core_1 | at libstd/panicking.rs:227
core_1 | 4: std::panicking::rust_panic_with_hook
core_1 | at libstd/panicking.rs:476
core_1 | 5: std::panicking::continue_panic_fmt
core_1 | at libstd/panicking.rs:390
core_1 | 6: rust_begin_unwind
core_1 | at libstd/panicking.rs:325
core_1 | 7: core::panicking::panic_fmt
core_1 | at libcore/panicking.rs:77
core_1 | 8: core::result::unwrap_failed
core_1 | at libcore/macros.rs:26
core_1 | 9: <core::result::Result<T, E>>::expect
core_1 | at libcore/result.rs:835
core_1 | 10: enigma_core_app::main
core_1 | at src/main.rs:27
core_1 | 11: std::rt::lang_start::{{closure}}
core_1 | at libstd/rt.rs:74
core_1 | 12: std::panicking::try::do_call
core_1 | at libstd/rt.rs:59
core_1 | at libstd/panicking.rs:310
core_1 | 13: __rust_maybe_catch_panic
core_1 | at libpanic_unwind/lib.rs:102
core_1 | 14: std::rt::lang_start_internal
core_1 | at libstd/panicking.rs:289
core_1 | at libstd/panic.rs:392
core_1 | at libstd/rt.rs:58
core_1 | 15: std::rt::lang_start
core_1 | at libstd/rt.rs:74
core_1 | 16: main
core_1 | 17: __libc_start_main
core_1 | 18: _start
The error is self-explanatory: the core binaries were compiled in Hardware mode, but no SGX device is available in the system (no /dev/isgx because it is running in simulation mode). The fix is straightforward, with the network up, execute:
$ docker-compose exec core /bin/bash
root@core:~# cd enigma-core/enigma-core
root@core:~/enigma-core/enigma-core# SGX_MODE=SW make DEBUG=1