First case - run `embark run` which starts a blockchain node, then manually kill the `geth` process. Would throw `{ [Error: connect ECONNREFUSED 127.0.0.1:8543] message: 'connect ECONNREFUSED 127.0.0.1:8543', code: -32603 }` error and ruins the dashboard.
Second case, 1) run `embark blockchain` 2) run `embark run` 3) kill `embark blockchain` throws the error `{ [Error: connect ECONNREFUSED 127.0.0.1:8543] message: 'connect ECONNREFUSED 127.0.0.1:8543', code: -32603 }` and ruins the dashboard.
The first case was solved by having the child blockchain process that spawns geth listen for geth exit, then kill itself.
The second case required updating of `eth-block-tracker` to v4.0.1 inside of the `embark-web3-provider-engine`. v4.0.1 was a major version update and introduced breaking changes. Those changes were handled inside of `embark-web3-provider-engine`, covered in **blocker** PR https://github.com/jrainville/provider-engine/pull/1.
We've introduced a regression in 6d75a4c6c4 where running
`embark run` breaks due to webserver options not being defined.
The reason this happened is because in the mentioned commit, we're checking given
webserver options with an identity check against `null` and only set dedicated
options, if that condition is true.
E.g.:
```
if (options.runWebserver !== null) {
webServerConfig.enabled = options.runWebserver;
}
```
Unfortunately, due to one of JavaScript design flaws, this condition behaves totally
different when no identity check is done, which causes Embark to break.
Meaning that in JavaScript:
```
undefined != null // false
undefined !== null // true
```
In other words, after the mentioned commit, the condition is true, resulting
in several webserver configurations to be set to `undefined`.
This conflicts at runtime when we compose webserver configurations out of user input
and Embark's defaults here: dc5de7c1b4/lib/core/config.js (L392)
In other words, since webserver config options are already set to `undefined`,
they don't get overridden by our config algorithm.
This commit ensures that webserver config values are not set when their values
are either `null` or `undefined`.
Removed a condition that detects if `mineWhenNeeded === true`. While reaslistically, the `defaultAccount` will only be populated when `mineWhenNeeded` is set, there is no need to check for this when falling back to the blockchain account.
When there is no account/password specified for swarm, there was an error shown in the logs (asking for address/password), and the swarm process would quit.
This PR changes the behaviour so that if a swarm address/password are not specified in the config, it attempts to use the blockchain address/password specified in `config/blockchain > account`. If `config/blockchain > account > address` doesn’t exist, the first account controlled by the node is used (provided by `web3.eth.getAccounts`, along with the password from `config/blockchain > account > password`.
Cargo in fact, bundles up subsequent tasks in to an array, so any tasks that are not immediately run get bundled in to another run later. This helps when lots of changes have been made in a short period of time.
Due to a process being spawned for every pipeline run, concurrency should remain around 3 to keep number of child processes from running away and allowing the CPU to stay on top.
For file changes that do not require a webpack run, ie HTML, the assets will still be copied to the output directory, but webpack will not run (as it’s too slow).