--restart-tries interaction with --kill-others

[mawhidby]

Hey there!

I'm trying to use concurrently on our CI server to run our integration tests. I'm running into a similar issue as that documented in #177, the difference being I'm also using restartTries to help ensure the server is up before the integration tests run. With this setup, the tests pass, successfully killing the server process, but it restarts the server because of restartTries.

concurrently --kill-others --success first --restart-tries 2 --restart-after 5000 "npm start" "npm test"

...

[1] npm test exited with code 0
--> Sending SIGTERM to other processes..
[0] 2018-12-14T03:31:02.097Z - warn:    Caught SIGTERM, exiting
[0] npm start exited with code SIGTERM
[0] npm start restarted

Any suggestions around this? I'm not sure if this is a supported use case, otherwise it seems like I'll need to modify our tests themselves to ensure the server is up and running

[mawhidby]

I guess one way around this would be to have restart-process.js explicitly check for the SIGTERM, but I'm not sure about the side-effects this may cause

handle(commands) {
        if (this.tries === 0) {
            return commands;
        }

        commands.map(command => command.close.pipe(
            take(this.tries),
            takeWhile(code => code !== 0 && code !== 'SIGTERM')
...

[runloop]

I ran into this exact issue. My solution was to write a script that delays the start of my other process long enough to ensure the server is up and running before it's executed.

[marklagendijk]

For starting a server and then running tests you can use start-server-and-test. If you need to run multiple servers you can use concurrently to start all of them.