andig

It could probably not in a general case- we simply don't know what config files you'd like to have watched. Or should we add any type of yaml?

This looks interesting for Linux: https://github.com/tsufeki/react-filesystem-monitor, I've also opened a ticket for OSX (https://github.com/tsufeki/react-filesystem-monitor/issues/2)

How shall we proceed with this one?

I think we used to have an interface for that on the bootstrap but removed it as it was lacking an actual use case. If you have a concrete suggestion...

Did you add the additional logging statements yourself?

Related to #470?

Ping @marcj I could imagine that degrading performance at higher concurrency lavels might be due to only having a single incoming process for all connections? What really discomforts me again...

@jcheron regarding your config you should adjust the number of workers with the level of parallelism. Each worker will only handle a single request at a time. Once you have...

Do you see any chance of adding logging in the bridge? Does the bridge even receive those stray requests or does the redponse somehow get stuck?

Then the next step would be to find out if the slave even receives the request or if it gets stuck in the master. Could you also find out if...