Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

have you seen naught? #3

Open
andrewrk opened this issue Aug 12, 2013 · 8 comments
Open

have you seen naught? #3

andrewrk opened this issue Aug 12, 2013 · 8 comments

Comments

@andrewrk
Copy link

https://github.com/superjoe30/naught

looks like we should merge projects, yes?

@spion
Copy link
Contributor

spion commented Aug 12, 2013

Yes, I've seen it. Its an excellent and well thought-out solution, but I don't think we could merge the projects right now, since I can't have recluster handling logging. This is because I use nac to handle logging, configuration and deployment.

I made this separation because I wanted nac to be able to handle all kinds of processes. Clustering needs are different in different situations: there is socket.io-compatible clustering using a proxy with up, there is simple clustering without any socket sharing for worker processes (e.g. stuff that uses rabbitmq), and then there are non-node processes which can't communicate using process.send, and I wanted nac to be compatible with all of them.

However I like that naught allows you to change environment variables on the fly. Have you considered not handling logging in naught?

@andrewrk
Copy link
Author

naught does not force its logging handling upon you. Simply avoid writing to stdout and stderr and no logs will be generated.

Thanks for explaining the difference. I'll put a link to this project in naught's readme and explain the use cases when you would want to use recluster instead of naught.

@spion
Copy link
Contributor

spion commented Aug 12, 2013

Yes, but I do want to handle stdout and stderr (e.g. in cases of uncaught exceptions), and I do handle them both with nac. So recluster simply pipes worker stdout/stderr to its own stdout/stderr, which are then captured by nac.

Basically what I meant was, is it possible to add something like
--nodaemon --stdout - --stderr -

... which would make naught run in non-daemon node and make stdout/stderr pass through allowing another process to handle them? If you did that, then there would be absolutely no need for recluster (except wanting to avoid IPC files, which is not much of a reason...)

@andrewrk
Copy link
Author

Yes I think it makes sense for naught to support those options.

@andrewrk
Copy link
Author

the stdout and stderr can be resolved like this: andrewrk/naught#18

--nodaemon is already an issue worth solving: andrewrk/naught#27

@andrewrk
Copy link
Author

oh, nvm about stdout and stderr. I understood what you said but then immediately forgot. sorry, bad short term memory. I think we can do what you said.

@sandinmyjoints
Copy link
Contributor

For the record, I also liked a lot of what I saw in naught, but want to keep control over stdout, because I am already handling logging and log rotation.

@andrewrk
Copy link
Author

andrewrk commented Jan 4, 2014

just a heads up - naught now supports --daemon-mode false as well as --stdout - and --stderr -.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants