Keepalive
baugarten opened this issue · comments
I'm looking at the documentation and I don't see any way to specify that we should restart the server on crash (such as in a production setting). Is there any way to do this?
Oh, interesting!
Is the scenario that you have some code that works fine when the server/task starts, but then you, for example, hit an end-point, a 500 is triggered, an error is thrown, and you want the server to come back up?
Sounds plausible & doable, and most importantly, testable!
:) the plan isn't really to error but for a variety of reasons a node server can crash in production (untested code path, out of memory, etc) so I would like to start my server knowing that if it dies for some strange reason, it will come back to life
Sent with my thumbs
On Jun 17, 2014, at 9:46 PM, Eric Clemmons notifications@github.com wrote:
Oh, interesting!
Is the scenario that you have some code that works fine when the server/task starts, but then you, for example, hit an end-point, a 500 is triggered, an error is thrown, and you want the server to come back up?
Sounds plausible & doable, and most importantly, testable!
—
Reply to this email directly or view it on GitHub.
Personally, I use forever (https://github.com/nodejitsu/forever) in production, since grunt nor grunt-express-server is really best suited for production environments.
My npm start is pretty much grunt build optimize && forever restart app/cluster.js.
Do you really want to have grunt running your process on production?
That's actually a good point. I don't. I'm going to go ahead and see myself out. Cheers!
I have no problem with restarting express via this task, but I wanted to be sure of the use-case before implementing it.
Thanks for the discussion!