[Glass] How to deal with timeouts, fastcgi and what is the expected behavior?

[Dale Henrichs]

I do think that nginx will just skip over gems that are no longer present.

daemontools/monit are the tools you would use to restart dead/unresponsive
gems


On Fri, Jul 18, 2014 at 8:58 AM, Mariano Martinez Peck <
marianopeck at gmail.com> wrote:

>
> If we assume that when a gem goes down it is normally that the process is
> aborted (rather than an unresponding http server), then it would be nice if
> nignx could check if the process is alive (using PID or whatever) in order
> to forward the request to another upstream rather than checking via http
> timeout ...
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.gemtalksystems.com/mailman/private/glass/attachments/20140718/40e18c4e/attachment.html>