Increassing timeouts

Eric Wong normalperson at
Tue Jan 15 19:44:08 UTC 2013

Manuel Palenciano Guerrero <mpalenciano at> wrote:
> Hello there
> We use Unicorn in all of our production/staging web-apps, and we are
> very happy with it. But we also have one of the apps talking to a
> web-service via SOAP, and this web-service now takes longer than usual
> processing requests. To solve this situation we have increased the
> time-outs for unicorn and nginx (proxy_read_timeout and
> proxy_connect_timeout) from 30 to 120, and everything works fine this
> way (although I don't like it).
> My question is: having increased the time-out to 120 means this app is
> now considered slow (or sleepy) and should be run by Rainbow ? or is
> 120 still ok for Unicorn ?

As usual, it depends.  If clients hitting other endpoints are able to
access your site, you're fine...

But seriously, who will wait 120s for a website to load?

More information about the mongrel-unicorn mailing list