From wyhaines at gmail.com Wed May 23 17:37:11 2007 From: wyhaines at gmail.com (Kirk Haines) Date: Wed, 23 May 2007 14:37:11 -0700 Subject: [Swiftiply-users] Breaking the ice -- deployment scenarios Message-ID: I figured I'd kick off traffic on the mailing list with a request for input. I want to put together a documentation section about various deployment scenarios. nginx -> evented_mongrel nginx -> swiftiply -> swiftiplied_mongrel nginx / swiftiply -> swiftiplied_mongrel etc.... Does anyone have any _in use_ configs that I could turn into examples for these or similar setups with other components (lighttpd, apache, etc...)? Thanks, Kirk Haines P.S. does anyone have a URL for the Sun hardware for open source development promotion? I'm having a heck of a time finding it. From ezmobius at gmail.com Wed May 23 17:52:04 2007 From: ezmobius at gmail.com (Ezra Zygmuntowicz) Date: Wed, 23 May 2007 14:52:04 -0700 Subject: [Swiftiply-users] Breaking the ice -- deployment scenarios In-Reply-To: References: Message-ID: On May 23, 2007, at 2:37 PM, Kirk Haines wrote: > I figured I'd kick off traffic on the mailing list with a request > for input. > > I want to put together a documentation section about various > deployment scenarios. > > nginx -> evented_mongrel > > nginx -> swiftiply -> swiftiplied_mongrel > > nginx / > swiftiply -> swiftiplied_mongrel > > etc.... > > Does anyone have any _in use_ configs that I could turn into examples > for these or similar setups with other components (lighttpd, apache, > etc...)? > > > Thanks, > > Kirk Haines > Kirk- Here is a production nginx.conf that can be used with evented_mongrel. It currently runs 3 mongrels on port 5000-5002 and is easily editable to add or remove mongrels. THis config does the right thing for rails and merb page caching and all that jazz. To use this config with swiftiply all you would need to do is use only one upstream server and point it to swiftiplied's front end port. http://pastie.caboo.se/63986 Cheers- -- Ezra Zygmuntowicz -- Lead Rails Evangelist -- ez at engineyard.com -- Engine Yard, Serious Rails Hosting -- (866) 518-YARD (9273) From wyhaines at gmail.com Fri May 25 02:34:32 2007 From: wyhaines at gmail.com (Kirk Haines) Date: Thu, 24 May 2007 23:34:32 -0700 Subject: [Swiftiply-users] Breaking the ice -- deployment scenarios In-Reply-To: References:

Message-ID: On 5/23/07, Ezra Zygmuntowicz wrote: > Here is a production nginx.conf that can be used with > evented_mongrel. It currently runs 3 mongrels on port 5000-5002 and > is easily editable to add or remove mongrels. THis config does the > right thing for rails and merb page caching and all that jazz. Thanks, Ezra. I'll get a common configurations section added to the documentation tomorrow. Anyone else have any non-nginx configs that they want to throw my way? Thanks, Kirk From cs327 at teknolabs.com Tue May 29 20:40:42 2007 From: cs327 at teknolabs.com (Cagan) Date: Wed, 30 May 2007 03:40:42 +0300 Subject: [Swiftiply-users] unable to get swiftiply working Message-ID: <275275D2-C66F-4550-8B0D-A46832268C5E@teknolabs.com> Hello, I was trying to get swiftiply with a mongrel-cluster working but no matter what I've tried I couldn't get it going. Here's the setup: SERVER 1: LoadBalancer (ULtraMonkey) SERVER 2: swiftiply + mongrel_cluster SERVER3: swiftiply + mongrel cluster I got the switfiply and mongrel cluster working but when I hit the website, I can't get through. My guess is a comm problem between the loadbalancer and the app servers due to some wrong swiftiply config I used. HEre's my config file: cluster_address: 0.0.0.0 cluster_port: 80 daemonize: true map: - incoming: - 127.0.0.1 outgoing: 127.0.0.1:30000 default: true I'd appreciate your help on this. Best, Cagan -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/swiftiply-users/attachments/20070530/f5867061/attachment.html From wyhaines at gmail.com Thu May 31 18:48:52 2007 From: wyhaines at gmail.com (Kirk Haines) Date: Thu, 31 May 2007 15:48:52 -0700 Subject: [Swiftiply-users] unable to get swiftiply working In-Reply-To: <275275D2-C66F-4550-8B0D-A46832268C5E@teknolabs.com> References: <275275D2-C66F-4550-8B0D-A46832268C5E@teknolabs.com> Message-ID: On 5/29/07, Cagan wrote: > Hello, Sorry for the delay in replying. In the middle of moving into a new home. > I got the switfiply and mongrel cluster working but when I hit the website, > HEre's my config file: > > cluster_address: 0.0.0.0 > cluster_port: 80 > daemonize: true > map: > - incoming: > - 127.0.0.1 > outgoing: 127.0.0.1:30000 > default: true It's probably that incoming declaration. Are the requests really to http://127.0.0.1/foo.html? The incoming config item is basically a virtual hosting config. It should have the domain name or names that actually appear in the requested URL. So, for example, if you app's url is http://foo.bar.com/index.html then you would have: - incoming: - foo.bar.com That's probably the problem. Kirk Haines From cs327 at teknolabs.com Thu May 31 19:07:03 2007 From: cs327 at teknolabs.com (Cagan) Date: Fri, 1 Jun 2007 02:07:03 +0300 Subject: [Swiftiply-users] unable to get swiftiply working In-Reply-To: References: <275275D2-C66F-4550-8B0D-A46832268C5E@teknolabs.com> Message-ID: <354B96AC-F53A-4B0F-8F93-AFF0CBEA8794@teknolabs.com> I tried it with the proper url (www.mondus.net)...But that didn't work either...I think the problem might be with mongrel. How do the workers know how to contact swiftiply? Thanks, Cagan On Jun 1, 2007, at 1:48 AM, Kirk Haines wrote: > On 5/29/07, Cagan wrote: >> Hello, > > Sorry for the delay in replying. In the middle of moving into a > new home. > >> I got the switfiply and mongrel cluster working but when I hit the >> website, >> HEre's my config file: >> >> cluster_address: 0.0.0.0 >> cluster_port: 80 >> daemonize: true >> map: >> - incoming: >> - 127.0.0.1 >> outgoing: 127.0.0.1:30000 >> default: true > > It's probably that incoming declaration. Are the requests really to > http://127.0.0.1/foo.html? > > The incoming config item is basically a virtual hosting config. It > should have the domain name or names that actually appear in the > requested URL. > > So, for example, if you app's url is http://foo.bar.com/index.html > then you would have: > > - incoming: > - foo.bar.com > > That's probably the problem. > > > Kirk Haines From sandrods at gmail.com Thu May 31 19:34:07 2007 From: sandrods at gmail.com (Sandro Duarte) Date: Thu, 31 May 2007 20:34:07 -0300 Subject: [Swiftiply-users] eventmachine problem Message-ID: <8133b8010705311634q2f51d0f1la01fbc354f96c923@mail.gmail.com> Hi... When I start swiftiply : swiftiply -c sw.conf I get a error like this: /usr/lib/ruby/gems/1.8/gems/eventmachine-0.7.2/lib/eventmachine.rb:453:in `start_tcp_server': no acceptor (RuntimeError) from /usr/lib/ruby/gems/1.8/gems/eventmachine-0.7.2/lib/eventmachine.rb:453:in `start_server ... sw.conf contains: cluster_address: 0.0.0.0 cluster_port: 80 daemonize: true map: - incoming: 127.0.0.1 outgoing: 127.0.0.1:8000 default: true Any ideas? -- Sandro Duarte -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/swiftiply-users/attachments/20070531/1f92332a/attachment.html