I've finally gotten lighttpd working for my production instance (SUSE 10.0 with SSL). (Had to install the fcgi dev toolkit, the SUSE 10 ruby-fcgi rpm, and the fcgi gem. I'm not positive all 3 were required, but that is what I did.) Now the devel instance of my app is broken. I'm not sure how to fix it. Previously I was using WEBrick for both devel and production. (their on the same box). I only have the lighttpd config file setup to support my production instance. Now when I try to run rails/app.devel/script/server it appears to try and run lighttpd. Unfortunately this fails. Do I need to update my lighttpd config to support devel, or somehow force script/server to launch WEBrick? Thanks Greg -- Greg Freemyer The Norcross Group Forensics for the 21st Century
on 2006-02-16 00:42
on 2006-02-16 01:12
On Wednesday, February 15, 2006, at 6:41 PM, Greg Freemyer wrote: > >support devel, or somehow force script/server to launch WEBrick? >http://lists.rubyonrails.org/mailman/listinfo/rails We tried to use one file for both production and development but it never worked very well. Right now, for DevLists.com there is config/ lighttpd.conf for development and config/lighttpd-devlists.conf for production. I like having app name in the production config file name -- we run multiple instances of lighttpd and this makes it easier to kill the right one when needed. Best Regards, Roustem.
on 2006-02-16 01:12
On Feb 15, 2006, at 3:41 PM, Greg Freemyer wrote: > > support devel, or somehow force script/server to launch WEBrick? > Rails@lists.rubyonrails.org > http://lists.rubyonrails.org/mailman/listinfo/rails > Greg- What version of l;ighttpd are you using? The recent versions ( >=1.4.9 ) cannot use relative paths in the config file. So the lighttpd.conf file in your config directory needs to be edited and add full absolute paths to your public dir and your dispatch.fcgi. Cheers- -Ezra Zygmuntowicz Yakima Herald-Republic WebMaster http://yakimaherald.com 509-577-7732 firstname.lastname@example.org
on 2006-02-16 01:19
On 16 Feb 2006 00:10:14 -0000, Roustem Karimov <email@example.com> wrote: > > > > > >Rails mailing list > right one when needed. > > Best Regards, Roustem. Thanks, I can do that easily enough for my standard instances (production, QA, devel). OTOH, with WEBrick I was creating little sandboxes all the time. I would simply make a copy of my devel setup and run script/server --port nnn, where nnn was a unique number. Is there a simple way to make script/server invoke WEBrick, or some other way to easily launch WEBrick manually? Thanks Greg -- Greg Freemyer The Norcross Group Forensics for the 21st Century
on 2006-02-16 01:22
On Wednesday, February 15, 2006, at 7:16 PM, Greg Freemyer wrote: >_______________________________________________ >Rails mailing list >Rails@lists.rubyonrails.org >http://lists.rubyonrails.org/mailman/listinfo/rails It is easy: ./script/server webrick Best Regards, Roustem.
on 2006-02-16 01:28
On 2/15/06, Ezra Zygmuntowicz <firstname.lastname@example.org> wrote: > > Now the devel instance of my app is broken. > > and run lighttpd. > > Forensics for the 21st Century > lighttpd.conf file in your config directory needs to be edited and > add full absolute paths to your public dir and your dispatch.fcgi. > > Cheers- > -Ezra Zygmuntowicz > Yakima Herald-Republic > WebMaster > http://yakimaherald.com > 509-577-7732 > email@example.com Thanks Ezra, That fixed my immediate issue and I see that "script/server webrick" can be used in my sandboxes. I think I'm set. I think I read that edge rails has a fix to be compatible with the relative path issue. That will be very handy as well. Greg -- Greg Freemyer The Norcross Group Forensics for the 21st Century