Ah, found the solution.  Looks like Matt's suggestion to note the use
of the full path in the doc has not been implemented yet.  Thank you,
Matt!

http://groups.google.com/group/capistrano/browse_thread/thread/8cd5faf6e1fa51e4/623b7ad44f1e12fd?lnk=gst&q=spin+restart#623b7ad44f1e12fd

On Feb 21, 5:00 pm, Mike Perham <[EMAIL PROTECTED]> wrote:
> Rails 2.0.2 with cap 2.1.0 on a RHEL linux server.
>
> deploy:stop and deploy:start works.  With restart though, it just
> seems to send a HUP signal to Mongrel via the reaper script, which
> reloads the code.  The problem seems straightforward to me:
>
> deploy   24828 41.0  2.7 128612 52620 ?        S    16:32   0:03 /usr/
> local/bin/ruby /usr/local/bin/mongrel_rails start -d -a 0.0.0.0 -p
> 6000 -P /var/www/rails/products/manage/frontend/releases/
> 20080221220935/tmp/pids/dispatch.6000.pid -e production -c /var/www/
> rails/products/manage/frontend/releases/20080221220935 -l /var/www/
> rails/products/manage/frontend/releases/20080221220935/log/mongrel.log
>
> Since it's using the absolute release path and not the current release
> path, the code it reloads will be the old build, not the current
> build.  Do we have something configured incorrectly?  Here's the spin
> script we use to start mongrel:
>
> #!/bin/sh
> script/process/spawner -p 6000 -i 5
>
> The problem seems to be in the spawer script which uses the rails
> initializer to construct the absolute release path (the -c parameter)
> we see above:
>
>   :rails_root  => File.expand_path(RELATIVE_RAILS_ROOT),
>
> which resolves to the absolute path.  Any ideas how to get
> deploy:restart working?
>
> mike
--~--~---------~--~----~------------~-------~--~----~
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/capistrano
-~----------~----~----~----~------~----~------~--~---

Reply via email to