Feature #108

Get backgroundrb and ferret restarting with capistrano

Added by yossarian over 3 years ago. Updated over 3 years ago.

Status:Closed Start:
Priority:High Due date:
Assigned to:yossarian % Done:

0%
Category:Core
Target version:0.4 - Rebel Yell (2b || !2b)
Resolution:

fixed

Description

The current deployment problems are making me nuts. It needs to be much smoother.

Here's one thing which may help, a shell script for starting backgroundrb in capistrano (from here):

#!/bin/sh

echo "Starting the backgroundrb server" 
echo "cd /var/www/xxx/current; rake backgroundrb:start" 
cd /var/www/xxx/current
log="/tmp/backgroundrb.@date +%s@" 
echo "Writing to $log" 
nohup rake backgroundrb:start > $log 2>&1
cat $log
rm -f $log

History

Updated by yossarian over 3 years ago

Here's a little tidbit regarding nohup + backgroundrb:
http://rubyforge.org/pipermail/backgroundrb-devel/2007-February/000718.html

Updated by yossarian over 3 years ago

Not a solution, but another interesting chunk of code:

# Confirm Backgroundrb is running and start if dead.
#
def prepare_backgroundrb
   begin
     # try to contact Backgroundrb in some way
     MiddleMan.gc!((Time.now-10.years).to_time)

   rescue
     begin
       @cd #{RAILS_ROOT} && script/backgroundrb start@
       sleep(10)
     rescue
       log.error("unable to start Backgroundrb: #{$!}")
     end
   end
end

Updated by yossarian over 3 years ago

  • Status changed from New to Closed
  • Resolution set to fixed

Ok, this seems to be working now.

Also available in: Atom PDF