I have what I thought would be a simple thing - we use backgroundrb to
watch a directory and process new files. We want to make sure that the
worker is running. I thought I could have a schedule that started the
worker every 5 minutes to make sure, but, it loads a new one each time
(not surprising).
Is there a way to make sure that there is one (and only one) of these
workers running at all times?
Hi Phil,
On Sat, 2009-09-05 at 04:41 -0700, phil wrote:
I have what I thought would be a simple thing - we use backgroundrb to
watch a directory and process new files. We want to make sure that the
worker is running. I thought I could have a schedule that started the
worker every 5 minutes to make sure, but, it loads a new one each time
(not surprising).
Is there a way to make sure that there is one (and only one) of these
workers running at all times?
You’ll probably have better luck with this on the backgroundrb list.
http://backgroundrb.rubyforge.org/community/
HTH,
Bill
On Sep 5, 1:32 pm, bill walton [email protected] wrote:
You’ll probably have better luck with this on the backgroundrb list.
http://backgroundrb.rubyforge.org/community/
HTH,
Bill
BackgrounDRb or no, creating a file containing the process’s PID, and
verifying that each time the process is started is the typical unixy
way to do this. How I do it:
def self.is_running?
if File.exists? pidfile
begin
Process.kill 0, pid
return true
rescue Errno::EPERM
puts “You don’t own this process.”
return true
rescue Errno::ESRCH #pid doesn’t exist anymore
delete_pidfile
end
end
false
end
def self.delete_pidfile
File.delete pidfile
rescue
end
def self.pid
File.read(pidfile).chomp.to_i rescue nil
end
def self.pidfile
“path/to/background_process.pid”
end
And of course just create the the appropriate file writing Process.pid
to it.