Hi all
I am working on a heavy engineering project. And the requirement is that
I needto do a lot of process before I store the data to the db.
Now the question is that I want to do the process that I am doing but I
want to do it in the background. What I actually mean to say is that…I
just want the user to get a msg that Records Saved. But in real I want
the process to go on in background.
There r multiple callbacks used in my system. Also I dont want to use
Starling or workling. pls suggest something where in my requirement is
satisfied. Since i dont want the user to wait until my process is going
on.
thanks
On 23 Feb 2009, at 10:38, Ankit P. wrote:
that…I
just want the user to get a msg that Records Saved. But in real I want
the process to go on in background.
There r multiple callbacks used in my system. Also I dont want to use
Starling or workling. pls suggest something where in my requirement is
satisfied. Since i dont want the user to wait until my process is
going
on.
Well one of the other background processors (bj, backgroundrb etc. )
might do the job
Fred
Spawn is dead easy (http://rubyforge.org/projects/spawn/). I used it
to great effect for sending emails while making the user think they
where all already sent. The others Frederick suggested work but are
more trouble than starling/workling.
Harm
On Feb 23, 12:14 pm, Frederick C. [email protected]
thanks for your reply but i dont want to run something like a cron job
or nething.
Since my method is very long i want it to run in background. So that the
user dont need to wait on that page while the process is going on.
Is there something where i can define that a particular method should
run in background.
Thanks
On 23 Feb 2009, at 11:44, Ankit P. wrote:
The way you do that reliably is to spawn another process to handle it.
You can do your thing in a thread but that pretty much requires that
you’re using a thread safe version of rails and has limitations given
ruby’s threading (eg your thread can block the whole mongrel)
Fred
Hi Ankit,
I found a solution that looks good and simple, see
http://devblog.imedo.de/2008/6/18/running-ruby-blocks-in-the-background.
I haven’t tried it yet.
Best Regards,
Onno
On Feb 23, 5:38 am, Ankit P. [email protected]
Ankit P. wrote:
I understand why you want to run it in the background but…what
happens if there is an error? I do not write perfect code or test 100%
of possible code paths or run on failproof hardware so there is always a
possibility of an error. You will seem to have the situation where the
user thinks everything was processed and saved into the db but that
might not be true. I would not be happy if that happened with my data.
Personally I would probably want to wait to see all was really
successful.