[Catalyst-dev] Re: [Catalyst] Race condition inCatalyst::Plugin::Session and Catalyst::Engine::Apache(possibly other engines too)

[Matt Pitts]

> -----Original Message-----
> From: Sergio Salvi [mailto:sergio.lists at salvi.ca]
> Sent: Thursday, December 11, 2008 11:51 AM
> To: Development of the elegant MVC web framework
> Subject: Re: [Catalyst-dev] Re: [Catalyst] Race condition
> inCatalyst::Plugin::Session and Catalyst::Engine::Apache(possibly
other
> engines too)
> 
> On Wed, Dec 10, 2008 at 9:43 AM, Sergio Salvi <sergio.lists at salvi.ca>
> wrote:
> > On Thu, Dec 4, 2008 at 2:12 PM, Matt Pitts <mpitts at a3its.com> wrote:
> >> I think this might be affecting my application as well -
> HTTP::Prefork
> >> *and* FastCGI instances using memcached for cache/session storage.
> >>
> >> I'm getting random "duplicate key" failures when trying to write
the
> >> sessionid to the Cart table and other strange session behavior. It
> just
> >> started happening a few weeks ago and I've done a lot of
> troubleshooting
> >> on it - thought it might be caused by this session save issue.
> >>
> >> Anybody know of the status of this?
> >
> > My apologies for the delay. I'll get my session patch committed
> today,
> > which will *not* fix the duplicate key issue -- that's another patch
> > I'm working on which will store "flash" inside the session row.
> >
> > Sergio
> >
> 
> Committed:
> http://dev.catalyst.perl.org/svnweb/Catalyst/browse/branches/Catalyst-
> Plugin-Session/
> 
> I'll soon merge the two patches into a single patch.
> 
> I've been successfully using the finalize_race_condition patch in
> production for about 2 months. The flash inside session patch is brand
> new, but all tests passed :)
> 
> Please try it out and let me know how it works for you.
> 

Thanks for getting this posted Sergio. Unfortunately, I already had your
previous patch applied against C::P::Session 0.19 which didn't help my
problem, but it didn't make things worse either. I'm going ahead and
applying your branch version (which appears to be against C::P::S 0.20),
so we'll see if it makes any difference. Sorry I can't provide much more
feedback right now.

v/r
-matt