SparkViewEngine spontaneous NullReferenceException while rendering view

[Ruud van Falier]

We are using the latest stable release of SparkViewEngine for
http://www.stribe.nl.
There is a problem that occurs -sometimes- while rendering the
homepage.
When you look at the homepage it will usually render perfectly fine,
but i noticed some exceptions in the EventLog from people visiting the
homepage.

The exception message is "Error executing child request for handler
'System.Web.Mvc.HttpHandlerUtil+ServerExecuteHttpHandlerAsyncWrapper'"
which means an exception occured while executing a ChildAction on the
controller.

So i ran the site in VS2010 debug mode and started a stresstest tool
that requests the homepage a few times per 10 seconds.
After running it for a few cycles of 10 seconds it breaks into my code
(see screenshot on http://www.stribe.nl/sparkerror_null.png ).

Output (which is the sparkview output buffer) is null.
When i continue running the code after the exception it will give the
exception about child request failing and then it renders the rest
fine (and next request the whole page just renders without problems
again)

This happens over and over again, but the times on which it happens
seem random, same as the location in code where it happens seems
random.
I have no idea what is going wrong here (it's one of those "it happens
sometimes" bugs) and also no way of analysing it further than i did
now.

Is there anyone with that maybe got this problem too or has any clue
on to what may be the problem?
Or anyone with ideas on how i can analyse this issue better?

[Ruud van Falier]

No one got any suggestions?
For the time being i've created a workaround by catching the "Error
executing child request" exception in Application_Error() and redirect
to the same page again (adding a counter to prevent infinite loop).
This works and prevents the user from ever seeing the error, but it's
not a solution of course :)

[Rob G]

Hi Ruud,

I'll try look into this when I get a chance - the next two weeks are pretty much out for me though. If you could email me directly, a small zip file with just enough in it to reproduce the error, that'll speed things up. I can run the stress test my end, but I'll need some example failing code otherwise it could take a lot long to reproduce.

All the best,

Rob

--
You received this message because you are subscribed to the Google Groups "Spark View Engine Dev" group.
To post to this group, send email to spar...@googlegroups.com.
To unsubscribe from this group, send email to spark-dev+...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/spark-dev?hl=en.

[Ruud van Falier]

Hi Rob,

I'll try to reproduce it in an isolated project and mail it to you.

Regards,
Ruud

On Jul 20, 9:54 am, Rob G <robertgreyl...@gmail.com> wrote:
> Hi Ruud,
>
> I'll try look into this when I get a chance - the next two weeks are pretty
> much out for me though. If you could email me directly, a small zip file
> with just enough in it to reproduce the error, that'll speed things up. I
> can run the stress test my end, but I'll need some example failing code
> otherwise it could take a lot long to reproduce.
>
> All the best,
> Rob
>
> On Tue, Jul 20, 2010 at 8:45 AM, Ruud van Falier

> <ruud.vanfal...@gmail.com>wrote:

>
>
>
> > No one got any suggestions?
> > For the time being i've created a workaround by catching the "Error
> > executing child request" exception in Application_Error() and redirect
> > to the same page again (adding a counter to prevent infinite loop).
> > This works and prevents the user from ever seeing the error, but it's
> > not a solution of course :)
>
> > --
> > You received this message because you are subscribed to the Google Groups
> > "Spark View Engine Dev" group.
> > To post to this group, send email to spar...@googlegroups.com.
> > To unsubscribe from this group, send email to

> > spark-dev+...@googlegroups.com<spark-dev%2Bunsubscribe@googlegroups .com>
> > .

[Ruud van Falier]

I can't reproduce it when i start a new project with just a simple
master and some dummy childactions / partial views.
If i ever get more time to look at it i will think about stripping
down the project until the exception dissapears, hopefully pinpointing
the exact spot of the problem this way.
But it's a huge project that took 8 months to build by several people
so it will cost alot of time.
And since the workaround is holding up the priority is somewhat low
now.

Still hoping that someone else pops up that recognizes the problem.
I feel so alone right now :))