nHibernate second-level cache with memcached and random invalid casts

Question

I have an MVC3 application using nHibernate and memcached as the second level cache provider. We've been intermittently (but much more frequently lately) getting weird casting issues. It happens randomly and invalidating the memcached cache will fix the problem for while. It only happens in our Production environment because we don't run memcached in our other environments. However I run memcached locally and have tried to get this to occur locally with no luck. We are using memcached 1.2.6 on Windows. Here's the stack trace. I know it's not going to be enough information to determine anything but if anyone has any ideas about how I could debug this that would be appreciated. I'm trying to get remote debugging on our Production machine but it's a busy time of year and risky. ``` Unable to cast object of type 'System.Int32' to type 'System.String'. Description: An unhandled exception occurred during the execution of the current web request. Please review the stack trace for more inform…

Accepted Answer

We've been encountering the same issue and we have pinpointed the problem to invalid cache entries in the second level cache. In our case the problem arises when we add or remove columns/properties from tables/entities that are cached in the second level cache and deploy the change. The code expects five columns (for argument's sake) whereas the cache stores six columns.

One strategy that we employed in our repository is to catch the exception and invalidate the cache. We then retry the get.

```
    public override T Get(int id)
    {
        try
        {
            return base.Get(id);
        }
        catch (InvalidCastException)
        {
            _sessionFactory.EvictEntity(typeof (T).Name);
            return Get(id);
        }
    }
```

Hope this helps!

nHibernate second-level cache with memcached and random invalid casts

Solution

Problem

Related problems