Neophytos - it's my ISP's filesystem that hold incoming mail, and they're aware of it. I imagine they've gotten hosed by some massive spam wave.
Dan ... yes, I used ns_cache. Individual timeouts are definitely a good idea but have to be used carefully. For instance, if you have a set of queries that build pages with content plus panels that link to further content not already displayed, then you'd better be caching a consistent snapshot otherwise the stars won't align properly. So I think there needs to be a mechanism beyond timeouts. For our system I implemented system-wide flushing to ensure that all the page pieces are displayed consistently ... not right for a generalized solution but as a (literally) last-day pre-launch hack gets the job done for us.
