> Why couldn't your exit policy just block the IPs of the journal sites?Because there's > 1000 of them (and each would be a /32). It was
discussed in another thread at the time, and the developers led me to
the conclusion that such hugely long exit policies were a bad idea.