we're not all SQL gurus in our team, so there's a good probability that someone may at one point issue queries that could cause problems that would affect our prod environment badly (badly as in huge sums of € lost). as far as i'm aware, there's a watchdog for scheduled queries that could keep a check on things, but as we've implemented our own extension for live-queries through persistent RPC connections, it seems the watchdog will not come into play there. before we implement our own watchdog system, maybe someone here has a better idea how to prevent such issues.