Contract's monitoring

[AuHau]

As part of the contract's deployment work, I ran into a question. We might have the capability to upgrade the contract's logic to fix a bug/exploit. One case is if the problem is reported to us or discovered by us, then we can act on it. What of a case if the logic is being exploited? It could be that somebody in the community notices and might report it to us, but it could be that it goes undetected for quite some time before someone reports it. This is mainly related to the business logic part as disappearing funds is fairly easy to detect. I am thinking cases like:

• unintended prolonging of requests
• modifying all sorts of counters (missed proof counters etc)

Should we monitor the contracts in some way? What metrics should be tracked?