With (Contract JARs) rolling upgrades the downgrade rule cannot be effectively check as the platform can't tell the difference between a transaction that's downgrading because of an attack, vs a transaction that's downgrading because Alice has upgraded but Bob hasn't yet. During a rolling upgrade we would expect state versions to fluctuate up and down as data gets read/written by a mix of nodes. With the feature as implemented Alice will upgrade and start trading with Bob. Bob will be able to read and process the states Alice sent him, but the moment he tries to consume such a state he will fail. This will result in cascading flow deaths and a hung business network the moment an upgrade starts.
- Delete information about hash->signature constraint migration. That
won't work for C4.0
- Rewrite parts of the api-contract-constraints.rst file to fix grammar
problems and reduce the weight of the page. Some detail isn't really
necessary at this point in the docs.
- Mention more things in the upgrade notes.
* Update changelog with changes to vault queries
* Move choice of relevancy state to QueryCriteria
* Ensure relevancy default is correct for all query types
* Remove superfluous imports
* Update documentation for queries using state relevancy
* Fix merge damage
* Revert change to use RELEVANT as the default and update docs
* Revert missed instance of relevancy change
* Rewrite of upgrade notes for vault state query updates
* Address review comments