You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Whenever Tendermint has a proposal and a precommit aggregation which holds 2f+1 votes for that proposal it should produce the proposal with the 2f+1 aggregation as proof as the decision value.
After adding a couple of tests on the nibhar/tendermint-tests branch it was observed that it does in fact not work when a proposal is received late.
For the scenario to not resolve there must be no subsequent updates to the aggregation in question which is the tested case on that branch. A subsequent update to the aggregation represents the it_accepts_late_polka test and will succeed.
The fix to this should be to check for decision criteria more proactively when receiving a proposal. It should be done in a way such that it does not matter whether or not the proposal was regularly received or as part of a response to a request.
Whenever Tendermint has a proposal and a precommit aggregation which holds 2f+1 votes for that proposal it should produce the proposal with the 2f+1 aggregation as proof as the decision value.
After adding a couple of tests on the
nibhar/tendermint-tests
branch it was observed that it does in fact not work when a proposal is received late.For the scenario to not resolve there must be no subsequent updates to the aggregation in question which is the tested case on that branch. A subsequent update to the aggregation represents the
it_accepts_late_polka
test and will succeed.The fix to this should be to check for decision criteria more proactively when receiving a proposal. It should be done in a way such that it does not matter whether or not the proposal was regularly received or as part of a response to a request.
Potentially related to #1533
The text was updated successfully, but these errors were encountered: