This is more robust in the event that staking rewards fails to compute for a minority of nodes. We had this happen on one node during testnet testing.
Details
Details
- Reviewers
Fabien - Group Reviewers
Restricted Project - Commits
- rABCe0776c77fd38: [avalanche] Remove requirement that local staking rewards have to be computed…
ninja check ./test/functional/test_runner.py abc_p2p_avalanche_contender_voting
Diff Detail
Diff Detail
- Repository
- rABC Bitcoin ABC
- Lint
Lint Not Applicable - Unit
Tests Not Applicable
Event Timeline
src/avalanche/processor.cpp | ||
---|---|---|
910 ↗ | (On Diff #53435) | This returns false if winners size is 0, so this whole diff looks like there should be no change in behavior to me. What am I missing ? |
src/avalanche/processor.cpp | ||
---|---|---|
910 ↗ | (On Diff #53435) | Since the winners.size() == 0 requirement was removed during promotion (the other callsite to setContenderStatusForLocalWinners), we know some contenders will always be polled once they are promoted. So now when staking rewards are computed, we just need to make sure that the local winners that had status changes are also polled. There is no need to addToReconcile if no winners were selected since no status changes will have occurred. |