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
It's possible for dqlite to request a raft_barrier through leader_barrier while a raft_barrier is already in progress.
It happens in one case where a new raft leader tries to commit a barrier entry to determine its commit index and at that moment dqlite tries to handle a query, notices its state machine is not up to date and fires a barrier.
I'm worried this could also lead to an avalanche of barriers on a busy system and raft will be more busy replicating barriers than actual transaction data.
It's possible for dqlite to request a
raft_barrier
throughleader_barrier
while araft_barrier
is already in progress.It happens in one case where a new raft leader tries to commit a barrier entry to determine its commit index and at that moment dqlite tries to handle a query, notices its state machine is not up to date and fires a barrier.
I'm worried this could also lead to an avalanche of barriers on a busy system and raft will be more busy replicating barriers than actual transaction data.
log
The text was updated successfully, but these errors were encountered: