The problem is MsgTimeoutNow always results in a new election, because
it explicitly ignores the pre_vote step. This results in integration
test flakiness whenever a follower calls promote_or_fail while the
leader's raft log is longer.

We've cherry-picked some fixes but didnt cut a release yet. Prepare
changelog.

The problem was that we were expecting last raft log record to be very
specific one but vshard rebalancing came and added some dml record. Use
pending_schema_change from pico property instead.

Close: https://git.picodata.io/picodata/picodata/picodata/-/issues/554

Arseniy Volynets authored 8 months ago and Dmitry Rodionov committed 8 months ago

- we didn't wait until replication catches up before
checking dml results on other instance. Fix this
by doing quorum read on the other instance