Erik Krogen

LGTM, great work @mkumar1984 !

cc @shardulm94 (thanks for the push towards constructing the path into a single message instead of using chained exceptions, the additional code changes were minimal and the resulting message is...

Pushed up new commits rebasing on latest changes. cc @cloud-fan @dongjoon-hyun in case either of you are interested.

Ping again to @MaxGekk @karenfeng @gengliangwang -- any thoughts?

Thanks for the suggestion @cloud-fan ! Good point about there many places where Spark trusts nullability. Here I am trying to target places where _user code_ could introduce a null....

@jzhuge The PR as-is seems to only contain the view catalog interface, not the other 3 facets that you mentioned. Is this intentional?

@steven-aerts just want to check if you are still working on this?

It looks like the current diff is disallowing in-progress edits during `catchupDuringFailover()`. But this isn't right, we _do_ want in-progress edits, since those can be durable and have been ack'ed...

> Before `catchupDuringFailover`, whatever active crashed or successfully changed to standby, the last segment in majority journalnode should be finalized. If the active crashed, then the segment won't be finalized,...

> If the active crashed, during standby starting active services, the standby will recover unclosed streams via `recoverUnclosedStreams`. So before `catchupDuringFailover`, the last segment should always closed. I don't think...