cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Aleksey Yeschenko (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-9917) MVs should validate gc grace seconds on the tables involved
Date Fri, 14 Aug 2015 10:46:46 GMT


Aleksey Yeschenko commented on CASSANDRA-9917:

{{max_hint_window_in_ms}} is misunderstood by almost everyone, it seems.

It only affects the decision to write a hint in the first place (down for long than the window?
stop writing hints).

1) Batchlog ignores {{max_hint_window_in_ms}} and always writes its hints, no matter what,
if a node is down
2) {{max_hint_window_in_ms}} does not affect replay decisions - it's ignored on replay - only
gc gs - turned into ttl - matters

It's most often true that if a node has been down for longer than {{max_hint_window_in_mx}},
it is going to have data missing, yes. But there are no guarantees that it being down for
shorter than that means it doesn't. Any number of nodes with hints for the target node could
go down for a time longer than hints' expiration time, die forever, lose their hints - you
name it.

And many outright disable hints (batchlog would still write them), or have ridiculously low

What I'm saying is that I don't like tying any correctness logic to {{max_hint_window_in_ms}}

> MVs should validate gc grace seconds on the tables involved
> -----------------------------------------------------------
>                 Key: CASSANDRA-9917
>                 URL:
>             Project: Cassandra
>          Issue Type: Bug
>            Reporter: Aleksey Yeschenko
>            Assignee: Paulo Motta
>              Labels: materializedviews
>             Fix For: 3.0.0 rc1
> For correctness reasons (potential resurrection of dropped values), batchlog entries
are TTLs with the lowest gc grace second of all the tables involved in a batch.
> It means that if gc gs is set to 0 in one of the tables, the batchlog entry will be dead
on arrival, and never replayed.
> We should probably warn against such LOGGED writes taking place, in general, but for
MVs, we must validate that gc gs on the base table (and on the MV table, if we should allow
altering gc gs there at all), is never set too low, or else.

This message was sent by Atlassian JIRA

View raw message