cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sylvain Lebresne (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-6630) counters++ handle duplicate cells in a mutation
Date Wed, 29 Jan 2014 19:42:10 GMT


Sylvain Lebresne commented on CASSANDRA-6630:

bq. replacing UnsortedColumns everywhere w/ ArrayBackedSortedColumns, but it is slightly tricky,
since we have to maintain backward compatibility.

Since neither thrift nor CQL guarantees that columns will be added in order so we'll need
an implementation that sorts on insert. And while we could bring back TreeMapBackedSortedColumns,
my preference would go to removing the assertion in ArrayBackedSortedColumns.add() so it doesn't
bitch when we add out of order columns since 1) that's trivial to do, 2) that's likely faster
in almost all case in practice (than TreeMapBackedSortedColumns at least) and 3) if we do
that we don't have to care about backward compatibility (at least if by that you mean "a pre-2.1
node sent me a mutation out of order").

bq. Guess I'll close this issue and open one for killing UnsortedColumns?

That or just rename this ticket, I don't mind either way but I do think that's the right thing
to do.

> counters++ handle duplicate cells in a mutation
> -----------------------------------------------
>                 Key: CASSANDRA-6630
>                 URL:
>             Project: Cassandra
>          Issue Type: Bug
>            Reporter: Aleksey Yeschenko
>            Assignee: Aleksey Yeschenko
>             Fix For: 2.1
>         Attachments: 6630.txt
> It's possible for a counter mutation to have several CounterUpdateCell-s with the same
cell name. Those should be summed up and not ignored.

This message was sent by Atlassian JIRA

View raw message