On Thu, Oct 9, 2014 at 3:29 PM, Robert Wille <[email protected]> wrote:

> What is the bad thing that happens (or can possibly happen) when a counter
> is deleted?
>

You get totally wacky counter state if you, later, re-create and use it.

https://issues.apache.org/jira/browse/CASSANDRA-6532
and
https://issues.apache.org/jira/browse/CASSANDRA-7346

Have some good discussion.

Is it safe to delete an entire row of counters?


Not unless :
a) you will never use that particular counter row again
OR
b) gc_grace_seconds has passed and you have repaired and run a major
compaction on every node, such that 100% of such tombstones have been
removed


> Is there any 2.0.x version of Cassandra in which it is safe to delete
> counters?


See above, 2.0 is the same as any other version for this behavior.

Is there an access pattern in which it is safe to delete counters in 2.0.x?
>

The summary of the above is that, in practice, it is only safe to delete
counters if you plan to never use that particular counter row again.

=Rob

Reply via email to