cassandra-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Robert Coli <rc...@eventbrite.com>
Subject Re: disk space issue
Date Thu, 02 Oct 2014 07:49:34 GMT
On Wed, Oct 1, 2014 at 6:17 AM, Ken Hancock <ken.hancock@schange.com> wrote:

> Major compaction is bad if you're using size-tiered, especially if you're
> already having capacity issues.  Once you have one huge table, with default
> settings, you'll need 4x that huge table worth of storage in order for it
> to compact again to ever reclaim your TTL'd data.
>

I agree that OP likely cannot major compact. But if he could, he could do
so and then run sstablesplit (most safely, with the node down) on the One
Big SSTable to avoid the problem you describe.

If the OP actually has SSTables full of TTLed garbage, he should use JMX to
run UserDefinedCompaction on them, one at a time.

He should also probably reconsider whether a database with immutable
datafiles is the ideal solution for his data he wants to keep for 7 days. ;D

=Rob

Mime
View raw message