cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Björn Hegerfors (JIRA) <>
Subject [jira] [Commented] (CASSANDRA-8406) Add option to set max_sstable_age in seconds in DTCS
Date Thu, 05 Feb 2015 21:35:35 GMT


Björn Hegerfors commented on CASSANDRA-8406:

[~krummas] well we did try max_sstable_age_days=1 in a new table, just the other day. This
table has a default_time_to_live at 1 day, and the hope was that SSTables older than that
would just disappear. That's not the case, and it may be timely for me to address this ticket
now: CASSANDRA-8359. With that fixed, I think that having max_sstable_age equal to the default_time_to_live
of the table is the obviously right setting, giving very efficient behavior. Why compact something
that's about to go away? If it works as well as I hope, the next logical step might be to
set max_sstable_age automatically when default_time_to_live is set. And if someone wants to
use default_time_to_live < 1 day, I don't see why they shouldn't be able to use max_sstable_age
< 1 day. So I do see a case for max_sstable_age being less than days. And for something
like this, having the setting in seconds seems most appropriate. But the floating point solution
is probably a viable compromise.

I suppose the main argument for having the setting in seconds is: why is default_time_to_live
in seconds when max_sstable_age is not?

> Add option to set max_sstable_age in seconds in DTCS
> ----------------------------------------------------
>                 Key: CASSANDRA-8406
>                 URL:
>             Project: Cassandra
>          Issue Type: Bug
>            Reporter: Marcus Eriksson
>            Assignee: Marcus Eriksson
>             Fix For: 2.0.13
>         Attachments: 0001-8406.patch, 0001-patch.patch
> Using days as the unit for max_sstable_age in DTCS might be too much, add option to set
it in seconds

This message was sent by Atlassian JIRA

View raw message