[zfs-discuss] zfs and fstrim
gordan.bobic at gmail.com
Wed Jan 20 09:41:18 EST 2016
There are a few options:
1.1) You don't
1.2) Leave untouched partition/space on a freshly secure-erased SSD so that
the SSD has more known untouched blocks to work with (under-provisioning),
and you don't worry about TRIM-ing.
2) Drop the SSD out of the array (make sure your redundancy is up to the
task), ATA secure erase it, re-add it, resilver. Space that isn't used will
be freshly trimmed. Repeat for all SSDs in the pool.
In most cases 1.1) is good enough as SSDs are internally adequately
overprovisioned to handle this reasonably well, especially ones based on
Sandforce controllers that internally compress and dedupe the data.
On Wed, Jan 20, 2016 at 2:17 PM, John Covici via zfs-discuss <
zfs-discuss at list.zfsonlinux.org> wrote:
> Hi. Since zfs does not support fstrim, how should I trim my pool? I
> have a pool with a number of datasets as legacy mountpoints on 2 ssd
> drives -- so do I do each one separately, or how to do this properly?
> Thanks in advance for any suggestions.
> Your life is like a penny. You're going to lose it. The question is:
> How do
> you spend it?
> John Covici
> covici at ccs.covici.com
> zfs-discuss mailing list
> zfs-discuss at list.zfsonlinux.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the zfs-discuss